Neoplasms of the Colon, Rectum, and Anus,

Neoplasms of the

Colon, Rectum, and Anus Second Edition Philip H. Gordon M.D., F.R.C.S. (C), F.A.C.S., F.A.S.C.R.S., Hon. F.R.S.M., Hon. F.A.C.G.B.I Professor of Surgery and Oncology, McGill University Director of Colon and Rectal Surgery Sir Mortimer B. Davis-Jewish General Hospital and McGill University Montreal, Quebec, Canada

Santhat Nivatvongs M.D., F.A.C.S., F.A.S.C.R.S., Hon. F.R.C.S.T. (Thailand), Hon. F.R.A.C.S. Consultant Surgeon and Professor of Surgery Mayo Clinic College of Medicine Rochester, Minnesota, U.S.A. With a special contribution by

Lee E. Smith, M.D.

Clinical Professor of Surgery, George Washington University Director, Section of Colon and Rectal Surgery, Washington Hospital Center, Washington, D.C.

ILLUSTRATORS Scott Thorn Barrows, C.M.I., F.A.M.I Director and Clinical Assistant Professor, Biomedical Visualization University of Illinois at Chicago Medical Center with the assistance of Carla Gunn, Gregory Blew, David Ehlert, Craig Kiefer, and Kim Martens

New York London

Informa Healthcare USA, Inc. 270 Madison Avenue New York, NY 10016 © 2007 by Informa Healthcare USA, Inc. Informa Healthcare is an Informa business No claim to original U.S. Government works Printed in the United States of America on acid‑free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number‑10: 0‑8247‑2959‑5 (Hardcover) International Standard Book Number‑13: 978‑0‑8247‑2959‑2 (Hardcover) This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or here‑ after invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copy‑ right Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978‑750‑8400. CCC is a not‑for‑profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging‑in‑Publication Data Gordon, Philip H. Neoplasms of the colon, rectum, and anus / Philip H. Gordon, Santhat Nivatvongs. ‑‑ 2nd ed. p. ; cm. Includes bibliographical references and index. ISBN‑13: 978‑0‑8247‑2959‑2 ISBN‑10: 0‑8247‑2959‑5 1. Colon (Anatomy)‑‑Cancer. 2. Rectum‑‑Cancer. 3. Anus‑‑Cancer. I. Nivatvongs, Santhat. II. Title. [DNLM: 1. Colonic Neoplasms. 2. Anus Neoplasms. 3. Rectal Neoplasms. WI 529 G64n 2006] RC280.C6G67 2006 616.99’4347‑‑dc22 Visit the Informa Web site at www.informa.com and the Informa Healthcare Web site at www.informahealthcare.com

2006049733]

I am once again deeply indebted to my wife Rosalie for her constant patience and understanding through the burden of this third edition. Her constant life-long support has made all my professional accomplishments possible. To my wonderful children Laurel and Elliot of whom I am extremely proud. My love and gratitude to all. —PHG

To my two angels Marisa and Nitara. Thinking of them makes me smile; talking to them recharges my energy. —SN

Preface gained. The new instrumentation of equipment that is available has been outlined including subjects such as handports and robotic surgery. Techniques of laparoscopic colectomy have been added. The results of laparoscopic colectomy, conversion rates, detailed morbidity, and mortality by disease process have been updated. Difficult situations such as obesity, inflammatory masses, and fistulas have been described. Quality of life and cost issues have been included. A major expansion of the complications, including incidence and prevention of complications with laparoscopic colectomy have been described. The book is replete with color illustrations and photographs adjacent to the text material rather than grouped at the beginning, middle, or end of the book. New illustrations have been added and others redrawn to conform to better understanding and improvement in operative technique. Each chapter is heavily referenced for those interested in further documentation. We hope we have accomplished our goal of summarizing the enormous body of knowledge published in the literature and share our personal experience and preferences with our readers. We strove for a book that strikes a balance of being authoritative and detailed without being so inclusive that somewhat irrelevant material and minutia are included. We sincerely hope our efforts will provide the practicing surgeon and surgeon in training, the appropriate information to permit them to provide a rational and up to date course of action to the ultimate benefit of each of their patients. Information in this text has been reprinted from the third edition of our comprehensive textbook Principles and Practice of Surgery for the Colon, Rectum, and Anus. We trust that non-colorectal specialists such as oncologic and general surgeons, radiologists, and others who diagnose and treat this kind of malignancy will find this book useful.

Colorectal carcinoma remains the second leading cause of death from malignancy. In the eight years that have elapsed since our last edition, the proliferation of information published in the surgical literature makes it necessary to update and elaborate on these developments. Highlights in the revision include new data regarding the incidence, prevalence, and trends in colorectal carcinoma. There is an update on the genetics of colorectal carcinoma in general and in particular HNPCC. Extensive discussion of the indications for and interpretation of genetic testing and the invaluable role of genetic counseling are described in detail. We believe this is a disease that, with appropriate screening, can for the most part be prevented and therefore suggestions for screening are made. The value and role of virtual colonoscopy is discussed. There is an update on the propriety of adjuvant therapy with its limitations and complications and possible fine tuning of indications for adjuvant therapy. There is also updated information on the treatment of recurrent metastatic carcinoma providing prognostic indicators for recurrence following therapy. A section on intra-luminal stenting for obstruction has been added and new information on the staging of rectal carcinoma. There is revised description of sphincter saving operations (pouch, coloplasty, coloanal anastomosis) and a discussion of total mesorectal excision with results of the use of this technique. There are updated results on the treatment of carcinoma of the rectum with a discussion of the propriety of the use of local excision of rectal carcinoma. The section on palliative management of patients with rectal carcinoma has been expanded and there is a discussion on the role of preoperative neoadjuvant treatment for rectal carcinoma. A new section on the management of presacral bleeding has been included. The laparoscopy chapter has been totally revised and expanded. The indications for laparoscopic colectomy have been revisited and expanded as newer technology has become available and increased experience has been

Philip H. Gordon Santhat Nivatvongs

v

Contents Cholecystectomy / 78 Diverticular Disease / 78 Activity and Exercise / 78 Other Factors / 79 Juvenile vs. Adult Carcinoma / Prospects for Prevention / 81

Dedication / iii Preface / v Part I: Colorectal Disorders

&

Pathology

1: Benign Neoplasms of the Colon and Rectum / 1 Santhat Nivatvongs

Clinical Features /

2: Large Bowel Carcinoma: Screening, Surveillance, and Follow-Up / 39 Santhat Nivatvongs

Detection of Early Colorectal Carcinoma / 39 Early Diagnosis of Colorectal Carcinoma / 40 What Is Screening? / 40 Who Should Be Screened? / 40 Screening People at Average Risk for Colorectal Carcinoma / 41 Screening People at Increased Risk for Colorectal Carcinoma / 43 New Screening Tests / 44 When to Stop Screening / 44 Surveillance / 45 Follow-Up after Curative Resection / 46 Other Primary Malignancies / 47 Summary / 47 References / 48

Classification / 52 Adenocarcinoma / 52 Incidence, Prevalence, and Trends / Epidemiology / 53

Complications

/

94

/

95

96

Obstruction / 96 Perforation / 97 Bleeding / 97 Unusual Infections Associated with Colorectal Carcinoma / 97

Diagnosis / 97 Investigations / 98 Occult Blood Testing / 98 Endoscopy / 98 Radiology / 98 Radioimmunodetection / 101 Cytology / 102 Blood Markers / 102

Treatment

/

104

Curative Resection / 104 Adjuvant Therapy / 117 Complicated Carcinomas / 123 Perforation / 128 Bleeding / 128 Obstructive Colitis / 128 Invasion of Adjacent Viscera / 129 Urinary Tract Involvement by Colorectal Carcinoma / 131 Primary Involvement of the Urinary Tract / 131 Bladder Involvement / 132 Ureteric Involvement / 132 Fistula / 132 Hydronephrosis / 133 Radiotherapy / 133 Unexpected Intraoperative Involvement / 133 Recurrent Colorectal Carcinoma / 133 Abnormal Renal Function / 133 Palliation / 133 Unresectable Carcinoma / 133 Palliative Resection / 133 Synchronous Carcinomas / 134 Synchronous Polyps and Carcinoma / 134

51

52

Age / 53 Sex / 53 Family History / 53 Site / 54 Geographic Distribution / 54 Race and Religion / 54 Occupation / 54

Etiology and Pathogenesis

87

Symptoms / 94 General and Abdominal Examinations / Digital Rectal Examination / 95 Extraintestinal Manifestations / 95 Synchronous Carcinomas / 95 Associated Polyps / 95 Other Associated Malignancies / 96

&

3: Malignant Neoplasms of the Colon / Philip H. Gordon

82

Macroscopic Appearance / 82 Microscopic Appearance / 83 Depressed Carcinoma / 85 Sentinel Lymph Node Mapping / Modes of Spread / 88 Site of Spread / 89 Staging / 89 Biology of Growth / 92

Polyps of Colon and Rectum / 2 Familial Adenomatous Polyposis / 16 Hemangiomas of Large Bowel / 30 Leiomyomas of Large Bowel / 31 Lipomas of Large Bowel / 32 References / 33

&

/

80

55

Polyp-Cancer Sequence / 55 Inflammatory Bowel Disease / 55 Genetics / 55 Dietary Factors / 72 Irradiation / 77 Ureteric Implantation / 77

Metachronous Carcinoma

/

134

Treatment of Metastatic Disease / 135

vii

viii

& CONTENTS

Carcinoma in Young Patients /

Radiology / 209

147

Postoperative Complications / 147 Results / 148 Prognostic Discriminants / 152 Clinical Features / 152 Pathologic Features / 160 Biochemical and Special Investigations /

Recurrent Disease

Preoperative Preparation / 213 Radical Extirpative Operations /

164

166

/

Follow-Up / 166 Incidence / 166 Contributing Factors / 166 Patterns / 167 Clinical Features / 168 Investigations / 168 Role of Carcinoembryonic Antigen / 169 Treatment / 170 Results of Reoperation / 175 Intestinal Obstruction Due to Recurrent Carcinoma /

Local Forms of Therapy Special Considerations

175

176

Incidence / 178 Clinical Features / 178 Pathology / 178 Imaging Procedures / 179 Chemical Activity / 179 Treatment / 180 Results / 180

/

253

Distal Margins / 253 Circumferential Margins / 254 Total Mesorectal Excision / 254 Radical Lymphadenectomy / 258 Concomitant Pelvic Organ Excision / 260 Palliative Therapy for Advanced Rectal Carcinoma / 261 Hartmann’s Procedure / 264 Unresectable Carcinoma of the Rectum / 264 High Ligation of Inferior Mesenteric Artery / 265 Marking the Rectum / 265

Adjuvant Therapy for Carcinoma of the Rectum / 266 Radiotherapy / 266 Chemotherapy / 274 Combination Chemoradiotherapy / Immunotherapy / 279 Summary / 280

Postoperative Complications Recurrent Disease / 280

180

/

Incidence / 180 Pathology / 180 Clinical Features / Treatment / 182 Results / 182

245

/

Rationale / 245 Procedures / 245

Colorectal Carcinoma Complicating Pregnancy / Ovarian Carcinoma Involving the Colon / 177 Malakoplakia and Colorectal Carcinoma / 177 Other Malignant Lesions / 177 Carcinoid / 177

Lymphoma

214

Assessment of Resectability / 214 Selection of Appropriate Operation / 214 Operative Procedures / 216 Postoperative Care / 236 Results / 236

274

/

280

Follow-up / 280 Incidence / 280 Factors Contributing to Recurrence / 281 Patterns of Recurrence / 281 Clinical Features / 282 Investigations / 282 Treatment of Recurrent Disease / 283 Results of Reoperation / 290

181

Sarcoma / 182 Squamous Cell Carcinoma / 183 Adenosquamous Carcinoma / 184 Plasmacytoma / 184 Melanoma / 184 Leukemic Infiltration / 185 Neuroendocrine Lesions of the Colorectum / 185 Medullary Carcinoma of the Colon / 185 Carcinosarcoma / 186 Schwannoma / 186 Angiosarcoma / 186 Choriocarcinoma / 186 Metastases from Other Sources / 187 References / 187

Other Malignant Lesions of the Rectum Carcinoid / 290

/

290

Clinical Presentation / 291 Investigation / 291 Pathology / 291 Treatment / 291 Results / 291

Lymphoma

/

292

Clinical Presentation / 292 Treatment and Results / 292

Sarcoma / 292 Gastrointestinal Stromal Tumor (GIST) Secondary Carcinoma / 294 Miscellaneous Neoplasms / 294 References / 294

/

293

&

4: Malignant Neoplasms of the Rectum / 207 Philip H. Gordon

Adenocarcinoma / 208 Mechanisms of Spread of Rectal Carcinoma / Direct Extension / 208 Transperitoneal Spread / 208 Implantation / 208 Lymphatic Spread / 208 Venous Spread / 208

Clinical Features /

209

Symptoms / 209 General and Abdominal Examination /

Investigations

/

209

Endoscopy / 209 Routine Laboratory Blood Work / 209

209

208

Part II: Anorectal Disorders

&

5: Perianal and Anal Canal Neoplasms / 305 Santhat Nivatvongs

Introduction / 305 Anatomic Landmarks / 305 Incidence / 306 Etiology and Pathogenesis / 306 Staging / 307 Screening for Anal Carcinoma Precursors / 307 Human Papilloma Virus Type 16 Vaccine / 307 Perianal Neoplasms (Anal Margin) / 309

CONTENTS &

Neoplasms of the Anal Canal References / 324

&

/

6: Transanal Techniques / Santhat Nivatvongs

315

Part III:

Minimally Invasive Surgery

&

7: Laparoscopic Colon and Rectal Surgery Lee E. Smith and Philip H. Gordon

327

Introduction / 327 Rectal Biopsy / 327 Electrocoagulation of Rectal Polyps / 328 Snare Polypectomy / 328 Transanal Excision of Rectal Adenoma / 328 Transanal Excision for Carcinoma of the Low Rectum / 332 Posterior Approach to the Rectum / 332 Transanal Endoscopic Microsurgery / 335 Electrocoagulation of Carcinoma of the Rectum / References / 338

Background and Rationale / 342 Indications / 343 Equipment and Instrumentation / 344 Operative Procedure / 348 Specific Colorectal Procedures / 353 Robotics / 363 Postoperative Care / 364 Results / 364 Laparoscopic Complications and Their Prevention / 383 References / 387 337

Index /

ix

391

/ 341

I: Colorectal Disorders

1

Benign Neoplasms of the Colon and Rectum Santhat Nivatvongs

& Polyps of Colon and Rectum, 2

Management, 23

Neoplastic Polyps, 2

Proctocolectomy with Ileostomy, 23 Proctocolectomy with Continent Ileostomy, 23 Colectomy with Ileorectal Anastomosis, 23 Risk of Carcinoma in the Retained Rectum, 24 Regression of Polyps, 25 Proctocolectomy with Ileal Pouch Procedure, 26

Adenomas, 2 Adenoma-Carcinoma Sequence, 3 Diagnosis of Large Bowel Adenomas, 4 Management of Benign Adenomas, 5 The Flat Polyp, 5 Why Remove a Polyp?, 6 Natural History of Untreated Large Bowel Adenomas, 6 What Happens to Smaller Adenomas?, 6 Management of Adenomas with Invasive Carcinoma, 7 Pedunculated Polyp with Invasive Carcinoma, 7 Sessile Polyp with Invasive Carcinoma, 7 Serrated Adenoma, 9 Clinical Importance of Serrated Adenoma, 9 Genetics, 9

Genetic Counseling and Testing, 26 Genetic Counseling, 27 Genetic Testing, 28

When to Screen and When to Operate?, 28 The Polyposis Registry, 29 The Countrywide Registry, 30 The Regional Registry, 30 The Tertiary Referral Center, 30 & Hemangiomas of Large Bowel, 30

Hamartomatous Polyps, 10

Classification, 30 Clinical Manifestations, 30 Diagnosis, 31 Treatment, 31

Juvenile Polyps and Juvenile Polyposis, 10 Juvenile Polyposis of Infancy, 10 Juvenile Polyposis in Childhood and Adult, 10 Peutz-Jeghers Syndrome, 11 Cronkhite-Canada Syndrome, 14 Cowden’s Disease, 14 Bannayan-Ruvalcaba-Riley Syndrome, 15

& Leiomyomas of Large Bowel, 31

Clinical Manifestations, 32 Pathology, 32 Origin, 32 Treatment, 32

Inflammatory and Lymphoid Polyps, 15 Hyperplastic Polyps, 15 & Familial Adenomatous Polyposis, 16

& Lipomas of Large Bowel, 32

Definition and Natural History, 16 Clinical Manifestations and Diagnosis, 16 Distribution of Polyps and Carcinomas, 17 Attenuated Familial Adenomatous Polyposis, 17

Clinical Manifestations, 33 Diagnosis, 33 Treatment, 33 & References, 33

Clinical Features, 17 Diagnosis and Genetic Test, 17 Surgical Management, 18

Molecular Genetics, 18 Extracolonic Expressions, 19 Endodermal Abnormalities, 19 Mesodermal Abnormalities, 21 Ectodermal Abnormalities, 22

1

2

& PART I: COLORECTAL DISORDERS

& POLYPS OF COLON AND RECTUM The word ‘‘polyp’’ is a nonspecific clinical term that describes any projection from the surface of the intestinal mucosa regardless of its histologic nature. Polyps can be conveniently classified according to their histologic appearance: 1. Neoplastic tubular adenoma, villous adenoma, and tubulovillous, adenoma and serrated adenoma 2. Hamartomatous—juvenile polyps, Peutz-Jeghers syndrome (PJS), Cronkhite-Canada syndrome, Cowden’s disease 3. Inflammatory—inflammatory polyp or pseudopolyp, benign lymphoid polyp 4. Hyperplastic

& NEOPLASTIC POLYPS Adenomas A neoplastic polyp is an epithelial growth composed of abnormal glands of the large bowel. A neoplastic polyp has been termed an adenoma and is classified according to the amount of villous component. Those with 0% to 25% villous tissue are classified as tubular adenomais, 25% to 75% as tubulovillous adenomas, and 75% to 100% as villous adenomas (1). Tubular adenomas (Fig. 1) account for 75% of all neoplastic polyps; villous adenomas (Fig. 2), 10%; and tubulovillous adenomas (Fig. 3), 15%. The villous growth pattern is most prominent in sessile large adenomas, particularly those located distally in the rectum. There remains considerable uncertainty as to the nature of villous growth, whether it is merely a manifestation of continued growth of tubular adenomas, or whether it is a distinct phenotype that may reflect an acquired genetic change. In favor of the former is the rarity of small villous adenomas and large purely tubular adenomas (1). Dysplasia describes the histologic abnormality of an adenoma according to the degree of atypical cells, categorized as low-grade (mild), moderate, and high (severe). Thus high-grade dysplasia designates a condition one step away from an invasive carcinoma. The frequency of highgrade dysplasia correlates with the size of the adenoma (Fig. 4). The term carcinoma-in-situ, or ‘‘intramucosal carcinoma’’ should be avoided, since it implies a biological potential for distant spread, which is unwarranted and could result in overtreatment (1).

FIGURE 1 & Tubular adenoma.

FIGURE 2 & Villous adenoma.

Neoplastic polyps are common. Since data on the clinical recording of adenomas may be biased due to selection of patients and diagnostic methods, most accurate epidemiologic data on adenomas are obtained from autopsy studies. In autopsy series adenomas are present in 34% to 52% of males and 29% to 45% of females over 50 years of age. Most adenomas (87–89%) are less than 1 cm in size (2,3). The number, but not the size, of adenomas, increases with age (2). Carcinomas are found in 0% to 4% (2–5). The National Polyp Study, a multicenter randomized clinical trial in the United States, included 3371 adenomas in 1867 patients detected by colonoscopy (6). This study gives valuable information regarding the natural history and characteristics of polyps: 66.5% of polyps were adenomas, 11.2% were hyperplastic, and 22.3% were classified as ‘‘other’’ (normal mucosa, inflammatory and juvenile polyps, lymphoid hamartomas, submucosal lipomas, carcinoids, and leiomyomas). The majority of the adenomas (69%) were in the left colon (Table 1). The sizes of the adenomas were
0.5 cm, 38% 0.6 to 1 cm, 37% and 1 cm, 25%. It is important to note that the size, the extent of villous component, and the increasing age are independent risk factors for high-grade dysplasia. The increased frequency of high-grade dysplasia in adenomas located distal to the splenic flexure is attributable mainly to increased size and villous component rather than to location per se. Multiplicity of adenomas affects the risk of high-grade dysplasia but is dependent on size and villous component and thus is not

FIGURE 3 & Tubulovillous adenoma; mixture of tubular and villous glands.

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

3

TABLE 2 & Relationship Between Size of Adenoma and Carcinoma Size (cm) < 0.5 0.6–1.5 1.6–2.5 2.6–3.5 > 3.5

Adenoma (No.)

Invasive Carcinoma (%)

5027 3519 1052 510 1080

0 2 19 43 76

Source: From Ref. 7.

FIGURE 4 & Relationship between adenoma size and frequency of dysplasia. Source: From Ref. 6.

an independent factor (6). Invasive carcinomas are uncommon in adenomas < 1 cm, and the incidence increases with and increased size of the adenomas (Table 2) (7,8).

Adenoma-Carcinoma Sequence The Observation The concept that carcinomas of the colon and rectum derived from benign adenoma was observed by Dukes (9) of St. Mark’s Hospital, London, in 1926. Jackman and Mayo (10) coined the term adenoma-carcinoma sequence in 1951. After decades of debates and challenges by those who believed that carcinoma of the colon and rectum derived de novo (11,12), the adenoma-carcinoma sequence has finally become widely accepted and currently is the rationale of the approach to the secondary prevention of colorectal carcinoma by colonoscopic polypectomy (1,13–16). Circumstantial evidence supporting the adenoma-carcinoma sequence abounds and explains the high concurrence rate of carcinoma and adenoma and the frequent findings of contiguous benign adenoma in the resected carcinoma (17). Numerous studies (most of which are retrospective), based on tumor registry reports, hospital records, pathology reports, surgical specimens, and colonoscopy show a coexistence of adenomas and adenocarcinomas of the colon and rectum ranging from 13% to 62% (18). The cumulative incidence curve of adenomas based on data from the Norwegian Cancer Registry precedes the corresponding incidence

curve of carcinomas by about five years (Fig. 5). It should be kept in mind that adenomas are first diagnosed and reported to the cancer registry simultaneously with the diagnosis of colorectal carcinoma, indicating a longer time span between the two types of lesions than the curve indicates. It is manifested also in the natural history of both familial adenomatous polyposis (FAP) and hereditary nonpolyposis colon cancer (HNPCC) syndrome. The latter was originally thought to offer support to the de novo school of thought but several studies have since demonstrated coexisting and contiguous adenomas associated with HNPCC carcinoma with a frequency similar to that observed with sporadic carcinomas (1). Due to the high prevalence of adenomas and the relatively far less frequent incidence of carcinomas, only a small proportion of adenomas give rise to carcinomas (20). Although the adenoma-carcinoma sequence concept has been favored by most authors as the main pathogenesis of colorectal carcinoma, the ‘‘de novo’’ origin of carcinoma developing from normal mucosa has received some attention in recent years as an alternative pathway (19). In support of this de novo theory, authors (21–23) reported early colorectal carcinomas without evidence of adjacent adenomatous cells. In the series reported by Stolte and Beckte (22) of 155 such lesions, 59% of the lessions were Polyponl and 34% were flat. However, proponents for the adenoma-carcinoma sequence may argue that these types of lessions are so aggressive that the infiltration destroys the adenomatous remuants. Muto et al. (24)

TABLE 1 & Distribution of Colorectal Adenomas Diagnosed by Colonoscopy Site

(%)

Cecum Ascending colon Hepatic flexure Transverse colon Splenic flexure descending colon Descending colon Sigmoid colon Rectum Total

8 9 4 10 4 14 43 8 100

Source: From Ref. 6.

FIGURE 5 & Cumulative incidence of colorectal adenomas and carcinomas recorded in the Norwegian Cancer Registry 1983 to 1985. Source: From Ref. 19.

4

& PART I: COLORECTAL DISORDERS

thought that all genetic alterations may take place rapidly, one after another, without a chance for morphologic changes to be expressed as seen in the adenoma-carcinoma sequence. They said, ‘‘until a specific responsible gene for de novo carcinoma is detected, de novo carcinoma arising directly from normal mucosa is only an imaginary entity. Until then, the term ’de novo’ carcinoma is better avoided and instead de novo-type carcinoma should be used.’’ Molecular Genetics Molecular genetic discoveries provide substantial support for the adenoma-carcinoma sequence concept (25). An adenoma represents an epithelial proliferation derived from a single cell (crypt). Its de-velopment occurs as a series of genetic mutations. The progression of colorectal epithelium from normal to adenoma to carcinoma can be simplified as in Figure 6. The initial step in colorectal carcinogenesis is the mutation in the adenomatous polyposis coli (APC) gene on chromosome 5q. The APC gene is inactivated, causing the affected cells to proliferate. These cells are thus primed for subsequent growth-enhancing mutation, which is more likely because of the increased rate of cell division. Hypomethylation of DNA has been identified as the next factor involved in colorectal carcinogenesis. Loss of methylation of CpG dinucleotides occurs in cells that are already hyperproliferative because of the inactivation of the APC gene. These changes produce a growth of the affected cells resulting in adenoma formation. Hypomethylation of DNA may be directly linked to the K-ras (Kirsten rate sarcoma virus) activation that enhances the dysplasia so that the neoplasia can progress. Because K-ras is an oncogene, thus mutation of one allele is enough to produce an effect. K-ras mutations can occur in the absence of APC gene mutations but, in this case, are usually limited to aberrant crypt foci (ACF) that do not progress to malignancy. In cells that have already suffered APC mutation (both alleles need two ‘‘hits’’), K-ras mutation will drive progression. Small adenomas tend to advance to intermediate adenomas. The transition from intermediate to advanced (or late) adenoma is associated with a distinct genetic alteration

on the long arm of chromosome 18. This alteration is correlated with the mutation of a gene that maps to 18q21, named deleted in colon cancer (DCC). Specific DCC mutation has been detected in a number of colorectal carcinomas and carcinomas that have lost the capacity to differentiate into mucus-producing cells that have uniformly lost DCC expression. The progress from advanced adenoma to carcinoma is frequently accompanied by loss of heterozygosity (i.e., mutation of one of two alleles) on chromosome 17p and mutation of the p53 gene that maps to 17p. These cumulative losses in tumor suppressor gene function accompanied by activation of dominant oncogenes drive the clonal expression of cells from the benign to the malignant site (25). A fuller account of molecular genetics of colon and rectal adenocarcinoma is provided in Chapter 23.

Diagnosis of Large Bowel Adenomas Clinically, there are two morphologic types of polyps, pedunculated and sessile. The pedunculated polyp has a stem lined with normal mucosa, called a stalk or a pedicle, and has the appearance of a mushroom (Fig. 7). A sessile polyp grows flat on the mucosa (Fig. 8). A pedunculated polyp rarely is > 4 cm in diameter, whereas a sessile polyp can encompass the entire circumference of the large bowel. Adenomas of the large bowel are usually asymptomatic and are frequently discovered during routine radiologic studies or endoscopic examinations. Bleeding per rectum is the most common finding if the polyp is situated in the rectum or sigmoid colon. A large pedunculated polyp in the lower part of the rectum may prolapse through the anus. A large villous adenoma may manifest as watery diarrhea; in rare instances it causes fluid and electrolyte imbalance. Intermittent abdominal pain from recurrent intussusception or spasm may occur with a large colonic polyp but is unusual. Mild anemia may follow chronic bleeding from an ulcerative polyp. With a small polyp, up to 8 mm, biopsy and electrocoagulation can be performed, preferably using a ‘‘hot’’ biopsy forceps for histopathologic examination. A large polyp should be completely snared or excised and sent for histopathologic

FIGURE 6 & A genetic model for the adenoma-carcinoma sequence. Tumorigenesis proceeds through a series of genetic alterations that accumulate. The histopathologic stages of colorectal tumor development are shown with increasing size and dysplasia until an invasive carcinoma is formed. Abbreviations: DCC, deleted in colon cancer; APC, adenomatous polyposis coli. Source: Modified from Ref. 26.

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

5

follow-up colonoscopy every three to five years is adequate. A large sessile polyp, particularly villous type, is prone to recur, and a follow-up check of the polypectomy site should be done every 3 to 6 months the first year, every 6 to 12 months the second year, and every year thereafter to the fifth year. Then colonoscopic examination every three to five years is appropriate.

FIGURE 7 & Pedunculated polyp.

examination. A biopsy of a large polyp does not represent the entire lesion and presents difficulty in the interpretation of an invasive carcinoma. Occasionally, biopsy may cause displacement of the gland into the submucosa and can be misinterpreted as an invasive carcinoma (27). This pseudoadenomatous invasion can also be caused by trauma from hard feces, repeated twisting of the stalk with subsequent ulceration of the surface (28).

Management of Benign Adenomas Colonoscopy has revolutionized the management of large bowel polyps. Most polyps throughout the entire colon and rectum can be excised through the colonoscope with minimal morbidity. At the present time, colonic resection or colotomy and polypectomy are reserved for cases in which colonoscopic polypectomy cannot be performed, such as lesions that are too large or too flat, or when the colonoscope cannot be passed to the site of the polyp. Most pedunculated polyps can be snared in one piece since the pedicles are rarely > 2 cm in diameter. Sessile polyps < 2 cm usually can be snared in one piece. Large sessile polyps should be snared piecemeal and in more than one session as appropriate. Excised polyps must be prepared properly and sectioned so that all the layers can be examined microscopically and the evidence of invasive carcinoma detected. Adenomas in the rectum present a unique situation. These lesions can be palpated with finger, suction, or endoscope. If there is no induration, the chance that a lesion is benign is 90% (29,30). There are a number of ways to remove a large adenoma in the rectum, including proctoscope or a colonoscope, per anal excision, trans anal endoscopic microsurgery and posterior proctotomy (see Chapter 19). Patients with a neoplastic polyp have a higher risk of developing another polyp; so follow-up colonoscopy is advised. After the colon and rectum are cleared of polyps,

FIGURE 8 & Sessile polyp.

The Flat Polyp In 1985, Muto et al. (31) called attention to a separate type of polyp called a ‘‘flat’’ adenoma. This type of polyp is unique in that it is usually small and flat, often with a central depression, and is difficult to detect with colonoscopy or even with the resected colon and rectal specimens. Ninety percent of flat adenomas are < 1 cm and more than half are less than 5 mm (32). The significance of flat adenomas is the high incidence of carcinomas, which occur in 6% of patients, even when the lesions are as small as 2 to 4 mm, and rapidly rise to 36% when the lesions are 9 to 10 mm. Approximately 10% of the adenomas in the Muto series were flat adenomas. They were most frequently located in the left colon and the rectum. Lynch et al. (33) found similar flat adenomas in patients who were members of the same kindred under study for HNPCC. Most of the lesions were in the right colon. The flat adenomas, originally thought to occur mostly among Japanese, have also been found in studies from Australia, Canada, and the United Kingdom (32). In a prospective study of 1000 executive patients attending for colonoscopy, flat or depressed lesions were examined by Rembacken et al. (34). Patients were not preselected and the indications were similar to other units in the United Kingdom. A flat adenoma was defined as mucosal elevations with a flat or slightly rounded surface and a height of less than half the diameter of the lesion. In practice, most flat adenomas were less than 2 mm in height and only very broad lesions were 5 mm high. During the examination, they used 0.2% indigo carmine dye, 3 to 6 mL, sprayed directly onto suspicious areas. Magnifying colonoscopy was also used. The authors identified 321 adenomas, 119 (37%) were flat and 4 (1%) appeared depressed. Fifty-four percent of the flat or depressed lesions were situated between splenic flexure and rectum. Seventy of the flat lesions (59%) were < 10 mm in size (mean, 5 mm) and 4% had early carcinoma (invasive into submucosa); 49 flat lesions (41%) were > 10 mm (mean, 21 mm), and 29% had early carcinoma. The mean size of the depressed lesions was 9 mm and three of four (75%) had early carcinoma, indicating their aggressiveness compared to other types of lesions. Rembacken et al. (34) suggested, ‘‘Western colonoscopists refuse training in the recognition of flat, elevated and depressed lesion in order to detect colorectal neoplasms in their early stages.’’ The readers should note that in this study, all of the patients had indications for colonoscopic examinations and not as a screening examination for low risk asymptomatic patients. In response to an editorial comment (35), Rembacken et al. wrote (34), ‘‘The use of indigo carmine dye is paramount to the detection of flat and depressed lesions and only takes a few seconds. Without the dye, it is difficult to evaluate non-polypoid

6

& PART I: COLORECTAL DISORDERS

lesions because they generally appear to be erythematous patches, easily mistaken for scope trauma. The magnifying colonoscope does not help in the initial recognition of lesions but allows the endoscopists to assess the crypt pattern and predict the histology.’’ Recent molecular analysis of such flat adenomas suggests that they are etiologically distinct from other polypoid adenomas (36). The mutation rate and the K-ras gene are both significantly reduced (16% in flat adenomas compared to 50% in ordinary colorectal adenomas) and do not occur in the same codons. The management of flat adenomas is the same as for sessile adenomas.

Why Remove a Polyp? It has generally been accepted that most colorectal carcinomas are derived from benign adenomas through the adenoma-carcinoma sequence. It takes about five years from a clean colon to the development of an adenoma and about 10 years from a clean colon to the development of invasive carcinoma (13). Thus, removal of an adenoma is prophylactic against the development of colorectal carcinoma. Gilbertsen (37), in a retrospective study, showed that removal of rectal polyps in patients under surveillance with yearly rigid proctosigmoidoscopy results in a lower than expected incidence of rectal carcinoma. This result was confirmed by Selby et al. (38) in a case-control study using rigid proctosigmoidoscopy; screening examination produced a 70% reduction in the risk of death from rectal and distal sigmoid carcinoma. The National Polyp Study also showed that colonoscopic polypectomy results in a lower than expected incidence of colorectal carcinoma (39). Most adenomatous polyps found on routine examination with rigid proctosigmoidoscopy, or through flexible sigmoidoscopy are small and have a minimal risk of harboring a carcinoma. Because we do not know, whether these small adenomas will continue to grow with eventual degeneration into an invasive carcinoma, their removal is logical provided it can be performed with minimal or no risk of complications. This approach also gives the opportunity to clear the colon and rectum and thus extends the follow-up time to several years. Another point of concern is whether the patient has a synchronous polyp or polyps more proximally and, if so, whether it is important to have it (or them) removed. The incidence of synchronous polyps beyond the reach of the rigid proctoscope or flexible sigmoidoscope is approximately 50% (6). However, most of these polyps are small and have little clinical significance. It is debatable whether a total colonoscopy should be performed in every person in whom a small polyp is found in the rectum or sigmoid colon. A small hyperplastic polyp frequently found in the rectum or sigmoid has no malignant potential, nor has it been shown to predict an adenoma in the proximal colon (40–42); therefore no further evaluation or follow-up is indicated. Church (43) studied diminutive (1–5 mm) and small (6–10 mm) adenomas of the colon and rectim and found that although the risk of invasive carcinoma was low (0.1 % and 0.2%, respectively) the risk of severe dysplasia was significant (4.4% and 15.6%, respectively). He advised a cold excision or a hot snare as appropriate (Table 3). A retrospective study using death from carcinoma as the end point, Atkin et al. (44) showed that the risk of

TABLE 3 & Risk of Diminutive and Small Adenomas

Size (mm)

No.

Severe Dysplasia (%)

Invasive Carcinoma (%)

1–5 (diminutive) 6–10 (small)

2066 418

4.4 15.6

0.1 0.2

No effect of age, site, or family history. Source: From Ref. 43.

development of carcinoma in the proximal colon is significant if the adenoma found in the rectum or sigmoid colon is > 1 cm, if the polyp has a villous component, and if there are multiple adenomas. The authors also found that if a tabular adenoma found in the rectum or sigmoid colon is
1 cm. This risk of carcinoma remote to these sites is insignificant.

Natural History of Untreated Large Bowel Adenomas A retrospective review of patients from the pre-colonoscopic era by Stryker et al. (45) analyzed 226 patients who had colonic polyps 10 mm in diameter and in whom periodic radiographic examination of the colon was elected over excision. Twenty-one invasive carcinomas were identified at the site of the index polyp at a mean follow-up of 108 months (range, 24–225 months). The risk of having a polyp 1 cm in size develop into an invasive carcinoma at 5, 10, and 20 years was 2.5%, 8%, and 24%, respectively. Further study of this same group of patients by Otchy et al. (46) revealed that the cumulative probability of developing an invasive metachronous carcinoma at a site different from the index polyp was 2% at five years, 7% at 10 years, and 12% at 20 years. Over a median duration of polyp surveillance of 4.8 years (range, 1–27 years), 11 (5%) of the index polyps disappeared, 129 (57%) had no growth noted, and 86 demonstrated growth. Forty-two of the 86 polyps (49%) had at least a twofold increase in size. Seventy-one of the 86 polyps were removed, and 24 (34%) were carcinomatous. Fifteen of the 86 polyps that increased in size were not removed, and none of these patients developed a carcinoma. Forty-three of the 129 polyps that did not grow were eventually removed. Five of those polyps had carcinoma and one of these patients also developed a metachronous carcinoma at a later date. In addition, two of the 43 patients developed a colon carcinoma in areas distant from the site of the index polyp. These data further support the recommendation for excision of all colonic polyps 10 mm in diameter and a periodic examination of the entire colon. Although this study has limitations inherent to any retrospective analysis, comparable prospective data are unlikely to be available in the future because of the widespread availability of colonoscopy and the compelling evidence to recommend the removal of neoplastic polyps. What Happens to Smaller Adenomas? Hofstad et al. (47) prospectively studied the growth of colorectal polyps. Colonoscopy was performed in 58 subjects. Polyps 10 mm were removed; polyps < 5 mm, and 5 to 9 mm were left behind for a follow-up study. Colonoscopy

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

was followed-up by one investigator once a year. On the third year, polyps were removed by snare or hot biopsy. The measurement of the polyps was performed by a measuring probe plus photography. On the third year, 7 of 58 patients had only hyperplastic polyps. Twenty-nine individuals had one adenoma, 17 individuals had two to three adenomas, 5 individuals had four to five adenomas. Twenty-five percent of all the adenomas were unchanged in size whereas 40% displayed growth and 35% showed regression or shrinking in size. Adenomatous polyps < 5 mm showed a tendency to growth, while the adenomas 5 to 9 mm showed a tendency to reduction in size. The hyperplastic polyps showed a similar pattern. There was a tendency to increase growth in the adenomatous polyps in the younger age groups reaching significance from initial examination to the third year and from the first to the second year of re-examination. Moreover, in the patients with four to five adenomas at the initial examination, the polyps showed larger growth than the polyps in patients with only one or two to three adenomas. There were no differences in polyp growth between the sexes. A similar prospective study by Bersentes et al. (48) on adenomas of the upper rectum or sigmoid colon, size 3 to 9 mm, showed no regression or consistent linear growth rates with a 2 year follow-up. In the study by Hofstad et al. (47), 86% of the individuals had at least one new polyp during the 3 years and 75% had at least one new adenoma. The newly discovered polyps were significantly smaller than the average size at initial examination. They were also more frequent in the proximal part of the colon (71%) than the polyps discovered at initial examination (38%). There were more new adenomas among those with more than four to five adenomas at initial examination, than those with one adenoma, reaching significance from initial examination to the first year of examination and from initial examination to third year. There were more new adenomas among patients  60 years of age than those < 60 years. No differences were found between the sexes.

Management of Adenomas with Invasive Carcinoma The term ‘‘invasive carcinoma’’ is applied only when the malignant cells have invaded the polyp, either sessile or pedunculated, partially or totally, through the muscularis mucosa into the submucosa. Carcinoma superficial to the muscularis mucosa does not metastasize and should be classified as atypia (rather than carcinoma in situ or superficial carcinoma) (13). For this type of lesion, complete excision is all that is necessary. Follow-up of these polyps is the same as for benign polyps. A polyp with invasive carcinoma or a malignant polyp is an early carcinoma. For the TNM classification, it is a T1NxMx. Local excision for a malignant polyp can be curative if the lession can be adequately excised and if the lession has not spread to the regional lymph nodes or if there are no distant metastaces. In 1985, Haggitt et al. (44) proposed a classification for polyps with adenocarcinoma according to the depth of invasion as follows (Fig. 9): Level 0—Carcinoma in situ or intramucosal carcinoma. These are not invasive.

7

FIGURE 9 & Anatomic landmarks of pedunculated and sessile adenomas. Source: From Ref. 49.

Level 1—Carcinoma invading through the muscularis mucosae into the submucosa but limited to the head of the polyp (i.e., above the junction between the adenoma and its stalk). Level 2—Carcinoma invading the level of the neck of the adenoma (junction between adenoma and its stalk). Level 3—Carcinoma invading any part of the stalk. Level 4—Carcinoma invading into the submucosa of the bowel wall below the stalk of the polyp but above the muscularis propria. By definition, therefore, all sessile polyps with invasive carcinoma are in level 4.

Pedunculated Polyp with Invasive Carcinoma Using Haggitt classification, the risk of lymph node metastasis for pedunculated polyp Haggitt level 1, 2, and 3 is low (49–52). For these lesions, a complete snaring or a transanal excision is adequate. A close follow-up examination with endoscopy to detect a local recurrence should be performed every 3 to 6 months for the first year. This period can be extended to every 6 to 12 months in the second year and to every year for the next 2 years. Thereafter, endoscopy every 3 years is adequate. There have been reports in the literature that undifferentiated carcinoma and invasion of the malignancy into the lymphatic or vascular channels have a high risk of lymph node metastasis. In such situations, bowel resection should be performed even though the invasion is limited to the head of the polyp (53–57). A pedunculated polyp with level 4 invasion is treated the same way as a sessile lesion. Sessile Polyp with Invasive Carcinoma The Haggitt classification has been widely used for pedunculated polyps with invasive carcinoma in the United States but it is not adequate for sessile lesions. In 1993, Kudo (58) classified the submucosal invasion of the sessile lesions into three levels (Fig. 10): Sm1—invasion into the upper third of the submucosa. Sm2—invasion into the middle third of the submucosa. Sm3—invasion into the lower third of the submucosa.

8

& PART I: COLORECTAL DISORDERS

FIGURE 10 & Incorporation of Haggitt classification to Sm system. Abbreviations: Sm1 ¼ Invasion into upper 1⁄3 of submucosa. Sm2 ¼ Invasion into middle 1⁄3 of submucosa. Sm3 ¼ Invasion into distal 1⁄3 of submucosa. Haggitt’s pedunculated levels 1, 2, 3, are all in Sm1; pedunculated level 4 can be Sm1, Sm2, or Sm3. Source: With permission from the Mayo Foundation.

The consensus workshop in Paris on November 30 to December 1, 2002, recommended Sm system for early carcinoma of the large bowel (59). The Sm system appears to be effective and practical. In the series by Nascimbeni et al. (60), the pathologist could evaluate the depth of invasion into Sm1, Sm2, and Sm3 in 97% of the cases. In fact, the Haggitt level for the pedunculated lesion can be incorporated into the Sm system (Fig. 10). The endoscopists must properly prepare the specimens and the pathologists must properly section them in order to examine the entire layers. Among the increased risk factors for early colorectal carcinoma reported in the literature are: lymphovascular invasion, poor differentiation, gender, extensive budding, micro-acinar structure, flat or depressed lesions, and depth of invasion in submucosa (61). Recent studies with multivariate analysis showed the independent risk of lymph node metastasis in early carcinoma of rectum to be: lymphovascular invasion and invasion into the depth of the submucosa (Sm3) (60,62). When the rectum was divided into three levels, Nascimbeni et al. (60) showed that the lower third of rectum had a high risk of lymph node metastasis. Some sessile lesions with invasive carcinoma < 2 cm in diameter in the colon, upper third and middle third of rectum can be adequately snared in one piece via colonoscopy. A microscopic free margin of at least 2 mm is considered adequate (53). A malignant lesion that is removed piecemeal requires further excision or resection. A sessile lesion that has high risk factors such as lymphovascular invasion and deep invasion into Sm3 level should have an oncologic resection. In case of a lower third rectal lesion, a full thickness transanal excision is required. Some authors advise postoperative radiation or chemoradiation (63–65); some series showed no advantage, or showed high recurrence rate (66,67). Benson et al. (67) reported a series of 21 patients with T1 adenocarcinoma of the lower third of the rectum (median 4 cm from anal verge) underwent radiation therapy without chemotherapy after a local transanal excision, the recurrence at five years was 39% and the disease-free survival at 5 years was 59%. A sessile polyp with invasive carcinoma or an early carcinoma of the low rectum is unique in that in spite of favorable histopathologic parameters, a full thickness transanal excision has a high recurrence rate from 7% to 29% and a cancer-specific 5-year survival of 74% to 95% (Table 4) (66,68–70). In a retrospective study by Nascimbeni

et al. (70) on the outcome comparing transanal local excision to oncologic resection for T1 carcinoma of low rectum revealed that for carcinoma of middle-third rectum or lower-third rectum, the 5-year and 10-year outcomes were significantly better for overall survival and cancer-free survival in the oncologic resection group. Local recurrence and distant metastasis were not significantly different. When it came to T1 carcinoma of the lower-third rectum, the authors showed the oncologic resection group had a trend of improved survival but was not statistically significant, possibly because of low statistical power from the small sample size. Suitable cases of T1 carcinoma of rectum for a transanal excision are uncommon. Some authors recommended to do it even fewer (69). Transanal excision for a sessile polyp with invasive carcinoma, or a T1 carcinoma of the low rectum has a three to fivefold higher risk of carcinoma recurrence compared with patients treated by radical resection (71). Waiting to perform a radical resection after a local recurrence is a poor choice. In most series, the cancer-free survival for salvage resection in these patients is 50% to 56% (68,69). On the other hand, an immediate radical resection after local excision (within 1 month), gives a better prognosis, 94% cancer-free survival at 10 years and is comparable to primary resection in a case-controlled comparison (72). In short, local excision for a sessile polyp with invasive carcinoma (T1) of the lower third of rectum has high local recurrence. It appears that the early lesion at this site is a locally disseminated disease. To improve the outcome, the recurrence rate has to be improved: options include doing more radical resection in young and good health patients;

TABLE 4 & Selected Series of Local Recurrence and Survival After

Transanal Excision for T1 Carcinoma of the Rectum

Institution University of Minnesota (68) Memorial Sloan-Kettering (66) Cleveland Clinic (69) Mayo Clinic (70)

No.

LR (%)

5-Yr Survival (%, CSS)

F-U (mo)

69

18

95

52

67

14

74

60

52 70

29 7

75 89

55 60

Abbreviations: LR, local recurrence; CSS, cancer-specific survival; F-U, follow-up.

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

9

finding a better adjuvant therapy; or finding better ways in selection of patients, such as molecular markers in the future.

Serrated Adenoma This is the term coined by Longacre and Fenoglio-Preiser in 1990 (73) to describe a new entity of mixed hyperplastic polyp/adenomatous polyp. In their study of 110 serrated adenomas, compared to 60 traditional adenomas and 40 hyperplastic polyps, they found that these lesions distributed throughout the colon and rectum, with a slight preponderance of large lesions ( > 1 cm) occurred in the cecum and appendix. There are two types of mixed epithelial polyps: one in which adenomatous and hyperplastic glands are mixed (Fig. 11A), and one in which the adenoma has a serrated appearance on microscopic examination (Fig. 11B). Microscopic examination of the lesions shows goblet cell immaturity, prominent architectural distortion, cytologically atypical nuclei, rare upper zone mitoses, and absence of a thickened collagen table (73,74). Grossly the lesion is flat and smooth; it may look like a plaque or thickened mucosa on colonoscopic examination (Fig. 12). This type of lesion can be easily missed on colonoscopy if the colon is overdistended stretching it flat or underdistended causing wrinkle on mucosa to mask it. Unlike the classic hyperplastic polyps that are small and

FIGURE 12 & Plaque-like serrated adenoma in transverse colon.

restricted to the rectum and rectosigmoid colon, serrated adenomas are larger and occur in both proximal and distal colon and rectum (75). Some of the individuals previously reported as having multiple hyperplastic polyps could instead have had multiple serrated adenomatous polyps (73).

Clinical Importance of Serrated Adenoma Based on the observation that 11% of serrated adenomas in the series of Longacre and Fenoglio-Preiser (73) contained foci of intramucosal carcinoma, it was surmised that an individual lesion would carry a significant malignant potential. Nevertheless, the rarity of serrated adenoma (0.6% of colorectal polyps) would minimize their contribution to the overall burden of colorectal malignancy (76). Torlakovic and Snover (74) reported six patients with serrated adenomatous polyposis. Each patient had at least 50 polyps, ranging from 0.3 to 4.5 cm in size, mostly sessile. Three patients had diffuse polyps, two patients had the polyps in the left colon, and one patient had them in the right colon. Four patients had carcinoma. Why are serrated adenomas infrequently observed in endoscopic practice? The answer may lie partly in under diagnosis and partly in their rapid evolution to carcinoma. The latter suggestion is supported by the demonstration of DNA microsatellite instability in mixed polyps and serrated adenomas and by analogy with the aggressive adenomas in hereditary nonpolyposis colorectal cancer (HNPCC) (76).

FIGURE 11 & (A) Mixed hyperplastic gland (red arrow) and adenomatous gland (black arrow). (B) Adenomatous gland with serrated appearance (arrow). Source: Courtesy of Thomas C. Smyrk, M.D., Mayo Clinic, Rochester, MN, U.S.A.

Genetics The known alterations include K-ras mutation, low and occasional high level microsatellite instability, 1pLOH, and methylation of HPP1/TPEF (a putative anti-adhesion molecule). Additional genetic alterations may be observed in neoplastic subclones occurring within or adjacent to hyperplastic polyps. These include loss of expression of MGMT or hMLH1 (77). Sporadic MSI-L and MSI-H carcinomas may evolve through the serrated adenoma pathway (76). The serrated adenoma pathway is likely to show marked molecular heterogeneity, but patterns are beginning to emerge. The view

10

& PART I: COLORECTAL DISORDERS

K-ras ACF

N K-ras HP

HP

N Meth BRAF

HP-like

N

Dys

MSI-H CRC

Dys

MSI-L CRC

hMLH1 HP-like

MGMT

FIGURE 13 & Possible key steps in three pathways to aberrant crypt foci (ACF), hyperplastic polyp, and hyperplastic-like polyp or small serrated adenoma. (See details in text.) Abbreviations: N, Normal mucosa; HP, hyperplastic polyp; Meth, methylation; BRAF, a gene encoding a kinase that is regulated by K-ras; DYS, dysplasia; MSI-H, microsatellite instability-high; MSI-L, microsatellite instability-low; CRC, colorectal carcinoma. Source: From Ref. 77.

that all, or even most, colorectal carcinomas are initiated by mutation of APC gene and evolve through the classical adenoma-carcinoma sequence may no longer be tenable. This understanding will surely transform our approach to the early detection and prevention of colorectal carcinoma (76). Figure 13 illustrated the possible key steps in three pathways to ACF, hyperplastic polyp, and hyperplastic-like polyp or small, serrated adenoma. The molecular steps that determine growth of ACF into hyperplastic polyp are not known. Colorectal carcinoma is envisioned to arise from hyperplastic-like polyps (or sessile serrated polyps) in which the earliest events might be BRAF mutation synergizing with a methylated and silenced pro-apoptotic gene. Subsequent methylation of hMLH1 or MGMT then predisposes to mutation, dysplastic change, and finally to malignancy that is frequently characterized by MSI-H or MSI-L status. K-ras mutation may substitute for BRAF in methylator pathways culminating in MSI-L and some MSS colorectal carcinomas (78). Management Serrated adenomas are neoplastic polyps. The treatment is the same as in adenomatous polyps.

& HAMARTOMATOUS POLYPS A hamartoma is a malformation or inborn error of tissue development characterized by an abnormal mixture of tissues endogenous to the part, with excess of one or more of these tissues. It may show itself at birth or by extensive growth in the postnatal period. Juvenile Polyps and Juvenile Polyposis Juvenile polyps characteristically occur in children, although they may present in adults at any age. This type of polyp is a hamartoma and is not pre-malignant. Macroscopically they are pink, smooth, round, and usually pedunculated. The cut section shows a cheeselike appearance from dilated cystic spaces. Microscopic pictures show dilated glands filled with mucus and an abnormality

FIGURE 14 & Juvenile or retention polyp. Note the Swiss-cheese appearance from dilated glands.

of the lamina propria, which has a mesenchymal appearance (Fig. 14). The muscularis mucosa does not participate in the structure of the polyp. Bleeding from the rectum is a common finding. A moderate amount of bleeding can occur if the polyp is auto-amputated, a phenomenon not seen in other types of polyps. Intussusception of the colon occasionally occurs if the polyp is large. Treatment is by excision or snaring through a colonoscope or a transanal excision. Juvenile polyposis is an entity characteristically and biologically distinct from solitary juvenile polyp or other polyposis. The condition was first observed by McCall et al. in 1964 (79). The term juvenile polyposis rather than juvenile polyposis coli is to be preferred as polyps are also found in the stomach and the small intestine (80). There are two types of juvenile polyposis: in infancy and in other variable age of onset (81).

Juvenile Polyposis of Infancy Juvenile polyposis of infancy is a rare form. No family history is found. The infant presents with diarrhea, either bloody or mucinous, anemia, protein-losing enteropathy, intussusception; rectal prolapse develops between 8 and 10 months of age and leads to significant morbidity (81,82). The entire gastrointestinal (GI) tract is usually affected; the prognosis depends on the severity and extent of GI involvement. Death occurs before the age of 2 years in severe cases (80). Surgery is indicated in cases of intussusception, or polypectomies in cases of rectal prolapse to reduce the leading point of the prolapse. Supportive care to replace fluid and electrolytes or total parenteral nutrition as indicated (82). Juvenile Polyposis in Childhood and Adult The majority of patients with juvenile polyposis manifest in their first or second decade but in 15% of patients, the diagnosis is delayed until they are adults. They usually present

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

11

with rectal bleeding and anemia. Family history of juvenile polyposis is found in 20% to 50% of patients. Various extracolonic abnormalities, described in 11% to 20% of cases, have included digital clubbing, pulmonary arteriovenous fistula, macrocephaly, alopecia, bony swellings, cleft lip, cleft palate, supernumerary teeth, porphyria, arteriovenous malformation affecting the skin, psoriasis, congenital heart disease, malrotation of the gut, abnormalities involving the vitello-intestinal duct, double renal pelvis and ureter, acute glomerulonephritis, undescended testes, and bifid uterus and vagina (80). Patients with juvenile polyposis usually have 50 to 200 colorectal polyps and a proportion have polyps in the stomach and small intestine. Some patients seem to have relatively few polyps, but these tend to be the parent of the prospectus. It is conceivable that the juvenile polyps are produced only within the first few decades and are subsequently lost through autoamputation. Thus, juvenile polyposis may be diagnosed when a relatively old and asymptomatic parent is screened colonoscopically and the smallest number of polyps found on this basis is 5 (81). Jass et al. (81) proposed a working definition of juvenile polyposis:

within a polyp, (ii) an adenoma showing no residual juvenile features (81). On the mechanism of polyp-cancer sequence in juvenile polyposis, Kinzler and Vogelstein (85) postulated, ‘‘an abnormal stroma can affect the development of adjacent epithelial cells is not a new concept. Ulcerative colitis is an autoimmune disease that leads to inflammation and cystic epithelium in the mucosa of the colon. Initially, the imbedded epithelium shows no neoplastic changes, but foci of epithelial neoplasia and progression to cancer eventually develops in many cases. The regeneration that occurs to replace damaged epithelium may increase the probability of somatic mutations in this abnormal microenvironment. The increased risk of cancer in juvenile polyposis syndrome and ulcerative colitis patients, therefore, seems primarily the result of an altered terrain for epithelial cell growth and can be thought of as a landscaper defect.’’

1. More than five juvenile polyps of the colorectum. 2. Juvenile polyps throughout the GI tract. 3. Any number of juvenile polyps with a family history of juvenile polyposis.

Management and Surveillance There is no good information about prophylactic colectomy or proctocolectomy to prevent occurrence of carcinoma. The decision on performing the operation should be dictated by the number and the site of the polyps. Polyps of the colon and rectum that are too numerous for colonoscopy and polypectomies should have an abdominal colectomy with ileorectal anastomosis (IRA) or proctocolectomy with ileal pouch-anal anastomosis (IPAA) or an ileostomy (80,86,90,91). In a series reported by Onsel et al. (90), 5 of 10 patients who underwent colectomy with IRA for juvenile polyposis required a subsequent proctectomy with a mean follow-up of 9 years (range 6–34 years). This and other studies suggest that proctocolectomy with ileoanal pouch procedure may be a better option as an initial operation (90,91). The proband and relatives of the first degree should be screened, probably starting in the later teen years, by upper and lower GI endoscopy. If this initial screen is negative, a follow-up endoscopy should be performed every 3 years (92). For patients who have had a colectomy or an ileoanal pouch, surveillance should be performed periodically (90,91). Howe et al. (93) recommended genetic testing as part of the workup. However, given the presumed genetic heterogeneity of this syndrome, failure to show a mutation in SMAD-4 does not support lengthening the surveillance interval to 10 years as they suggested (90).

On the other hand, Giardiello et al. (83) suggested that the patients with as few as three juvenile polyps should undergo screening for colorectal neoplasm. A Precancerous Condition Although there is no evidence that isolated juvenile polyp could be malignant, it is now well established that juvenile polyposis is a precancerous condition (81,84–87). The risk of GI malignancy in affected members of juvenile polyposis kindred exceeds 50% in a series of kindred reported by Howe et al. (84). In a classic paper on juvenile polyposis, Jass et al. (81) studied 87 patients with juvenile polyposis recorded in the St. Mark’s Polyposis Registry, including 1032 polyps and 18 patients with colorectal carcinoma. They found that about 20% of juvenile polyps did not conform to the classical description. Grossly, they formed lobular mass (instead of spherical). These atypical juvenile polyps also revealed relatively less lamina propria and more epithelium than that found in the more typical variety and often adopted a villous or papillary configuration. Epithelial dysplasia occurred in both typical and atypical juvenile polyps but was very much more frequent in the latter. Nearly 50% of the atypical juvenile polyps showed some degree of dysplasia; these resembled adenomatous dysplasia. Eighteen patients in this series had colorectal adenocarcinoma with a mean age of 34 years (range 15–59 years). A high proportion of carcinomas were mucinous and/or poorly differentiated and this is in accord with case reports from other authors. There is little direct information on the histogenesis of carcinoma in juvenile polyposis. Dysplasia has been shown to occur in two forms: (i) a focus of adenomatous change

Genetics Juvenile polyposis is an autosomal dominant condition (84). The germ-line mutation is in the gene SMAD-4 (also known as DPC-4), located on chromosome 18q21.1 (88,89).

Peutz-Jeghers Syndrome PJS is a rare autosomal dominant disease characterized by GI hamartomatous polyposis and mucocutaneous pigmentation. It was originally described by Peutz in 1921 but was not clearly identified until attention was brought to it by Jeghers, McKusick, and Katz (94) in 1949. The syndrome comprises of melanin spots of buccal mucosa arid lips; the face and digits may be involved to a variable extent, but mouth pigmentation is the sine qua non of this portion

12

& PART I: COLORECTAL DISORDERS

known genetic linkage to STK11, testing carries a sensitivity of 70%. In sporadic cases, genetic testing has sensitivity ranging from 30% to 67%. A significant proportion of familial and sporadic Peutz-Jeghers cases may result from mutations in genes other than STK11 (96).

FIGURE 15 & Peutz-Jeghers polyp. Note Christmas-tree appearance from branching of muscularis mucosa.

of the syndrome. The presence of polyps in the small bowel is a constant finding of this syndrome, but the stomach, colon, and rectum also may be involved. The characteristic Peutz-Jeghers polyp has an abnormal muscularis mucosa branching into the lamina propria, giving the appearance of a Christmas tree (Fig. 15). Diagnosis Giardiello et al. (95) defined a definitive diagnosis of PJS by the presence of histologically confirmed hamartomatous polyps, plus at least two of the following:

1. Family history of the syndrome. 2. Labial melanin deposits. 3. Small bowel polyposis. The diagnosis is ‘‘probable’’ if two of the three clinical criteria described above are present but without histopathological verification of hamartomatous polyps (95). Genetic testing may then be used to confirm the diagnosis (96). For patients without a family history of PJS, definitive diagnosis depends upon the presence of two or more histologically verified Peutz-Jeghers type hamartomatous polyps (97). For patients with a first-degree relative with PJS, the presence of mucocutaneous hyperpigmentation is sufficient for presumptive diagnosis (96). Genetics To date, the only identifiable mutations causing PJS affect the serine/threonine-protein kinase 11 (STK11, also known as LKB1) gene, located on chromosome 19p13.3. Although PJS is inherited in an autosomal dominant manner, up to 25% of documented cases are not familial. These sporadic cases are felt to be due to de novo mutations in STK11 or low penetrance variance (96). Genetic testing for STK11 mutations is available but they have variable sensitivity. In familial cases with a

High Risk of Cancers It is a well-known fact that patients with PJS have high risk of developing cancer in many parts of the body. However, the risk varies depending on how the studies are undertaken. Giardiello et al. (98) conducted an individual patient metaanalysis to determine the relative risk (RR) of malignancy in patients with PJS compared with general population. The authors used strict criteria for the analysis. Searches of MEDLINE EMBASE, and referenced articles yielded 94 articles. Only six publications which consisted of 210 individuals qualified for the study. The results showed that the RR for all carcinomas was 15.2. A statistically significant increase of RR was noted for: esophagus (57.0), stomach (213.0), small intestine (520.0), colon (84.0), pancreas (132.0), lung (17.0), breast (15.2), uterus (16.0), ovary (27.0). There was no risk for testicular or cervical malignancy. The cumulative risk for all malignancy was 93% from age 15 to 64 years old. Carcinoma in Peutz-Jeghers Polyps Ordinarily, hamartomatous polyps should not degenerate into malignancy. However, there have been reports of invasive adenocarcinoma in Peutz-Jeghers polyps of the small and large intestine, although the risk is not high. Giardiello et al. (95) did not detect invasive carcinoma within hamartomatous polyps in any of their patients. The polyps containing hamartomatous, adenomatous, and malignant components have been observed in PeutzJeghers polyps of the small and large intestine (99–103). Spigelman et al. (102) surveyed 72 patients registered with PJS at St. Mark’s Polyposis Registry. Four patients had nine carcinomas in hamartomatous polyps in stomach, duodenum, jejunum, and colon. This observation suggests that a hamartomatous, adenomatous, and carcinomatous progression may be important in the development of malignancy in Peutz-Jeghers polyps. Genetic analysis showed that STK11/LKB1 acts as a tumor suppressor gene and may be involved in the early stages of PJS carcinogenesis (104,105). The results suggest that Peutz-Jeghers related carcinoma have different molecular genetic alteration compared with those found in sporadic GI carcinomas (103). Peutz-Jeghers–Like Mucocutaneous Pigmentation Characteristic mucocutaneous pigmentation is often the clinical clue that heralds the diagnosis of PJS. The melanotic or lentiginous pigmented macules are dark brown, blue, or blue-brown and located on the vermillion border of the lips ( > 90%), buckle mucosa, digits, and occasionally on the periorbital, auricular, perianal, and vulva skin (106). The relevance of PJS-like hyperpigmentation in the absence of other features of PJS is not known. Boardman et al. (106) coined the terms isolated melanotic mucocutaneous pigmentation (IMMP). To ascertain the risk of malignancy for patients with IMMP, they identified a group of individuals with

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

mucocutaneous melanotic macules indistinguishable clinically from PJS hyperpigmentation but who did not manifest the other phenotypic characteristics of PJS. To distinguish those patients with possible or definite PJS from those with pigmentation only, the authors applied the diagnostic criteria of Giardiello et al. (95) to define definite PJS. Patients who had PJS-like oral hyperpigmentation only and none of the other criteria of PJS were classified as IMMP. Of 60 patients who had the diagnosis of PJS or PJS-like pigmentation were identified through the patient registry of the Mayo Clinic from 1945 to 1996. Twenty-six unrelated patients were identified with IMMP. There were 16 men and 10 women. The results showed that 10 individuals developed 12 noncutaneous malignancies including breast (n ¼ 1), cervical (n ¼ 3), endometrial (n ¼ 3), renal (n ¼ 1), lung (n ¼ 2), colon (n ¼ 1), and lymphoma (n ¼ 1). The median age of diagnosis of noncutaneous malignancy was 47 years (range 33–84 years); this compared to a median age of carcinoma in the general population of 68 years. In their previous review of carcinoma risk in PJS patients, the median age at diagnosis of carcinoma was 38 years (range 16–59 years) (105). The mean interval from the identification of the pigmentation to the development of carcinoma in IMMP patients was 24.2 years, compared to a mean latency period of 19.9 years in PJS patients (103,106). Although the magnitude and gender associations of carcinomas in patients with IMMP and PJS are remarkably similar, the authors detected no alterations in the LKB1 among IMMP patients. Is IMMP an entity distinct from PJS? The overlap in the two conditions of phenotypic pigmentary features and the increased risk of malignancy, specifically of the breast and gynecologic tract in women, support the notion that they might share a common genetic origin. Though none of nine individuals with IMMP had mutations in LKB1, 14% to 42% of patients with definite PJS lack LKB1 mutations, suggesting that another yet to be identified gene or genes may be responsible for cases of both PJS and IMMP not caused by LKB1 mutations (106). Based on the increased RR for gynecologic and breast carcinomas that they detected in their patient population of IMMP, the authors recommend following current screening guidelines for gynecologic and breast carcinoma with thorough evaluation of PJS-like pigmentation. They recommended examination of the GI tract at age 20 years in asymptomatic individuals with PJS-like hyperpigmentation. Screening Given the multitude of carcinomas that these patients are susceptible, aggressive screening protocols are recommended. Upper and lower GI endoscopies are indicated for any adolescent or adult suspected of having PJS. Radiographic studies should also be used to screen for distal small intestinal polyps. Pelvic ultrasound of females and gonadal examination in young men is also recommended. An at-risk, but unaffected relative is a first-degree relative of an individual with PJS who does not meet clinical criteria for PJS. Guidelines for surveillance of affected patients also apply to these at-risk family members. The current guideline for carcinoma screening is summarized in Table 5.

13

Management of Peutz-Jeghers Polyps The clinical course of PJS is characterized by asymptomatic periods interspersed with complications such as abdominal pain, intussusception often leading to frank intestinal obstruction, and hemorrhage that is often occult. Small bowel obstruction is the presenting complaint in half of the cases, and exploratory celiotomy due to polyp-induced complications occurs commonly and may do so at quite short intervals (107). Because this problem is coupled with the significant risk of malignancy in the polyps, the surgical approach is now more aggressive. The current approach is to operate on the patient if the small intestinal polyps are larger than 1.5 cm (107,108). Endoscopic resection of Peutz-Jeghers polyps throughout the small intestine at double-balloon enteroscopy without exploratory celiotomy has been reported to be successful (109). However, in general, an enteroscopy is performed at the time of exploratory celiotomy with polypectomy or resection of the small bowel (110,111). The indications for surgery included obstructing or intussuscepting polyps, polyps larger than 1.5 cm identified radiologically, or smaller polyps associated with iron deficiency anemia (111). In order to achieve more complete polyp clearance, Edwards et al. (111) analyzed their experience of using intraoperative enteroscopy in conjunction with explore celiotomy. The enteroscope was introduced through an enterotomy at the site of polypectomy for the largest polyps. Depending on the size of the polyps, snare polypectomy, electrocoagulation, or biopsies were performed. In their experience of 25 patients, enteroscopy identified 350 polyps not detected by palpation or transillumination of the bowel by an operating light. All the polyps were removed. There was one early complication of a delayed small bowel perforation at the site of a snare polypectomy that resulted in an urgent reoperation but no long-term sequelae. No patient in this group had required operative polypectomy within four years of polyp clearance by intraoperative enteroscopy, compared with registry data of 4 of 23 patients who had more than one exploratory celiotomy within a year. It appears that intraoperative enteroscopy for PJS improves polyp clearance without the need for additional enterotomies and may help to reduce the frequency of exploratory celiotomy (112). TABLE 5 & Screening Recommendations for Peutz-Jeghers

Syndrome

Organs

Age to Begin

Interval (yr)

Procedure

Colon Gastrointestinal tract Pancreas

25 10

2 2

Colonoscopy Upper endoscopy

30

1–2

Breast

20

Uterus

20

2 1 1

Cervix Testicular

20 10

1 1

Endoscopic ultrasound Transabdominant ultrasound Mammography Self-breast exam Transvaginal ultrasound Endometrial biopsy Pap smear Physical exam, ultrasound if clinically indicated

Source: From Ref. 96.

14

& PART I: COLORECTAL DISORDERS

Cronkhite-Canada Syndrome Cronkhite-Canada syndrome is characterized by generalized GI polyposis associated with alopecia, cutaneous pigmentation, and atrophy of fingernails and toenails (Onychotrophia). It was first deducted in two patients and described by Cronkhite and Canada in 1955 (113). Etiology The etiology is unknown. There is no familial inheritance pattern and no associated gene or mutation has been identified (114). Clinical Presentations Diarrhea is a prominent feature of this syndrome, accounting for 46 of 55 patients in the series of Daniel et al. (115). The cause of diarrhea is unknown. Nardone et al. (116) reported a case of Cronkhite-Canada syndrome associated with achlorhydria and hypergastrinemia causing direct gastric wall invasion by gram-negative Campylobacter pylori. This may explain the diarrhea in those patients. Hair loss was noted in 49 of 55 patients. In most patients, hair loss took place simultaneously from the scalp, eyebrows, face, axillae, pubic areas and extremities, but in some only loss of scalp hair was described (115). Nail changes were reported in 51 of 55 patients. In most of them, the nails showed varying degrees of dystrophy, such as thinning and splitting, and partial separation from the nail bed (onycholysis). Complete loss of all finger and toenails (onychomadesis), over a period of several weeks, was also noted in some patients (115). Hyperpigmentation was present in 45 of 55 patients, ranging from a few millimeters to 10 cm in diameter. The distribution of pigmentary skin changes could be anywhere, including extremities, face, palms, soles, neck, back, chest, scalp, and lips (115). Other manifestations include nausea, vomiting, weakness, weight loss, abdominal pain, numbness and tingling of extremities (115). Electrolyte disturbances are a prominent feature and appear to reflect malabsorption and losses from the GI tract. Total serum protein is also found to be low in most patients due to excessive enteric protein loss (115). The Polyps From radiologic, endoscopic, and autopsy data, the stomach and large intestine were involved in 53 of 55 cases. The actual frequency of small bowel involvement would be inaccurate because the small bowel X-rays and biopsies were not performed in every case, in the series of Daniel et al. (115). From the autopsy data, the number of polyps were greatest in the duodenum, less in the jejunum and proximal ileum, and again increased in the terminal ileum (115). The polyps consist of cystic dilatation of the epithelial tubules similar to that of juvenile polyps, but the lesions are usually smaller and do not show marked excess of lamina propria (115,117). Risk of Malignancy in Polyps The true incidence of GI carcinoma in CronkhiteCanada syndrome is unknown. In the review of literature by Daniel et al. (115) in 55 cases, they found six cases of carcinoma of the colon and/or rectum, including one case of carcinoma of the stomach. Some of these carcinomas

were multiple. Watanabe et al. (118) reported a case of Cronkhite-Canada syndrome associated with triple gastric carcinoma. Histopathologic examination revealed that the polyp underwent malignant transformation without an adenoma component. Management There has been no specific treatment. The management is symptomatic and the correction of any deficiencies. A complete spontaneous remission has been reported (119). Bowel resection is reserved for cases in which complications such as carcinoma, bleeding, intussusception, and rectal prolapse develop (115). Surgery is not usually performed for improvement of protein-losing gastroenteropathy because the protein losing is usually not localized (120). Hanzawa et al. (120) reported a patient with Cronkhite-Canada syndrome with numerous polyps in the stomach, duodenum, and from cecum to transverse colon. The patient had severe hypoproteinemia and peripheral edema, unresponsive to conservative treatment including elemental diet and hyperalimentation. Scintigraphy with technetium TC99m-labeled human albumin (121,122) demonstrated a protein-losing region in the ascending colon. An ileo-right colectomy was performed. After the operation, the protein-losing enteropathy stopped; the ectodermal changes improved, and other polyps that was a secondary cause to malnutrition regressed.

Cowden’s Disease Cowden’s disease is an uncommon familial syndrome of combined ectodermal, endodermal, and mesodermal hamartomas. The disease was named after the family name of the propositus by Lloyd and Dennis in 1963 (123). Eighty percent of patients present with dermatologic manifestations, such as keratosis of extremities, the most common being a benign neoplasm of the hair shaft: a trichilemmoma. If a patient is diagnosed with more than one trichilemmoma, consideration should be given to the diagnosis of Cowden’s disease. The second most common area of involvement is the central nervous system. Cowden’s disease in concert with cerebella gangliocytomatosis is referred to as the Lhermitte-Duclos syndrome. Approximately 40% of affected individuals have macrocephaly as a component of the syndrome. Only 35% of patients who meet the diagnostic criteria for Cowden’s disease have GI polyposis (124). Polyps in patients with Cowden’s disease are small, typically < 5 mm in diameter. Microscopic features are consistent with hamartomas, characterized by disorganization and proliferation of the muscularis mucosa with minimally abnormal overlying mucosa (125). Genetics Most patients with Cowden’s disease have been shown to subsume germ-line mutations in the PTEN gene located at 10q22 (126). PTEN is a tumor suppressor gene which has been shown to be involved with other forms of carcinoma such as familial thyroid carcinoma, inherited breast carcinoma, prostatic carcinoma, and malignant melanoma (124). Neoplastic Risk The majority of patients with Cowden’s disease will have some form of benign thyroid or breast disease. In addition,

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

the projected lifetime risk of thyroid malignancy is 10% and of breast malignancy is approximately 30% to 50% (127–129). There has been no reported increased risk of invasive GI malignancy to date (124). Management and Surveillance Screening and surveillance for breast malignancies should include a schedule of monthly breast self-examinations. Clinical examination should be undertaken annually, beginning in the late teen years or as clinically warranted by symptoms. Mammography should be implemented at the age of 25. Although no specific recommendations for thyroid surveillance have been published, annual screening by clinical examination should begin in the late teen years or as symptoms warrant. A thyroid ultrasound may be used in parallel every 1 to 2 years (124). GI polyposis should be addressed by endoscopic surveillance. Although no definitive increased risk of colorectal carcinoma has been documented, the syndrome is rare; thus, the true risk may be unrecognized (124).

Bannayan-Ruvalcaba-Riley Syndrome This disease encompasses three previously described disorders: Bannayan-Zonana syndrome, Riley-Smith syndrome, and Ruvalcaba-Myhre-Smith syndrome. In 1960, Riley and Smith noted an autosomal dominant condition in which macrocephaly with a slowed cycle motor development, pseudopapilledema, and multiple hemangiomas were observed (130). In 1971, Bannayan noted the congenital combination of macrocephaly with multiple subcutaneous and visceral lipoma as well as hemangiomas (131). In 1980, Ruvalcaba described two males with macrocephaly, hamartomatous intestinal polyposis, and pigmentary spotting of the penis (132). Given the clinical similarities between the conditions and the autosomal dominant pattern of inheritance, geneticists began to accept the notion of combining the disorders into a single entity as BannayanRuvalcaba-Riley syndrome (96). The syndrome gene is located at chromosome 10q23 (133). Intestinal polyposis affects up to 45% of these patients. Usually multiple hamartomatous polyps are identified with the majority limited to the distal ileum and colon, though they may be seen throughout the GI tract. Histologically, they appear similar to the juvenile polyposis-type polyp (96).

15

or schistosomal), resulting in partial loss of mucosa, leaving remnants or islands of relatively normal mucosa. Radiologically, both the acute and chronic forms appear similar. Distinction can be made with the proctosigmoidoscope, but in the chronic stage a biopsy may be necessary to distinguish the condition from familial polyposis. Inflammatory polyps are not premalignant, and their presence in no way influences the potential malignant status of the patient with ulcerative colitis, a development that remains related to the extent, age of onset, and duration of disease. That these polyps are not premalignant in ulcerative colitis is relative; the potential carcinomatous status of the pseudopolyp in this condition is no more or less than that of the adjacent mucosa (134). Benign lymphoid polyps are enlargements of lymphoid follicles commonly seen in the rectum. They may be solitary or diffuse. Their cause is unknown. Lymphoid polyps must not be confused with familial adenomatous polyposis (FAP). The histologic criteria set out by Dawson et al. (135) for the diagnosis of benign lymphoid polyps are as follows: the lymphoid tissue must be entirely within the mucosa and submucosa; there must be no invasion of the underlying muscle coat; at least two germinal centers must be present; and if the rectal biopsy fails to include the muscle coat and no germinal centers are seen, the diagnosis is inconclusive.

& HYPERPLASTIC POLYPS Hyperplastic polyps, also known as metaplastic polyps, are nonneoplastic polyps commonly found in the rectum as small, pale, and glassy mucosal nodules. Most are 3 to 5 mm located predominately in the left colon (136), although larger ones can be seen in the more proximal part of the colon. Histologic differentiation from neoplastic polyps presents no problem. The characteristic picture is a sawtooth appearance of the lining of epithelial cells, producing a papillary outline (Fig. 16). There is no nuclear dysplasia and thus no potential for malignancy. Despite the colonoscopic findings of more adenomas than hyperplastic polyps, autopsies from Hawaii, Finland,

Genetics Bannayan-Ruvalcaba-Riley syndrome is an autosomal dominant condition and, like Cowden’s disease, appears to be associated with genetic alterations in the PTEN gene (133). Neoplastic Risk There has been no increased risk of colorectal carcinoma, other GI malignancies, or extraintestinal malignancy documented in these patients (124).

& INFLAMMATORY AND LYMPHOID POLYPS Inflammatory polyps, or pseudopolyps, may look grossly like adenomatous polyps. However, microscopic examination shows islands of normal mucosa or mucosa with slight inflammation. They are caused by previous attacks of any form of severe colitis (ulcerative, Crohn’s, amebic, ischemic,

FIGURE 16 & Hyperplastic polyp. Note typical sawtooth appearance of the surface epithelium with a papillary appearance.

16

& PART I: COLORECTAL DISORDERS

and England demonstrate hyperplastic polyps in excess upto threefold over adenomas, with the great majority of them occurring in the sigmoid colon and rectum. In contrast adenomas are distributed fairly evenly along the length of the large bowel (137). The possibility of hyperplastic polyps serving as markers for adenomas has been raised in some colonoscopic data. It is clear, though, that the predictive value of the hyperplastic polyp is low, and the clinical usefulness of the marker must be critically questioned (137). Hyperplastic polyposis is a relatively new entity. The following criteria for hyperplastic polyposis have been proposed: (i) at least histopathologically diagnosed hyperplastic proximal to the sigmoid colon, of which two are greater than 10 mm in diameter; (ii) any number of hyperplastic polyps occurring in proximal to the sigmoid is an individual who has a first-degree relative with hyperplastic polyposis; (iii) more than 30 hyperplastic polyps of any size, but distributed throughout the colon (137). Although Williams et al. (138) found no association between hyperplastic polyposis and colorecal carcinoma, some of the polyps contained mixture of hyperplastic and adenomatous elements which nowadays would have been classified as serrated adenomas. Subsequent case reports and small series recorded the presentation of colorectal carcinoma in patients with hyperplastic polyposis (139,140). Colorectal carcinoma complicating hyperplastic polyposis is characterized by early age at onset multiplicity, frequent location in proximal colon, and greater likelihood of showing the molecular phenotype known as DNA microsatellite instability– high (MSI–H). The association between colorectal carcinoma and hyperplastic polyposis does not prove that carcinomas orginate within hyperplastic polyposis. Adenomas might coexist with hyperplastic polyposis and might be the precursors of colorectal carcinomas, or these polyposis are infact serrated adenomas. Little has been reported on the risk of metachronous adenomas in patients with hyperplastic polyps. Benson et al. (141) examined data from two large randomized colorectal chemoprevention trials for possible associations of hyperplastic polyps and adenomatous polyps with subsequent development of these lesions. Of the 1794 patients randomized in two trials, 1583 completed two follow-up colonoscopies, and are considered in their analysis. They computed rates of incidence on hyperplastic polyps and adenomas over the three-year follow-up after the first surveillance examination with polyp status (type and number) at that examination as predictors. During the threeyear follow-up, 320 (20%) had one or more hyperplastic polyps detected, and 564 (36%) had one or more adenomas. Patients with hyperplastic polyps at the first surveillance examination had a higher risk of any hyperplastic polyp recurrence on follow-up than those without hyperplastic polyps (odds ratio 3.67). Similarly, patients with adenomas at the first surveillance examination had a higher risk of adenoma recurrence than those without adenomas (odds ratio 2.08). However, the presence of hyperplastic polyps at the first surveillance examination was not significantly associated with adenoma occurrence during follow-up, nor was the presence of adenoma significantly associated with subsequent hyperplastic polyp occurrence.

& FAMILIAL ADENOMATOUS POLYPOSIS & DEFINITION AND NATURAL HISTORY Familial adenomatous polyposis (FAP) is an inherited, nonsex-linked and Mendelian–dominant disease characterized by the progressive development of hundreds or thousands of adenomatous polyps throughout the entire large bowel. The clinical diagnosis is based on the histologic confirmation of at least 100 adenomas (Fig. 17). However, with the widespread practice of family counseling and the genetic testing, this number of adenomas is no longer rigidly applied. In the absence of a family history of FAP, the number 100 or more is still good to entertain the diagnosis. The important feature of the disease is the fact that one or more of these polyps will eventually develop into an invasive adenocarcinoma unless a prophylactic proctocolectomy is undertaken. The disease has high penetrance, with a 50% chance of development of the disease in the affected family. Approximately 20% of patients with FAP have no family history and their condition represents spontaneous mutation (142). The term ‘‘FAP’’ is now used to replace the term ‘‘familial polyposis coli’’ because the disease also affects other organs. The older terms Gardner’s syndrome, familial polyposis of the gastrointestinal (GI) tract, familial multiple polyposis, and many other names should be avoided. The incidence of FAP is one in 7000 live births (143). Although the disease is congenital, there is no evidence that adenomas have ever been present at birth. In his extensive experience with the St. Mark’s Hospital, London, Polyposis Registry, Bussey (144) summarized the natural course of FAP in the average untreated patient as follows: Age Age Age Age Age

of appearance of adenomas: of onset of symptoms: of diagnosis of adenomas: of diagnosis of carcinoma: at death from carcinoma:

25 years 33 years 36 years 39 years 42 years

& CLINICAL MANIFESTATIONS AND DIAGNOSIS Symptoms usually do not develop until there is a fullblown development of polyposis. Bleeding from the rectum and diarrhea are the most common symptoms. The diagnosis is made by endoscopic examination of the colon and rectum or by barium enema studies. It must be confirmed by histologic findings of adenomatous polyps.

FIGURE 17 & Numerous small adenomatous polyps of the colon and rectum in a patient with familial adenomatous polyposis.

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

Only occasionally are tubulovillous adenomas found and villous adenomas are rare. The smallest possible microadenoma consists of only a single crypt, obviously not visible by examination with the naked eye (145). The average age at which the disease is diagnosed is 36 years. The adenomas actually appear much earlier, as is seen by comparison with the age of diagnosis in family members called for examination. In this group of patients, the average age is 24 years. Nearly two of three patients (65%) who were present because of symptoms already have carcinoma. The average age of colorectal carcinoma in these patients is 39 years, compared with 65 years in the normal population. Since most of the polyps in FAP are small, the best methods of diagnosis are colonoscopy and biopsy. A complete colonic examination has become important since rectal sparing has been reported, even when adenocarcinoma is present in the proximal colon (146).

& DISTRIBUTION OF POLYPS AND CARCINOMAS Although the rectum is almost invariably involved with polyps, the number of polyps in each segment of the colon and rectum varies from person to person. In general, the left colon has a higher density of polyps than the right colon (144). In any one patient, the polyps vary in size from barely visible mucosal nodules 1 or 2 mm in diameter, to up to 1 cm or larger. In some patients and families, the adenomas are mostly small, while in others they are large. Most patients with FAP have myriads of polyps, frequently up to 5000 (147). In a series from Denmark, the risk of developing carcinoma was highest in the rectum, followed by the sigmoid colon (Table 6) (147). & ATTENUATED FAMILIAL ADENOMATOUS POLYPOSIS This is a variant of FAP that has only relatively recently been recognized (148,149). The majority of patients who were present with between 1 and 50 adenomas, primarily located proximal to the splenic flexure and often morphologically flat. The polyps are diagnosed at the mean age of 44 years, and carcinomas at the mean age of 56 years. Thus, diagnosis of polyps and carcinomas in attenuated familial adenomatous polyposis (AFAP) is generally 10 to 15 years later than in FAP. However, because these data are based on when these lesions are detected and not necessarily on when they arise, the true age of development of polyps and carcinomas in AFAP is unclear. Certainly, lack of recognition of AFAP by patients and by physicians results

TABLE 6 & Distribution of Colorectal Carcinoma in 109 Propositions

Right colon Transverse colon Descending colon Sigmoid colon Rectum Total Source: From Ref. 147.

No. of Carcinomas

(%)

8 6 8 31 77 130

6 5 6 24 59 100

17

in fewer patients presenting for voluntary surveillance, perhaps contributing to a delay in diagnosis in these patients (150).

Clinical Features A striking feature of AFAP is the variability in number of polyposis within members of the same kindred. Some affected members have few polyps, while others have several hundred. This variability presents difficulties in classifying members of the same kindred as AFAP or FAP. Similar to FAP, colorectal carcinomas in patients with AFAP are generally accompanied by synchronous adenomas (150). The extracolonic manifestations in AFAP are similar to FAP. Church (151) argues that AFAP is not a distinct clinical entity. It is not distinct generally because a large number of different APC mutations can be expressed as AFAP. It is not distinct clinically because patients with fewer than 100 adenomas may have FAP, HNPCC, or multiple sporadic adenomas. It is not even distinct in a familial sense, because members of AFAP family may vary widely in the severity of their polyposis. The definition of AFAP, multiple but fewer than 100 synchronous colorectal adenomas, is arbitrarily one that suffers from its imposition of a finite number on a disease with a spectrum of subtle variations. He would rather regard AFAP as some patients with FAP with a mild expression of the colonic polyposis. This mild form of the colonic disease is most common with mutations at either end of the gene, and in many cases, the polyps are predominantly right sided. However, the underlying disease remains FAP. Diagnosis and Genetic Test The clinical diagnosis of AFAP is more difficult than that of classic FAP because of the wide variability of phenotypic expression, and overall lack of awareness of this syndrome. In addition, screening with flexible sigmoidoscopy, the recommended modality for classic FAP, is inadequate because the majority of colonic lesions in patients with AFAP are right-sided. For asymptomatic at-risk individuals belonging to known FAP or AFAP kindreds, genetic testing should be ideally performed between the ages of 10 and 15 years to determine the presence or absence of an APC mutation. A baseline colonoscopy and esophagoduodenoscopy at the time of genetic testing or by the age of 15 years should be performed (151). In patients with true-negative APC test results (a mutation has been demonstrated in an affected member but not in an at-risk member), a colonoscopy should be performed at the time of genetic testing or by the age of 15. Although the protein truncation test (PTT) is nearly 100% accurate in this setting, endoscopic evaluation serves as confirmation of a negative test. Because polyps occur later in AFAP individuals than in classic FAP, a second colonoscopy at age 20 should be considered to detect late, appearing polyps. If both examinations are negative, no further surveillance is necessary, and the patient may undergo future colorectal carcinoma screening as an average-risk individual (150). Church (151), however, has the opinion that people who test negative when their affected relatives test positive should be recognized that they do not have FAP and can be excluded from surveillance.

18

& PART I: COLORECTAL DISORDERS

Surgical Management Patients with AFAP are at increased risk for the development of colorectal carcinoma, although the exact risk remains unknown at this time. They do not have the near certainty of developing colorectal carcinoma that classic patients with FAP have. Thus, the indications for prophylactic colectomy differ between these two entities. In patients with few adenomas, colonoscopic polypectomy is sufficient to clear the affected bowel segments. When multiple polyps are clustered within a single segment of the colon, especially the cecum, resection may be the safest option. When resection is required, a total abdominal colectomy can be performed with an IRA. Because the rectal segment is generally uninvolved in these patients, total proctocolectomy with IPAA does not seem to be required. The rectal segment does need continued surveillance because this mucosa is still at risk. Total abdominal colectomy with IRA may also be required in patients who are difficult to examine fully by colonoscopy and, thus, unable to undergo proper surveillance (151). AFAP has two forms: patients with mutations at the five prime of APC are at minimal risk for desmoid disease, whereas patients with mutation in exon 15 are at high risk. This risk of desmoids, often manifest in other relatives who have had an operation, may encourage deferment of surgery. The alternative to colectomy, endoscopic polypectomy with or without chemoprevention, is risky especially when the patient has been shown to carry a germline APC mutation. Colonoscopic surveillance does not prevent carcinoma in all patients with HNPCC, the same can be applied to AFAP; this must be reserved for truly compliant patients who realize the risks (151). & MOLECULAR GENETICS Using genetic-linkage analysis, it has been determined that FAP is caused by a mutation in the tumor suppressor gene APC located on the long arm of chromosome 5q21–22. The term FAP is not used to describe this gene because familial amyloidotic polyneuropathy takes historical precedence in the genetic literature (152). The genetic alterations found in the FAP patient’s colon and rectal carcinoma are similar to those noted in sporadic carcinoma, except that an APC

mutation is already present constitutionally at birth (a germline mutation). There are correlations between the location of the APC mutation and the clinical phenotype. Figure 18 (153) shows the correlation between the APC genotype and the clinical phenotype. The 15 exons of the APC gene are shown. The locations of germ-line mutations associated with specific clinical phenotypes indicated by the dark horizontal lines. Thirty-four mutations causing AFAP have been reported to date; these are clustered either at the five prime ends (before codon 436) or at the three prime end (after codon 1596) of the APC gene. In contrast, mutations causing classic FAP are located in the central region, and mutations between codons 1250 and 1464 are associated with particularly severe polyposis. Abdominal desmoid tumors are more likely in persons with mutations between codons 1445 and 1578. The molecular mechanisms that explain why certain APC mutations result in a classic phenotype and others in an attenuated phenotype are currently being elucidated. Most models are predicated on the ‘‘two-hit hypothesis’’— which states that both alleles of APC must be inactivated in order to initiate tumorigenesis. In Figure 19 (153), both copies of chromosome 5 are shown. In classic FAP (panel A), the biallelic inactivation of APC is typically achieved by the combination of an inherited germ-line mutation in one allele (black X) and a chromosomal deletion of the remaining wild-type allele; this is called loss of heterozygosity. In some cases, the germ-line APC mutation (red X) can result in the production of a protein that can inhibit the activity of the wild-type protein (white X). This dominant negative effect functionally results in biallelic inactivation. In AFAP (panel B), the mechanism of APC inactivation is different. Germ-line mutations involved in AFAP may lead to the formation of alternative APC proteins that are initiated from an internal translation site that is located distal to the truncating mutation. This alternative APC protein does have functional activity. Because of this residual gene activity, an additional ‘‘hit’’ is necessary to fully inactivate APC (panel C). This third ‘‘hit’’ is indicated by the blue X. The second hit is often an intragenic mutation (green X) that inactivates the wild-type APC allele, rather than a large

FIGURE 18 & Correlation between the APC genotype and the clinical phenotype. (See details in text.) Source: From Ref. 153.

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

19

some indications that the location of the APC mutation itself may have an effect on the phenotype, although conclusive evidence for this proposal is lacking (155).

Endodermal Abnormalities Gastric Polyps With improved survival rates following colorectal resection, gastric polyps or upper GI lesions have become increasingly important because of the risk of malignant change in duodenal polyps. The introduction of flexible endoscopy has provided more ready access to the upper GI tract, although at present the course of the disease is not precisely known (156). The prevalence of gastric polyps ranges from 34% to 100%; most of them are hyperplastic type in the fundus of the stomach, and a few adenomatous types have been reported in the anturm (157,158). Of the 102 patients screened prospectively, in the series of Spigelman et al. (159), 56 had gastric polyps. Gastric fundus polyps were small (mean 4.7 mm) and multiple, whereas antral polyps, when present, were larger (mean 6.4 mm) and less numerous. Only 6 of 73 patients who had gastric biopsy revealed adenoma. When gastric adenomas are present they seem to be in patients who have duodenogastric reflux, in an area, exposed to bile (160).

FIGURE 19 & Mechanisms of inactivation of the APC gene in classic and attenuated familial adenomatous polyposis (FAP). (See details in text.) Source: From Ref. 153.

chromosomal deletion as in classic FAP. The red X represents the inherited APC mutation (153).

& EXTRACOLONIC EXPRESSIONS In 1951, Gardner (154) reported finding osteomatosis, epidermoid cysts, and fibromas of the skin, a triad in FAP known as Gardner’s syndrome. The detection of identical mutations in individuals with FAP and Gardner’s syndrome helps confirm that at the genetic level they are variants of a common entity (155). The disease affects the whole body, involving tissues derived from all three germ layers (145). Factors that contribute to the extracolonic manifestations are unresolved. Modifying genetic factors (e.g., other genes or different genetic backgrounds) or environmental variables probably play a role in the final phenotype. Likewise, the role that APC plays in the development of various extracolonic neoplasms and manifestations remains to be defined. There have been

Duodenal Polyp and Carcinoma In most series, duodenal adenomas occur in more than 90% of FAP patients, particularly in the periampullary region (161–163). The macroscopic appearance of duodenal polyps is very different to that of colonic polyps. The number of the former varies from invisible to over 100. They may present as multiple discrete adenomas (1–10 mm in diameter) or as flat confluent plagues. Sometimes no lesion can be seen and the only clinical abnormality is a prominent ampulla, or the mucosa may appear pale and seem to have a white covering which cannot be removed by rubbing. Biopsy of apparently normal mucosa frequently showed microadenomas (160). The lifetime risk of adenoma in FAP patients is high. Mutations downstream from coden 1051 seem to be associated with severe periampullary adenomas (164). Spigelman et al. (159) staged the duodenal polyposis according to polyp number, polyp size, and histologic type. The criteria provided a four-stage scoring system (Table 7). The classification allows estimation of the severity of duodenal polyposis. In a prospective study conducted by Domizio et al. (165), over 102 asymptomatic FAP patients were screened TABLE 7 & Staging of Duodenal Polyposis Grade Points Criteria Polyp number Polyp size (mm) Histology Dysplasia

1

2

3

1–4 1–4 Tubular Mild

5–20 5–10 Tubulovillous Moderate

> 20 > 10 Villous Severe

Note: Stage 0, 0 point; stage I, 1–4 points; stage II, 5–6 points; stage III, 7–8 points; stage IV, 9–12 points. Source: From Ref. 159.

20

& PART I: COLORECTAL DISORDERS

with side-viewing video endoscope, duodenal polyps were found to be multiple in two-thirds of patients, with onethird of patients having more than 20. The average size of a duodenal polyp was 9 mm, but they can be much bigger (2 cm). Duodenal polyps were almost always adenomas. Duodenal adenomas were tabular in architecture in about 70% of patients, tubulovillous in 20% and villous in 10%. Duodenal polyps were not seen in approximately 10% of patients. In just over one-half of these patients, micro adenomas were subsequently found. Presumably, if more random biopsies had been taken, more patients with microadenomas would have been found. Only 10% of patients had stage IV duodenal polyposis, while just under 20% had stage I disease and the remaining, 35% each, had stage II or stage III duodenal polyposis. Those with stage IV disease were older than the rest, implying that duodenal polyposis is a progressive disorder. Advanced duodenal disease might be a marker for the presence of gastric adenomas, as nearly all those with gastric adenomas had stage III or stage IV duodenal polyposis in their series (165). The risk of duodenal carcinoma in FAP has increased more than 100 times that of the normal population. Of 222 FAP patients who had a colectomy and IRA at St. Mark’s Hospital between 1948 and 1990 (inclusive), duodenal carcinoma accounted for 11 deaths, more than twice the number of deaths attributed to carcinoma of the rectal stump (166). A retrospective survey based on 10 polyposis registries in the Leeds Castle Polyposis Group showed duodenal and periampullary carcinoma in 30% of 1225 patients (167). The major causes of death in a series of 36 FAP patients treated with prophylactic proctocolectomy or colectomy with an IRA at the Cleveland Clinic were desmoid tumor (31%), periampullary carcinoma (22%), and rectal carcinoma (8%) (168). The cause of death from extracolonic diseases is also higher than carcinoma of colon and rectum in recent decades in the series of Belchetz et al. (169). One of the most difficult problems is the treatment of duodenal adenomas. Bile has been implicated in the pathogenesis of duodenal polyps in patients with FAP. FAP bile has been shown to contain an excess of carcinogens able to form DNA adducts. DNA adducts are chemical modification of DNA, formed by covalent binding of electrophilic carcinogens to DNA, which are implicated in the initiation of carcinogenesis because when they are left unrepaired they can lead to mutations. Modification of the action of these carcinogens may reduce the adduct load to the duodenum and so decrease actual duodenal polyp number. However, in the double-blind randomized placebocontrolled trial conducted by Wallace et al. (170), 26 patients with FAP were randomly assigned to ranitidine, 300 mg daily, or placebo for six months after baseline endoscopy. The result showed that acid suppression therapy does not seem to improve duodenal polyposis. Celecoxib (Celebrex1) has been shown to reduce the number of duodenal polyps. Phillips et al. (171) conducted a randomized, double-blind, placebo controlled study of celecoxib, 100 mg twice daily (n ¼ 34), or 400 mg twice daily (n ¼ 32), versus placebo (n ¼ 17), given orally twice daily for six months to patients with FAP associated with duodenal polyposis. Efficacy was assessed qualitatively by blinded review of shuffled endoscopy videotapes comparing the

extent of duodenal polyposis at entry and at six months and quantitatively by measurement of the percentage change in duodenal area covered by discrete and plaquelike adenomas from photographs of high- and low-density polyposis. The results showed a statistically significant effect of 400 mg twice daily celecoxib compared with placebo treatment. Overall, patients taking celecoxib, 400 mg twice daily, showed a 15.5% reduction in involved areas compared with a 1.4% for placebo. The authors suggested that celecoxib might have been indicated in patients with established duodenal disease, particularly when it is severe. However, it is harder to justify its use in patients with lesser duodenal disease, as progression to duodenal carcinoma in these patients with an earlier onset of the disease is unusual. Some authors successfully eradicated few adenomas of duodenum in FAP patients but the number of these patients had been too small to judge its efficacy (172,173). In general, endoscopic snaring or thermal contact can be performed only in patients with few small lesions as an initial treatment, or in patients who are not a candidate for major surgery. The natural history of untreated duodenal and ampullary adenomas in patients with FAP has been studied by Burke et al. (174). One hundred fourteen FAP patients who had two or more surveillance examinations were followed for a mean of 51 months (range 10–151 months). Duodenal polyps progressed in size in 26% (25 of 95), number in 32% (34 of 106), and histology in 11% (5 of 45) of patients. Morphology and histology of the main duodenal papilla progressed in 14% (15 of 110) and 11% (12 of 105) of patients, respectively. A minority of FAP patients had progression of endoscopic features and histology of duodenal polyps or the main duodenal papilla when followed over 4 years. An endoscopic surveillance interval of at least 3 years may be appropriate for the majority of untreated patients with FAP. An operation is indicated if polyps showed villous change, severe dysplasia, rapid growth, and induration at endoscopic probing (162). Duodenectomy or local excision is not preferred because of very high recurrences and complications (161,175–178). A pancreatoduodenectomy, particularly duodenectomy with preservation of pancreas, for patients with severe duodenal polyposis or patients who already had carcinoma appears to be the best option (176,179–181). The lifetime risk of duodenal adenomas approaches 100% (182). Recommendations concerning the age of initiation of upper tract surveillance are not uniform. Some propose that screening for upper GI disease should start at the time of FAP diagnosis. The National Comprehensive Cancer Network, after review of all case reports of duodenal carcinoma in FAP patients, recommended a baseline upper GI endoscopic examination at 25 to 30 years of age. In general, recommendations include stage 0 every 4 years; stage I every 2 to 3 years; stage II every 2 to 3 years; stage III every 6 to 12 months with consideration for surgery; and stage IV strongly consider surgery (182). Polyps in the Small Bowel Adenomas have been detected in the ileum following colectomy and IRA and also in Kock’s pouch following

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

proctocolectomy. A small number of cases of malignant neoplasms in the small bowel in association with FAP have been recorded; the risk of developing such a lesion appears minimal. Lymphoid polyps have also been noted in FAP, both in the small bowel and colon. Histological confirmation should be undertaken because presentation may mimic FAP (156).

Mesodermal Abnormalities Desmoid Tumors Patients with mutation between codons 1445 and 1578 frequently developed desmoid tumors (183). Desmoid tumors are benign tumors arising from fibroaponeurotic tissue. It is not known whether they are true neoplasms or the result of a generalized fibroblast abnormality; there is increasing evidence to support the former theory (184). Although a benign disease, desmoid tumors are focally invasive. They do not metastasize but can be lethal because of aggressive growth with pressure and erosion causing obstruction of the small bowel (Fig. 20). A report from the Finnish Polyposis Registry included 202 FAP patients, of whom 169 underwent colectomy. Desmoids were observed in 29 patients (14%): 15 (7%) in the mesentery, 10 (5%) on the abdominal wall, and four (2%) in other sites. The cumulative lifetime risk is 21.0%, 1.5%, 3.0%, 8.9%, 16.0%, and 18.0% at ages of 10, 20, 30, 40 and 50 years, respectively (185). Clark et al. (186) studied desmoid tumors from St. Mark’s Polyposis Registry Database. Eighty-eight patients had 166 desmoids (median age 32). Eighty-three patients (50%) had the tumor intra-abdominally, with 88% in the small bowel mesentery; 80 patients (48%) had the tumor on the abdominal wall, with 39% in surgical scars; three patients (2%) had the tumor extra-abdominally (chest wall, intra-thoracic). All (82%) but 16 patients had already undergone abdominal surgery. The behavior of desmoids in FAP ranges from rapid growth with symptoms resulting from visceral compression to a more indolent course, or even spontaneous regression. Plaque-like thickening of the small bowel mesentery and the peritoneum that does not amount to a discrete mass

FIGURE 20 & Large desmoid tumor of mesentery causing partial small bowel obstruction. Source: Courtesy of Roger R. Dozois, M.D., Mayo Clinic, Rochester, MN, U.S.A.

21

has been described as a relatively common finding at celiotomy in patients with FAP undergoing surgery (187,188). Hartley et al. (187) studied the natural history of these lesions that were incidentally found on celiotomy. A total of 266 patients underwent abdominal surgery for FAP. Incidental intra-abdominal desmoid tumors were identified in 34 patients, eight at the index surgery and 26 at receliotomy (median 130 months from the index procedure, range, 23–364 months). Intra-abdominal desmoids identified at the time of index surgery influenced the intended procedure in one of eight cases (6 cm mass in mesentery precluded IPAA). Intra-abdominal desmoids identified at second celiotomy influenced the intended surgery procedure in 10 of 26 cases (38%), including one for Kock pouch, two for IPAA, two difficult pouch reach, two prevented covering stoma, one iliac vein surgery, and two bypass only. Desmoid reaction was found in one of the index and 11 of the reexplore celiotomy group. This type of lesion was not an obstacle to the planned surgery. Desmoid reaction or mesenteric fibromatosis is precursor lesion for subsequent desmoid formation. However, the risk for progression in any individual case is likely to be small (188). Phillips (189) recommended computed tomography (CT) scan of any FAP patients before planned second major surgery, whether for pouch conversion or management of duodenal polyposis. The most common symptom in patients with intraabdominal desmoids is a painful abdominal mass (50%). The rest have a painless mass or no palpable mass. The pain is usually caused by bowel obstruction. Other causes of the pain are ureteric obstruction, direct pressure effects of the tumor, or hemorrhage into the tumor (190). The preferred investigation is the CT scan, which permits serial observation of the tumor (156,191). Magnetic resonance imaging (MRI) has been shown to provide adequate images of intra-abdominal soft tissue tumors while sparing the patient’s exposure to ionizing radiation (156,192). Intra-abdominal desmoids are a difficult clinical challenge. Their tendency to recur (65–85%) after removal has encouraged a conservative approach to management (190). Operation should be preferred for patients in whom lifethreatening complications have occurred as a result of local invasion (156). In many patients, operation is unavoidable. Middleton and Phillips of St. Mark’s hospital (193) were forced to remove a large intra-abdominal desmoid tumor from four patients (three patients had FAP). Three had complete excision their desmoids and all remained well with no recurrence at a median follow-up of 12 (range 7–14) months. Eight of 22 patients who underwent resection of their intra-abdominal desmoids at St. Mark’s died in the postoperation period (186). Church (190) cautioned that survey for a large intra-abdominal desmoid is technically extremely difficult, demanding high levels of skill and support. It should only be done in setting of a major medical center. On the other hand, abdominal wall and other superficial desmoids can be often cured by wide excision, especially when the tumor is small. The tumors are extremely resistant to radio therapy and cytotoxic chemotherapy (156). Encouraging reports have appeared following treatment with sulindac, with or without tamoxifen. Church and Others (190,194) suggested treatment of intra-abdominal desmoid tumors in FAP with sulindac, 150 mg twice daily. If the tumor continues to grow

22

& PART I: COLORECTAL DISORDERS

as shown by clinical observation and CT scan, add tamoxifen, 80 mg/day. If the tumor stabilizes, continue the medications but reduce the tamoxifen dose after 6 months and then gradually discontinue therapy. If the desmoid keeps growing or is still symptomatic, consider chemotherapy. If an intra-abdominal desmoid is discovered during operation and can be resected with a minimum of small bowel and low risk of complications, proceed. If complete excision is impossible, obtain tissue for histologic and estrogen-receptor assays. Using an antisarcoma regimen consisting of doxorubicin and dacarbazine, Mo¨slein and Dozois (195) used this regimen in nine patients with FAP-related desmoid tumor and it led to complete regression in four patients and partial regression in five patients. Poritz et al. (196) treated eight patients with desmoid tumors and FAP who had inoperable GI obstruction and/or uncontrolled pain. The regimen consisted of doxorubicin and dacarbazine followed by carboplatin and dacarbazine. Follow-up at a mean of 42 months in seven patients revealed two patients achieved complete remission after the therapy. Four patients achieved a partial remission after completing or some of the chemotherapy regimen; of these, three remained at stable remission, whereas the other was lost to follow-up. There were two recurrences that required further therapy; one of these patients was treated with further chemotherapy, which induced a second remission, and the other was treated with pelvic exenteration and has subsequently died. This cytotoxic regimen should be considered only for patients with fast-growing, life-threatening mesenteric desmoid tumors. Church et al. (197) developed a staging system that can be applied to the management of FAP-related desmoid tumor: Stage I: Asymptomatic, not growing. Asymptomatic desmoids are usually small and are found incidentally either during exploratory celiotomy or on CT scan performed for unrelated reasons. Such tumors can be observed, or, at the most, a relatively nontoxic medication such as nonsteroidal anti-inflammatory drugs (NSAIDs) may be prescribed. If a stage I desmoid is found incidentally at surgery and it is easily resectable without the removal of a significant amount of bowel, resection is appropriate. Stage II: Symptomatic and 10 cm or less in maximum diameter, not growing. Small desmoids that are causing symptoms (including bowel or ureteral obstruction) need therapy even if they are not obviously growing. If they are resectable with minimal sequelae, then resection is best. If the tumor is unresectable, the addition of tamoxifen or raloxifene to a NSAID offers the possibility of a quicker and more consistent response with low risk of side effects. Stage III: Symptomatic and 11 to 20 cm, or asymptomatic and slowly growing desmoids. Larger, symptomatic (including bowel or ureteric obstruction) desmoids, or desmoids that are slowly increasing in size (<50% increase in diameter in 6 months), need active treatment. Here the choices include NSAID, tamoxifen, raloxifene, and vinblastine/methotrexate. Antisarcoma chemotherapy can be given if the tumor continues to grow despite less toxic agents. Stage IV: Symptomatic, > 20 cm, or rapid growth, or complicated desmoids. These are the worse desmoids: Large, or growing rapidly (> 50% increase in diameter within 6 months), these cause life-threatening complications

such as sepsis, perforation or hemorrhage. Here treatment is an urgent necessity and the possibilities are major exenterative surgery likely to result in significant loss of bowel, antisarcoma therapy, and radiation. The authors noted that one of the most important uses of the staging system is to allow prospective trials of the various treatment options, some of which are inappropriate for certain stages of tumor (197). Death from desmoid tumors is caused by either direct effects, such as erosion into a blood vessel or sepsis from an enteric fistula, or to secondary effects as the result of desmoid surgery (190). Osteomas Osteomas may occur in any bone but they are most commonly located on the facial skeleton, particularly the mandible. These tumors are benign but may cause symptoms following local growth. They are sometimes identified before the diagnosis of FAP is made (156). Teeth Teeth are derived from both mesoderm and ectoderm. Dental abnormalities, distinct from osteomas of the jaw, have been described in from 11% to 80% of individuals with FAP. Although slightly less frequent than osteomas of the jaw, their frequency and the fact that they may appear at an early age are sufficient reasons to make them diagnostically useful. The lesions in question are impactions, supernumerary or absent teeth, fused root of first and second molars, and unusually long and tapered roots of posterior teeth (145).

Ectodermal Abnormalities Eye Lesions Although the presence of pigmented lesions of the fundus was noted in a patient with the signs of Gardner’s syndrome by Cabot (198) in 1935, it was not until 1980, when Blair and Trempe (199) recorded pigmented lesions in three affected members of a kindred with Gardner’s syndrome, that the possibility of using this lesion as a marker for FAP was suggested. The abnormality is considered to be congenital hypertrophy of retinal pigment epithelium (CHRPE). The occurrence of CHRPE is restricted to APC mutation in codons 463 to 1444 (183). The examination is made by indirect ophthalmoscopy, following instillation of 1% tropicamide for pupil dilatation. It normally appears as a round or oval pigmented lesion with a surrounding pale halo (Fig. 21). Microscopic examination shows that CHRPE is a hamartoma. The acronym, while not strictly correct, has become accepted and has continued in use (156). The incidence of CHRPE in patients with FAP varies widely

FIGURE 21 & Fundus photograph of the pigmented lesion of the retina (CHRPE). Source: Courtesy of Helmut Buettmer, M.D., Mayo Clinic, Rochester, MN, U.S.A.

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

from 50% to 79% (145,156) compared with 7% and 5% in atrisk groups and in age- and sex-matched individuals, respectively. The sensitivity of CHRPE in FAP is 79% and specificity is 95% (156). The presence of CHRPE in a person who is a member of a kindred of a patient manifesting FAP suggests that he or she has inherited the gene. The absence of the lesion does not, however, indicate that the person has not inherited the gene (156). Epidermoid Cyst In patients with FAP, cysts may be found on the limbs, face, and scalp. In the general population they occur predominantly on the back. Leppard and Bussey (200) found epidermoid cysts in 53% of a series of 74 patients affected by FAP. Perhaps the most significant finding related to epidermoid cysts is their rarity in childhood in any condition other than FAP; skin cysts may be evident before the development of colorectal polyps (201). Leppard and Bussey (200) recommend that a child with an epidermoid cyst undergo sigmoidoscopy after the age of 14 years but before attaining 30 years of age. Brain Neoplasms In 1949 Crail (202) reported a case of synchronous cerebellar medulloblastoma, colonic polyposis (approximately 100 polyps), and papillary carcinoma of the thyroid. This report did not receive widespread recognition at that time (203). In 1959 Turcot et al. (204) described two siblings who presented with colonic polyposis at the ages of 13 and 15 years, respectively. Both patients went on to develop a glioblastoma of the frontal lobe and a medullary medulloblastoma, respectively. The first sibling also had a chromophobe adenoma of the pituitary gland. The authors suggested that these neoplasms might be another extracolonic manifestation of FAP. These associations bear the name Turcot’s syndrome. Hamilton et al. (205) studied Turcot’s syndrome at the molecular level. Fourteen families with Turcot’s syndrome were identified. Germ-line mutations in the APC gene characteristic of FAP were evaluated, as well as DNA replication errors and germ-line mutations in nucleotide mismatch-repair genes characteristic of HNPCC. Genetic abnormalities were identified in 13 of 14 registry families. Germ-line APC mutations were detected in 10. The predominant brain tumor in these 10 families was medulloblastoma (11 of 14 patients, or 79%), and the RR of cerebellar medulloblastoma in patients with FAP was 92 times that in the general population. In contrast, the type of brain tumor in the other four families was glioblastoma multiforme. Germ-line mutations in the mismatch-repair gene hMLH1 or hMMS2 were found in two families. In the study of the APC gene in 14 families that included at least one affected member, 12 families classified as having polyposis, mutations were found in 10 (83%). All the mutated genes encoded truncated variance of the APC protein, as is true of the vast majority of patients with FAP. The mutations were heterogeneous in type and location, and there was no association between specific mutations and the development of brain tumors. Two families with polyposis and both families without polyposis had no identifiable germ-line APC mutations. Analysis of the hMSH2, hMLH1, hPMS1, and hPMS2 mismatch-repair genes, which are mutated in HNPCC, was carried out in the three patients with neoplasms that

23

contained replication errors. Two had germ-line alterations: the hPMS2 gene was mutated in one patient and hMLH1 was mutated in another patient. No germ-line APC mutations were detected in these three patients. The authors concluded that the association between brain neoplasms and multiple colorectal adenomas might result from two distinct types of germ-line defects: mutation of the APC gene or mutation of the mismatchrepair gene. Molecular diagnosis may contribute to the appropriate care of affected patients. Review of literature by Matsui et al. (206) and Itoh et al. (207) revealed that there were 131 documented cases in the medical literature. Of the 35 cases that they considered having true Turcot’s syndrome, the average age of death was 20.3 years. Most (76%) died from brain malignancy, 16% died from colorectal carcinoma, and the rest from other causes. Death at young age has made it difficult to determine whether the mode of inheritance is autosomal recessive or autosomal dominant. A more complete list of the extracolonic manifestations of FAP is summarized in the Box 1 (203).

& MANAGEMENT Histologic verification of adenomas is essential so that confusion with familial juvenile polyposis, hyperplastic polyposis, pseudopolyposis, and lymphoid polyposis is avoided. Performing total colonoscopy arid biopsy is the best choice. Virtually all patients with FAP will develop carcinoma of the colon and rectum by age 40. For this reason, patients with FAP should have prophylactic colectomy. At present, there are several surgical options, and each has advantages and disadvantages. Proctocolectomy with Ileostomy This procedure removes all the disease, but an obvious side effect is the creation of a permanent ileostomy, which is not well accepted by most patients, particularly the young ones. With other alternatives, especially colectomy with IRA and proctocolectomy with an ileoanal pouch procedure, and proctocolectomy with ileostomy are seldom chosen by patients. However, if there is a carcinoma in the rectum or in patients with a desmoid tumor of the small bowel mesentary, a proctocolectomy should be performed. Proctocolectomy with Continent Ileostomy This procedure was a popular option in the 1970s. An ileal reservoir with a nipple valve is created from the terminal ileum and is brought out as an ileostomy. Its advantage over the conventional ileostomy is that an ileostomy bag is not required. The pouch must be evacuated four to six times a day with a catheter. Because of the frequent extrusion of the nipple valve, which results in incontinence, use of the procedure has been limited to a small number of patients. The IPAA now has largely replaced this procedure. Colectomy with Ileorectal Anastomosis This procedure minimizes the risk of development of carcinoma upto only the last 12 to 15 cm of the rectum. Patients require lifelong close follow-up, at least once or twice a year, with electrocoagulation of the polyps as indicated. It

24

& PART I: COLORECTAL DISORDERS

BOX 1 & Extracolonic Manifestations of Familial Adenomatous Polyposis Ectodermal Origin Epidermoid cyst Pilomatrixoma Tumors of central nervous system Congenital hypertrophy of the retinal pigment epithelium

Mesodermal Origin Connective tissue Fibroma Fibrosarcoma Desmoid tumors Diffuse fibrosis mesenteric retroperitoneum Excessive intra-abdominal adhesion ?Lipoma Bone Osteoma Exostosis Sclerosis Dental Dentigerous cyst Odontoma Supernumerary teeth Unerupted teeth Lymphoid Hyperplasia of ileum

Endodermal Origin Adenomas Stomach Duodenum Hepatopancreatobiliary system Small intestine Endocrine tissue Adrenal cortex (adenomas) Thyroid gland ?Parathyroid ?Pituitary ?Pancreatic islets Carcinomas Stomach Duodenum Hepatobiliary system Small intestine Thyroid gland Adrenal gland Fundic gland polyp Hepatoblastoma

Source: From Ref. 203.

should be selected for patients in whom the rectum is not carpeted with polyps and who are willing to return for follow-up. The main advantages of this choice are that it is a relatively simple procedure familiar to most surgeons and it has excellent functional results. Controversy exists about the risk of developing carcinoma in the remaining rectum after colectomy and IRA. The risk varies from series to series, from 0% at the Cleveland Clinic (208) to an overall 32% at the Mayo Clinic (209). The discrepancy is not clear, but it appears that the risk of developing carcinoma increases with time (210). The unusually high incidence of carcinoma in the retained rectum in the Mayo Clinic series prompted Bess et al. (209) to reanalyze the series 10 years later, with the aim of identifying factors that may contribute to the risks. The following variables do not cause an increase in the risks: male versus female, number of colonic polyps (100 vs.
100), family history of polyposis, age at the time of surgery (40 years vs.
40 years), and level of ileorectal or ileosigmoid anastomosis (15 cm vs.
15 cm). Factors that contribute to an increase in carcinoma risk include the number of preoperative rectal polyps (the risk is increased if there are more than 20 polyps in the remaining rectum) and colonic carcinoma resected at or before colectomy. Of significance is the cumulative risk of carcinoma in the St. Mark’s series, increasing from 10% at 20 years follow-up to more than 30% at 35 years (210).

Risk of Carcinoma in the Retained Rectum The series from St. Mark’s (210) showed that until the age of 50 years, the cumulative risk of carcinoma in the IRA is reasonably low at 10%, increasing sharply to 29% by the age of 60 years. This means that surveillance of the retained rectum in older patients must either be improved or the patients should undergo restorative proctocolectomy in earlier middle age. Nugent and Phillips (211) recommend a flexible videoendoscopy at fixed intervals of four months for all patients at risk over the age of 45 years who wish to retain the rectum. A similar age-dependent

rectal carcinoma risk is also reported by Heiskanen and Jarvinen (212) 3.9%, 12.8%, and 25.7% at 40, 50, and 60 years, respectively, and rectal excision rates of 9.5%, 26.3%, and 44.0%, respectively. The authors also reported a cumulative rectal carcinoma risk of 4.0%, 5.6%, 7.9% and 25.0% at 5, 10, 15, and 20 years, respectively, after the ileorectal anastomosis (IRA). These findings have raised the question of the justification of ileorectal anastomosis (IRA) as the primary treatment of FAP. The planned strategy of two prophylactic operations, first a colectomy with IRA at an earlier age (perhaps 20 years of age), and then a restorative proctectomy at the age of 45 years, doubles the risks of the operations and may also increase the risk of desmoid tumors. Furthermore, a second-stage restorative proctectomy may not result in perfect functional outcome, or is impossible in cases of pelvic fibromatous adhesions or desmoids (212). The previous recommendation by Bess et al. (209) in 1980 is still a good one, that the colon and rectum should be removed but the rectum may be retained if there are fewer than 20 polyps in the rectum. Heiskanen and Jarvinen (212) now favor proctocolectomy and ileoanal pouch procedure as the primary operation for FAP. A more detailed view of the cumulative rectal carcinoma incidence is shown on Table 8. A close follow-up of patients with IRA is essential. The retained rectum should be examined, preferably with a flexible sigmoidoscope, once a year or sooner, depending on the number of polyps that should be electrocoagulated. Any time the number of the polyps become too numerous or too large to be safely removed, a proctectomy should be considered. In a series reported by Penna et al. (220) of 148 patients with an IRA, 29 required a secondary proctectomy: 16 because the rectal polyps were too numerous, 8 because the patients wished to discontinue regular surveillance, and 3 because of the discovery of rectal carcinoma (one was Dukes’ A 3 years after the IRA, one Dukes’ B 14 years after, and one Dukes’ C 17 years later). An IPAA was

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

TABLE 8 & Selected Series on Risk of Rectal Carcinoma After Colectomy and Ileorectal Anastomosis

Authors Bess et al. (209) Bulow (147) Sarre et al. (213) DeCosse et al. (214) Nugent and Phillips (211) Iwama and Mishima (215) Heiskanen and Jarvinen (212) Jenner and Levitt (216) Bjork et al. (217) Bertario et al. (218) Church et al. (219)

No. of Patents

Years After IRA

Rectal Carcinoma Rate (%)

143 58 133 294 224 342 100 55 195 371 62

19 10 20 25 25 15 20 10 25 20 15

32 13 12 13 15 24 25 13 24 23 13

Abbreviation: IRA, ileorectal anastomosis.

successfully performed in all but three patients who had pelvic desmoid tumors. In a series reported by Nugent and Phillips (211) of 224 patients with IRA, 22 patients developed carcinoma of the rectum. Nine were Dukes’ A, four were Dukes’ B, and nine were Dukes’ C. These carcinomas developed despite a close follow-up; 14 of 22 patients were last examined less than 6 months before. It is essential to realize that despite the close follow-up, surveillance cannot always prevent rectal carcinoma (221). A Place for Ileorectal Anastomosis Colectomy with IRA has its strong support. Phillips and Spigelman (222) reason that the ileoanal pouch procedure has high morbidity and does not give perfect functional results and that patients can still succumb to other related diseases of FAP. Bulow et al. (223) studied 659 patients undergoing IRA. The data was obtained from the National Polyposis Registries in Denmark, Finland, the Netherlands, and Sweden. They found that chronologic age was the only independent risk factor of developing rectal carcinoma. The risk of secondary proctectomy was higher in patients with mutation in codon 1250 to 1500 than outside this region. None of the 18 patients with AFAP (mutation in codon 0–200 or greater than 1500) had a secondary proctectomy. Church et al. (224) found that the risk of rectal carcinoma after IRA was strongly related to the severity of colorectal polyposis at presentation. Bertario et al. (218) found independent predictors of rectal carcinoma after IRA in the FAP with mutation between codon 1250 and 1464 (RR ¼ 4.4). It is, therefore, reasonable to perform colon resection with IRA in young patients with few adenomas (less than 20 rectal adenomas, less than 1000 colonic adenomas) and in FAP with mutation in codon 0 to 200 or greater than 1500 (223,224). These patients must understand their responsibility to have a periodic surveillance with an endoscopy, and that they may require a proctectomy in the future.

Regression of Polyps A temporary spontaneous regression or disappearance of polyps in the rectum after IRA is a common observation. This gives some comfort to clinicians, a hope that perhaps

25

the risk of developing carcinomas can be minimized as well. A study of the effect of colectomy and IRA on rectal mucosal proliferation in FAP showed a significant reduction in rectal mucosal cell proliferation. However, the mechanism is unknown (225). Sulindac has also been found to markedly reduce epithelial cell proliferation and significant polyp regression, including dysplastic reversion (226,227). This observation must be guarded, since Spagnesi et al. (228) observed the persistence of abnormal rectal mucosal proliferation after sulindac therapy, despite the reduction of number of polyps. Winde et al. (229) conducted a prospective, controlled, nonrandomized Phase II dose-finding study from sulindac given rectally, and looked at the molecular mechanism by which sulindac worked. The study group (n ¼ 28) and the control group (n ¼ 10) underwent colectomy and IRA, with repeated proctoscopy with endoluminal ultrasound and biopsies every three months. The treatment group was given sulindac suppositories, 150 mg twice daily, for three months. Visible improvement was followed by a dose reduction to 50 mg daily. Worsening of the polyps required changing to the initial dose level. The results showed that all patients responded to sulindac after 24 weeks (at the latest). Complete reversion was reached with 50 mg/day in 78% of patients. Twenty-two percent had partial reversions of adenomas at latest re-examination and there was no influence on upper GI tract adenomas. There was a permanent antiproliferative effect (Ki-67) of low-dose sulindac, significant blocking of ras mutation activation, and a significant difference of untreated and treated mucosa in mutant p53 content. The follow-up was 4 years. The authors concluded that low dose antiproliferative sulindac therapy is highly effective in adenoma reversion in FAP patients. Sulindac shows influence on tumor-suppressor genes and on apoptosis markers. All cases with relapse represented by newly developed flat mucosal elevations respond to dose increases. Giadiello et al. (230) conducted a randomized, doubleblind, placebo-controlled study looking whether sulindac can prevent adenoma rather than causing regression of the polyps. The study consisted of 41 young subjects (age range, 8 to 25 years) who were genotypically affected with familial FAP but phenotypically unaffected. Subjects received either 75 or 150 mg of sulindac orally twice a day or identical-appearing placebo tablets for 48 months. The number and size of new adenomas and the side effects of therapy were evaluated every four months for four years, and the levels of five major prostaglandins were serially measured in biopsy specimens of normal-appearing colorectal mucosa. The results after 4 years of treatment showed the average rate of compliance exceeded 76% in the sulindac group, and mucosal prostaglandin levels were lower in this group than in the placebo group. During the course of the study, adenomas developed in 9 of 21 subjects (43%) in the sulindac group and 11 of 20 patients in the placebo group (55%). There were no significant differences in the mean number or size of polyps between the groups. The authors concluded that standard doses of sulindac did not prevent the development of adenomas in subjects with FAP. Evidence that sulindac has a short-lived effect on established polyps in patients with FAP has been reported.

26

& PART I: COLORECTAL DISORDERS

The rate of regression of adenomas was greater after 6 months of sulindac treatment than after nine months has been reported in some patients who had undergone IRA. Long-term use of sulindac resulted in the development of resistance to this medication (230). Moreover, colorectal carcinoma has developed in rectal segment in patients with FAP during maintenance therapy with sulindac (231–233). The lack of efficacy of primary chemoprevention could have been due to resistance to sulindac. Most results do not provide support for the use of NSAIDs such as sulindac for the primary treatment of FAP. Prophylactic colectomy remains the treatment of choice to prevent colorectal carcinoma in patients with this disorder (230).

Proctocolectomy with Ileal Pouch Procedure The advantage of the ileal pouch procedure for FAP is its total eradication of the colonic disease. The inconvenience and a small risk of perforation of a long-term or lifetime follow-up of the rectum with proctoscopy or flexible sigmoidoscopy and electrocoagulation in patients who undergo colectomy with IRA thus is eliminated. The argument against its use as the procedure of choice is the fact that it is a more extensive procedure, with high potential for complications, especially sepsis and fecal incontinence more and more authors and medical centers favor the IPAA as the primary treatment for patients with FAP (234,235). Although IPAA is a longer, more bloody, and more complex operation with a longer hospital stay than IRA, no significant difference is found between the two groups in terms of complication rate and quality of life in a series of teenagers. IPAA also has few effects on life activities that are especially important at this age (235,236). The same is true in adult series (234,237,238). In the series of Soravia et al. (239), although IRA had a significantly better functional outcome with regard to nighttime continence and perianal skin irritation, functional results otherwise, and the quality of life were similar, they favor IPAA over IRA because of lower long-term failure rate. With concern about the increasing risk of carcinoma in the retained rectum, which is 25% at 20 years after the IRA, Heiskanen and Jarvinen (212) the IPAA for their patients with FAP as the primary treatment. This approach eliminates the need for a proctectomy, after an IRA. Proctocolectomy with IPAA for FAP cannot be considered the end of the patient’s management. In these patients, adenomas and carcinomas can develop in the ileal pouch, at the anastomosis, or in the anal transitional zone. Church (240) reviewed these problems, searching the MEDLINE database for studies reporting ileoanal pouch adenomas, ileal pouch-anal anastomotic carcinomas, and ileal pouch carcinomas in patients with FAP. Reports of adenomas in Kock pouches and in Brooke ileostomies in the setting of FAP were also included. The primary end points of the study were the time between pouch (or ileostomy) construction and the diagnosis of neoplasia, the age of the patients at the diagnosis of neoplasia, and the severity of the neoplasia. The results showed that 18 studies reporting pouch neoplasia, 15 with adenomas (98 patients) and three with carcinomas (three patients). Three prospective studies showed that the incidence of the pouch adenomas increases

with time of follow-up and that the severity of the polyposis varies. The median time from pouch construction to diagnosis of pouch adenomas was 4.7 years and the range was 0.5 to 12 years. A prospective review at the Cleveland Clinic showed a rate of 28% for adenomas 3.5 years after stapled IPAA and 14% at four years after hand-sewn IPAA (241). Similar data had been reported from 97 FAP patients who had an IPAA in a multicenter study by van Duijvendijk et al. (242). At a median follow-up of 78 months (range, 25–137 months), 13 of these patients had adenomas at the anastomosis. The risk for anastomotic adenomas after double-stapled IPAA was 31%, three times that after mucosectomy and hand-sewn IPAA (10%). There were six studies, reporting eight patients with carcinoma at the IPAA, diagnosed a median of eight years after pouch construction (range, 3 to 20 years). One-half of the carcinomas were locally advanced (T4) and one-half were not (T1 or T2). One-half followed stapled anastomosis and one-half were after mucosectomy. There were eight case reports of carcinoma described in an ileostomy in patients with FAP. The median time for ileostomy construction to the ileostomy carcinomas was 25 years (range, 9–40). Pouch adenomas are difficult to manage endoscopically because of the thin ileal mucosa and the way it is tethered to the submucosa and underlying muscle. Their propensity to occur on suture lines also makes endoscopic treatment difficult. There are no large studies of endoscopic treatment of pouch adenomas so risks and complication rates are not established. The prospect of coagulating or snaring tens or hundreds of polyps is a concern, however. Pouch excision for uncontrollable polyposis has previously been reported (242) but would be a difficult operation in many patients, usually resulting in an ileostomy and risking complications such as impotence, retrograde ejaculation, ureteric injury, and worsening female fecundity. Chemoprevention of pouch polyposis is, therefore, an attractive alternative. The role of nonsteroidal anti-inflammatory drugs (NSAIDs) in suppressing colorectal adenomas in FAP is established, and there are anecdotal reports of their use in pouch polyposis. While chemoprevention of colorectal adenomas in FAP using sulindac works only partially and for a limited time, its effectiveness in the ileal pouch has not been systematically studied (240). Those caring for patients with FAP need to make sure that endoscopic surveillance is continued after IPAA. The endoscopy has to be accurate and the endoscopists must be aware of the increasing risk of pouch and anastomotic neoplasia. Church (240) recommended pouchoscopy to be done yearly for life, initially to look for anastomotic adenomas and then later to check for pouch adenomas. Once neoplasia is seen, appropriate treatment needs to be determined. This includes transanal excision of residual, adenoma-bearing anal transitional zone, transanal polypectomy for isolated large (1 cm) pouch adenomas, or sulindac (150 mg twice daily) for multiple (>10) pouch adenomas.

& GENETIC COUNSELING AND TESTING The gene responsible for FAP, known as APC, is located on chromosome 5q21. Mutations of the APC gene have been found in patients with FAP. These are often insertions,

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

deletions, and nonsense mutations that lead to frameshifts and/or premature stop codons in the resulting transcript of the gene. It is not yet clear how the subsequent truncated protein product causes adenomas to form. Capitalizing on the nature of these mutations has led to the development of molecular genetic tests for FAP (243). It is not carcinoma but the predisposition to carcinoma that is inherited (244). Genetic testing can capture the opportunity for surgeons to prevent the development of the carcinoma.

Genetic Counseling Genetic counseling is a dynamic communication process between the patient and the counselor who provides education and support within a multidisciplinary team (245). Patients and their families must understand the natural history of the disease, the involvement of other blood relatives, and that carcinomas of the large bowel as well as other organs are, in most cases, preventable.

27

The decision to undergo genetic testing is a personal one based on informed consent. The elements of informed consent include information on the gene being tested and the implications, limitations, and impact of results for the person being tested and two other family members. It should be emphasized that genetic testing is voluntary and the patients need to be aware of alternatives (245). The first phase in determining whether genetic testing is appropriate is genetic counseling. This stage educates the patient about the role of inherited causes for developing carcinoma, determines their risk for a malignancy, and provides screening recommendations. The objective is effectively accomplished by understanding the patient’s perspective. Education and discussing the benefits and risks of genetic testing through each patient’s viewpoint promotes autonomy and lays the foundation for informed consent. An overall general summary of the genetic counseling and testing process is described in Figure 22 (245).

FIGURE 22 & General summary of the genetic counseling and testing process. Abbreviation: CRC, colorectol carcinoma. Source: From Ref. 245.

28

& PART I: COLORECTAL DISORDERS

Genetic Testing The most important recent development in the management of families of patients with FAP is the use of predictive gene testing. In this context, both children and adults at 50% risk could benefit from genetic testing because of the resultant reduction of uncertainty regardless of test outcome, modifications in screening guidelines for those who do not have the mutant gene, and increased compliance with screening regimens in those who do have the mutant gene (243). Because approximately 96% of the mutations in FAP lead to a truncated protein, it has become routine to use the in vitro synthesized-protein assay (IVSP). Sometimes this is called a PTT, for mutation detection. When a truncated protein is identified in the assay, it is possible to localize the mutation to a specific segment of the gene and then use DNA sequencing to determine the mutated nucleotides. However, the use of IVSP as the sole genetic test in FAP misses approximately 20% of APC mutations. Another screening technique is based on analysis of electrophoretic migration of small segments of the wild type and mutant gene (SSCA). The sequential use of two molecular diagnostic tests has become a common practice; a simple and less expensive screening technique (of high sensitivity and moderate specificity) followed by a definitive test of high sensitivity, usually DNA sequencing (246). When the mutation in the family is known from one or more affected members previously studied, only one test, usually IVSP, need be performed, as the expression pattern of the mutation is already established for that family. A positive result in such a test is considered a ‘‘mutation-positive’’ result, and the patient can be counseled as recommended. When the mutation in the family is not known, IVSP is performed. In the majority of cases, mutation will be found in the APC gene that can be further characterized by DNA sequencing. Again, a ‘‘mutationpositive’’ result is obtained leading to the appropriate genetic counseling to the patient. Other options include DNA sequencing and linkage analysis (which tests whether FAP is associated with particular markers in or near APC but requires large families). Even by combining two or more techniques, it is not possible to achieve 100% sensitivity, because the mutations may not be in the coding region of the gene or because a few FAP kindreds do not exhibit linkage to chromosome 5q. Therefore, in such cases, a ‘‘no mutation detected’’ result must not be interpreted as a ‘‘negative test’’ result, with very important considerations for counseling the patient (246). Implications of a ‘‘Mutation-Positive Test’’ Result (245) Approximately 75% of individuals who carry an APC germline truncating mutation manifest an adenomatous polyp by age 20. These premalignant lesions inevitably evolve to malignancy, and if untreated, the risk for developing colorectal carcinoma is virtually 100% by age 40. Because polyposis often starts before puberty, flexible sigmoidoscopy beginning at puberty is recommended as a screening procedure by most authors. Mutation carriers require surveillance by gastroduodenoscopy for extracolonic neoplasms in the upper GI tract. Other variant manifestations in FAP include osteomas, cutaneous cysts, and CHRPE. A phenotype-

genotype correlation exists. For example, CHRPE is often present if the mutation is located in exon 9–15e of the APC gene. For families who carry mutation in this region, the presence of CHRPE is a diagnostic indication of FAP. When a mutation has been identified in the family, direct gene testing of relatives who have not yet been clinically assessed will distinguish between those who carry the mutation and those who do not. However, about 30% of FAP cases are caused by new mutations in the APC gene. In these cases, the parents will not carry this mutation and are not at risk of FAP; only descendants of the proband are at 50% risk. Implications of a ‘‘Negative Test’’ Result versus ‘‘No Mutation Detected’’ Result (246) A negative test result is given if the patient does not carry the mutation that is known to exist in their family. In this instance, family members are not at increased risk for developing colorectal carcinoma compared with the general population and should follow guidelines for carcinoma surveillance for that group. If the patient is affected with FAP and complete coding sequence analysis of the APC gene fails to identify a mutation, this could mean that the APC gene is not responsible for the patient’s diagnosis. This is possible because locus heterogeneity has been reported in FAP; not all FAP kindreds are linked to chromosome 5q. In this case, a ‘‘no mutation detected’’ result, APC gene testing has no predictive value for asymptomatic at-risk relatives. In these families, first-degree relatives should continue colorectal surveillance annually between the ages of 12 and 25 years, every other year between the ages of 25 and 35 years, and every third year between the ages of 35 and 50 years. Family members who have not developed multiple adenomatous polyps on colonoscopic examination by the age of 50 years are assumed to be unaffected by FAP. Caution should be used in interpreting genetic test results when a mutation is not detected, because these results can be misinterpreted as a negative result. This in turn could lead to controversy in the guidelines in the follow-up care for those with a ‘‘no mutation detected’’ result. A third scenario with a ‘‘no mutation detected’’ result occurs when an unaffected family member of an APC kindred is tested without testing an affected family member. In such a case, the result does not mean that the unaffected person is not at risk for FAP, because it has not been determined that a mutation in APC is present in the family. It is important that the patient not be falsely reassured, because other mechanisms may inactivate the APC gene, or other genes may be involved. Summary on interpretation of APC genetic testing is in Figure 23 (245).

& WHEN TO SCREEN AND WHEN TO OPERATE? Gene test results can change the risk for FAP from a priori 50% to essentially 0% or 100%. Presymptomatic genetic testing removes the necessity of annual screening of those at-risk individuals who do not have the gene, and probably improves compliance in those who do. No change in conventional screening guidelines for colon polyps is recommended for those whose presymptomatic DNA diagnosis indicates that they have the FAP-causing gene. These individuals should have annual colon and rectal

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

29

FIGURE 23 & Summary interpretation of APC genetic testing. Abbreviations: CRC, colorectal carcinoma; FAP, familial adenomatous polyposis; FDR, first degree relative; F-U, follow-up; IVSP, in vitro synthesized protein assay (protein truncation test); pt, patient. Source: From Ref. 245.

examinations with at least a flexible sigmoidoscopy beginning at approximately 10 or 11 years’ of age (243). The age for the start of screening varies from series to series. The St. Mark’s series began at 14 years of age or older (226). Follow-up surveillance for extracolonic neoplasms is also indicated. At this time, the patients should also be counseled to prepare for an eventual prophylactic colectomy and should be given genetic counseling about the risk of FAP for future offspring. Timing of operation depends on the number of polyps found in the colon and rectum. The natural history of the disease, even though it is highly variable between individual cases, suggests that operation should be performed by 25 years of age, although most of the St. Mark’s patients had a colectomy by the age of 20 years. The youngest FAP patient with a carcinoma in the St. Mark’s registry is a 17-year-old girl (226). For those who have not inherited the APC mutation, colon screening can be significantly reduced to three or fewer time points: at ages 18, 25, and 35 years. These time points are selected to provide the clinician with a management margin that accommodates false negative results from laboratory error and infrequent phenomena, such as tissue mosaicism or de novo mutations. The individual’s lifetime risk of colorectal carcinoma becomes the general population’s risk of approximately 3%, and colon carcinoma screening should resume again around age 50, according to conventional guidelines. These individuals can also be assured that their offspring will not be at risk for FAP (243).

& THE POLYPOSIS REGISTRY The aims of a polyposis registry are to ensure efficient care of patients and their families and to promote and carry out the research that will advance the knowledge of FAP for physicians and surgeons. Data held in a standardized format on computers assist the day-to-day of those in the registry as well as allow for speedy analysis (244). A registry deals not only with the patient but also with the patient’s family members. Family pedigrees are collected and regularly updated. A counseling team plays an important part of the registry. Early diagnosis, using modern and reliable techniques, timely operation to prevent the onset of carcinoma, and continued surveillance after operation for early detection and treatment of associated carcinomas are key features of a successful registry. The impact of screening on carcinoma incidence in patients with FAP at St. Mark’s Registry is shown in Figure 24. It has a remarkable success in preventing the development of colorectal carcinomas (244). The Leed’s Castle Polyposis Group in Kent, United Kingdom, was established in June 1985. The aims of this gathering were to discuss the problems facing those who cared for polyposis patients and to establish an international cooperative organization. In 1992, there were 51 such centers worldwide. The group seeks furtherknowledge of the etiology, clinical features, prevention, and treatment of FAP in all of its manifestations (244). As outlined by Church and McGannon (247), there are three types of FAP registries.

30

& PART I: COLORECTAL DISORDERS

of developing colorectal or other carcinoma because they have a personal or family history of colorectal carcinoma. Setting up a successful polyposis registry is a large project requiring large sums of funds, space equipment, and personnel. Table 9 shows the organization of a successful registry (247).

& HEMANGIOMAS OF LARGE BOWEL

FIGURE 24 & The results of screening by the St. Mark’s Polyposis Registry. Source: From Ref. 244, p. 9.

The Countrywide Registry In this type of registry, cases of FAP from the entire country are collected into one central registry. The registry staff need not undertake direct patient care but must collect and collate data from everywhere. Examples of countrywide registries are those in Denmark, Sweden, Holland, and Singapore.

The Regional Registry This type fits well with very large or widespread populations or those without a uniform health care system. Regional registries are miniatures of countrywide registries in which families are collected from a defined region, for example, an area health board or a county. Parish, city, and county records contain extensive information with which to flush out family data, and the registry develops lines of communication with hospitals and physicians in its area. Registry staff need not commit to patient care but can offer specialized counseling and testing services. Regional registries are common in the United Kingdom.

The Tertiary Referral Center This type develops around one or two physicians of repute who work in a tertiary referral medical center and who have a special interest in FAP. Families are accumulated and a patient base is built. Patients may come from all over a country for treatment and return to the registry for surveillance. An example of such a registry is the Cleveland Clinic Foundation. Minnesota Colorectal Cancer Initiative (248) is a community-based Colorectal Cancer Registry run by two physician groups and the University Cancer Center. It is a not-for-profit organization financially supported by healthcare organizations, pharmaceutical companies, a consulting firm, and other practice groups. It is a model of effective collaboration between an academic tertiary referral center and community healthcare providers. Enrollment is not limited to individuals with an established diagnosis of a hereditary colorectal cancer syndrome, but is most likely to benefit people who may be at increased risk

Hemangiomas are rare vascular lesions of the large bowel. More than 200 cases have been documented in the literature from 1931 to 1974, with more than 50% involving the rectum (249). Hemangiomas of the large bowel are congenital, developing from embryonic sequestration of mesodermal tissue. However, whether they represent a simple vascular malformation (hamartoma) or a neoplasm is unsettled.

& CLASSIFICATION Capillary hemangiomas consist of a network of fine, newly formed, closely packed capillaries, with a distinct hyperplastic endothelial lining (250). They are usually encountered at operation and are found mainly in the small intestine, appendix, and perianal skin. This type of hemangioma accounts for 10% of all colon and rectal hemangiomas (249). Cavernous hemangiomas are large, thin-walled vessels. The stroma has little connective tissue and muscle. & CLINICAL MANIFESTATIONS The great majority of patients present early in life with recurrent, painless, and sometimes massive hemorrhage. Nine of 15 cases in the report of Londono-Schimmer et al. (251) were first seen at a hospital before the age of 5 years. The age of onset, type, frequency, and severity of bleeding are related to the size, number, and type of vascular malformations. Capillary hemangiomas are characterized by slow, persistent bleeding, producing dark stools and anemia. Cavernous hemangiomas characteristically present early with moderate to severe, painless hemorrhage. The bleeding recurs and increases in severity with each subsequent episode. In severe cases of profuse and unremitting bleeding, an urgent or emergency bowel resection is necessary. This propensity to bleed has been attributed to the lack of muscular and supporting connective tissue in the wall and around the vessels. Anemia is common and accounts for 43% of patients in the series presented by Allred and Spencer (250). Bowel obstruction accounts for 17% of the TABLE 9 & Staff for a Successful Registry Physicians Gastroenterologist Colorectal surgeon Medical geneticist Pathologist Source: From Ref. 247.

Administrative

Research

Secretary Statistician Computer technician Financial counselor Marketing

Molecular geneticist Lab technician Research fellows Study assistant/ research nurse

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

31

cases. Diarrhea, tenesmus, rectal prolapse, and sometimes constipation can occur.

& DIAGNOSIS Hemangiomas of the large bowel are often misdiagnosed because of the lack of awareness of the clinical features. They are frequently mistaken for hemorrhoids, inflammatory bowel diseases, polyps, carcinomas, and other entities (249). At St. Mark’s Hospital, 8 of 10 patients had at least one hemorrhoidectomy before the age of 20 years (252). Flexible sigmoidoscopy is the most important and most useful part of the workup. The mucosa has a deep blue or dull red color and appears wine or plum colored. The mucosa is usually nodular, with or without dilated viens (Fig. 25). Chronic inflammatory changes may mimic proctitis. Usually there is no mucosal ulceration. Hemorrhoids and ulcerative proctocolitis are the most common misdiagnoses. Plain abdominal films are useful. Calcification within the venous plexus of the bowel wall repre sents a sequela of thrombosis within the neoplasm caused by perivascular inflammation and sluggish blood flow. Phleboliths also accompany heman giomas at other locations along the GI tract and in venous plexuses of the uterine broad ligament, the urinary bladder, the prostate, and the spleen. Approximately 50% of adults with intestinal hemangiomas are noted to have phlebolith clusters on plain films (249). In a patient with an extensive hemangioma, barium enema studies reveal a characteristic scalloped contour of submucosal masses, causing narrowing of the colonic and rectal lumen. It can be differentiated from other neoplastic masses by its compressibility and a longer segment of involvement than polyps or carcinomas. A CT scan of the pelvis may show a large and markedly thickened rectal wall and phleboliths (if present) (Fig. 26). Selective inferior mesenteric arteriography is valuable to assist in determining the extent of the disease, particularly the less extensive hemangiomas in which localization at operation may be difficult. The delayed phase of arteriography may show large venous pooling within the cavernous hemangiomas (253). Arteriography is also helpful in detecting any abnormal blood vessels in the pelvis. Hypofibrinogenemia or afibrinogenemia that is caused by a consumptive coagulopathy is occasionally seen in these patients. Isotope studies with 51Cr-labeled platelets suggest sequestration within the hemangiomas. Resection

FIGURE 26 & Computed tomography scan of pelvis shows calcification of submucosal vascular plexus in diffuse hemangioma of rectum (arrows).

of the hemangiomas has resulted in complete reversal of the coagulopathy (249).

& TREATMENT Until 1971 abdominoperineal resection was the only definitive treatment for diffuse hemangiomas of the anorectum. Since then, the St. Mark’s group has performed a rectosigmoid resection with preservation of the pelvic floor and sphincter muscles. The entire rectum is mobilized and transected at the level just above the levator ani muscle. The anorectal mucosa and submucosa from the dentate line are stripped from underlying muscle up to the level of the transection. The sigmoid colon or the descending colon is then brought through the denuded anorectum and anastomosed to the dentate line. Londono-Schimmer et al. (251) reported success in all 15 patients using this technique and concluded that it is the treatment of choice. Telander et al. (254) successfully performed this operation in four children with rectal hemangiomas associated with Kleppel-Trenaunay syndrome, a congenital venous anomaly manifested by extensive hemangiomatous malformation within the pelvis and extremity, hypertrophy of the extremity, and atypical varicosities. Coppa et al. (253), on the other hand, reported that three of eight patients who underwent coloanal sleeve anastomosis had recurrent episodes of rectal bleeding similar to preoperative symptoms, most likely caused by involvement of the anal canal by diffuse cavernous hemangiomas. Ten of those patients who underwent abdominoperineal resection had no recurrent bleeding. Many other treatments have been reported in literature, all of which are temporary measures. They include colostomy, injection of sclerosing agents, radium implantation, irradiation, ligation of superior rectal arteries, embolization of inferior mesenteric artery, rectal packing, and cryosurgery. The place for the use of the laser in treating diffuse colon and rectal hemangiomas has not been established, but it is doubtful that it would give a permanent cure.

& LEIOMYOMAS OF LARGE BOWEL FIGURE 25 & Diffuse hemangioma of rectum. Note thickening and irregular mucosa. Source: Courtesy of Roger R Dozois, M.D., Mayo Clinic, Rochester, MN, U.S.A.

Leiomyomas of the large bowel are rare neoplasms of the smooth muscle of the bowel. In a review of 160 cases of smooth muscle neoplasms of the GI tract by He et al. (255) only four cases involved the rectum, and none was in the

32

& PART I: COLORECTAL DISORDERS

colon or the anal canal. One hundred thirty-one patients with benign gastrointestinal (GI) smooth muscle neoplasms were treated at the Massachusetts General Hospital between 1963 and 1987: 8% in the esophagus, 61% in the stomach, 19% in the small intestine, 5% in the colon, and 6% in the rectum (256). Kusminsky and Bailey (257) reviewed the world literature from 1959 to 1979 and found 79 cases of leiomyoma of the rectum. Leiomyomas are most commonly found in patients 50 to 59 years old and are rare or absent in children. Miettinen et al (258) studied all the mesenchymal neoplasms involving the rectum and anus coded as leiomyoma, leiomyosarcoma, smooth muscle neoplasm, schwannomas, neurofibromas, nerve sheath, stromal neoplasm. They used immunonohistochemistry. Antibodies to the antigens were: KIT (CD1147), CD34 a-smooth muscle, desmin D33, keratin 18 and 19, neurofilaments, sglial fibrillary acidic protein, and S-100 protein. The results showed a total of 133 anorectal GI stromal tumors (GIST). Of these, 50 tumors had been originally diagnosed as lymphosarcomas, 29 as smooth muscle tumors of uncertain malignant potential, 21 as leiomyomas, and 3 as GIST. Only three patents were diagnosed as leiomyoma. All three patients were women, age 25, 29, and 38 years. Tumor size was 1.5, 2, and 4 cm in diameter. All tumors were positive for desmin, a-smooth muscle acting, or both. They were negative for CD117 and CD34. One should be cautious to interprete the diagnosis of leiomyomas in the older literature where the study for GIST were not performed.

& CLINICAL MANIFESTATIONS Leiomyomas of the large bowel usually occur in the rectum and less commonly in the anal canal. The most common finding is the presence of a mass, usually detected as an incidental finding during rectal or proctosigmoidoscopic examination. Most patients with leiomyomas are asymptomatic, but some may present with constipation, bleeding, anorectal pain, and a pressure sensation in the rectum. & PATHOLOGY Leiomyomas may be pedunculated or sessile, submucosal, extrarectal, or dumbbell shaped to involve both intramural and extramural positions (Fig. 27). They may be in the abdomen or may protrude into the retrorectal space. In the anal canal, they may be found submucosally but are most commonly found in the intersphincteric position with a variable relationship to the sphincter muscles (257). Leiomyomas are firm, rounded, sharply circumscribed neoplasms, but they have no definite capsule microscopically. There are few reliable criteria for separating benign from malignant growths. Most malignant neoplasms are judged by metastases and may not be recognized until months or years later. Some pathologists accept more than two mitoses revealed under high power as evidence of malignancy. Morgan et al. (256) concluded that symptomatic gastric and small intestinal leiomyomas with more than two mitoses per 50 high-power field should be suspicious for a malignant potential. Whether this statement can be applied to leiomyomas of the large intestine is unknown. Others do not accept these criteria and do

FIGURE 27 & Leiomyoma of the rectum of a patient with ulcerative colitis.

not attempt to differentiate between benign and malignant growths at the time of examination (259).

& ORIGIN Smooth muscle neoplasms of the large bowel may arise from the muscularis mucosa or the muscularis propria. The division into neoplasms arising from the muscularis mucosa of the rectum, muscularis propria, or internal sphincter is based primarilon microscopic appearance to indicate the layer from which the neoplasm appears to arise. Histologically, all neoplasms arising from the muscularis mucosae of the rectum appear totally benign and arid, while the others vary in degree of differentiation. Of the 48 cases at St. Mark’s Hospital, 26 arose from muscularis mucosa, 18 from the muscularis propria, and four from the internal sphincter muscle (259). & TREATMENT Since it may not be known which neoplasm is malignant, the key to successful management is complete removal of the lesion. Walsh and Mann (259) reported local excision of 26 neoplasms arising from the muscularis mucosa and four neoplasms arising from the internal sphincter with no recurrence. However, local excision of 18 lesions arising from the muscularis propria resulted in a 60% recurrence rate. Of significance is that the incidence of recurrence after local excision bears no relationship to the degree of histologic differentiation. It appears that for neoplasms arising from the muscularis propria and those > 5 cm in diameter, a radical approach is indicated, that is, bowel resection or even abdominoperineal resection (257). Vorobyov et al. (260) presented their experience of 36 patients with rectal leiomyomas at the Moscow Proctology Institute. The authors recommend electro-excision through the endoscope for a neoplasm <1 cm in diameter. A transanal excision can be performed for a lesion <3 cm in diameter. When the lesion is >5 cm, a resection is advised.

& LIPOMAS OF LARGE BOWEL Lipomas of the large bowel are uncommon fatty neoplasms. The Mayo Clinic experience reveals 91 cases of large bowel lipomas during the period from 1976 to 1985 (261).

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

Approximately 90% of the large bowel lipomas are in the submucosa and 10% in the subserosa (262). In order of frequency, the most common sites are the cecum, ascending colon, and sigmoid colon, with one-third to one-half located in the right colon (261,263) Rare cases of submucosal lipomatous polypors of the colon have been reported (264,265). In one of the reports (265) the number of polyps was between 700 and 1000 (size 1–9 mm) through out the entire colon. A subtotal colectomy was performed in this case because of the frequent bowel movement and inability to rule out neoplastic polyps. The specimen did showed a combination of tabular adenomas, estimated fewer than 60 in number.

& CLINICAL MANIFESTATIONS A lipoma < 2 cm rarely causes symptoms. A large lipoma can cause abdominal pain from partial bowel obstruction. Colicky pain may be caused by intermittent intussusception. Lipomas are the most common benign neoplasms that cause intussusception in adults. Bleeding can occur if the mucosa overlying the lipoma is ulcerated, which is usually caused by chronic intussusception. & DIAGNOSIS In the Mayo Clinic series 46% of the large bowel lipomas were discovered incidentally in the specimens removed for other diseases. Eleven percent of them were resected because of a neoplasm suspected as being a carcinoma (263). Barium enema reveals a filling defect in the colon or rectum. A lipoma appears smooth and radiolucent and changes in shape on compression during examination. Despite these characteristics, the correct radiologic diagnosis is not always made. A CT scan is helpful in confirming the presence of fatty tissue in the mass (266,267). Colonoscopy can confirm the diagnosis in most cases. The lipoma has a characteristic yellowish submucosal soft mass. When pressed with the tip of a biopsy forceps, the lipoma is depressed; this is known as the ‘‘cushion sign.’’ A biopsy must include the submucosal layer but seldom is necessary for diagnosis. An ulcerated lipoma can be confused with an adenoma or an adenocarcinoma. Lipomas of the large bowel should not be forgotten in the differential diagnosis of all large bowel neoplasms. & TREATMENT Small and asymptomatic lipomas of the large bowel do not require removal. Colonoscopic polypectomy of a large lipoma should be avoided since a tremendous amount of heat is required to cut through the lipoma because of its high water content. The risk of transmural burn, perforation, or bleeding outweighed the benefits in seven lipomas removed by colonoscopic polypectomy by Pfeil et al. (268); three had perforations (size, 4.2  0.7 cm). The authors warned that a lipoma >2 cm is at the greatest risk for perforation. Symptomatic lipomas should be excised by colotomy or a limited colon resection (269).

& REFERENCES 1. O’Brien MJ. Colorectal adenomas: concepts and controversies. Semin Colon Rectal Surg 1992; 3:195–206. 2. Rickert RR, Averbach O, Garfinkel L, Hammond EC, Frasca JM. Adenomatous lesions of the large bowel: an autopsy survey. Cancer 1979; 43:1847–1857.

33

3. Williams AR, Balasooriya BA, Day DW. Polyps and cancer of the large bowel: a necropsy study in Liverpool. Gut 1982; 23:835–842. 4. Vatn MH, Stalsberg H. The prevalence of polyps of the large intestine in Oslo: an autopsy study. Cancer 1982; 49:819–825. 5. Eide TJ, Stalsberg H. Polyps of the large intestine in northern Norway. Cancer 1978; 42:2839–2848. 6. O’Brien MJ, Winawer SJ, Zauber AG, et al. The National Polyp Study. Patient and polyp characteristics associated with high-grade dysplasia in colorectal adenomas. Gastroenterology 1990; 98:371–379. 7. Nusko G, Mansmann U, Partzsch U, et al. Invasive carcinoma in colorectal adenomas: multivariate analysis and adenoma characteristics. Endoscopy 1997; 29:626–631. 8. Muto T, Bussey HJ, Morson BC. The evolution of cancer of the colon and rectum. Cancer 1975; 36:2251–2270. 9. Dukes C. Simple tumors of the large intestine and their relationship to cancer. Br J Surg 1926; 13:720–733. 10. Jackman RJ, Mayo CW. The adenoma-carcinoma sequence in cancer of the colon. Surg Gynecol Obstet 1951; 93:327–330. 11. Castleman B, Krickstein HI. Do adenomatous polyps of the colon become malignant? N Engl J Med 1962; 267:469–475. 12. Ackerman LV, Spratt JS, Fischel E. Do adenomatous polyps become cancer? Gastroenterology 1963; 44:905–907. 13. Morson BC. The polyp-cancer sequence in the large bowel. Proc R Soc Med 1974; 67:451–457. 14. Jass JR. Do all colorectal carcinomas arise in pre-existing adenomas?. World J Surg 1989; 13:45–51. 15. Nelson RL. Diet and adenomatous polyp risk. Semin Colon Rectal Surg 1991; 2:262–268. 16. Armitage NC. Intervention studies in adenoma patients. World J Surg 1991; 15:29–34. 17. Kronborg O, Fenger C. Clinical evidence of the adenoma–carcinoma sequence. Eur J Cancer Prevention 1999; 8:573–586. 18. Tierney RP, Ballantyne GH, Modlin IM. The adenoma to carcinoma sequence. Surg Gynecol Obstet 1990; 171:81–94. 19. Eide TJ. Natural history of adenomas. World J Surg 1991; 15:3–6. 20. Eide TJ. Risk of colorectal cancer in adenoma-bearing individuals within a defined population. Int J Cancer 1986; 38:173–176. 21. Shimoda T, Ikegami M, Fujisaki J, Matsui T, Aizawa S, Ishikawa E. Early colorectal carcinoma with special reference to its development de novo. Cancer 1989; 64:1138–1146. 22. Stolte M, Beckte B. Colorectal mini–de novo carcinoma: a realty in Germany too. Endoscopy 1995; 27:286–290. 23. Bedenne L, Faivre J, Boutron MC, Piard F, Cauvin JM, Hillon P. Adenoma– carcinoma sequence or ‘‘de novo’’ carcinogenesis? A study of adenomatous remnants in a population-based series of large bowel cancers. Cancer 1992; 69:883–888. 24. Muto T, Nagawa H, Watanabe T, Masaki T, Sawada T. Colorectal carcinogenesis. Historical review. Dis Colon Rectum 1997; 40(suppl):580–585. 25. Church JM, Williams BR, Casey G. Molecular Genetics and Colorectal Neoplasia. A Primer for the Clinician. New York: Igaku-Shoin Medical Publishers, 1996. 26. Nivatvongs S, Dorudi S. Colorectal polyps and their management. In: Williams NS, ed. Colorectal Cancer. London: Churchill Livingstone, 1996:39–54. 27. Dirschmid K, Kiesler J, Mathis G, Beller S, Stoss F, Schobel B. Epithelial misplacement after biopsy of colorectal adenomas. Am J Surg Pathol 1993; 17:1262–1265. 28. Fenoglio-Preiser CM. Colonic polyp histology. Semin Colon Rectal Surg 1991; 2:234–245. 29. Nivatvongs S, Nicholson JD, Rothenberger DA, et al. Villous adenoma of the rectum: the accuracy of clinical assessment. Surgery 1980; 87:549–551. 30. Galandiuk S, Fazio VW, Jagelman DG, et al. Villous and tubulovillous adenomas of the colon and rectum. A retrospective review, 1964–1985. Am J Surg 1987; 153:41–46. 31. Muto T, Kamiya J, Sawada T, et al. Small ‘‘flat adenoma’’ of the large bowel with special reference to its clinicopathologic features. Dis Colon Rectum 1985; 28:847–851. 32. Muto T, Watanabe T. Flat adenomas and minute carcinomas of the colon and rectum. Perspect Colon Rectal Surg 1993; 6:117–132. 33. Lynch HT, Smyrk T, Lanspa SJ, et al. Flat adenomas in a colon cancer-prone kindred. J Natl Cancer Inst 1988; 80:278–282. 34. Rembacken BJ, Fujii T, Cairms A, et al. Flat and depressed colonic neoplasms: a prospective study of 1000 colonoscopies in UK. Lancet 2000; 355:1211–1214. 35. Radaelli F, Minoli G. Editorial comments. Gastrointest Endosc 2001; 53: 689–691. 36. Minamoto T, Sawaguchi K, Mai M, Yamashita N, Sugimura T, Esumi H. Infrequent K-ras activation in superficial-type (flat) colorectal adenomas and adenocarcinomas. Cancer Res 1994; 54:2841–2844. 37. Gilbertsen VA. Proctosigmoidoscopy and polypectomy in reducing the incidence of rectal cancer. Cancer 1974; 34(suppl):936–939. 38. Selby JV, Friedman GD, Quesenberry CP Jr, Weiss NS. A case-control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 1992; 326:653–657. 39. Winawer SJ, Zauber AG, Ho MN. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med 1993; 329:1977–1981.

34

& PART I: COLORECTAL DISORDERS

40. Sciallero S, Constantini M, Bertinelli E, et al. Distal hyperplastic polyps do not predict proximal adenomas. Results from a multicentric study of colorectal adenomas. Gastrointest Endosc 1997; 46:124–130. 41. Rex DK, Smith JJ, Ulbright TM, Lehman GA. Distal colonic hyperplastic polyps do not predict proximal adenomas in asymptomatic average-risk subjects. Gastroenterology 1992; 102:317–319. 42. Provenzale D, Garret JW, Condon SE, Sandler RS. Risk for colon adenomas in patients with rectosigmoid hypoplastic polyps. Ann Intern Med 1990; 113: 760–763. 43. Church JM. Clinical significance of small colorectal polyps. Dis Colon Rectum 2004; 47:481–485. 44. Atkin WS, Morson BC, Cuzick J. Long-term risk of colorectal cancer after excision of rectosigmoid adenomas. N Engl J Med 1992; 326:658–662. 45. Stryker SJ, Wolff BG, Culp CE, Libbe SD, Ilstrup DM, Mac-Carty RL. Natural history of untreated colonic polyps. Gastroenterology 1987; 93:1009–1013. 46. Otchy DP, Ransohoff DF, Wolft BG, et al. Metachronous colon cancer in persons who have had a large adenomatous polyp. Am J Gastroenterol 1996; 91:448–454. 47. Hofstad B, Vatri MH, Anderson SN, et al. Growth of colorectal polyps: redetection and evaluation of unresected polyps for a period of three years. Gut 1996; 39:449–456. 48. Bersentes K, Fennerty MB, Sampliner RE, Harinder SG. Lack of spontaneous regression of tubular adenomos in two years of follow-up. Am J Gastroenterol 1997; 92:1117–1120. 49. Haggitt RC, Glotzbach RE, Soffer EE, Wruble LD. Prognostic factors in colorectal carcinomas arising in adenomas: implications for lesions removed by endoscopic polypectomy. Gastroenterology 1985; 89:328–336. 50. Nivatvongs S, Rojanasakul A, Reiman HM, et al. The risk of lymph node-metastasis in colorectal polyps with invasive adenocarcinoma. Dis Colon Rectum 1991; 34:323–328. 51. Pollard CW, Nivatvongs S, Rojanasakul A, Reiman HM, Dozois RR. The fate of patients following polypectomy alone for polyps containing invasive carcinoma. Dis Colon Rectum 1992; 35:933–937. 52. Kyzer S, Begin LR, Gordon PH, Mitmaker B. The care of patients with colorectal polyps that contain invasive adenocarcinoma.Endoscopic polypectomy or colectomy.. Cancer 1992; 70:2044–2050. 53. Muto T, Sawada T, Sukihara K. Treatment of carcinoma in adenomas. World J Surg 1991; 15:35–40. 54. Morson BC, Whiteway JE, Jones EA, Macrae FA, Williams CB. Histopathology and prognosis of malignant colorectal polyps treated by endoscopic polypectomy. Gut 1984; 25:437–444. 55. Richards WO, Webb WA, Morris SJ, et al. Patient management after endoscopic removal of the cancerous colon adenoma. Ann Surg 1987; 205:665–672. 56. Coverlizza S, Risio M, Ferrari A, Fenoglio-Preiser CM, Rossini FP. Colorectal adenomas containing invasive carcinoma. Pathologic assessment of lymph node metastatic potential. Cancer 1989; 64:1937–1947. 57. Cooper HS, Deppisch LM, Gourley WK, et al. Endoscopically removed malignant colorectal polyps: clinicopathologic correlations. Gastroenterology 1995; 108:1657–1665. 58. Kudo S. Endoscopic mucosal resection of flat and depressed types of early colorectal cancer. Endoscopy 1993; 25:455–461. 59. Participants in the Paris Workshop. The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach and colon. Gastrointest Endosc 2003; 58:S3–S43. 60. Nascimbeni R, Burgart LJ, Nivatvongs S, Larson DR. Risk of lymph node metastasis in T1 carcinoma of the colon and rectum. Dis Colon Rectum 2002; 45:200–206. 61. Nivatvongs S. Surgical management of malignant colorectal polyps. Surg Clin North Am 2002; 82:959–966. 62. Kikuchi R, Takano M, Takagi K, et al. Management of early invasive colorectal cancer. Risk of recurrence and clinical guidelines. Dis Colon Rectum 1995; 38:1286–1295. 63. Lamont JP, McCartz TM, Digan RD, Tulanon P, Lichliter WE. Should locally excised T1 rectal cancer receive adjuvant chemoradiation? Am J Surg 2000; 180:402–406. 64. Bouvet M, Milas M, Giacco GG, Cleary KR, Janjan NA, Skibber JM. Predictors of recurrence after local excision and postoperative chemoradiation therapy of adenocarcinoma of the rectum. Ann Surg Oncol 1999; 6:26–32. 65. Bailey HR, Huval WV, Max F, Smith KW, Butts DR, Zamora LF. Local excision of carcinoma of the rectum for cure. Surgery 1992; 111:555–561. 66. Paty PB, Nash GM, Baron P, et al. Long-term results of local excision for rectal cancer. Ann Surg 2002; 236:522–530. 67. Benson R, Wong CS, Cummings BJ, et al. Local excision and postoperative radiotherapy for distal rectal cancer. Int J Radiat Oncol Biol Phys 2001; 50:1309–1316. 68. Mellgren A, Sirivongs P, Rothenberger DA, Madoff RD, Garcia-Aguilar J. Is local excision adequate therapy for early rectal cancer? Dis Colon Rectum 2000; 43:1064–1074. 69. Madbouly KM, Remzi FM, Erkek BA, et al. Recurrence after transanal excision of T1 rectal cancer: should we be concerned? Dis Colon Rectum 2005; 48: 711–721.

70. Nascimbeni R, Nivatvongs S, Larson DR, Burgart LJ. Long-term survival after local excision for T1 carcinoma of the rectum. Dis Colon Rectum 2004; 47: 1773–1779. 71. Bentrem DJ, Okabe S, Wong WD, et al. T1 adenocarcinoma of the rectum. Transanal excision or radical surgery? Ann Surg 2005; 242:472–479. 72. Hahnloser D, Wolff BG, Larson DW, Ping J, Nivatvongs S. Immediate radical resection after local excision of rectal cancer: an oncologic compromise? Dis Colon Rectum 2005; 48:429–437. 73. Longacre TA, Fenoglio-Preiser CM. Mixed hyperplastic adenomatous polyps serrated adenomas. A distinct form of colorectal neoplasia. Am J Surg Pathol 1990; 14:524–537. 74. Torlakovic E, Snover DC. Serrated adenomatous polyposis in humans. Gastroenterology 1996; 110:748–755. 75. Torlakovic E, Skovlund E, Snover DC, Torlakovic G, Nesland JM. Morphologic reappraisal of serrated colorectal polyps. Am J Surg Pathol 2003; 27:65–81. 76. Jass JR. Serrated route to colorectal cancer: back street or super highway? J Pathol 2001; 193:283–285. 77. Jass JR. Pathogenesis of colorectal cancer. Surg Clin N Am 2002; 82:891–904. 78. Jass JR. Hyperplastic polyps and colorectal cancer: is there a link? Clin Gastroenterol Hepatol 2004; 2:1–8. 79. McCall I, Bussey HJ, Veale AM, Morson BC. Juvenile polyposis coli. Proc R Soc Med 1964; 57:896–897. 80. Desai DC, Neali KF, Talbot IC, Hodgson SV, Phillips RK. Juvenile polyposis. Br J Surg 1995; 82:14–17. 81. Jass JR, Williams CB, Bussey HJ, Morson BC. Juvenile polyposis—a precancerous condition. Histopathology 1988; 13:619–630. 82. Sachatello CR, Hahn IS, Carrington CB. Juvenile gastrointestinal polyposis in a female infant: report of a case and review of the literature of a recently recognized syndrome. Surgery 1974; 75:107–114. 83. Giardiello FM, Hamilton SR, Kern SE, et al. Colorectal neoplasia in juvenile polyposis or juvenile polyps. Arch Dis Child 1991; 66:971–975. 84. Howe JR, Mitros FA, Summers RW. The risk of gastrointestinal carcinoma in familial juvenile polyposis. Ann Surg Oncol 1998; 5:751–756. 85. Kinzler KW, Vogelstein B. Landscaping the cancer terrain. Science 1998; 280:1036–1037. 86. Giardiello FM, Hamilton SR, Kern SE. Colorectal neoplasia in juvenile polyposis or juvenile polyposis. Arch Dis Child 1991; 66:971–975. 87. Jarvinen H, Franssila KO. Familial juvenile polyposis coli; increased risk of colorectal cancer. Gut 1984; 25:792–800. 88. Howe JR, Ringold JC, Summers RW, Mitros FA, Nishimura DY, Stone EM. A gene for familial juvenile polyposis maps to chromosome 18q21.1. Am J Hum Genet 1998; 62:1129–1136. 89. Howe JR, Roth S, Ringold JC, et al. Mutations in the SMAD4/DPC 4 gene in juvenile polyposis. Science 1998; 280:1086–1088. 90. Onsel M, Church JM, Remzi FH, Fazio VW. Colonic surgery in patients with juvenile polyposis syndrome: a case series. Dis Colon Rectum 2005; 48: 49–56. 91. Scott-Conner C, Hausmann M, Hall TJ, Skelton DS, Anglin BL, Subramony C. Familial juvenile polyposis: patterns of recurrence and implications for surgical management. J Am Coll Surg 1995; 181:407–413. 92. Wirtzfeld DA, Petrelli NJ, Rodriguez-Bigas MA. Hamartomatous polyposis syndrome: molecular genetics, neoplastic risk, and surveillance recommendations. Ann Surg Oncol 2001; 8:319–327. 93. Howe JR, Ringold JC, Hughes JH, Summers RW. Direct genetic testing for SMAD4 mutations in patients at risk for juvenile polyposis. Surgery 1999; 126:162–170. 94. Jeghers H, McKusick VA, Katz KH. Generalized intestinal polyposis and melanin spots of the oral mucosa, lips and digits. A syndrome of diagnostic significance. N Engl J Med 1949; 241:993–1005. 95. Giardillo FM, Walsh SB, Hamilton SR, et al. Increased risk of cancer in PeutzJeghers syndrome. N Engl J Med 1987; 316:1511–1514. 96. Schreibman IR, Baker M, Amos C, McGarrity TJ. The hamartomatous polyposis syndromes: a clinical and molecular review. Am J Gastroenterol 2005; 100:476–490. 97. Tomlinson IP, Houlston RS. Peutz-Jeghers syndrome. J Med Genet 1997; 34:1007–1011. 98. Giardiello FM, Brensinger JD, Tersmette AC, et al. Very high risk of cancer is familial Peutz-Jeghers syndrome. Gastroenterology 2000; 119:1447–1453. 99. Foley TR, Mcgarrity TJ, Abt AB. Peutz-Jeghers syndrome: a clinicopathologic survey of the ‘‘Harrisburg Family’’ with a 49-year follow-up. Gastroenterology 1988; 95:1535–1540. 100. Narita T, Eto T, Ito T. Peutz-Jeghers syndrome with adenomas and adenocarcinomas in colonic polyps. Am J Surg Pathol 1987; 11:76–81. 101. Perzin KH, Bridge MF. Adenomatous and carcinomatous changes in hamartomatous polyps of the small intestine (Peutz-Jeghers syndrome): report of a case and review of the literature. Cancer 1982; 49:971–983. 102. Spigelman AD, Murday V, Phillips RK. Cancer and Peutz-Jeghers syndrome. Gut 1989; 30:1588–1590. 103. Boardman LA, Thibideau SN, Schaid DJ, et al. Increased risk for cancer in patients with Peutz-Jeghers syndrome. Ann Intern Med 1998; 128: 896–899.

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

104. Entius MM, Keller JJ, Westerman AM, et al. Molecular genetic alterations in hamartomatous polyps and carcinomas of patients with Peutz-Jeghers syndrome. J Clin Pathol 2001; 54:126–131. 105. Gruber SB, Entius MM, Petersen GM, et al. Pathogenesis of adenocarcinoma in Peutz-Jeghers syndrome. Cancer 1998; 58:5267–5270. 106. Boardman LA, Pittelkow MR, Couch FJ, et al. Association of Peutz-Jeghers like mucocutaneous pigmentation with breast and gynecologic carcinomas in women. Medicine 2000; 79:293–298. 107. Pedersen IR, Hartvigsen A, Hansen BF, Toftgaard C, Konstantin-Hansen K, Bullow S. Management of Peutz-Jeghers syndrome. Experience with patients from the Danish polyposis register. Int J Colorect Dis 1994; 9:177–179. 108. Spigelman AD, Phillips RK. Peutz-Jeghers syndrome. In: Phillips RK, Spigelman AD, Thomson JP, eds. Familial Adenomatous Polyposis and Other Polyposis Syndromes. London: Edward Arnold, 1994:188–202. 109. Ohmiya N, Taguchi A, Shirai K, Mabuchi N, et al. Endoscopic resection of Peutz-Jeghers polyps throughout the small intestine at double-balloon enteroscopy without laparotomy. Gastrointest Endosc 2005; 61:140–147. 110. Oncel M, Remzi FH, Church JM, Connor JT, Fazio VW. Benefits of "clean sweep" in Peutz-Jeghers patients. Colorectal Dis 2004; 6:332–335. 111. Edwards DP, Khosraviani K, Stafferton R, Phillips RK. Long-term results of polyp clearance by intraoperative enteroscopy in the Peutz-Jeghers Syndrome. Dis Colon Rectum 2003; 46:48–50. 112. Oncel M, Remzi FH, Church JM, Connor JT, Fazio VW. Benefits of "clean sweep" in the Pertz-Jeghers patients. Colorectal Dis 2004; 6:332–335. 113. Cronkhite LN Jr, Canada NJ. Generalized gastrointestinal polyposis. An unusual syndrome of polyposis, pigmentation, alopecis and onjehotrophia. N Engl J Med 1955; 252:1011–1015. 114. Hurt S, Mutch MG. The genetic of other polyposis syndromes. Semin Colon Rect Surg 2004; 15:158–162. 115. Daniel ES, Ludwig SL, Lewin KJ, Ruprecht RM, Rajacich GM, Schwale AD. The Croukhite-Canada syndrome. An analysis of clinical and pathologic features and therapy in 55 patients. Medicine 1982; 61:293–309. 116. Nardone G, D’Armiento F, Carlomagno P, Budillon G. Cronkhite Canada syndrome: case report with some features not previously described. Gastrointest Endosc 1990; 36:150–151. 117. Rappaport LB, Sperling HV, Stavirides A. Colon cancer in Cronkhite-Canada syndrome. J Clin Gastroenterol 1986; 8:199–202. 118. Watanabe T, Kudo M, Shirane H, et al. Cronkhite-Canada syndrome associated with triple gastric cancers: a case report. Gastrointest Endosc 1999; 50:688–691. 119. Russell DM, Bhathal PS, St John DJ. Complete remission in Cronkhite-Canada syndrome. Gastroenterology 1983; 85:180–185. 120. Hanzawa M, Yoshikawa N, Tezuka T, et al. Surgical treatment of CronkhiteCanada syndrome associated with protein-losing enteropathy. Report of a case. Dis Colon Rectum 1998; 41:932–934. 121. Divgi CR, Lisann NM, Yeh SD, Benua RS. Technitium-99 m albumin scintigraphy in the diagnosis of protein-losing enteropathy. J Nucl Med 1986; 27: 1710–1712. 122. Tseng KC, Sheu BS, Lee JC, Tsai HM, Chiu NT, Dai YC. Application of Technetium-99 m-labeled human serum albumin scan to assist surgical treatment of protein-losing enteropathy in Cronkhite-Canada syndrome: report of a case. Dis Colon Rectum 2005; 48:870–873. 123. Lloyd KM II, Dennis M. Cowden’s disease: a possible new symptom complex with multiple system involvement. Ann Intern Med 1963; 58:136–142. 124. Wirtzfeld DA, Petrelli NJ, Rodriguez-Bigus MA. Hamartomatous polyposis syndrome: moleculer genetics, neoplastic risk, and surveillance recommendations. Ann Surg Oncol 2001; 8:319–327. 125. Carlson GJ, Nivatvongs S, Snover DC. Colorectal polyps in Cowden’s disease (multiple hamartoma syndrome). Am J Surg Pathol 1984; 8:763–770. 126. Nelen MR, Padberg GW, Peeters EA, et al. Localization of the gene for Cowden disease to chromosome 10q22–23. Nat Genet 1996; 13:114–116. 127. Hanssen AM, Fryns JA. Cowden syndrome. J Med Genet 1995; 32:117–119. 128. Starink TM, van der Veen JP, Arwet F, et al. The Cowden syndrome: a clinical and genetic study in 21 patients. Clin Genet 1986; 29:222–233. 129. Longy M, Lacombe D. Cowden disease. Report of a family and review. Ann Genet 1996; 39:35–42. 130. Riley HD, Smith WR. Macrocephaly, pseudopapilledema and multiple hemangioma. Pediatrics 1960; 26:293–300. 131. Bannayan GA. Lipomatosis, angiomatosis, and macrocephaly. A previously undescribed congenital syndrome. Arch Pathol 1971; 92:1–5. 132. Ravalcaba RHA, Myhre S, Smith DW. Sotos syndrome with intestinal polyposis and pigmentary changes of the genitalia. Clin Genet 1980; 18:413–416. 133. Zigman AF, Lavine JE, Jones MC, et al. Localization of the Bannayan-RileyRuvalcaba syndrome gene to chromosome 10q23. Gastroenterology 1997; 113:1433–1437. 134. Morson BC, Bussey HJ. Predisposing causes of intestinal cancer. Curr Probl Surg 1970:1–46. 135. Dawson IM, Cornes JS, Morson BC. Primary malignant lymphoid tumors of the intestinal tract: report of 37 cases with a study of factors influencing prognosis. Br J Surg 1961; 49:80–89. 136. Khan A, Shrier I, Gordon PH. The changed histologic paradigm of colorectal polyps. Surg Endosc 2002; 16:436–440.

35

137. Jass JR. Nature and clinical significance of colorectal hyperplastic polyp. Semin Colon Rectal Surg 1991; 2:246–252. 138. Williams GT, Arthur JF, Bussey HJ, Morson BC. Metaplastic polyps and polyposis of the colorectum. Histopathology 1980; 4:155–170. 139. Hyman NH, Anderson P, Blasyk H. Hyperplastic polyposis and the risk of colorectal cancer. Dis Colon Rectum 2004; 47:2101–2104. 140. Jeevaratnam P, Cottier DS, Browett PJ, Van De Water NS, Pokos V, Jass JR. Familial giant hyperplastic polyposis predisposing to colorectal cancer: a new hereditary bowel cancer syndrome. J Pathol 1996; 179:20–25. 141. Bensen SP, Cole BF, Mott LA, Baron JA, Sandler RS, Halle R. Colorectal hyperplastic polyps and risk of recurrence of adenomas and hyperplastic polyps. Lancet 1999; 354:1873–1874. 142. Rustin RB, Jagelman DG, McGannon E, Fazio VW, Lavery IC, Weakley FL. Spontaneous mutation in familial adenomatous polyposis. Dis Colon Rectum 1990; 33:52–55. 143. Bulow S. The Danish polyposis registry. Dis Colon Rectum 1984; 27:351–355. 144. Bussey HJ. Familial Polyposis Coli. Family Studies, Histopathology, Differential Diagnosis and Results of Treatment. Baltimore: Johns Hopkins University Press, 1975. 145. Talbot IC. Pathology. In: Phillips RK, Spigelman AD, Thomson JP, eds. Familial Adenomatous Polyposis and Other Polyposis Syndromes. London: Edward Arnold, 1994:15–25. 146. Perry RE, Christensen MA, Thorson AG, Williams T. Familial polyposis: colon cancer in the absence of rectal polyps. Br J Surg 1989; 76:744. 147. Bulow S. The risk of developing rectal cancer after colectomy and ileorectal anastomosis in Danish patients with polyposis coli. Dis Colon Rectum 1984; 27:726–729. 148. Spiro L, Olschwang S, Groden J, et al. Alleles of the APC gene: an attenuated form of familial polyposis. Cell 1993; 75:951–957. 149. Lynch HT, Smyrk T, McGinn T, et al. Attenuated familial adenomatous polyposis (AFAP). A phenotypically and genotypically distinctive variant of FAP. Cancer 1995; 76:2427–2433. 150. Hernegger GS, Moore HG, Guillem JG. Attenuated familial adenomatous polyposis. An evolving and poorly understood entity. Dis Colon Rectum 2002; 45:127–136. 151. Church JM. Editorial. Dis Colon Rectum 2002; 45:134–135. 152. Hodgson SV, Spigelman AD. Genetics. In: Phillips RK, Spigelman AD, Thomson JP, eds. Familial Adenomatous Polyposis and Other Polyposis Syndromes. London: Edward Arnold, 1994:26–35. 153. Chung DC, Mino H, Shannon RM. Case 34–2003: a 45-year-old woman with a family history of colonic polyps and cancer. N Engl J Med 2003; 349:1750–1760. 154. Gardner EJ. A genetic and clinical study of intestinal polyposis, a predisposing factor for carcinoma of the colon and rectum. Am J Hum Genet 1951; 3:167–176. 155. Powell SM. Clinical applications of molecular genetics in colorectal cancer. Semin Colon Rectal Surg 1995; 6:2–18. 156. Campbell WJ, Spence RA, Parks TG. Familial adenomatous polyposis [review]. Br J Surg 1994; 81:1722–1733. 157. Utsunomiya J, Maki T, Iwama T, et al. Gastric lesions familial polyposis coli. Cancer 1974; 34:745–754. 158. Arcello PW, Asbun AJ, Veidenheimer MC, et al. Gastroduodenal polyps in familial adenomatous polyposis. Surg Endosc 1996; 40:418–421. 159. Spigelman AD, Williams CB, Talbot IC, Domizio P, Phillips RK. Upper gastrointestinal cancer in patients with familial adenomatous polyposis. Lancet 1989; 2:783–785. 160. Wallace MH, Phillips RK. Upper gastrointestinal disease in patients with familial adenomatous polyposis. Br J Surg 1998; 85:742–750. 161. Bulow S, Alm T, Fausa O, Hulterantz R, Jarvinen H, Vasen H. DAF Project Group. Duodenal adenomatosis in familial adenomatous polyposis. Int J Colorect Dis 1995; 10:43–46. 162. Spigelman AD. Familial adenomatous polyposis and the upper gastrointestinal tract. Semin Colon Rectal Surg 1995; 6:26–28. 163. Church JM, McGannon E, Hull-Boiner S, et al. Gastroduodenal polyps in patients with familial adenomatous polyposis. Dis Colon Rectum 1992; 35:1170–1173. 164. Bjork J, Akerbrant H, Iselius L, et al. Periampullary adenomas and adenocarcinomas in familial adenomatous polyposis: cumulative risks and APC gene mutations. Gastroenterology 2001; 121:1127–1135. 165. Domizio P, Talbot IC, Spigelman AD, Phillips RK, Williams CB. Upper gastrointestinal tract pathology in familial adenomatous polyposis: results from a prospective study of 102 patients. J Clin Pathol 1990; 43:738–743. 166. Kashiwagi H, Spigelman AD, Debuiski HS, Talbot IC, Phillips RK. Surveillance of ampullary adenomas in familial adenomatous polyposis [letter]. Lancet 1994; 344:1582. 167. Jagelman DG, DeCosse JJ, Bussey HJ. the Leeds Castle Polyposis Group Upper gastrointestinal cancer in familial adenomatous polyposis. Lancet 1988; 1:1149–1151. 168. Arvanitis ML, Jagelman DG, Fazio VW, Lavery IC, McGannon E. Mortality in patients with familial adenomatous polyposis. Dis Colon Rectum 1990; 33:639–642. 169. Belchetz LA, Berk T, Bapat BV, Cohen Z, Gallinger S. Changing causes of mortality in patients with familial adenomatous polyposis. Dis Colon Rectum 1996; 39:384–387.

36

& PART I: COLORECTAL DISORDERS

170. Wallace MH, Forbes A, Beveridge IG, et al. Randomized, place co-controlled trial of gastric acid-lowering therapy on duodenal polyposis and relative adduct labeling in familial adenomatous polyposis. Dis Colon Rectum 2001; 44:1585–1589. 171. Phillips RK, Wallace MH, Lynch PH, et al. The FAP study group. A randomized, double blind, placebo controlled study of celecoxib, a selective cyclooxygenase inhibitor on duodenal polyposis in familial adenomatous polyposis. Gut 2002; 50:857–860. 172. Perez A, Saltzman JR, Carr-Locke DL, et al. Benign nonampullary duodenal neoplasms. J Gastrointest Surg 2003; 7:536–541. 173. Alarcon FJ, Burke CA, Church JM, van Stolk RW. Familial adenomatous polyposis. Efficacy of endoscopic and surgical treatment for advanced duodenal adenomas. Dis Colon Rectum 1999; 42:1533–1536. 174. Burke CA, Beck GJ, Church JM, van Stolk RW. The natural history of untreated duodenal and ampullary adenomas in patients with familial adenomatous polyposis followed in an endoscopic surveillance program. Gastrointest Endosc 1999; 49:358–364. 175. Soravia C, Bork T, Haber G, Cohen Z, Gallinger S. Management of advanced duodenal polyposis in familial adenomatous polyposis. J Gastrointest Surg 1997; 1:474–478. 176. Penna C, Bataille N, Balladur P, Tiret E, Parc R. Surgical treatment of severe duodenal polyposis in familial adenomatous polyposis. Br J Surg 1998; 85:665–668. 177. de Vostot Nederveen Cappel WH, Jarvinen HJ, Bjork J, Berk T, Griffioen G, Vasen HF. Worldwide survey among polyposis registries of surgical management of severe duodenal adenomatosis in familial adenomatous polyposis. Br J Surg 2003; 90:705–710. 178. Morpurgo E, Vitale GC, Galandick S, Kimberling J, Ziegher C, Polk HC Jr. Clinical characteristics of familial adenomatous polyposis and management of duodenal adenomas. J Gastrointest Surg 2004; 8:559–564. 179. Mackey R, Walsh RM, Chung R, et al. Pancreas–sparing duodenectomy is effective management for familial adenomatous polyposis. J Gastrointest Surg 2005; 9:1088–1093. 180. Farnell MB, Sakorafas GH, Sarr MG, et al. Villous tumors of the duodenum: reappraisal of local vs extended resection. J Gastrointest Surg 2000; 4:13–23. 181. Ruo L, Coit DG, Brennam MF, Guillem JG. Long-Term follow-up of patients with familial adenomatous polyposis undergoing pancreaticoduodenal surgery. J Gastrointest Surg 2002; 6:671–675. 182. Brosens LA, Keller JJ, Offerhaus GJ, Goggins M, Giardiello FM. Prevention and management of duodenal polyps in familial adenomatous polyposis. Gut 2005; 54:1034–1043. 183. Giardiello FM, Petersen GM, Piantadosi S, et al. APC gene mutations and extraintestinal phenotype of familial adenomatous polyposis. Gut 1997; 40: 521–525. 184. Clark SK, Phillips RK. Desmoids in familial adenomatous polyposis [review]. Br J Surg 1996; 83:1494–1504. 185. Heiskanen I, Jarvinen HJ. Occurrence of desmoid tumors in familial adenomatous polyposis and results of treatment. Int J Colorect Dis 1996; 11:157–162. 186. Clark SK, Neale KF, Landgrebe JC, Phillips RK. Desmoid tumors complicating familial adenomatous polyposis. Br J Surg 1999; 86:1185–1189. 187. Hartley JE, Church JM, Gupta S, Mcgannon E, Fazio VW. Significance of incidental desmoids identified during surgery for familial adenomatous polyposis. Dis Colon Rectum 2004; 47:334–340. 188. Lofti AM, Dozois RR, Gordon H, et al. Mesenteric fibromatosis complicating familial adenomatous polyposis: predisposing factors and results of treatment. Int J Colorect Dis 1989; 4:30–36. 189. Phillips RK. Editorial. Dis Colon Rectum 2004; 47:339–340. 190. Church JM. Desmoid tumors in patients with familial adenomatous polyposis. Semin Colon Rectal Surg 1995; 6:29–32. 191. Middleton SB, Clark SK, Matravers P, Katz D, Reznek R, Phillips RK. Stepwise progression of familial adenomatous polyposis-associated desmoid precursor lesions demonstrated by a novel CT scoring system. Dis Colon Rectum 2003; 46:481–485. 192. Healy C, Reznek RH, Clark SK, Phillips RK, Armstrong P. MR appearances of desmoid tumors in familial adenomatous polyposis. Am J Roentgenol 1997; 169:465–472. 193. Middleton SB, Phillips RK. Surgery for large intra-abdominal desmoid tumors (Editorial, Church JM). Dis Colon Rectum 2000; 43:1759–1763. 194. Bulow S. Sulindac and tamoxifen in the treatment of desmoid tumors in patients with familial adenomatous polyposis. Colorect Dis 2001; 3:266–267. 195. Mo¨slein G, Dozois RR. Desmoid tumors associated with familial adenomatous polyposis. Perspect Colon Rectal Surg 1998; 10:109–126. 196. Poritz LS, Blackstein M, Berk T, Gallinger S, Melevd RS, Cohen Z. Extended follow-up of patients treated with cytotoxic chemiotherapy for intra-abdominal desmoid tumors. Dis Colon Rectum 2001; 44:1268–1273. 197. Church J, Berk T, Buman BM, et al. Staging intra-abdominal desmoid tumors in familial adenomatous polyposis: a search for a uniform approach to a troubling disease. Dis Colon Rectum 2005; 48:1528–1534. 198. Cabot RC. Case 21061. N Engl J Med 1935; 212:263–267. 199. Blair N, Trempe CL. Hypertrophy of the retinal pigment epithelium associated with Gardner’s syndrome. Am J Ophthalmol 1980; 90:661–667.

200. Leppard B, Bussey HJ. Epidermoid cysts, polyposis coli and Gardner’s syndrome. Br J Surg 1975; 62:387–393. 201. Campbell WJ, Spence RA, Parks TG. The role of congenital hypertrophy of the retinal pigment epithelium in screening for familial adenomatous polyposis. Int J Colorect Dis 1994; 9:191–196. 202. Crail HW. Multiple primary malignancies arising in the rectum, brain and thyroid. Report of a case. US Navy Med Bull 1949; 49:123–128. 203. Bret MC, Hershman MJ, Glazer G. Other manifestations of familial adenomatous polyposis. In: Phillips RK, Spigelman AD, Thomson JP, eds. Familial Adenomatous Polyposis and Other Polyposis Syndromes. London: Edward Arnold, 1994:143–158. 204. Turcot J, Despre´s JP, St. Pierre F. Malignant tumors of the central nervous system associated with familial polyposis of the colon. Dis Colon Rectum 1959; 2:465–468. 205. Hamilton SR, Liu B, Parsons RE, et al. The molecular basis of Turcots syndrome. N Engl J Med 1995; 332:839–847. 206. Matsui T, Hayashi N, Yao K, et al. A father and son with Turcot’s syndrome: evidence for autosomal dominant inheritance. Report of two cases. Dis Colon Rectum 1998; 41:797–801. 207. Itoh H, Hirata K, Ohsato K. Turcot’s syndrome and familial adenomatous polyposis associated with brain tumor: review of related literature. Int J Colorect Dis 1993; 8:87–94. 208. Gingold BS, Jagelman DG, Turnbull RD. Surgical management of familial polyposis and Gardner’s syndrome. Am J Surg 1979; 137:54–56. 209. Bess MA, Adson MA, Elveback LR, Moertel CG. Rectal cancer following colectomy for polyposis. Arch Surg 1980; 115:460–467. 210. Bussey HJ, Eyers AA, Ritchie SM, Thomson JP. The rectum in adenomatous polyposis. The St. Mark’s policy. Br J Surg 1985; 72(suppl):S29–S31. 211. Nugent KP, Phillips RK. Rectal cancer risk in older patients with familial adenomatous polyposis and an ileorectal anastomosis: a cause for concern. Br J Surg 1992; 79:1204–1206. 212. Heiskanen I, Jarvinen HJ. Fate of the rectal stump after colectomy and ileorectal anastomosis for familial adenomatous polyposis. Int J Colorect Dis 1997; 12:9–13. 213. Sarre RG, Jagelman DG, Beck GJ, et al. Colectomy with ileorectal anastomosis for familial adenomatous polyposis: the risk of rectal cancer. Surgery 1987; 101:20–26. 214. DeCosse JJ, Bulow S, Neale K, the Leeds Castle Polyposis Group. Rectal cancer risk in patients treated for familial adenomatous polyposis. Br J Surg 1992; 79:1372–1375. 215. Iwama T, Mishima Y. Factors affecting the risk of rectal cancer following rectum-preserving surgery in patients with familial adenomatous polyposis. Dis Colon Rectum 1994; 37:1024–1026. 216. Jenner DC, Levitt S. Rectal Cancer following colectomy and ileorectal anastomosis for familial adenomatous polyposis. Aust N Z J Surg 1998; 68:136–138. 217. Bjork JA, Akerbrant HI, Iselius L, Hultcrantz RW. Risk factors for rectal cancer morbidity and mortality in patients with familial adenomatous polyposis after colectomy and ileorectal anastomosis. Dis Colon Rectum 2000; 43: 1719–1725. 218. Bertario L, Russo A, Radice P, et al. Genotype and phenotype factors as determinants for rectal stump cancer in patients with familial adenomatous polyposis. Ann Surg 2000; 231:538–543. 219. Church J, Burke C, McGannon E, Pastean O, Clark B. Risk of rectal cancer in patients after colectomy and ileorectal anastomosis for familial adenomatous polyposis. A function of available surgical options. Dis Colon Rectum 2003; 46:1175–1181. 220. Penna C, Kartheuser A, Parc R, et al. Secondary proctectomy and ileal pouchanal anastomosis for familial adenomatous polyposis. Br J Surg 1993; 80: 1621–1625. 221. Heiskanen I, Matikainen M, Hiltunen KM, Rintala R, Jarvinen HJ. Colectomy and ileorectal anastomosis or restorative for proctocolectomy for familial adenomatous polyposis. Colorect Dis 1999; 1:9–14. 222. Phillips RK, Spigelman AD. Can we safely delay or avoid prophylactic colectomy in familial adenomatous polyposis? Br J Surg 1996; 83:769–770. 223. Bulow C, Vasen H, Jarvinen H, Bjork J, Bisgaard ML, Bulow S. Ileorectal anastomosis is appropriate for a subset of patients with familial adenomatous polyposis. Gastroenterology 2000; 119:1454–1460. 224. Church J, Burke C, McGannon E, Pasteam O, Clark B. Predicting polyposis severity by proctoscopy: how reliable is it? Dis Colon Rectum 2001; 44: 1249–1254. 225. Farmer KC, Phillips RK. Colectomy with ileorectal anastomosis lowers rectal mucosal cell proliferation in familial adenomatous polyposis. Dis Colon Rectum 1993; 36:167–171. 226. Karen P, Nugent MA, Northover J. Total colectomy and ileorectal anastomosis. In: Phillips RK, Spigelman AD, Thomson JP, eds. Familial Adenomatous Polyposis and Other Polyposis Syndromes. London: Edward Arnold, 1994:80–91. 227. Winde G, Schmid KW, Schfegel W, Fischer R, Osswald H, Bunte H. Complete reversion and prevention of rectal adenomas in colectomized patients with familial adenomatous polyposis by rectal low-dose sulindac maintenance treatment. Dis Colon Rectum 1995; 38:813–830.

CHAPTER 1: BENIGN NEOPLASMS OF THE COLON AND RECTUM &

228. Spagnesi MT, Tonelli F, Dolara P, et al. Rectal proliferation and polyp occurrence in patients with familial adenomatous polyposis after sulindac treatment. Gastroenterology 1994; 106:362–366. 229. Winde G, Schmid KW, Brandt B, Mu¨ller O, Osswald H. Clinical and genomic influence of sulindac on rectal mucosa in familial adenomatous polyposis. Dis Colon Rectum 1997; 40:1156–1169. 230. Giadiello FM, Yang VW, Hylind LM, et al. Primary chemoprevention of familial adenomatous polyposis with sulindac. N Engl J Med 2002; 346:1054–1059. 231. Thorson AG, Lynch HT, Smyth TC. Rectal cancer in FAP patients after sulindac. Lancet 1994; 343:180. 232. Yang VW, Geiman DE, Hubbard WC, et al. Tissue prostanoids as biomarkers for chemoprevention of colorectal neoplasia: correlation between prostanoid synthesis and clinical response in familial adenomatous polyposis. Prostaglandins Lipid Mediat 2000; 60:83–90. 233. Niv Y, Fraser M. Adenocarcinoma in the rectal segment in familial polyposis coli is not prevented by sulindac therapy. Gastroenterology 1994; 107:854–857. 234. Kartheuser AH, Parc R, Penna CP, et al. Ileal pouch-anal anastomosis as the first choice operation in patients with familial adenomatous polyposis: a 10year experience. Surgery 1996; 119:615–623. 235. Ziv Y, Church JM, Oakley JR, McGannon E, Fazio VW. Surgery for the teenager with familial adenomatous polyposis: ileo-rectal anastomosis or restorative proctocolectomy?. Int J Colorect Dis 1995; 10:6–9. 236. Pare YR, Moslein G, Dozois Pamberton JH, Woeff BG, King JE. Familial adenomatous polyposis. Results after ileal pouch-anal anastomosis in teenagers. Dis Colon Rectum 2000; 43:893–902. 237. Ambrose WL, Dozois RR, Pemberton JH, Beart RW, Ilstrup DM. Familial adenomatous polyposis: results following ileal pouch-anal anastomosis and ileorectostomy. Dis Colon Rectum 1992; 35:12–15. 238. Setti-Carraro P, Nicholls RJ. Choice of prophylactic surgery for the large bowel component of familial adenomatous polyposis [review]. Br J Surg 1996; 83:885–892. 239. Soravia C, Klein L, Berk T, Oconnoroz BI, Cohen Z, McLeod RS. Comparison of ileal pouch-and anastomosis and ileorectal amastomosis in patients with familial adenomatous polyposis. Dis Colon Rectum 1999; 42:1028–1034. 240. Church J. Ileoanal pouch neoplasia in familial adenomatous polyposis: an underestimated threat. Dis Colon Rectum 2005; 48:1708–1713. 241. Remzi FH, Church JM, Bast J, et al. Mucosectomy vs stapled ileal pouch-and adenomatous polyposis. Functional outcome and neoplasia control. Dis Colon Rectum 2001; 44:1590–1596. 242. van Duijvendijk P, Varen HF, Bertario L, et al. Cumulative risk of developing polyps or malignancy at the ileal pouch-anal anastomosis in patients with familial adenomatous polyposis. J Gastrointest Surg 1999; 3:325–330. 243. Petersen GM. Genetic counseling and predictive genetic testing in familial adenomatous polyposis. Semin Colon Rectal Surg 1995; 6:55–60. 244. Spigelman AD, Thomson JP. Introduction, history and registries. In: Phillips RK, Spigelman AD, Thomson JP, eds. Familial Adenomatous Polyposis and Other Polyposis Syndromes. London: Edward Arnold, 1994:3–14. 245. Wong N, Lesko D, Rahelo R, Pinsky L, Gordon PH, Foulkes W. Genetic counseling and interpretation of genetic tests in familial adenomatous polyposis and hereditary nonpolyposis colorectal cancer. Dis Colon Rectum 2001; 44:271–279. 246. Rabelo R, Foulkes W, Gordon PH, et al. Role of molecular diagnostic testing in familial adenomatous polyposis and hereditary nonpolyposis colorectal cancer familiar. Dis Colon Rectum 2001; 44:437–446. 247. Church JM, McGannon E. A polyposis registry: how to set one up and make it work. Semin Colon Rectal Surg 1995; 6:48–54.

37

248. Rotherberger DA, Dalberg DI, Leininger A. Minnesota colorectal cancer initiative: successful development and implementation of a community-based colorectal cancer registry. Dis colon Rectum 2004; 47: 1571–1577. 249. Lyon DT, Mantia AG. Large bowel hemangiomas. Dis Colon Rectum 1984; 27:404–414. 250. Allred HW, Spencer RJ. Hemangiomas of the colon, rectum, and anus. Mayo Clin Proc 1974; 49:739–741. 251. Londono-Schimmer EE, Ritchie JK, Hawley PR. Coloanal sleeve anastomosis in the treatment of diffuse cavernous hemangioma of the rectum: long-term results. Br J Surg 1994; 81:1235–1237. 252. Jefferey PI, Hawley PR, Parks AG. Colo-anal sleeve anastomosis in the treatment of diffuse cavernous hemangioma involving the rectum. Br J Surg 1976; 63:678–682. 253. Coppa GF, Eng K, Localio SA. Surgical management of diffuse cavernous hemangioma of the colon, rectum, and anus. Surg Gynecol Obstet 1984; 159:17–22. 254. Telander RL, Ahlquist D, Blanfuss MC. Rectal mucosectomy: a definitive approach to extensive hemangiomas of the rectum. J Pediatr Surg 1993; 28:379–381. 255. He LJ, Wang BS, Chen CC. Smooth muscle tumors of the digestive tract: report of 160 cases. Br J Surg 1988; 75:184–186. 256. Morgan BK, Compton C, Talbert M, Gallagher WJ, Wood WC. Benign smooth muscle tumors of the gastrointestinal tract. Ann Surg 1990; 211:63–66. 257. Kusminsky RE, Bailey W. Leiomyomas of the rectum and anal canal: report of 6 cases and review of the literature. Dis Colon Rectum 1977; 20:580–599. 258. Miettinen M, Furlong M, Sarlomo-Rikala M, Burke A, Sobin LH, Lasota J. Gastrointestinal stromal tumors, intramural leiomyomas and leiomyosarcomas in the rectum and anus. A clinicopathologic, immunohistochemical, and molecular genetic study of 144 cases. Am J Surg Path 2001; 25:1121–1133. 259. Walsh TH, Mann CV. Smooth muscle neoplasms of the rectum and anal canal. Br J Surg 1984; 71:597–599. 260. Vorobyov GI, Odaryule TS, Kapulter LL, Shelygin YA, Kornyak BS. Surgical treatment of benign, myomatous rectal tumors. Dis Colon Rectum 1992; 35:328–331. 261. Taylor B, Wolff BG. Colonic lipomas. Report of two unusual cases and review of the Mayo Clinic experience, 1976–1985. Dis Colon Rectum 1987; 30:888–893. 262. Gordon RT, Beal JM. Lipoma of the colon. Arch Surg 1978; 113:897–899. 263. Castro EL, Stearns MW. Lipoma of the large intestine: a review of 45 cases. Dis Colon Rectum 1972; 15:441–444. 264. Brouland JP, Poupard B, Nemeth J, Valleur P. Lipomatous polyposis of the colon with multiple lipomas of peritoneal folds and giant diverticulosis. Dis Colon Rectum 2000; 43:1767–1769. 265. Santos-Briz, Garcia JP, Gonzalez C, Colina F. Lipomatous polyposis of the colon. Histopathology 2001; 38:81–83. 266. Zhang H, Cong JC, Chen CS, Qiao L, Liu ES. Submucous colon lipoma: a case report review of the literature. World J Gastroenterol 2005; 11:3167–3169. 267. Liessi G, Pavanello M, Cesari S, Dell’Antonio C, Avventi P. Large lipomas of the colon: CT and MR findings in three symptomatic cases. Abdom Imaging 1996; 21:150–152. 268. Pfeil SA, Weaver MG, Abdul-Karim FW, Yang P. Colonic lipomas: outcome of endoscopic removal. Gastrointest Endosc 1990; 36:435–438. 269. Chung YF, Ho YH, Nyam DC, Leong AF, Seow-Choen F. Management of colonic lipomas. Aust N Z J Surg 1998; 68:133–135.

2

Large Bowel Carcinoma: Screening, Surveillance, and Follow-Up Santhat Nivatvongs

& DETECTION OF EARLY COLORECTAL CARCINOMA

& Detection of Early Colorectal Carcinoma, 39 & Early Diagnosis of Colorectal Carcinoma, 40

Data in 2005 showed that there were 145,290 estimated new cases of colon and rectal carcinomas in the United States, and of these, 40,340 (28%) were carcinomas of the rectum. Of this 56,290 (combined colon and rectum) will die, a death rate of 39%. In the United States, between 1991 and 2001, the death rate from carcinoma of colon and rectum has improved by 5% in men, and 4% in women (1). Survival for colon and rectal carcinoma is closely related to the clinical and pathologic stage of the disease at diagnosis. Data from the German Multicenter Study in colorectal carcinoma showed five-year survival rates in stages I, II, III, and IV as 76%, 65%, 42%, and 16%, respectively (surgical mortality not excluded) (2). A good starting point for an analysis of the pattern of colorectal carcinoma can be seen in data from the National Cancer Data Base (NCDB) (3) (Table 1). It shows the stages at diagnosis in different age groups and ethnicity. These data show the better stage in patients over 70 years for both carcinoma of colon and rectum. Ethnicity is another significant factor. African-Americans are less likely to present with stage I or II than are non-Hispanic whites. Hispanics are also slightly less likely to be diagnosed with stage I or II colorectal carcinoma than non-Hispanic whites. However, Asians have a pattern of stage and diagnosis similar to that of non-Hispanic whites (data not shown), and the sample of native Americans is too small to draw any conclusion about the stage at presentation. The data from NCDB suggest that age and ethnicity are important factors to consider in the diagnosis of colorectal carcinoma (3). It has become clear that if the disease can be detected at an early stage, the overall prognosis can be improved, with another benefit that some colorectal carcinomas can even be prevented. Most colorectal carcinomas are asymptomatic until a late stage, when some partial obstruction occurs, causing abdominal pain or change in bowel habits. Although carcinoma of the colon and rectum bleeds occasionally and unpredictably, it may be possible to diagnose it in an early stage by examining for occult blood in the stool. Through many observations and studies, including current knowledge of molecular genetics of colorectal carcinoma, the natural history of colorectal carcinoma starts with one crypt. The numerous mutations of genes slowly give rise to a small polyp and then progresses to an invasive carcinoma that eventually metastasizes. The National Polyp Study (NPS) showed that it takes about 10 years for the development of an invasive carcinoma from a ‘‘clean’’ colon (4). This lengthy, stepwise natural history provides a window of opportunity for detecting early carcinoma and

& What Is Screening?, 40 & Who Should Be Screened?, 40 & Screening People at Average Risk for

Colorectal Carcinoma, 41 Fecal Occult Blood Test, 42 Flexible Sigmoidoscopy, 42 Combined FOBT and Flexible Sigmoidoscopy, 42 Colonoscopy, 42 Double-Contrast Barium Enema (DCBE), 43 & Screening People at Increased Risk for

Colorectal Carcinoma, 43 Family History of Colorectal Carcinoma or Adenomatous Polyp, 43 Genetic Syndromes, 43 Detection of Second Malignancies, 43 & New Screening Tests, 44

Computed Tomography Colonography (CTC, Virtual Colonoscopy), 44 Fecal DNA Testing, 44 & When to Stop Screening, 44 & Surveillance, 45

After Removal of Adenomatous Polyps, 45 In People with Inflammatory Bowel Disease, 45 In People with Ureterosigmoidostomy, 45 & Follow-Up After Curative Resection, 46

Investigation to Detect Recurrences and Metastases, 46 Metachronous Carcinomas and Polyps, 47 & Other Primary Malignancies, 47 & Summary, 47 & References, 48

39

40

& PART I: COLORECTAL DISORDERS

TABLE 1 & Distribution of Adenocarcinoma of the Colon and

Rectum by American Joint Commitee on Cancer (AJCC) Stage, Age at Diagnosis, and Ethnicity of Patients Diagnosed in 1993 at Hospitals Participating in the National Cancer Data Base Distribution by Stage (%) Category and Site Age Colon < 50 50–59 60–69 70–79
80 Subtotal Rectum < 50 50–59 60–69 70–79
80

I–II

III

IV

Total (%)

Subtotal Rectum Non-Hispanic white Hispanic African-American Subtotal

No. of Cases

& WHAT IS SCREENING? 44 47 53 57 59

27 27 26 25 25

29 26 21 19 16

100 100 100 100 100

2503 4130 9112 12,871 8661 37,277

50 53 58 59 62

33 30 27 26 22

17 17 15 15 16

100 100 100 100 100

Subtotal Ethnicity Colon Non-Hispanic white Hispanic African-American

carcinoma and detecting early-stage carcinomas reduces mortality from the disease. Finally, screening benefits outweigh its harms. The various ways of screening for colorectal carcinoma all have cost-effectiveness ratios comparable to those of other generally accepted screening tests (10).

1338 2406 4273 4579 2238 14,834

55 51 48

25 29 27

20 20 25

100 100 100

32,463 680 3087 36,230

58 53 54

27 27 25

15 20 21

100 100 100

11,731 372 950

Screening identifies individuals who are more likely to have colorectal carcinoma or adenomatous polyps from among those without signs or symptoms of disease (10). It is the use of simple, affordable, and acceptable tests to identify a subgroup of the at-risk population more likely to have a clinically significant lesion or abnormality (11). Screening, which refers primarily to a population approach, has been used interchangeably with early detection. Case finding refers to early detection on an individual basis. These terms refer to the identification of individuals with an increased probability of having colorectal neoplasia (4). The goal of screening for colorectal carcinoma is to reduce mortality from the disease. It is important to note that once the screening results are positive, a complete investigation of the entire colon and rectum is mandated, preferably by CT colonography, total colonoscopy, or with flexible sigmoidoscopy and barium enema, to identify colorectal polyps or carcinomas. Screening should be accompanied by efforts to optimize the participation of patients and healthcare providers, and to remind patients and providers about the need for rescreening at recommended intervals (12).

13,053

Source: From Ref. 3.

removing malignant polyps (Fig. 1). Thus screening strategy can be directed toward detecting early carcinoma to reduce morbidity and mortality as well as removing premalignant polyps to reduce the incidence of colorectal carcinoma.

& EARLY DIAGNOSIS OF COLORECTAL CARCINOMA Colorectal carcinoma fulfills all the criteria for justified screening. First, it is common and serious: it is the second leading cause of death from carcinoma in the United States, affecting men and women equally. Treatment of patients with advanced colorectal carcinoma is largely unsuccessful. In 2005 the overall survival rate was 61% (1). Second, various screening tests have been shown to achieve accurate detection of early-stage colorectal carcinomas (5–9). Third, evidence from controlled trials and case-control studies suggests varying degrees of persuasiveness that removing adenomatous polyps reduces the incidence of colorectal

& WHO SHOULD BE SCREENED? Most people in the United States are not currently screened for colorectal carcinoma. Seeff et al. (13) analyzed data from the National Health Interview Survey which demonstrated that less than half of the U.S. population age
50 years reported undergoing hemoccult occult blood test (FOBT) and endoscopy (sigmoidoscopy, colonoscopy, or proctoscopy) within recommended time intervals. A lack of awareness by the respondent of the need for the test and a lack of recommendation by the physician for the test to be performed were found to be the most commonly reported barriers to undergoing the test. Lack of physician recommendation clearly was an important barrier; among persons who reported undergoing no colorectal carcinoma testing or none recently, only 5% reported that a physician had recommended colorectal carcinoma testing. Approximately 75% of all new cases of colorectal carcinoma occur in people with no known predisposing factors for the disease. Incidence increases with age, beginning around 40 years (14). People with no predisposing factors are considered to be at average risk for colorectal

FIGURE 1 & The natural history of colorectal carcinoma. Source: From Ref. 4.

CHAPTER 2: LARGE BOWEL CARCINOMA: SCREENING, SURVEILLANCE, AND FOLLOW-UP &

41

and low in fiber, excessive alcohol consumption, and sedentary lifestyle (15). Screening people at average risk for colorectal carcinoma is different from screening people at high risk. Clinicians should determine an individual patient’s risk status well before the earliest potential initiation of screening. The individual’s risk status determines when screening should be initiated, and what tests and frequency are appropriate (Fig. 3) (12). Risk stratification can be accomplished by asking several questions aimed at uncovering the risk factors for colorectal cancer (12). They are as follows:

FIGURE 2 & Annual new cases of colorectal carcinoma in men and women
50 years of age with no special risk factors. Abbreviations: FAP, familial adenomatous polyposis; FH, family history; HNPCC, hereditary nonpolyposis colon cancer; IBD, inflammatory bowel disease. Source: From Ref. 10.

carcinoma. People with a family history of colorectal carcinoma (i.e., one or more parents, siblings, or children with the disease) but without any apparent defined genetic syndrome account for most of those at high risk (15%– 20%). Hereditary nonpolyposis colon cancer (HNPCC) accounts for 4–7% of all cases and familial adenomatous polyposis (FAP) about 1%. The remainder, about 1%, are attributed to a variety of uncommon conditions: chronic ulcerative colitis, Crohn’s colitis, Peutz-Jeghers syndrome, and familial juvenile polyposis, in which the colorectal carcinoma risk is elevated but is not as high as in HNPCC and FAP (Fig. 2) (10). Other risk factors that should be kept in mind include older age, a diet high in saturated fats

1. Has the patient had colorectal carcinoma or an adenomatous polyp? 2. Does the patient have an illness (e.g., inflammatory bowel disease) that predisposes him or her to colorectal carcinoma? 3. Has a family member had colorectal carcinoma or an adenomatous polyp? If so, how many, was it a firstdegree relative (parent, sibling, or child) and at what age was the carcinoma or polyp first diagnosed? A positive response to any of these questions should prompt further efforts to identify and define the specific condition associated with increased risk.

& SCREENING PEOPLE AT AVERAGE RISK FOR COLORECTAL CARCINOMA Men and women at average risk should be offered screening with one of the following options beginning at 50 years. A study of a screening colonoscopy in 40–49 years old people confirmed that colorectal carcinomas are uncommon at this age group, supporting the recommendation that screening in average risk people begin at age 50 years (16).

FIGURE 3 & The algorithm for colorectal carcinoma screening. Abbreviations: AVG, average; HNPCC, hereditary nonpolyposis colorectal cancer; FAP, familial adenomatous polyposis. Source: From Ref. 12.

42

& PART I: COLORECTAL DISORDERS

The rationale for presenting multiple options is that no single test has unequivocal superiority, and that giving patients a choice allows them to apply personal preferences and may increase the likelihood that screening will occur. The strategies are not equal with regard to evidence of effectiveness, magnitude of effectiveness, risk, or up-front costs. Reviewing the rationale section for each screening test below will provide clinicians with information that they can use in presenting the relative effectiveness of each test to patients. These tests are recommended by the American Gastroenterological Association (12) as well as the American Cancer Society guidelines (17). The following excerpts are heavily taken from the American Gastroenterological Association, presented by Winawer et al. (12).

& FECAL OCCULT BLOOD TEST This test offers yearly screening with fecal occult blood test (FOBT) using Guaiac-based test with dietary restriction (confined to red meat alone by waiting three days before developing the test) or an immunochemical test without dietary restriction. Two samples from each of three consecutive stools should be examined without rehydration. The patients with a positive test on any specimen should be followed up with colonoscopy (12). The American Gastroenterological Society recommend yearly testing because it is more effective than screening every two years. Rehydration is not recommended. Newer Guaiac-based and immunochemical tests are available that have improved sensitivity and appear to maintain acceptable specificity. Dietary restrictions during testing are commonly recommended to reduce the false-positive rate for the more sensitive Guaiac-based tests but are not necessary for the immunochemical and less sensitive Guaiac-based tests. With longer (18 years) follow-up in the Minnesota trial, FOBT screening every other year was found to reduce colorectal cancer mortality by 21% (18), a rate consistent with the results of the biennial screening in the two European trials (7,8). The incidence of colorectal carcinoma was also reduced in the screened group (19). A systematic review of three clinical trials (5–7,19) has shown that a restricted diet does not reduce the positivity rate for the older, less-sensitive Guaiac-based tests and that very restricted diets may reduce compliance rates (20). Disadvantages of FOBT are that currently available tests for occult blood fail to detect many polyps and some carcinomas. Also, most people who test positive will not have colorectal neoplasia (have a false-positive test result) and, thus, will undergo the discomfort, cost, and risk of colonoscopy without benefit. Colonoscopy is recommended for all those with a positive FOBT because it was a diagnostic procedure used throughout most of the trials, and because it is substantially more accurate than doublecontrast barium enema for the detection of both small carcinomas and adenomas (21). & FLEXIBLE SIGMOIDOSCOPY Screening with flexible sigmoidoscopy is recommended every five years. Case-controlled studies have reported that sigmoidoscopy was associated with reduced mortality for colorectal carcinoma (22–24). Colon carcinoma risk in the

area beyond the reach of the sigmoidoscope was not reduced. A five-year interval between screening examinations is a conservative choice. It is supported by the observation that a reduction in colorectal carcinoma deaths related to screening sigmoidoscopy was present up to 10 years from the last screening examination (22), and that repeat colonoscopy five years after a negative colonoscopy found few instances of advanced neoplasia (25), and follow-up of a cohort of patients after polyp excision showed that development of advanced neoplasia was rare up to five years after a negative colonoscopy (26). The interval is shorter than for colonoscopy because flexible sigmoidoscopy is less sensitive than colonoscopy. Several studies have shown that the prevalence of proximal advanced adenomas in patients without distal adenomas is in the 2–5% range (27–30). A flexible sigmoidoscopy followed by colonoscopy if a polyp was found would have identified 70–80% of patients with advanced proximal neoplasia (28). In one randomized controlled trial, screening sigmoidoscopy followed by colonoscopy when polyps were detected was associated with an 80% reduction in colorectal carcinoma incidence (31).

& COMBINED FOBT AND FLEXIBLE SIGMOIDOSCOPY FOBT is done every year combined with flexible sigmoidoscopy every five years. When both tests are performed, the FOBT should be done first because a positive result is an indication for colonoscopy, obviating the need for the sigmoidoscopy examination. The effectiveness of this combined screening strategy in reducing mortality has never been studied directly in a randomized trial. It is likely that the combination of both screening methods is more effective than either method of screening alone for several reasons: FOBT may be less sensitive for distal colon lesions (32), case-controlled studies report screening FOBT and sigmoidoscopy each are associated with reduced colorectal carcinoma mortality after controlling for the other (22,33), and a nonrandomized controlled trial reported a 43% reduction (which was not statistically significant) in colorectal carcinoma deaths in people screened with FOBT and sigmoidoscopy relative to sigmoidoscopy alone (6). The disadvantage of the FOBT/sigmoidoscopy strategy is that people incur the inconvenience, cost, and complications of both tests with an uncertain gain in effectiveness. & COLONOSCOPY Colonoscopy is offered every 10 years. Although there are no studies evaluating whether screening colonoscopy alone reduces the incidence or mortality from colorectal carcinoma in people at average risk, several lines of evidence support the effectiveness of screening colonoscopy (12). There is direct evidence that screening sigmoidoscopy reduces colorectal carcinoma mortality (22,23), and colonoscopy allows more of the large bowel to be examined. Colonoscopy has been shown to reduce the incidence of colorectal carcinoma in two cohort studies of people with adenomatous polyps (26,34). Colonoscopy permits detection and removal of polyps and biopsy of carcinoma throughout the colon. However, colonoscopy involves greater cost, risk, and inconvenience to the patient than other screening tests, and not all examinations visualize

CHAPTER 2: LARGE BOWEL CARCINOMA: SCREENING, SURVEILLANCE, AND FOLLOW-UP &

the entire colon. The added value of colonoscopy over sigmoidoscopy screening, therefore, involves a tradeoff of incremental benefits and harms (12). Choice of a 10-year interval between screening examinations for average-risk people (if the preceding examination is negative) is based on estimates of the sensitivity of colonoscopy and the rate at which advanced adenomas develop. The dwell time from the development of adenomatous polyps to transformation into carcinoma is estimated to be at least 10 years on average (26,35). In two large prospective studies of screening colonoscopy, about half of patients with advanced proximal neoplasms had no distal colonic neoplasms (28,29). Similarly, a prospective study of distal colon findings in a cohort of average-risk persons with carcinoma proximal to the splenic flexure found that 65% had no neoplasm distal to the splenic flexure (36). A randomized controlled trial of sigmoidoscopy with follow-up colonoscopy for all patients with polyps compared with no screening demonstrated a significant reduction in colorectal carcinoma incidence in the screened patients (31). A cohort of 154 asymptomatic average-risk persons with negative screening colonoscopies had < 1% incidence of advanced neoplasms at a second colonoscopy five years later (25), lending support to the recommended interval of 10 years. Two colonoscopy studies suggested that flat and depressed adenomas account for 22% and 30% of adenomas (37,38), and one report suggests that dye spraying is necessary to not miss these lesions (37). However, the precise prevalence and clinical significance of flat adenomas is uncertain. By the end of 2000, the U.S. Medicare had decided to reimburse for colonoscopy screening. Colonoscopy as the best primary screening test began to be discussed and to be advocated by some gastroenterology organizations (39).

& DOUBLE-CONTRAST BARIUM ENEMA (DCBE) This test is recommended every five years. There are no randomized trials evaluating whether screening DCBE reduces the incidence or mortality from colorectal carcinoma in people at average risk of the disease. The sensitivity of DCBE for large polyps and carcinomas is substantially less than with colonoscopy, the procedure does not permit removal of polyps or biopsy of the carcinomas, and DCBE is more likely than colonoscopy to identify artifacts and other findings (such as stool) as polyps. Patients with an abnormal barium enema need a subsequent colonoscopy. DCBE is included as an option because it offers an alternative (albeit less sensitive) means to examine the entire colon, it is widely available, and it detects about half of large polyps, which are most likely to be clinically important. Adding flexible sigmoidoscopy to DCBE is not recommended in the screening setting. A five-year interval between DCBE examinations is recommended because DCBE is less sensitive than colonoscopy in detecting colonic neoplasms. In a prospective study of DCBE in a surveillance population with a spectrum and prevalence of disease similar to a screened population, DCBE detected 53% of adenomatous polyps 6–10 mm in size, and 48% of those >1 cm in size compared with colonoscopy (21). In a nonrandomized study of 2193 consecutive colorectal carcinoma cases in community practice, the sensitivity for carcinoma was 85% with DCBE and 95% with colonoscopy (40).

43

& SCREENING PEOPLE AT INCREASED RISK FOR COLORECTAL CARCINOMA Screening high-risk people can take several forms. Patients can begin screening at an earlier age if polyps and carcinomas arise at an earlier age, they can be screened more frequently if the evolution from small polyps to carcinoma is more rapid, they can be screened by tests that reach the right colon if the carcinoma occurs more proximally, or they can be screened with more sensitive methods, such as colonoscopy or DCBE rather than FOBT or sigmoidoscopy. Patients already found to have adenomatous polyps are at increased risk for colorectal carcinoma and are candidates for surveillance rather than screening (10).

& FAMILY HISTORY OF COLORECTAL CARCINOMA OR ADENOMATOUS POLYP This group consists of individuals having one or more firstdegree relatives (parent, sibling, or child) with colorectal carcinoma or adenomatous polyps diagnosed at age < 60 years. There is significant evidence that carcinomas arise at an earlier age in these people than in average-risk persons. In effect, the risk of a 40-year-old person with a family history of colorectal carcinoma is comparable to that of an average-risk 50-year-old person (41) Screening colonoscopy should be started at age 40 years or 10 years younger than the earliest diagnosis in their family, whichever comes first, and repeated every five years (12). Colorectal carcinoma screening recommendations for people with familial or inherited risk is in Table 2. The lifetime risk of colon carcinoma in people with family history of colon carcinoma is in Table 3. & GENETIC SYNDROMES See Chapter 22 for screening of familial adenomatous polyposist (FAP) and Chapter 23 for screening of HNPCC. & DETECTION OF SECOND MALIGNANCIES Some patients have a higher incidence of a second malignancy than the normal population. Analysis of the Utah Cancer Registry, which documented more than 35,000 carcinomas, revealed that Utah men with one carcinoma have a 1.2 times greater likelihood of developing another carcinoma and Utah women have a 1.5 times greater likelihood than did other persons in the Utah population of the same race, sex, and age who have not had a previous malignancy. In particular, men with primary carcinoma of the colon and rectum have a higher incidence of developing a second carcinoma in the colon, rectum, prostate, or bladder (42). A complete colonic examination with colonoscopy or flexible sigmoidoscopy combined with barium enema to check metachronous lesions should be performed every five years. At present, the glycoprotein, prostate-specific antigen (PSA) is the most useful marker available for the diagnosis and management of prostate carcinoma. However, PSA is prostate specific but is not sufficiently specific to be used alone as a screening test for prostate carcinoma. PSA is also produced by normal prostatic tissue and can indicate the presence of benign prostatic hyperplasia (BPH).

44

& PART I: COLORECTAL DISORDERS

TABLE 2 & Colon Carcinoma Screening Recommendations for

People with Familial or Inherited Risk Familial Risk Category First-degree relative affected with colorectal carcinoma or an adenomatous polyp at age
60 years, or two second-degree relatives affected with colorectal carcinoma Two or more first-degree relativesa with colon carcinoma, or a single first-degree relative with colon carcinoma or adenomatous polyps diagnosed at an age < 60 years, One second-degree or any third-degree relativeb,c with colorectal carcinoma Gene carrier or at risk for familial adenomatous polyposisd Gene carrier or at risk for HNPCC

Screening Recommendation Same as average risk but starting at age 40 years

Colonoscopy every five years, beginning at age 40 years or 10 years younger than the earliest diagnosis in the family, whichever comes first Same as average risk

Sigmoidoscopy annually, beginning at age 10–12 yearse Colonoscopy, every 1–2 years, beginning at age 20–25 years or 10 years younger than the earliest case in the family, whichever comes first

a

First-degree relatives include parents, siblings, and children. Second-degree relatives include grandparents, aunts, and uncles. c Third-degree relatives include great-grandparents and cousins. d Includes the subcategories of familial adenomatous polyposis, Gardner syndrome, some Turcot syndrome families, and AAPC. e In AAPC, colonoscopy should be used instead of sigmoidoscopy because of the preponderance of proximal colonic adenomas. Colonoscopy screening in AAPC should probably begin in the late teens or early 20s. Abbreviations: HNPCC, hereditary nonpolyposis colon cancer; AAPC, attenuated adenomatous polyposis coli. Source: From Ref. 12. b

The combination of the PSA test and digital rectal examination provides reliable early detection of prostatic carcinoma. These should be performed annually (43). For a 55-year-old man presenting to a physician’s office, it is prudent to obtain a serum PSA concentration and perform a digital rectal examination. If both are normal, the patient should be followed with an annual TABLE 3 & Familial Risk

Familial Setting General population risk in the U.S. One first-degree relative with colon carcinomaa Two first-degree relatives with colon carcinomaa First-degree relative with colon carcinoma diagnosed at 50 years One second- or third-degree relative with colon carcinomab,c Two second-degree relatives with colon carcinomab One first-degree relative with an adenomatous polypa a

Approximate Lifetime Risk of Colon Carcinoma 6% 2–3-fold increased 3–4-fold increased 3–4-fold increased About 1.5-fold increased About 2–3-fold increased About 2-fold increased

First-degree relatives include parents, siblings, and children. Second-degree relatives include grandparents, aunts, and uncles. c Third-degree relatives include great-grandparents and cousins. Source: From Ref. 12. b

evaluation. If the results of the digital rectal examination are unremarkable but the serum PSA level is mildly elevated (range, 4.1–10.0 mg/L), transrectal ultrasonography should be performed (44). Cytologic examination of urine for exfoliated cells should be performed annually, if indicated. In women, a second carcinoma is more likely to occur in the colon, rectum, cervix, uterus, or ovary (44). Thus a complete large bowel examination should be performed every five years. Mammograpny, pelvic examination, and a Papanicolaou smear test should be performed as part of a routine annual checkup as appropriate.

& NEW SCREENING TESTS There are a couple emerging screening tests that are not yet ready for mass screening but have promising potential.

& COMPUTED TOMOGRAPHY COLONOGRAPHY (CTC, VIRTUAL COLONOSCOPY) At the time of this writing, computed tomography colonography (CTC) is not yet ready for a mass screening test for colorectal carcinoma. It is used as a backup for an incomplete colonoscopy, or for patients who are not suitable for a colonoscopy. However, the advances in technology, techniques, and clinical studies have progressed rapidly. It will be just a matter of time that CTC will become another option for colorectal carcinoma screening (see Chapter 3). & FECAL DNA TESTING Fecal DNA testing is based on the idea that, because carcinoma is a disease of mutations that occur as tissue evolves from normal to adenoma to carcinoma, those mutations should be detectable in stool (39). Preliminary reports that persons with advanced carcinoma have detectable DNA mutations in stool (45) provided the basis for a large study, using a panel of 21 mutations, in more than 4000 asymptomatic persons who received screening colonoscopy, fecal DNA testing, and FOBT with Hemoccult II (46). The DNA marker panel, including mutations in APC, K-ras, and p53, showed a sensitivity of 52% for colorectal carcinoma and specificity of 94% (40). Such stool-based testing is appealing because it is noninvasive, requires no special colonic preparation, and has the capability of detecting neoplasia throughout the entire length of the colon (47). Because the DNA alterations in colorectal carcinoma are heterogeneous, future assays will need to detect mutations in the number of genes in addition to what we have known at this time. The future of such an approach would seem promising if sensitivity could be increased by additional markers such as methylation and if cost could be reduced (39,47).

& WHEN TO STOP SCREENING There is no direct evidence as to when screening should stop, but indirect evidence supports stopping screening in people nearing the end of life. Polyps take at least 10 years to progress to carcinoma, and screening to detect polyps

CHAPTER 2: LARGE BOWEL CARCINOMA: SCREENING, SURVEILLANCE, AND FOLLOW-UP &

may not be in the patient’s best interest if he or she is not expected to live at least that long. Also, screening and diagnostic tests are, in general, less well tolerated by elderly people. Therefore there will come a time in most peoples’ lives when the rigors of screening and diagnostic evaluation of positive tests are no longer justified by the potential to prolong life. The age at which to stop screening depends on the judgment of the individual patient and his or her clinician, taking into account the lead time between screening and its benefits and the patient’s life expectancy (10).

& SURVEILLANCE Surveillance is the monitoring of people known to have colon or rectal disease.

45

polypectomy were randomized to undergo surveillance colonoscopy either one and three years or only three years after polypectomy. The two groups showed no difference in the proportion of detected adenomatous polyps with advanced pathology (3% in both groups) (48). This suggests that the first follow-up screening after polypectomy can be deferred for at least three years. The study also showed that if the results of the first surveillance colonoscopy are negative, subsequent examinations are highly unlikely to reveal further adenomatous polyps. There have been no reported studies of surveillance after polypectomy using barium enema and no reported studies comparing surveillance with barium enema versus colonoscopy (10). New evidence supports the concept that colonoscopic polypectomy reduces subsequent colorectal carcinoma incidence (12). A study of postpolypectomy surveillance demonstrated a 66% reduction in colorectal carcinoma incidence, similar to the previous report of the NPS (34). There is no direct evidence related to when to stop surveillance. As with screening, the age at which surveillance should stop depends on the judgment of patients and their clinicians, taking into account the patient’s medical history and comorbidity. The characteristics of the polyps removed and the results of follow-up examinations should also be taken into account.

& AFTER REMOVAL OF ADENOMATOUS POLYPS The main options for surveillance are colonoscopy and DCBE. The best evidence of the effectiveness of surveillance is from colonoscopy. In the National Polyp Study (NPS), a cohort of 1418 patients who had undergone complete colonoscopy and removal of one or more adenomatous polyps from the colon or rectum were followed up for an average of 5.9 years per patient with periodic colonoscopy. After adjusting for age, sex, and polyp size, rates of carcinoma were 76–90% lower than expected when compared with three reference groups (from published reports) who had not undergone surveillance. The study used reference groups as controls, with the assumption that patients undergoing polypectomy would have experienced the same incidence of carcinoma as the reference populations who have not undergone polypectomy (Fig. 4). The optional frequency of surveillance was also studied in the NPS. All patients who had undergone prior

& IN PEOPLE WITH INFLAMMATORY BOWEL DISEASE The primary goal of surveillance for colorectal carcinoma in patients with inflammatory bowel disease is to detect moderate to severe dysplasia and early carcinoma rather than polyps. Surveillance should be performed in patients with a history of the disease for about eight years, after which time the risk of carcinoma starts to appear. It is debatable whether surveillance alone is a reliable indicator of prophylactic proctocolectomy (see details in Chapter 26).

FIGURE 4 & The observed and expected incidences of colorectal carcinoma in a National Polyp Study cohort after having undergone colonoscopic polypectomy. Abbreviations: NPS, National Polyp Study; SEER, Surveillance, Epidemiology, and End Results (Program). Source: From Ref. 26.

& IN PEOPLE WITH URETEROSIGMOIDOSTOMY Ureterosigmoidostomy has been replaced for the most part by ileocystoplasty or the ileal conduit and therefore few patients have this condition. Recently it has been found that patients with ureterocystoplasty or ileal conduit have an equally high risk of developing carcinoma as patients with ureterosigmoidostomy. It takes an average 18 years (range, 5–29 years) for patients with ileocystoplasty or ileal conduit to develop malignancy. Most of the carcinomas or adenomas are at the anastomotic line. At one time in the experimental animal model (rats), it was believed that fecal matter was needed in the urinary stream to develop neoplasia. Later studies disputed this concept and the recent findings of malignancy in ileocystostomy or iteal conduit also confirm that urine alone can cause the malignancy. The pathogenesis is unclear but it appears in animal studies (rats) and the findings in humans that nitrosamines produced by gram negative bacteria are probably part of the mechanism in carcinogenesis (49). Patients with ureterosigmoidostomy should have a flexible sigmoidoscopy or colonoscopy. The examinations should begin on the 10th anniversary of the original operation and should be repeated annually (46). If a polyp is found at the anastomotic line, endoscopic removal is not advisable because most

46

& PART I: COLORECTAL DISORDERS

of them are situated at the site of the ureterosigmoidostomy (50,51). If possible, the patient should have an alternative of urinary diversion (51,52).

& FOLLOW-UP AFTER CURATIVE RESECTION Patients with a colorectal carcinoma should have a colonoscopy before the operation to rule out synchronous neoplasms. If the colon is obstructed preoperatively, colonoscopy should be performed approximately six months after operation. If this or a complete preoperative examination is normal, subsequent colonoscopy should be offered after three years; and then, if normal, every five years (12). The goals of follow-up are to detect local recurrences, distant metastasis, metachronous carcinomas and adenomas, and detection of other primary carcinomas. The rationale for postoperative follow-up of colorectal carcinoma is based on the assumption that detection of a recurrence in an asymptomatic patient indicates carcinoma at an earlier stage that can be more effectively treated than those detected at a more advanced symptomatic stage. In general, the investigation includes history and physical examination, endoscopy, carcinoembryonic antigen (CEA), liver function tests, ultrasonography and/or computed tomography (CT) scan, and chest x-ray films. Approximately 30–50% of patients who undergo curative resection for carcinoma of the colon and rectum have a recurrence (53). In the series by Sugarbaker et al. (54), 85% of patients who develop recurrence do so within 2.5 years, and all sites of recurrence develop at approximately the same period of time, with a median of 17 months. The series by Pihl et al. (55) revealed that 50% of liver metastases are clinically obvious within 21–22 months of treatment, compared with 22–34 months for pulmonary metastases. Recurrences within the abdominal cavity are noted in 84% of all recurrences; 15% have recurrences involving distant metastases only (54). Approximately 35% of patients have liver metastasis. Of these, the liver is the only site of metastasis in approximately 20%. Between 10% and 22% have pulmonary metastases and approximately 10% of these patients have metastasis isolated to the lungs (53).

& INVESTIGATION TO DETECT RECURRENCES AND METASTASES Careful history taking and physical examination provide an effective means of surveillance. Several prospective studies showed that patient symptoms and physical signs can provide the first indication of recurrent carcinoma in 21–48% of those with advanced disease (56,57). In a study by Beart et al. (58) 168 patients who had undergone colorectal resection for cure were followed at least every 15 weeks for up to four years. In 41 of 48 patients, symptoms developed before detection by physical examination or biochemical and radiologic investigations. The symptoms included coughing, abdominal and pelvic pain, change in bowel habit, rectal bleeding, and malaise. Physical examination is a frequent indicator of recurrence but is not as sensitive as the presence of symptoms. All patients with positive physical findings were found to have symptoms or a positive CEA test. Recurrent carcinoma that is symptomatic or that can

be detected on physical examination is likely to be advanced and not curable (56). Alkaline phosphatase seems to be the most sensitive liver function test (58). However, it also has a high false positive rate and, therefore, has little predictive value if used alone (56). Because chest x-ray films are noninvasive and inexpensive, most clinicians order them annually. Any patient with a questionable or positive lesion should undergo a chest CT scan. A CT scan of the abdomen is not performed routinely, because the results are somewhat unreliable. Its limitation in detecting early pelvic recurrence is compounded by the fact that postoperative artifacts in the pelvis may persist for as long as two years. However, a CT scan is invaluable for assessing symptomatic patients and for confirming the clinical impression of recurrences (59). A CT scan to detect liver metastases is more accurate, with 85–90% sensitivity; this is similar to ultrasonography and magnetic resonance imaging (MRI). When CT scan is doubtful, positron emission tomography (PET) can be helpful (60). A systemic review and meta-analysis of randomized trials suggests a survival benefit associated with performing CT (every 3–12 months) (61). However, others have reported that CT scan did not increase the number of curative hepatectomies (62). The track record for detecting curable pelvic or local relapse is also disappointing (63–65). When adjuvant radiation therapy is used, related posttreatment changes further complicate the interpretation of such imaging (66). CEA is a sensitive marker for identifying recurrence or metastases. CEA level is elevated most commonly in patients with hepatic metastases; 95% of these patients have increased plasma CEA levels. On the other hand, in patients with local abdominal or pelvic recurrences, 17–25% have normal CEA levels. Of the 70 patients with recurrences, 89% have elevated plasma CEA levels before the recurrences are detected by any other means. However, CEA does not always detect recurrences or metastases at a resectable stage. In the series reported by Beart et al. (58), 48 patients with elevated CEA were explored and only one patient was thought to have a potentially resectable lesion. On the other hand, in another series of 146 asymptomatic patients with elevated CEA, evidence of recurrence was found in 95%, and 58% of these were resectable for potential cure (67). In an uncontrolled study by Moertel et al. (68), using a patient population entered into a large National Surgical Adjuvant Trial, the patients were accrued from a variety of settings ranging from small community clinics to large universities and are probably representative of carcinoma in most practices nationwide. Every 12 weeks of the first year, every four months for the second year, and every six months thereafter, they have evaluations supplemented by hemoglobin and chemical analysis of the blood and chest radiographs. At 24 and 48 weeks and then annually, they underwent either proctoscopy, colon radiography, or total colonoscopy. Performance of the CEA test was optional, according to the usual practice of the responsible physician. Of a total of 1216 patients with resected carcinoma of the colon, 1017 (84%) had CEA monitoring. Among 417 monitored patients with recurrence, 59% had

CHAPTER 2: LARGE BOWEL CARCINOMA: SCREENING, SURVEILLANCE, AND FOLLOW-UP &

a preceding elevation of CEA concentration. Of 600 patients without recurrence, 16% showed a false positive result. CEA testing is most sensitive for detecting hepatic or retroperitoneal metastases and relatively insensitive for local, pulmonary, or peritoneal involvement. Surgical explorations were performed in 115 patients with CEA elevations, and 47 recurrences, usually hepatic, were resected with curative intent. On the other hand, 38 patients with normal CEA concentrations and 23 patients not monitored also underwent such resection—usually for a pulmonary or local recurrence. Of all CEA-monitored patients, 2.3% were alive and disease free more than one year after the salvage operation (2.9% of those with CEA elevations and 1.9% of those with no elevations). Of patients with no CEA monitoring, 2% were also alive and disease free more than one year after the salvage operation. The authors concluded that carcinoma cures attributable to CEA monitoring are, at best, infrequent. It is questionable whether this small gain justifies the substantial cost in dollars and the physical and emotional stress that this intervention may cause patients. Although intensive postoperative follow-up programs, including CEA evaluation, do identify recurrence earlier and thus result in higher resectable rates, they do not translate into a higher survival. Ohlsson et al. (63) of Sweden conducted a prospective randomized study investigating the value of intense postoperative follow-up compared with no follow-up in 107 patients followed up from 5.5–8.8 years. The study showed no difference in survival. A similar but larger prospective randomized study was conducted by Kjeldsen et al. (64) of Denmark. It consisted of 597 patients who also showed no improvement in overall survival or in disease-related survival. Richard and McLeod (53) critically examined the literature on postoperative follow-up of patients with colorectal carcinoma through a Medline search for articles published from 1966 through February 1996. The report included randomized controlled clinical trials, cohort studies, and descriptive studies. From the findings of their review, the authors believe that there is inconclusive evidence either to support or refute the value of follow-up surveillance programs to detect recurrent colorectal carcinoma. The authors pointed out many flaws in most of the studies, including small sample sizes, patient bias, and variations in follow-up protocols. The authors estimated that even if postoperative surveillance were effective, the survival benefit would not be > 10%. On the other hand, in a meta-analysis of 3283 patients in seven nonrandomized studies, Bruinvels et al. (69) found a 9% better overall survival rate in intensively followed patients, including more asymptomatic recurrences and more operations for recurrence. Surgeons and patients should understand the limitations of an intensive postoperative follow-up to detect recurrences and metastases. The follow-up regimen should be discussed and individualized to fit each patient’s needs.

& METACHRONOUS CARCINOMAS AND POLYPS The goal of surveillance after colorectal resection for carcinoma is to clear the adenomas if this has not been done preoperatively and to detect metachronous carcinoma. The incidence of metachronous carcinoma of the large

47

bowel is approximately 2%–4% within 3–20 years (70–72). Periodic examination of the large bowel with colonoscopy every three to five years provides the most accurate results. An alternative is to use flexible sigmoidoscopy combined with DCBE, possible CT colography in the near future. There are no controlled studies of the effectiveness of surveillance strategies in this situation. Available information suggests that the metachronous carcinomas have biological behavior that is not different from initial carcinomas except in increased frequency of occurrence.

& OTHER PRIMARY MALIGNANCIES The detection of other primary carcinomas is not important until the patient has achieved long-term survival. This fact has usually been ignored, unrealized, or forgotten. Enblad et al. (42) analyzed the occurrence of a second primary malignant disease in 38,166 patients with carcinoma of the colon and in 23,603 patients with carcinoma of the rectum, as reported to the Swedish Cancer Registry between 1960 and 1981. The overall relative risk (RR) of developing a second primary malignant disease was significantly increased both after carcinoma of the colon (women, RR ¼ 1.4; men, RR ¼ 1.3) and rectum (women, RR ¼ 1.4; men, RR ¼ 1.3). The increased risk of secondary primary diseases occurs in the stomach, small intestine, ovary, endometrium, cervix, breast, kidney, bladder, and prostate (Table 4).

& SUMMARY Despite the plethora of publications on the subject of follow-up strategies for patients who have undergone curative resection for colorectal carcinoma, there is no general agreement on the optimal follow-up program. The efficacy of aggressive and intensive surveillance remains TABLE 4 & The Relative Risk of Secondary Primary Malignancy Site of Second Malignancy Stomach Small intestine (carcinoid, adenocarcinoma) Kidney Bladder Prostate Ovary Endometrium Cervix Breast

Sex

Primary Lesions Primary Lesion of Colon (RR) of Rectum (RR)

M F M

1.3 1.0 10.2

1.0 1.3 3.8

F M F M F M F F F F

5.1 1.5 1.5 1.5 2.0 1.3 3.0 1.7 1.1 1.3

2.0 1.2 1.3 1.3 1.7 1.3 1.5 1.7 1.4 1.3

Abbreviation: RR, relative risk. Source: From Ref. 42.

48

& PART I: COLORECTAL DISORDERS

TABLE 5 & Summary of Guideline for Colorectal Surveillance after Primary Surgery with Curative Intent Test or Procedure

ESMO (73)

ASCO (74)

ASCRS (75)

History and physical examination

Every 3–6 months for first 3 Three times per year for History is taken. Clinical years; annually thereafter 2 years exam only in patients with suspicious symptoms

Fecal occult-blood test

Restricted to patients with suspicious symptoms Restricted to patients with suspicious symptoms Restricted to patients with suspicious symptoms

Liver function tests CEA

ONTARIO’S (76)

No routine testing

Not recommended

For Stage IIB and III, every 6 months for 3 years; annually for an additional 3 years Not addressed

No routine testing

No routine testing

Not addressed

Every 2–3 months for Stage II or III for 2 years or longer. Only in patients who can undergo liver resection No routine testing. Order when CEA is elevated or symptoms suggestive of pulmonary metastasis For patients who have not received pelvic irradiation. Direct imaging of rectum periodically. For patients who have received pelvic irradiation, direct imaging is not suggested Every 3–5 years

Three times per year for 2 years

For Stage IIB and III, every 6 months for 3 years; annually for additional 3 years

No routine testing

For Stage IIB and III, every 6 months for 3 years; annually for additional 3 years Not addressed

Chest X-ray

Restricted to patients with suspicious symptoms

Flexible sigmoidoscopy and endoscopic ultrasound

Every 6 months for 2 years in carcinoma distal sigmoid or rectum. Endorectal ultrasound

Colonoscopy and barium enema CT abdomen

Every 5 years Restricted to patients with suspicious symptoms

Not routine

Not routine

Ultrasonography of the abdomen

Of liver every year for 3 years

Not addressed

Not addressed

Periodic anastomotic examination

Every 3 years

Annually if there is a polyp; 3–5 years if no polyp For Stage IIB and III, every 6 months for 3 years; annually for additional 3 years For Stage IIB and III, every 6 months for 3 years; annually for additional 3 years

Abbreviations: ASCO; American Society Clinical Oncology; ASCRS, American Society of Colon and Rectal Surgeons; CEA, carcinoembryonic antigen; CT, computed tomography; ESMO, European Society for Medical Oncology.

a topic of intense debate. Some data support the contention that surveillance leads to increased rates of early detection and resection of recurrence but this has not necessarily always translated into improved survival. What constitutes the ideal or even preferred follow-up program remains controversial, and this is clearly reflected by the lack of uniformity among practicing clinicians. It is important that patients understand the limitations of the postoperative follow-up no matter how minimal or extensive it is. They should actively participate in the decision-making of the follow-up plan. Guidelines from some major societies are helpful on this issue. See Table 5 (73–76).

& REFERENCES 1. Jemal A, Murray T, Warel B, et al. Cancer Statistic 2005. CA Cancer J Clin 2005; 55:10–30. 2. Hermanek P, Sobin LH. Colorectal carcinoma. Hermanek P, Gaspodarowicz MK, Henson DE, Hutter RVP, Sobin LH, eds. Prognostic Factors in Cancer. HCII. International Union Against Cancer. Berlin: Springer-Verlag, 1995:64–79. 3. Jessup JM, Menck HR, Fremgen A, Winchester DP. Diagnosing colorectal carcinoma: Clinical and molecular approaches. CA Cancer J Clin 1997; 47:70–92. 4. Winawer SW, Zauber AG, Stewart E, O’Brien MJ. The natural history of colorectal cancer. Opportunities for intervention. Cancer 1991; 67:1143–1149.

5. Mandel JS, Bond JH, Church TR, et al. Reducing mortality from colorectal cancer by screening for fecal occult blood. N Engl J Med 1993; 328:1365–1371. 6. Winawer SJ, Flehinger BJ, Schottenfeld D, Miller DG. Screening for colorectal cancer with fecal occult blood testing and sigmoidoscopy. J Natl Cancer Inst 1993; 85:1311–1318. 7. Kronborg O, Fenger C, Olsen J, Jorgensen OD, Sondergaard O. Randomized study of screening for colorectal cancer with fecal-occult blood test. Lancet 1996; 348:1467–1471. 8. Hardcastle JD, Chamberlain JO, Robinson MHE, et al. Randomized controlled trial of fecal-occult-blood screening for colorectal cancer. Lancet 1996; 348: 1472–1477. 9. Kewenter J, Brevinge H, Engaras B, Haglind E, Ahren C. Results of screening, rescreening and follow up in a prospective randomized study for detection of colorectal cancer by fecal occult blood testing. Results for 68,308 subjects. Scand J Gastroenterol 1994; 29:468–473. 10. Winawer SJ, Fletcher H, Miller L, et al. Colorectal cancer screening: Clinical guidelines and rationale. Gastroenterol 1997; 112:594–642. 11. Bond JH. Screening for colorectal cancer: Confuting the refuters. Gastrointest Endosc 1997; 45:105–109. 12. Winawer S, Fletcher R, Rex D, et al. Colorectal cancer screening and surveillance: Clinical guidelines and rationale—update based on new evidence. Gastroenterol 2003; 124:544–560. 13. Seeff LC, Nadel MR, Klabunde CN, Thompson T, Shapiro JA, Vernon SW, Coates RJ. Patterns and predictors of colorectal cancer test use in the adult U.S. population. Cancer 2004; 100:2093–2103. 14. Surveillance, epidemiology and results (SEER) program, 1973–1992. Bethesda, Maryland: National Cancer Institute. 15. Potter JD, Slattery ML, Bostick RM, Gapstur SM. Colon cancer: A review of the epidemiology. Epidemiol Rev 1993; 15:499–545. 16. Imperiale TF, Wagner MS, Liney, Larkin GN, Rogge JD, Ransohoff DF. Results of screening colonoscopy among persons 40 to 49 years of age. N Engl J Mech 2002; 346:1781–1785.

CHAPTER 2: LARGE BOWEL CARCINOMA: SCREENING, SURVEILLANCE, AND FOLLOW-UP &

17. Smith RA, Cokkinides V, Eyre H. American cancer society guidelines for early detection of cancer, 2005 CA cancer J Clin 2005; 55:31–44. 18. Mandel JS, Church TR, Ederer F, Bond JH. Colorectal cancer mortality: effectiveness of biennial screening for fecal occult blood. J Nah Cancer Inst 1999; 91:434–437. 19. Mandel JS, Church TR, Bond JH, et al. The effect of fecal occult-blood screening on the incidence of colorectal cancer. N Engl J Med 2000; 343:1603–1607. 20. Pignone M, Campbell MK, Carr C, Phillipe C. Meta-analysis of dietary restriction during fecal occult blood testing. Eff Clin Pract 2001; 4:150–156. 21. Winower SJ, Stewart ET, Zauber AG, et al. For the National Polyp Study Work Group: A comparison of colonoscopy and double contrast barium enema for surveillance after polypectomy National Polyp Study Work Group. N Engl J Med 2000; 342:1766–1772. 22. Selby JV, Friedman GO, Quesenberry CP Jr, Weiss NS. A case-control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 1992; 326:653–657. 23. Newcomb PA, Norfleet RG, Storer BE, Surawicz T, Marcus PM. Screening sigmoidoscopy and colorectal cancer mortality. J Natl Cancer Inst 1992; 84: 1572–1575. 24. Muller AD, Sonnenberg A. Protection by endoscopy against death from colorectal cancer. Arch Intern Med 1995; 155:1741–1748. 25. Rex DK Cummings OW, Helper DJ, et al. Five year incidence of adenomatous adenomas after negative colonoscopy in asymptomatic average-risk persons. Gastroenterol 1996; 111:1178–1181. 26. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. The National Polyp Study Work Group. N Engl J Med 1993; 329:1977–1981. 27. Levin TR, Palitz A, Grossman S, et al. Predicting advanced proximal colonic neoplasia with screen sigmoidoscopy. JAMA 1999; 2818:1611–1617. 28. Lieberman DA, Weiss DG, Bond JH, Ahnen DJ, Garewal H, Chejfec G. Use of colonoscopy to screen symptomatic adults for colorectal cancer. Veterans Affair Cooperative Group. N Engl J Med 2000; 343:162–168. 29. Imperiale TF, Wagner DR, Lin CY, Larkin GN, Rogge JD, Ransohoff DF. Risk of advanced proximal neoplasms in asymptomatic adults according to the distal colorectal findings. N Engl J Med 2000; 343:169–174. 30. Farraye FA, Wallace M. Clinical significance of small polyps found during screening with flexible sigmoidoscopy. Gastrointest Endorc Clin NAM 2002; 12:41–51. 31. Thiis-Evensen E, Hoft GS, Sauer J, Langmark F, Majak BM, Vatn MH. Population-based surveillance by colonoscopy: Effect on the incidence of colorectal cancer. Scand J Gastroenternal 1999; 34:414–420. 32. Jorgensen OD, Kronborg O, Fenger C. A randomized study of screening for colorectal cancer using fecal occult blood testing: Results after 13 years and seven biennial screening rounds. Gut 2002; 50:29–32. 33. Selby JV, Friedman GD, Quesenberry CP Jr., Weiss NS. Effect of fecal occult blood testing on mortality from colorectal cancer. A case-control study. Ann Intern 1993; 118:1–6. 34. Citarda F, Tomaselli G, Capocaccia R, Barcherini S, Crespi M. Efficacy in standard clinical practice of colonoscopic polypectomy in reducing colorectal cancer incidence. Gut 2001; 48:812–815. 35. Hofstad B, Vatri M. Growth rate of colon polyps and cancer. Gastrointest Endosc Clin N Anm 1997; 7:345–363. 36. Rex DK, Chak A, Vasudeva R, et al. Prospective determination of distal colon findings in average-risk patients with proximal colon cancer. Gastrointest Endosc 1999; 49:727–730. 37. Rembacken BJ, Fujii T, Cairus A, et al. Flat and depressed colonic neoplasms: A prospective study of 1000 colonoscopies in the UK. Lancet 2000; 355:1211–1214. 38. Saitohy, Waxman I, West AB, et al. Prevalence and distinctive biologic features of flat colorectal adenomas in a North American population. Gastroentrol 2001; 120:1657–1665. 39. Ransohoff DF. Colon cancer screening in 2005: status and challenges. Gastroenterol 2005; 128:1685–1695. 40. Rex DK, Rahmani EY, Haseman JH, Lemmei GT, Kaster S, Buckley JS. Relative sensitivity of colonoscopy and barium enema for detection of colorectal cancer in clinical practice. Gastroenterol 1997; 112:17–23. 41. Fuchs CS, Giovannucci EL, Colditz GA, Hunter DJ, Speizer FE, Willet NC. A prospective study of family history and risk of colorectal cancer. N Engl J Med 1994; 331:1669–1674. 42. Enblad P, Adami HO, Glimelius B, Krusemo W, Pahlman L. The risk of subsequent primary malignant diseases after cancers of the colon and rectum. Cancer 1990; 65:2091–2100. 43. Barry MJ. Prostate-specific-antigen testing for early diagnosis of prostate cancer. N Engl J Med 2001; 344:1373–1377. 44. Oesterling JE. Prostate-specific antigen. Improving its ability to diagnose early prostate cancer. JAMA 1992; 264:2236–2238. 45. Ahlquist DA, Skoletsky JE, Boynton KA, et al. Colorectal cancer screening by detection of altered human DNA in stool: Feasibility of a multitarget assay panel. Gastroenterol 2000; 119:1219–1227.

49

46. Imperiale TF, Ransohoff DF, Itz Kowitz SH, Turnbull BA, Ross ME. Fecal DNA versus fecal occult blood for colorectal-cancer screening in an average-risk population. N Engl J Med 2004; 351:2704–2714. 47. Walsh JME, Terdiman JP. Colorectal cancer screening. Scientific Review. JAMA 2003; 289:1288–1296. 48. Winawer SJ, Zauber AG, O’Brien MJ, et al. Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps. N Engl J Med 1993; 328:901–906. 49. Filmer RB, Spencer JR. Malignancies in bladder augmentations and intestinal conduits. J Urol 1990; 143:671–678. 50. Woodhouse CRJ. Guidelines for monitoring of patients with ureterosigmoidostomy. Gut 2002; 51(suppl V):V15–V16. 51. Hurlstone DP, Wells JM, Bhala N, McAlindon ME. Ureterosigmoid anastomosis: risk of colorectal cancer and implications for colonoscopists. Gastrointest Endosc 2004; 59:248–254. 52. Azimuddin K, Khubchandani IT, Stasik JJ, Rosen L, Riether RD. Neoplasia after ureterosigmoidostomy. Dis Colon Rectum 1999; 42:1632–1638. 53. Richard CS, McLeod RS. Follow-up of patients after resection for colorectal cancer: A position paper of the Canadian Society of Surgical Oncology and the Canadian Society of Colon and Rectal Surgeons. Can J Surg 1997; 40:90–100. 54. Sugarbaker PH, Gianola FJ, Dwyer A, Neuman NR. A simplified plan for follow-up of patients with colon and rectal cancer supported by prospective studies of laboratory and radiologic test results. Surgery 1987; 102:79–87. 55. Pihl E, Hughes ESR, McDermott FT. Lung recurrence after curative surgery for colorectal cancer. Dis Colon Rectum 1987; 30:417–419. 56. Kelly CJ, Daly JM. Colorectal cancer. Principles of postoperative follow-up Cancer 1992; 70:1397–1408. 57. Goldberg RM, Fleming TR, Tangen CM, et al. Strategies for identifying resectable recurrence and success rates after resection. Ann Intern Med 1998; 129:27–35. 58. Beart RW Jr, Metzger PP, O’Connell MJ, Schutt AJ. Postoperative screening of patients with carcinoma of the colon. Dis Colon Rectum 1981; 24:585–588. 59. Zheng G, Eddleston B, Schofield PF, Johnson RJ, James RD. Computed tomographic scanning in rectal carcinoma. J R Soc Med 1984; 877:915–920. 60. Tzimas GN, Koumanis DJ, Meterissian S. Positron emission tomography and colorectal carcinoma: An update. JACS 2004; 198:645–652. 61. Renchan AG, Egger M, Saunders Mp, O’Dwyer ST. Impact on survival of intensive follow up after curative resection for colorectal cancer: Systematic review and meta-analysis of randomized trials. BMJ 2002; 384:813–818. 62. Schoemaker D, Black R, Giles L, Toouli J. Yearly colonoscopy, liver CT, and chest radiography do not influence 5-year survival of colorectal cancer patients. Gastroenterol 1998; 114:7–14. 63. Ohlsson B, Breland W, Ekberg H, Graffner H, Tranberg KG. Follow-up after curative surgery for colorectal carcinoma: Randomized comparison with no follow-up Dis Colon rectum 1995; 38:619–626. 64. Kjeldsen BJ, Kronborg O, Fenger C, Jorgensen OD. A prospective randomized study of follow-up after radical surgery for colorectal cancer. Br J Surg 1997; 84:666–669. 65. Pietra N, Sarli L, Costi R, Ouchemi C, Grattarola M, Peracchia A. Role of followup in management of local recurrences of colorectal cancer: A prospective randomized trial. Dis Colon Rectum 1998; 41:1127–1133. 66. Pfister DG, Benson III Al B, Somerfield MR. Surveillance strategies after curative treatment of colorectal cancer. N Engl J Med 2004; 350:2375–2382. 67. Martin EW Jr., Minton JP, Carey LC. CEA-directed second look surgery in the asymptomatic patient after primary resection of colorectal carcinoma. Ann Surg 1985; 202:310–317. 68. Moertel CG, Fleming TR, Macdonald JS, Haller DG, Laurie JA, Tangen C. An evaluation of the carcinoembryonic antigen (CEA) test for monitoring patients with resected colon cancer. JAMA 1993; 270:943–947. 69. Bruinvels DJ, Stiggelbous AM, Kievit J, van Houwelingen HC, Habberna JDF, van de Velde CJH. Follow-up of patients with colorectal cancer. A metaanalysis. Ann Surg 1994; 219:174–182. 70. Tornqvist A, Ekelund G, Leandoer, L. Early diagnosis of metachronous colorectal carcinoma. Aust N Z J Surg 1981; 51:442–445. 71. Luchtefeld MA, Ross DS, Zander JD, Folse JR. Late development of metachronous colorectal cancer. Dis Colon Rectum 1987; 30:180–184. 72. Herald RJ, Bussey HJR. Clinical experiences at St. Mark’s Hospital with multiple synchronous cancers of the colon and rectum. Dis Colon Rectum 1975; 18:6–10. 73. ESMO minimum clinical recommendations for diagnosis, adjuvant treatment and follow-up of colon cancer. Ann Oncol 2001; 12:1053–1054. 74. Benson III Al B, Desch CE, Flynn PJ, et al. 2000 update of American Society of Clinical Oncology Colorectal Cancer Surveillance guidelines. J Clin Onc 2000; 18:3586–3588. 75. The Standards Practice Task Force. The American Society of Colon and Rectal Surgeons. Practice parameters for the surveillance and follow-up of patients with colon and rectal cancer. Dis Colon Rectum 2004; 47:807–817. 76. Figueredo A, Rumble RB, Maroun J, et al. Follow-up of patients with curatively resected colorectal cancer: a practice guideline. BMC Cancer 2003; 3:26–38.

3

Malignant Neoplasms of the Colon Philip H. Gordon

& CLASSIFICATION, 52

& Diagnosis, 97

& ADENOCARCINOMA, 52

& Investigations, 98

Occult Blood Testing, 98 Endoscopy, 98 Radiology, 98 Radioimmunodetection, 101 Cytology, 102 Blood Markers, 102

& Incidence, Prevalence, and Trends, 52 & Epidemiology, 53

Age, 53 Sex, 53 Family History, 53 Site, 54 Geographic Distribution, 54 Race and Religion, 54 Occupation, 54

& Treatment, 104

Curative Resection, 104 Adjuvant Therapy, 117 Radiotherapy, 117 Chemotherapy, 118 Immunotherapy, 123

& Etiology and Pathogenesis, 55

Polyp-Cancer Sequence, 55 Inflammatory Bowel Disease, 55 Genetics, 55 Dietary Factors, 72 Irradiation, 77 Ureteric Implantation, 77 Cholecystectomy, 78 Diverticular Disease, 78 Activity and Exercise, 78 Other Factors, 79 Juvenile vs. Adult Carcinoma, 80 Prospects for Prevention, 81

Complicated Carcinomas, 123 Obstruction, 123 Perforation, 128 Bleeding, 128 Obstructive Colitis, 128 Invasion of Adjacent Viscera, 129 Urinary Tract Involvement by Colorectal Carcinoma, 131 Unresectable Carcinoma, 133 Palliative Resection, 133 Synchronous Carcinomas, 134 Synchronous Polyps and Carcinoma, 134 Metachronous Carcinoma, 134 Treatment of Metastatic Disease, 135 Liver, 135 Simultaneous Colorectal and Hepatic Resection, 137 Lung, 142 Liver and Lung, 144 Ovary, 145 Bone, 145 Brain, 145 Peritoneal Carcinomatosis, 146 Other Metastatic Disease, 147 Carcinoma in Young Patients, 147

& Pathology, 82

Macroscopic Appearance, 82 Microscopic Appearance, 83 Depressed Carcinoma, 85 Sentinel Lymph Node Mapping, 87 Modes of Spread, 88 Site of Spread, 89 Staging, 89 Biology of Growth, 92 & Clinical Features, 94

Symptoms, 94 General and Abdominal Examinations, 95 Digital Rectal Examination, 95 Extraintestinal Manifestations, 95 Synchronous Carcinomas, 95 Associated Polyps, 95 Other Associated Malignancies, 96

& Postoperative Complications, 147 & Results, 148 & Prognostic Discriminants, 152

Clinical Features, 152 Pathologic Features, 160 Biochemical and Special Investigations, 164

& Complications, 96

Obstruction, 96 Perforation, 97 Bleeding, 97 Unusual Infections Associated with Colorectal Carcinoma, 97

51

52

& PART I: COLORECTAL DISORDERS

& Recurrent Disease, 166

Follow-Up, 166 Incidence, 166 Contributing Factors, 166 Patterns, 167 Clinical Features, 168 Investigations, 168 Role of Carcinoembryonic Antigen, 169 Treatment, 170 Results of Reoperation, 175 Intestinal Obstruction Due to Recurrent Carcinoma, 175 & Colorectal Carcinoma Complicating Pregnancy, 176 & Ovarian Carcinoma Involving the Colon, 177 & Malakoplakia and Colorectal Carcinoma, 177 & OTHER MALIGNANT LESIONS, 177 & Carcinoid, 177

Incidence, 178 Clinical Features, 178 Pathology, 178 Imaging Procedures, 179 Chemical Activity, 179 Treatment, 179 Results, 180 & Lymphoma, 180

Incidence, 180 Pathology, 180 Clinical Features, 181 Treatment, 182 Results, 182 & Sarcoma, 182 & Squamous Cell Carcinoma, 183 & Adenosquamous Carcinoma, 184 & Plasmacytoma, 184 & Melanoma, 184 & Leukemic Infiltration, 185 & Neuroendocrine Lesions of the Colorectum, 185 & Medullary Carcinoma of the Colon, 185 & Carcinosarcoma, 186 & Schwannoma, 186 & Angiosarcoma, 186 & Choriocarcinoma, 186 & Metastases from Other Sources, 187 & References, 187

CLASSIFICATION Malignancies of the large intestine assume a major importance because of their frequency in the general population. The types of malignancies that may occur can be classified as follows: 1. 2. 3. 4. 5. 6.

Adenocarcinoma Carcinoid Lymphoma Sarcoma Squamous cell carcinoma Plasmacytoma

DiSario et al. (1) reported the review of a populationbased registry with complete ascertainment. There were 7422 colorectal carcinomas—4900 (66%) colonic and 2522 (34%) rectal. The breakdown of the 222 (3%) nonadenocarcinoma malignancies was squamous, 75 (34%); malignant carcinoids, 74 (33%); transitional cell-like, 37 (17%); lymphomas, 25 (11%); sarcomas, nine (4%); and melanomas, two (0.9%). Although not stated, the transitional cell-like (including cloacogenic and basaloid carcinomas) and the squamous cell carcinomas (including epidermoid carcinomas) were almost certainly of anal canal origins.

ADENOCARCINOMA

& INCIDENCE, PREVALENCE, AND TRENDS Colorectal carcinoma is the fourth most common internal malignancy; it is second only to carcinoma of the lung as a cause of carcinoma death (2). It was estimated that in 2004, there would be 146,940 new cases of colorectal carcinoma in the United States (estimated to be 11% of all malignancies) and that 56,600 deaths would result from this disease (2). Colorectal carcinoma incidence rates have stabilized since the mid-1990s in males and females. For the same year, it was estimated that there would be 19,100 new cases of colorectal carcinoma in Canada and that 8300 individuals would die of the disease (3). There has been a slight increase in incidence in both men and women annually since 1997 (3). The mortality has continued to decline in both sexes but more so among women. Consensus is emerging internationally about the benefits of population-based screening for colorectal carcinoma. This is under consideration in Canada at both provincial and national levels. However, casual screening is already prevalent in Canada and may have contributed to the most recent increased incidence and decreased mortality rates. At birth, the probability of eventually developing a colorectal carcinoma in women is 5.5% in the United States and 6.5% in Canada and in men, 5.9% and 6.5%, respectively. The probability of dying of the disease in the United States is 2.3% and in Canada is 3.2% in women and in men 2.4% in the United States and 3.6% in Canada. In the United States, enormous data have been collected in the ongoing SEER program and the following information has been obtained from that database (4). In the last 30 years, the incidence of carcinoma of the colon and

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

rectum has been relatively stable with a 0.8% annual increase from 1975 to 1985, a 1.8% annual decrease from 1985 to 1995, a 1.2% annual increase from 1995 to 1998, and a 2.9% annual decrease from 1998 to 2001. In absolute terms, the age-adjusted incidence has dropped from 59.5 per 100,000 in 1975 to 51.8 per 100,000 in 2001, the greatest decline in the last few years of reporting. Mortality rates have shown a decline over this period with a 0.5% annual decrease from 1975 to 1984 and a 1.9% annual decrease from 1984 to 2001. The largest decrease has been noted with rectal carcinoma 1998 to 2001 when a 3.9% annual decline was reported. Age-adjusted mortality rates for colon and rectal carcinoma have dropped from 28.1 per 100,000 in 1975 to 20.0 per 100,000 in 2001. The 5-year relative survival rates have improved from 49.8% in 1974–1976 to 63.4% in 1995–2000. The stage distribution of cases during the 1995 to 2000–reporting period was localized (39%), regional (38%), distal (19%), and unstaged (5%) with the corresponding 5-year relative survival rates at 89.9%, 67.3%, 9.6%, and 35.2%, respectively, with an overall survival of 63.4%. Hayne et al. (5) examined the trends in colorectal carcinoma in England and Wales over the last 30 years. Agestandardized incidence, mortality, and survival rates for colorectal carcinoma based on data from the National Cancer Intelligence Centre at the Office for National Statistics were calculated and trends assessed. Between 1971 and 1997, the total number of cases of colorectal carcinoma increased by 42%. The site distribution of the colorectal carcinomas between 1971 and 1974 was rectum (38%), sigmoid (29%), cecum (15%), transverse colon and flexures (10%), ascending colon (5%), and descending colon (3%). Between 1971 and 1997, the direct age-standardized incidence increased by 20% in males and by 5% in females. The direct age-standardized mortality fell by 24% in males and by 37% in females. Age-standardized relative 5-year survival in adults improved from 22% to 27% for patients diagnosed during 1971 to 1975 to over 40% for those diagnosed during the period of 1991 to 1993. The 5-year survival has improved substantially but rates are still below those in comparable countries, elsewhere in Europe, and in the United States.

& EPIDEMIOLOGY An extensive and comprehensive review of the worldwide information on colon carcinoma was collated by Correa and Haenszel (6). Much of the following information was extracted from their excellent review.

& AGE Carcinoma of the large intestine is predominantly a disease of older patients, with the peak incidence being in the seventh decade. However, it must be borne in mind that the disease can occur at virtually any age and may be seen in patients in their twenties and thirties (7). It has been estimated that only 5% of colorectal carcinomas occur in patients who are younger than 40 years of age (8). & SEX It was estimated that during 2005 in the United States 71,820 men and 73,470 women would develop carcinoma

53

of the colon and rectum. In Canada, corresponding numbers were 10,600 and 9,000. In the United States it was estimated that 28,540 men and 27,750 women would die from the disease in 2005 (2). In Canada corresponding numbers were 4,500 and 3,900 (3). Carcinoma of the colon and rectum ranks third as a cause of death from carcinoma, surpassed by lung and equalled by prostate carcinoma in men and lung and breast carcinoma in women (2).

& FAMILY HISTORY There have been many reports that indicate an increased incidence of colorectal carcinoma in first-order relatives of patients who have suffered from the disease. In a prospective study of 32,085 men and 87,031 women, Fuchs et al. (9) found that the age-adjusted relative risk of colorectal carcinoma for men and women with affected first-degree relatives, when compared with those without a family history of the disease, was 1.72. The relative risk among study participants with two or more firstdegree relatives was 2.75. For participants younger than 45 years, who had one or more affected first-degree relatives, the relative risk was 5.37. Slattery et al. (10) assessed the risk of developing multiple primaries after a diagnosis of colon carcinoma and determined the impact that having a family history of carcinoma has on carcinoma risk. Data from the Utah Cancer Registry and the Utah Population Database were used. A cohort of 2236 patients with first primary colon carcinomas was observed for the subsequent development of additional carcinomas. The authors observed a greater than expected incidence of colon, rectal, and pancreatic carcinomas among the cohort. The standardized incidence ratios were 2.77, 2.26, and 2.38, respectively. Having a family history of colon or rectal carcinoma did not greatly influence risk of having a multiple primary carcinoma. However, there was a trend toward increased risk of pancreatic carcinoma (hazard ratio, 1.99), and bladder carcinoma (hazard ratio, 2.35) among patients with a family history of rectal carcinoma. The authors also observed that the risk of uterine carcinoma in a cohort was positively associated with a family history of uterine carcinoma, risk of breast carcinoma was positively associated with a family history of breast carcinoma, and risk of prostate carcinoma was positively associated with a family history of prostate carcinoma. To compare the risk in relatives of patients with colorectal carcinoma diagnosed at different ages, Hall et al. (11) studied two cohorts of patients, 65 diagnosed when they were younger than 45 years of age and 212 patients of all ages. The overall relative risk of colorectal carcinoma in first-degree relatives was 5.2 in the first group and 2.3 in the second group. The cumulative incidence of colorectal carcinoma for relatives of the young cohort rose steeply from 40 years, reaching 5% at 50 years and 10% at 70 years, compared with the older group, reaching 5% at 70 years and 10% at 80 years. St. John et al. (12) conducted a case-control family study of 7493 first-degree relatives and 1015 spouses of 523 case-control pairs to determine the relative risk of developing carcinoma. The authors found an odds ratio of 1.8 for one and 5.7 for two affected relatives. The risk

54

& PART I: COLORECTAL DISORDERS

TABLE 1 & Estimated Relative and Absolute Risk of Developing

Colorectal Carcinoma

Family History No family history One first-degree relative with colorectal carcinoma More than one first-degree relative with colorectal carcinoma One affected first-degree relative diagnosed with colorectal carcinoma before age 45 One first-degree relative with colorectal adenoma

Relative Risk

Absolute Risk by Age 79

1 2.3

4%a 9%b

4.3

16%b

3.9

15%b

2.0

8%b

a

Data from SEER database. Source: From Ref. 13. The absolute risks of colorectal carcinoma (CRC) for individuals with affected relatives was calculated using the relative risks for CRC and the absolute risk of CRC by age 79.

b

to parents and siblings was 2.1 times greater; 3.7 for patients diagnosed before 45 years and 1.8 times greater for patients diagnosed at the age of 45 years or older. The cumulative incidence was 11.1%, 7.3%, and 4.4% among relatives 55 years and older, between 45 and 54 years, and younger than 45 years, respectively. The most recent summary has shown the risk to be increased 2- to 4-fold (Table 1) (13). Furthermore people who have a first-degree relative with colorectal carcinoma are estimated to have an average onset of colorectal carcinoma about 10 years earlier than people with sporadic colorectal carcinoma (9).

& SITE The distribution of carcinoma in the various segments of the large bowel has been the subject of several detailed clinical studies. Each of these studies has shown that over the past 50 years there has been a gradual shift in the location of carcinomas from the rectum and left colon toward the right colon. Our initial account reported a dramatic shift from the rectum to the right colon (14). Our subsequent study has determined that the left-to-right progression has continued (15). Reasons for this shift to the right are not entirely clear. A review of patterns in different countries has revealed an increase in the incidence of colon carcinoma with a corresponding decrease in rectal carcinoma (6). Such findings imply that methods for the early detection and screening of large bowel carcinoma should be directed at the entire colon rather than being limited to the distal 25 cm of the large intestine. Qing et al. (16) in a comparison between American and Chinese patients found lesions in 36.3% of white patients versus 26.0% of Asian patients, while carcinomas of the rectum were found in 63.7% of white patients and 74% of Asian patients. The rightward shift has continued in recent decades in the United States (17) and Japan (18). & GEOGRAPHIC DISTRIBUTION There is a wide variation of the incidence of colorectal carcinoma in different countries. In general, countries of the Western world have the highest incidence of colorectal carcinoma, and these include Scotland, Luxembourg, Czechoslovakia, New Zealand, Denmark, and Hungary.

Countries with the lowest incidence include India, El Salvador, Kuwait, Martinique, Poland, and Mexico. The United States and Canada hold an intermediate position (2,6). In large countries extending over a wide range of latitudes, there may be considerable regional differences that mimic international variations (6). It has been suggested that low-risk populations have a relatively increased incidence of right-sided carcinomas while relatively high-risk communities have an increased risk of left-sided malignancies (6). There is an increased risk of large bowel carcinoma in urban populations when compared to rural populations. The incidence of colorectal carcinoma in Japanese Americans is higher than in Japanese individuals living in Japan. The children of these immigrants have an incidence approximating that of the general U.S. population. The effects of environmental exposure and food habits can be exemplified by a notable occurrence in Israel. Israelis who were born in Europe or North America run roughly 2.5 times, the risk of bowel carcinoma than those born in North Africa or Asia. After their arrival in Israel, the incidence becomes similar (6).

& RACE AND RELIGION Black Americans who once enjoyed a lower incidence of colorectal carcinoma than their white counterparts now suffer a similar incidence of the disease (6), but the 5-year survival rate for African-Americans is significantly lower than for whites (19). The risk of large bowel malignancy in American Indians is less than half that for whites in the United States. Individuals of Mexican extraction born in the United States also experience a lower risk for large bowel carcinoma. With respect religion, Jews in the United States have a higher incidence of colorectal carcinoma, while Mormons and Seventh Day Adventists have a lower rate than the general U.S. population (6). Ashkenazi Jews have a lifetime colorectal carcinoma risk of 9% to 15%, which differs strikingly from the 5% to 6% colorectal carcinoma risk for non-Ashkenazi members of the general western populations (20). The lower incidence in Mormons has been attributed to their prohibition of the use of tobacco and alcohol. Self-reported or perceived religiousness has been determined to be a protective factor in the development of colorectal carcinoma (a relative risk of 0.7) (21). & OCCUPATION Vobecky, Devroede, and Caro (22) observed an increased relative risk of three (i.e., a threefold increase) in the incidence of colorectal carcinoma in individuals working in factories that produce synthetic fibers. In the authors’ review of the literature, they found that other workers who were at greater risk of developing large bowel carcinoma included metallurgy workers handling chlorinated oil, manufacturers of transport equipment, weavers, firemen, those working with asbestos or coke by-products, and those working in copper smelters. de Verdier et al. (23) found elevated relative risks of colon carcinoma among male petrol station and/or automobile repair workers (2.3) and men exposed to asbestos (1.8), while elevated relative risks of rectal carcinoma were found among men exposed to soot (2.2), asbestos (2.2), cutting fluids and/or oils (2.1), and combustion gases from coke, coal, and/or wood (1.9).

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

A meta-analysis by Homa et al. (24) suggested that exposure to amphibole asbestos may be associated with colorectal carcinoma but exposure to serpentine asbestos is not. Another study failed to find an association between asbestos exposure and carcinoma of the colon and rectum (25). Cumulative exposure to organic solvents, dyes, or abrasives also may contribute to an increased risk of colorectal carcinoma (26,27). Workers involved in the manufacture of polypropylene also exhibit an increased incidence of colorectal carnicoma (28), but this risk was more recently reported not to exist (29). Workers with intense exposure to ethyl acrylate and methyl methacrylate for 3 years have an increased risk of colon carcinoma two decades later (30).

& ETIOLOGY AND PATHOGENESIS As with other malignancies, neither the etiology nor the pathogenesis of carcinoma of the colon is known. A number of factors have been considered important in its causation, and certain clinical conditions are considered precursors of carcinoma and will be detailed here.

& POLYP-CANCER SEQUENCE Considerable evidence has accumulated to suggest that most, if not all, carcinomas develop from a precursor polyp, a situation known as the polyp-cancer sequence. This sequence is described in detail in Chapter 22. & INFLAMMATORY BOWEL DISEASE Although colorectal carcinoma, complicating ulcerative colitis and Crohn’s disease, only accounts for 1% to 2% of all cases of colorectal carcinoma in the general population, it is considered a serious complication of the disease and accounts for approximately 15% of all deaths in inflammatory bowel disease patients. Patients with universal ulcerative colitis, having a more severe inflammation burden and risk of the dysplasia-carcinoma cascade especially those who have had the condition for more than 10 years and those patients who experienced onset in childhood, without doubt are at increased risk of developing carcinoma of the colon or rectum. Lennard-Jones et al. (31) reported that the incidence of colorectal carcinoma (in 22 patients among 401 patients with extensive ulcerative colitis followed over 22 years) was 3%, 5%, and 13% at 15 years, 20 years, and 25 years, respectively. For the 17 patients developing colorectal carcinoma during supervised surveillance (344 patients) Dukes’ staging was A or B in 12 patients. In half the carcinoma patients under surveillance, dysplasia signaled the associated carcinoma found only after colectomy in the operative specimens. Others have confirmed the increased risk (32). Of 3117 patients with ulcerative colitis followed for up to 60 years through the Swedish Cancer Registry, the relative risk of colorectal carcinoma was 5.7 (nonsignificant for proctitis), 2.8 for left-sided disease, and 14.8 for pancolitis (33). Recent figures suggest that the risk of colon carcinoma for people with inflammatory bowel disease increases by 0.5% to 1.0% yearly, 8 to 10 years after diagnosis (34). Considering the chronic nature of the disease, it is remarkable that there is such a low incidence of colorectal

55

carcinoma in some of the population-based studies, and possible explanations have to be investigated. One possible carcinoma protective factor could be treatment with 5-aminosalicylic acid preparations (5-ASAs) (34). Adenocarcinoma of the small bowel is extremely rare, compared with adenocarcinoma of the large bowel. Although only few small bowel carcinomas have been reported at sites of involvement with Crohn’s disease, the number was significantly increased in relation to the expected number (34). The incidence of colorectal carcinoma in patients with Crohn’s disease has been reported as being four to 20 times greater than the general population (32). In a study of 1656 patients with Crohn’s disease, Ekbom et al. (35) indicated relative risks for colon carcinoma of 3.2 (in Crohn’s ileocolitis) to 5.6 (in Crohn’s colitis only). With the onset of any Crohn’s colitis before the patient was 30 years of age, the relative risk was 20.9, but only 2.2 when diagnosed after age 30.

& GENETICS In the last decade and a half, there has been an explosive increase in knowledge about the molecular biology of carcinoma. Publications are legion but often difficult to understand. This not withstanding, the past decade has been witness to unprecedented progress in the comprehension of the basic mechanisms involved in the genesis of colorectal carcinoma. In his outstanding review of the molecular biology of colorectal carcinoma, Allen (36) attempted to make the subject understandable to the clinician, and the following dissertation draws heavily from that review. Molecular Biology The codes that control production of protein enzymes and form the basic information needed for life itself are found within the cell nucleus as long strands of deoxyribonucleic acid (DNA) molecules that are, in turn, composed of four nucleotides: adenine (A), guanine (G), thymine (T), and cytosine (C). Under normal circumstances, adenine only pairs with thymine (A:T), and guanine pairs with cytosine (G:C). As a result of base pairing, cellular DNA forms the familiar stepladder configuration that is twisted into a double helix and supercoiled into microscopically visible structures called chromosomes. Long DNA sequences are subdivided into smaller segments called genes, each of which contains the information needed for a single protein. Genes are composed of hundreds or thousands of nucleotides. In humans, the entire genetic code, termed the genome, is composed of approximately 3 billion nucleotides organized into approximately 100,000 genes contained within 23 pairs of chromosomes (total, 46). One chromosome in each pair is inherited from the mother and one from the father. Thus each gene has another similar (but not identical) gene called an allele on the complementary chromosome. Genes can act in a dominant or recessive fashion. For dominant genes, one allele assumes the responsibility for producing the protein and the other allele remains dormant. The sequence of nucleotides within cellular chromosomes is reproduced faithfully and is passed down from generation to generation during cell division. A ‘‘normal’’ rate of mutation is estimated to be one mistake in every

56

& PART I: COLORECTAL DISORDERS

10 billion base pairs copied. To correct replication errors, a repair mechanism is dependent on genes called mismatch repair genes. Malignant transformation appears to result from the accumulation of mutations within genes that are critical to cell growth and differentiation caused either by an increase in the mutational rate or because the DNA repair process is compromised. A carcinoma is the end result of four to 12 genetic changes that convey a growth advantage to the mutated cells. During the initiation phase, there is an increase in the mutational rate of DNA. Mutations in some genes become incorporated into an individual’s genome and are passed from generation to generation. These ‘‘germline mutations’’ may occur in genes related to carcinoma and, as a result, cause hereditary carcinoma. Other mutations, termed ‘‘somatic,’’ cause a sporadic carcinoma. Knudson (37) proposed that inherited carcinomas arise in individuals with germline mutations of one allele of a recessively acting carcinoma gene, after which only one additional somatic alteration is needed to inactivate the gene and initiate carcinogenesis. Sporadic carcinomas require two somatic mutations (or allelic loss).

Mechanisms of Gene Action Three major categories of genes have been implicated in carcinoma development: (1) oncogenes, (2) tumor suppressor genes, and (3) mismatch repair genes (Table 2). When a proto-oncogene (a normal human growth-related gene) becomes abnormally activated, it drives the cell through the cell cycle facilitating clonal proliferation and is known as an oncogene. Oncogenes act in a dominant fashion because alteration of only one allele is necessary to produce a cellular effect. Oncogenes, however, do not tell the entire story, because only 20% of human carcinomas carry oncogene alterations. Other genes called tumor suppressor genes can halt the cell cycle even when oncogenes are altered. Tumor suppressor genes act in a recessive manner and promote carcinoma only when they are inactivated by allelic loss or mutations in both alleles. If cells cannot repair DNA damage, tumor suppressor genes such as p53 drive the cell into a suicide mode called apoptosis. A tumor suppressor gene critical to colorectal carcinoma was described on TABLE 2 & Genes Known to Be Involved in Development of

Colorectal Carcinoma Type

Name

Oncogen Tumor suppressor gene

K-ras APC DCC p53 MCC TGF-b-RII hMLH1 hMSH2 hPMS1 hPMS2 hMSH6 hMSH3 Fat acetylation p450 genes, etc.

Mismatch repair gene

Others (currently of theoretical importance only) Source: From Refs. 13,36.

Chromosome 12 5 18 17 5 3 3 2 2 7 2 5 Many

chromosome 5—the adenomatous polyposis coli (APC) gene. It was found to contain an inherited mutation causing truncation of the protein product. Somatic mutations of APC are found early during the neoplastic process in most polyps and carcinomas. The latest genes found to be related to carcinogenesis are called mismatch repair genes, which are needed for cells to repair DNA replication errors and spontaneous base pair loss. The six DNA mismatch repair genes found in humans to date are hMSH2 (chromosome 2p16), hMLH1 (chromosome 3p21), hPMS1 (chromosome 2q31–33), hPMS2 (chromosome 7q11), hMSH6 (chromosome 2p16), and hMSH3 (chromosome 5q11.2-q13.2). When both copies of these genes are inactivated, DNA mismatch repair is defective, and the cell exhibits an increased frequency of mistakes in DNA replication, thereby accelerating the progression to oncogenesis. The first four genes are regarded to contribute to hereditary nonpolyposis colorectal carcinoma (HNPCC) in 31%, 33%, 2%, and 4%, respectively (38). The founder mutation MSH2
1906G > C is also considered an important cause of HNPCC in the Ashkenazi Jewish population (39). This pathogenic mutation accounting for 2% to 3% of colorectal carcinoma in those whose age at diagnosis is less than 60 years is highly penetrant and accounts for approximately one-third of HNPCC in Ashkenazi Jewish families that fulfill the Amsterdam criteria. This founder mutation MSH2
1906 was found in 8% of 1342 individuals (0.6%) of those of Ashkenazi descent with colorectal carcinoma. A subsequent study (40) sought to characterize the proportion of individuals of Ashkenazi heritage with very early-onset colon carcinoma (diagnosed at age 40 or younger) that could be attributed to MSH2
1906G > C detected the mutation in 3 of the 41 samples (7.14%) of patients who had colorectal carcinoma diagnosed at age 40 years or younger. The incidence is significantly greater than the 8 in 1345 (0.6%) observed for cases of colorectal carcinoma in Ashkenazi Jews not selected for age. These results suggest that consideration for testing for the MSH2
1906G > C mutation should be included in the evaluation of Ashkenazi Jewish individuals diagnosed with early onset of colon carcinoma. hMSH2 and hMLH1 accounted for 63% of kindreds meeting international diagnostic criteria (41). A recent review by Peltonaki (42) cited germline mutations in one of four major HNPCC-associated mismatch repair genes (MLH1, MSH2, MSH6, and PMS2) detected in up to 70% to 80% of such families. More than 400 different predisposing mismatch repair gene mutations are known with approximately 50% effecting MLH1, about 40% MSH2, and about 10% MSH6 (42). The share of PMS2 is less than 5%. The newly identified human mismatch repair gene MLH3 may account for a small percentage of HNPCC. A germline mutation in PMS1 was originally reported in an HNPCC-like family but there is presently no evidence of PMS1 as an HNPCC predisposition gene. The available data on two additional components of mismatch repair, exonuclease 1 (EXO1) and DNA polymerase, are too limited to allow any reliable assessment of their role in HNPCC predisposition (42).

Genetic Pathways to Colorectal Carcinoma Traditionally, carcinoma is seen as a three-step process of initiation, promotion, and progression. Colorectal carcinoma

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

57

FIGURE 1 & Molecular pathway to colorectal carcinoma. The two arrows associated with the box labeled "genetic factors" illustrate the difference between initiation of a sporadic vs. hereditary colon carcinoma. The arrow leading from the box labeled "genetic factors" to "initial mutation" illustrates inheritance of a germline mutation capable of initiating neoplasia. APC ¼ Adenomatous polyposis coli; LOH ¼ loss of heterozygosity; DCC ¼ deleted in colon carcinoma; MMR ¼ mismatch repair; RER ¼ replication error; TGF-b ¼ transforming growth factor-b. Source: From Ref. 36.

is a genetically heterogeneous disease, and a series of genetic events has been described in the evolution of colorectal carcinoma. The initiation stage (from the beginning to the first mutation) involves a complex (and poorly understood) interplay between environmental factors and host susceptibility (Fig. 1). Specific environmental factors are known to modify colorectal carcinoma (Fig. 2). For patients with hereditary colorectal carcinoma, the influence of environmental factors is small compared with the power of the underlying genetic mutations. Thus the risk of initiating colorectal carcinoma is substantially higher in hereditary conditions (100% in patients with polyposis syndromes and approximately 85% in HNPCC). A number of early reports tried to relate genetic importance to the etiology of colorectal carcinoma. Burt

FIGURE 2 & Environmental factors that may contribute to altering colonic mucosa to produce a field effect that precedes initiation of neoplasia. Colonic cells respond to each environmental factor based on the genotype of DNA loci associated with metabolic pathways that relate to the various dietary constituents. Genetic polymorphisms are thought to play a role in determining how individuals respond at a cellular level to various environmental factors. Source: From Ref. 36.

et al. (43) examined the inheritance of susceptibility to colonic polyps and carcinomas in a large pedigree with multiple cases of colorectal carcinoma. The authors’ analysis suggested that the observed excess of discrete adenomatous polyps and colorectal carcinoma was the result of an inherited autosomal-dominant gene for susceptibility rather than an inherited recessive gene for susceptibility. Solomon, Voss, and Hall (44) examined colorectal carcinomas for loss of alleles on chromosome 5. Using a special probe that maps to chromosome 5q, the authors demonstrated that at least 20% of carcinomas lose one of the alleles present in matched normal tissue. They suggested that becoming recessive for this gene may be a critical step in the progression of a relatively high proportion of colorectal carcinomas. No deletions were found in any other chromosome, which indicates that the loss from chromosome 5 is nonrandom. Law et al. (45) reported allelic losses in chromosomes 17 and 18 to be more frequent in colorectal carcinoma than losses on chromosome 5. It is now believed that the mutation that initiates colonic neoplasia is found in one of two gene loci. The 5q21 loci contains the APC gene, which is altered in more than 70% of all neoplastic lesions. Other polyps and carcinomas demonstrate micro-satellite instability, a hallmark of mismatch repair gene mutations. Depending upon which type of gene has been inactivated, one of two pathways to colorectal carcinoma is followed (see Fig. 1). In the first pathway, APC gene inactivation leads to a pathway termed loss of heterozygosity (LOH). Approximately 70% to 80% of colorectal carcinoma develops through the LOH pathway following inactivation of the APC gene. The genes involved in the LOH pathway include K-ras, DCC, and p53 in addition to APC (Fig. 3). Germline APC mutations initiate the neoplastic process in patients with familial adenomatous polyposis (FAP) and endow all colonic crypt stem cells with a high risk for clonal proliferation. A large body of evidence supports the concept of a multistep process that typically develops over decades and appears to require at least seven genetic events for

58

& PART I: COLORECTAL DISORDERS

FIGURE 3 & LOH pathway to colorectal carcinoma. Source: From Ref. 36.

completion. But even single altered genes can result in disease (e.g., FAP, HNPCC). In 1990 Fearon and Vogelstein (46) published the now-classic genetic model for colorectal carcinogenesis. The authors proposed a genetic series of events that corresponded to the apparent ordered sequence from a benign to a malignant lesion in histopathologic recognizable stages. They postulated that colorectal carcinoma arises´ as a result of mutational activation of oncogenes coupled with the mutational inactivation of tumor suppressor genes. Their original suggestion was that there must be mutations in at least four or five genes, but it is now believed to be at least seven for the formation of a carcinoma. Although the genetic alterations often occur according to a preferred sequence, the orderly sequence (see Fig. 2 and Fig. 3) rarely occurs in any individual carcinoma. It is the total accumulation of genetic alterations rather than their order that is responsible for determining the biologic properties of the carcinoma. The cascade of events described by Fearon and Vogelstein (46) begins with a loss or mutation the FAP gene on chromosome 5q, resulting in a change from normal epithelium to hyperproliferative epithelium. One of these hyperproliferating cells gives rise to a small adenoma in which the genome is hypomethylated. The next event involves activation of the K-ras oncogene on the chromosome 12p mutation to form the intermediate adenoma. Unlike oncogenes, tumor suppressor genes are expressed in a recessive manner. Therefore, both allelic copies must be lost or inactivated by point mutations for phenotypic expression to occur. Usually the DCC gene on chromosome 18q is next to be deactivated or lost, and results in the development of a late adenoma. The final genetic alteration found consistently in colorectal carcinoma is loss and/or mutation of the p53 tumor suppressor gene on chromosome 17p. The p53 gene

is altered in 50% of all human carcinomas and in 70% of colorectal carcinomas. Further genetic alterations are required for the development of metastases, subsequently believed to involve the loss of heterozygosity of the Nm23 gene (47). While there is no obligatory sequence of mutations in the pathway from normal mucosa through adenoma to carcinoma, there is clearly an association of certain types of mutations in specific oncogenes or tumor suppressor genes with early and late states of transformation. This multistep pathway can be observed in sporadic and inherited colorectal carcinoma. Many other genes, such as MCC, TGF-b, Rb, and Myc, have been implicated in the genesis of colorectal carcinoma. Further study of the molecular events will undoubtedly lead to a better understanding of the multistep carcinogenesis and the relative importance of each. Mismatch repair gene defects initiate an entirely different sequence of events known as the replication error (RER) pathway. These pathways lead to carcinomas that are biologically quite different. This second pathway to colorectal carcinoma is found in approximately 20% of carcinomas. The RER pathway is similar in both patients with HNPCC and those who develop a spontaneous RER carcinoma. Patients with HNPCC inherit a single defective allele of a mismatch repair gene and require an additional somatic mutation to inactivate the second allele. Spontaneous carcinomas develop after two somatic events inactivate the relevant gene. In either case, inactivation leads to a marked increase in replication errors. As errors accumulate in microsatellites, malfunction of genes that contain or are near affected microsatellites mav occur (Fig. 4). Aaltonen et al. (48) found the RER positive phenotype in 77% of colorectal carcinomas from HNPCC patients compared with only 13% of patients with sporadic carcinoma. How do DNA mismatch repair defects cause carcinoma? The mismatch repair gene defect increases the risk of malignant transformation of the cells, which may ultimately result from the disruption of one or several

FIGURE 4 & Genetic alterations often occur according to a preferred sequence. Source: From Ref. 36.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

anticarcinogenic functions of the mismatch repair genes. Peltomaki recently summarized these (42). First, malfunction of the mismatch repair system is associated with decreased genomic stability, which may manifest itself as highly elevated rates of subtle mutations (MSI) throughout the genome. Second, although mismatch repair–deficient cells typically have a diploid or near-diploid DNA content, loss of heterology-dependent suppression of recombination in these cells may promote gene conversion and expose tumor suppressor genes in analogy to loss of heterozygosity, or allow chromosomal translocations to occur. Furthermore, increased mutational inactivation of genes involved in DNA double-strand break repair may contribute to an elevated degree of chromosomal aberrations in mismatch repair–deficient cells. Third, besides anonymous microsatellite sequences, critical genes may be affected with mutations, conferring a growth advantage on the cells. Typical ‘‘target’’ genes include those involved in growth suppression, apoptosis, or signal transduction. Fourth, there is evidence that failed protection against endogenous or exogenous DNA damage and the ensuing persistence of mutagenic or premutagenic lesions may contribute to genomic instability/MSI in intestinal cells. There are examples of non-FAP, non-HNPCC hereditary colorectal carcinomas; a family not meeting the Amsterdam criteria with no deficiency of DNA mismatch repair in the malignancy of the proband was found to have a germline mutation of the TGFBR2 gene that encodes the type II TGF-b receptor; loss of the remaining normal allele was observed in the carcinoma (49). Thus, other biologic mechanisms may underlie non-FAP, non-HNPCC hereditary colorectal carcinoma. The transition from normal epithelium to adenoma to carcinoma is associated with acquired molecular events. A recent update on the genetic pathway to carcinoma has become available (13). At least five to seven major molecular alterations may occur when a normal epithelial cell progresses in a colonal fashion to carcinoma. There are at least two major pathways by which these molecular events can lead to colorectal carcinoma. About 85% of colorectal carcinomas are due to events that result in chromosomal instability and the remaining 15% are due to events that result in microsatellite instability. Key changes in chromosomal instability carcinomas include widespread alterations in chromosome number (aneuploidy) and detectable losses at the molecular level of portions of chromosome 5q, chromosome 18q, and chromosome 17p; and mutation of the KRAS oncogene. The important genes involved in these chromosome losses are APC (5q), DCC/MADH2/MADH4 (18q), and TP53 (17p), respectively, and chromosome losses are associated with instability at the molecular and chromosomal level. Among the earliest events in the colorectal carcinoma progression pathway is loss of the APC gene, which appears to be consistent with its important role in predisposing persons with germline mutations to colorectal neoplasms. Acquired or inherited mutations of DNA damage repair genes also play a role in predisposing colorectal epithelial cells to mutations. Not every carcinoma acquires every mutation, nor do mutations always occur in a specific order. The key characteristics of microsatellite instability carcinomas are largely intact chromosome complement, but acquisition of defects in DNA repair, such that mutations that may occur in important carcinoma-associated

59

genes are allowed to persist. These types of carcinomas are detectable at the molecular level by alterations in repeating units of DNA that occur normally throughout the genome, known as a DNA microsatellite. Mitotic instability of microsatellites is the hallmark of microsatellite instability carcinomas. Microsatellite instability, the hallmark of HNPCC, occurs in approximately 15% to 25% of sporadic colorectal carcinomas. According to international criteria, a high degree of microsatellite instability (MSI-H) is defined as instability at two or more of five loci or 30% to 40% of all microsatellite loci studied, whereas instability at fewer loci is referred to as MSI-low (MSI-L). Colorectal carcinomas with MSI-H encompass a group of malignancies with a predilection for the proximal colon, that have diploid DNA content, that are high grade, that are associated with female sex; and have better survival. These features distinguish MSI-H carcinomas from those without widespread MSI; that is, MSI-L or microsatellite-stable (MSS) carcinomas. A majority of MSI-H colon carcinomas are caused by inactivation of MLH1. Whereas the MSI-L subset of colon carcinomas is as equally prevalent as the MSI-H group, immunohistochemical and mutation studies have found no involvement in MLH1, MSH2, MSH6, or MSH3 in the former carcinomas. The clinicopathological features do not seem to distinguish this group from MSS colon carcinomas either.

Clinical Relevance of Basic Genetic Knowledge An effort by Allen (36) to reclassify the subtypes of colorectal carcinoma based on their molecular pathogenesis and inheritance pattern is depicted in Table 3. Prior to the definition of molecular pathways to colorectal carcinoma, proximal carcinomas were known to have normal cytogenetics, diploid DNA content, slower growth, less frequent metastasis, and a better prognosis compared to distal carcinomas. In addition, the frequency of extracolonic carcinomas had been found to be higher in both patients with colorectal carcinoma and their first-degree relatives when the index colon carcinoma was proximal. Most carcinomas that arise in the distal colon develop along the LOH pathway, while most proximal carcinomas are RER (Fig. 5). Clinical characteristics of LOH carcinomas indude a propensity for left-sided location (80%), aneuploidy, a polyp-to-carcinoma ratio of 20:1, and a total developmental period of 7 to 10 years. The exception to this observation is the rare adenomatous polyposis syndrome variant called hereditary flat adenoma syndrome (HFAS) or attenuated adenomatous polyposis coli (AAPC) syndrome associated with APC mutations. In this syndrome the inherited point mutation is found upstream from FAP mutations within exons 1 to 4. This subtle difference between FAP and AAPC mutations (sometimes within 10 base pairs of each other) results in dramatically different phenotypes. Patients with AAPC have few polyps (usually fewer than 10); the polyps are diminutive, flat, and located in the proximal colon; and a high proportion of polyps progress to carcinomas and they develop late compared with other hereditary neoplasias related to APC mutations. RER carcinomas tend to develop proximal to the splenic flexure ( > 70%), have a normal (diploid) DNA content, and carry a better prognosis compared to LOH carcinomas. Molecular analysis of resected carcinomas has led to recognition that LOH carcinomas carry a worse prognosis,

60

& PART I: COLORECTAL DISORDERS

TABLE 3 & Classification of Colorectal Carcinoma Based on Molecular Pathogenesis, Genetic Pattern, and Clinical Features

Genetic Pattern

Total Colorectal Carcinoma (%)

LOH Sporadic

35

Familial

25

Inherited (polyposis syndromes) FAP Gardner’s syndrome Turcot’s syndrome HFAS/AAPC

1 to 3

RER Sporadic

20

Familial

6

Inherited (HNPCC) Lynch syndrome I Lynch syndrome II Muir-Torre Turcot’s syndrome

Clinical Features Distal carcinomas (70%), aneuploid DNA, no family history of polyps or colorectal carcinoma, age of colorectal carcinoma older than 60 yr Distal carcinoma, aneuploid, family history of polyps or colorectal carcinoma in several relatives, age of colorectal carcinomas 50 to 60 yr More than 100 polyps, early onset of disease (polyps, 10 to 25 yr; colorectal carcinoma, 30 to 40 yr)(except HFAS/AAPC) Upper gastrointestinal polyps and carcinoma, retinal findings Desmoid neoplasms, bone abnormalities Medulloblastoma Small flat adenomas of proximal colon, usually fewer than 10, late age of onset (50 yr or older), gastric fundic polyps Proximal carcinomas (70%), diploid DNA, better prognosis than LOH carcinomas, age of colorectal carcinoma older than 60 yr Proximal carcinomas, diploid DNA, family history of colorectal carcinoma or polyps, age of colorectal carcinoma 50 to 60 yr

10 Colorectal carcinoma only, proximal carcinomas (70%), diploid, 40% have synchronous or metachronous colorectal carcinoma, age of colorectal carcinoma 40 to 45 yr Lynch I plus carcinoma of endometrium, ovaries, pancreas, stomach, larynx, urinary system, small bowel, bile ducts (vary with families) Lynch syndromes plus skin lesions Glioblastomas

DNA ¼ Deoxyribonucleic acid; FAP ¼ familial adenomatous polyposis; HFAS/AAPC ¼ hereditary flat adenoma syndrome/attenuated adenomatous polyposis coli; HNPCC ¼ hereditary nonpolyposis colorectal carcinoma; LOH ¼ loss of heterozygosity; RER ¼ replication error pathway. Source: From Ref. 36.

stage-for-stage, than RER carcinomas. To date, the most important molecular marker for prognosis appears to be the DCC gene locus on chromosome 18. Classification based on genetic patterns and specific syndromes reveals three major forms: sporadic, familial, or inherited. The difference appears to be the method by which the initiating mutation occurs. For sporadic carcinomas, there is a period during which environmental factors influence colon mucosa and eventually alter it so that clonal growth can occur. Familial colorectal carcinoma is defined by patients who have several family members with colon or rectal carcinoma but who do not fit a recognized inherited pattern. Familial colorectal carcinoma may include as many as 30% of patients with colorectal carcinoma when both carcinoma and polyps are included in the pedigree. Explanations for familial clustering could be either shared environment, shared genetics, or both.

Liu et al. (50) divided carcinomas with microsatellite instability into three classes. The first subset of colorectal carcinomas occurs in patients without a strong family history of colorectal carcinoma (sporadic cases). These account for 12% to 15% of the total colorectal carcinomas in the United States. The second class comprises colon, endometrial, and ovarian carcinomas that develop in patients with a family history of these forms of carcinoma (HNPCC). Virtually all carcinomas that develop in these patients exhibit microsatellite instability. The third class of RER carcinomas includes a variable fraction of several types of neoplasms, including those of the lung, breast, and pancreas. The magnitude and prevalence of the microsatellite alterations in this class are generally less pronounced than in RER colorectal carcinomas. The results of Liu et al. led to three major conclusions. First, carcinogenesis associated with MMR gene defects usually results from the inactivation of

FIGURE 5 & The distribution of neoplastic lesions within the colon is depicted by percentages within the colonic diagram. The ratio of LOH to RER carcinomas is given for the proximal and distal colon in addition to the locations of most lesions found in various hereditary syndromes. Source: From Ref. 36.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

both alleles of the relevant gene. Second, a significant fraction of sporadic RER carcinomas (four of seven in study) arises from mutations in genes other than the four that account for most HNPCC cases. Finally, most sporadic RER carcinomas are not associated with germline mutations of the known match repair genes. In their review, DeFrancisco and Grady (51) cited the causative genes of HNPCC in decreasing frequency of occurrence to include MLH1, MSH2, MSH6, MLH3, PMS1, PMS2, TGFBR2, and EXO1. Peltomaki summarized the phenotypic features associated with germline mutations manifesting a predisposition to HNPCC (42). MLH1 is mostly associated with typical HNPCC. Approximately 30% of mutations are of the missense type whose phenotypic manifestations may vary. MSH2 is also mostly associated with typical HNPCC. Extracolonic carcinomas may be more common than in MLH1 mutation carriers. It is a major gene underlying the Muir-Torre syndrome. MSH6 is associated with typical or atypical HNPCC. It is often characterized by late onset, frequent occurrence of endometrial carcinoma, distal location of colorectal carcinomas, and low degree of MSI in carcinomas. PMS2 is also associated with typical or atypical HNPCC. The penetrance of mutations may vary. It is a major gene underlying Turcot’s syndrome. MLH3 is seen mostly in atypical HNPCC. It may be characterized by distal location of colorectal carcinomas and variable degrees of MSI in carcinomas. EXO1 is mostly seen in atypical HNPCC. It may be associated with MSI in carcinomas. Lawes et al. (52) reviewed the world literature to determine the clinical importance and prognostic implications of microsatellite instability and sporadic carcinoma. In clinical studies, colorectal carcinomas demonstrating MSI respond better to chemotherapy while in-vitro studies using MSI positive cell lines show resistance to radiotherapy and chemotherapy. They concluded MSI may be a useful genetic marker in prognosis and could be an influential factor in deciding treatment options.

Clinical Syndromes For many years, FAP was considered the only hereditary variety of colon carcinoma. Now, three types of hereditary colorectal carcinomas are recognized: adenomatous polyposis syndromes (Chapter 22), HNPCC, and familial colorectal carcinoma, a group in which families exhibit aggregation of colorectal carcinoma, and/or adenomas, but with no identifiable hereditary syndrome. Approximately 1% to 3% of all colorectal carcinomas are due to hereditary adenomatous polyposis syndromes that include FAP, Gardner’s syndrome, Turcot’s syndrome, and HFAS/AAPC, all of which begin with germline mutations in the APC gene. HNPCC HNPCC is inherited as an autosomal dominant pattern with almost complete penetrance. It is estimated that from 0.5% to 6.0% of all colorectal carcinomas can be attributed to HNPCC (38,53–55). Based on clinical criteria only, Peltomaki (42) noted the estimated incidence of HNPCC varies between 0.5% and 13% of the total colorectal carcinoma burden. Strictly molecular approaches based on the identification of germline mutation carriers among newly diagnosed colorectal patients

61

whose carcinomas showed MSI have arrived at lower estimates—0.3% to 3% of the total colorectal carcinoma burden. HNPCC has four major subtypes: (i) Lynch type I (sitespecific nonpolyposis colorectal carcinoma); (ii) Lynch type II (formerly called the carcinoma family syndrome), in which carcinomas occur in the colon and related organs (endometrium, ovaries, stomach, pancreas, and proximal urinary tract, among others); (iii) Muir-Torre syndrome, associated with multiple benign and malignant skin neoplasms, sebaceous gland adenomas, carcinomas, and keratoacanthomas (56), and (iv) a variant of Turcot’s syndrome (brain neoplasms). HNPCC-related colon carcinomas begin with mutations in the mismatch repair genes and are RER positive. Clinical criteria were established to confirm the diagnosis of HNPCC at a meeting of the International Collaborative Group on HNPCC in Amsterdam in 1990 (54). The group agreed that minimum criteria should include (i) at least three relatives with histologically verified colorectal carcinoma, one of whom should be a first-degree relative of the other two (those with FAP should be excluded); (ii) at least two successive generations should be affected; and (iii) in one of the relatives, colorectal carcinoma should have been diagnosed when the patient was younger than 50 years of age. These criteria have not proved to be comprehensive, therefore, in 1999, the International Collaborative Group developed revised criteria and these are known as Amsterdam-II criteria. They are essentially the same except that there should be at least three relatives with HNPCCassociated carcinoma (colorectal, endometrial, small bowel, ureter, or renal pelvis). Even this extension of clinical features fails to identify some families with germline mismatch repair gene mutations so the Bethesda guidelines for testing of colorectal carcinomas for microsatellite instability have been developed. They include: 1. Individuals with carcinoma in families that meet the Amsterdam criteria. 2. Individuals with two HNPCC-related carcinomas, including synchronous and metachronous colorectal carcinomas or associated extracolonic malignancies (endometrial, ovarian, gastric, hepatobiliary or small bowel carcinoma, or transitional cell carcinoma of the renal pelvis or ureter). 3. Individuals with colorectal carcinoma and a first-degree relative with colorectal carcinoma and/or HNPCC-related extracolonic malignancy and/or a colorectal adenoma; one of the malignancies diagnosed at age less than 45 years, and the adenoma diagnosed at less than 40 years. 4. Individuals with colorectal or endometrial carcinoma diagnosed at age less than 45 years. 5. Individuals with right-sided colorectal carcinoma with an undifferentiated pattern (solid/cribriform) on histopathology diagnosed at less than 45 years. (Note: solid/cribriform defined as poorly differentiated or undifferentiated carcinoma composed of irregular, solid sheets of large eosinophilic cells and containing small gland-like spaces.) 6. Individuals with signet-ring-cell-type colorectal carcinoma diagnosed at age less than 45 years. (Note: composed of 50% signet-ring cells.) 7. Individuals with adenomas diagnosed at age less than 40 years.

62

& PART I: COLORECTAL DISORDERS

TABLE 4 & The Revised Bethesda Guidelines for Testing Colorectal

Carcinomas for Microsatellite Instability (MSI) Carcinomas from individuals should be tested for MSI in the following situations: 1. Colorectal carcinoma diagnosed in a patient who is less than 50 yr of age 2. Presence of synchronous, metachronous colorectal, or other HNPCC-associated malignanciesa regardless of age 3. Colorectal carcinoma with the MSI-Hb histologyc diagnosed in a patient who is less than 60 yr of aged 4. Colorectal carcinoma diagnosed in one or more first-degree relatives with an HNPCC-related neoplasm, with one of the neoplasms being diagnosed under age 50 yr 5. Colorectal carcinoma diagnosed in two or more first- or seconddegree relatives with HNPCC-related malignancies regardless of age a

HNPCC-related malignancies include colorectal, endometrial, stomach, ovarian, pancreas, ureter and renal pelvis, biliary tract, and brain (usually glioblastoma as seen in Turcot syndrome) neoplasms, sebaceous gland adenomas, and keratoacanthomas in Muir-Torre syndrome, and carcinoma of the small bowel. b MSI-H ¼ microsatellite instability-high in carcinomas refers to changes in two or more of the five National Cancer Institute–recommended panels of microsatellite markers. c Presence of carcinoma infiltrating lymphocytes, Crohn’s-like lymphocytic reaction, mucinous/signet-ring differentiation, or medullary growth pattern. d There was no consensus among the Workshop participants on whether to include the age criteria in guideline 3 above; participants voted to keep less than 60 yr of age in the guidelines.

In 2002 another workshop was held at the NCI in Bethesda and the following revised guidelines were created (Table 4) (57). It is estimated that 20% to 25% of population-based cases of colorectal carcinoma meet the Bethesda criteria. It is suggested that for all patients who meet these criteria, a search for MSI is indicated. To establish the most effective and efficient strategy for the detection of MSH2/MLH1 gene carriers in HNPCC, Pinol et al. (58) conducted a prospective multicenter nationwide study (the EPICOLON study) in 20 hospitals in the general community in Spain of 1222 patients with newly diagnosed colorectal carcinoma. Microsatellite instability testing and MSH2/MLH1 immunostaining were performed in all patients regardless of age, personal, or family history, and carcinoma characteristics. Patients whose carcinoma exhibited microsatellite instability and/or lack of protein expression underwent MSH2/MLH1 germline testing. The revised Bethesda guidelines were fulfilled by 23.5% of patients and 7.4% had a mismatch repair deficiency, with the carcinoma exhibiting either microsatellite instability or loss of protein expression. Germline testing identified mutations in 0.9% in either MSH2 or MLH1 genes. Strategies based on either microsatellite instability testing or immunostaining previous selection of patients according to the revised Bethesda guidelines were the most effective (sensitivity, 81.8% and 81.8%; specificity, 98.0% and 98.2%; and positive predictive value, 27.3% and 29.0%, respectively) to identify MSH2/MLH1 gene carriers. They concluded the revised Bethesda guidelines are the most discriminating set of clinical parameters (OR 33.3). In conclusion, MSI and IHC analysis using antibodies against MLH1, MSH2, PMS2, and MSH6 appeared to be equally effective for the identification of mutation carriers.

Despite its name, polyps are a feature of HNPCC, and a review of the literature revealed a polyp incidence in 8% to 17% of first-degree relatives during colonoscopic screening (38). Carriers of a MMR defect develop adenomas more frequently than controls. The adenomas identified in carriers are larger, and a significantly higher proportion showed histologic features that are associated with a high risk of malignant degeneration, such as a high degree of dysplasia and the presence of more extensive villous architecture (59). A relatively high proportion of patients develop colorectal carcinoma within 3 years after a clean colonoscopy and this suggests that the adenoma-carcinoma sequence is accelerated and that the progression from adenoma to carcinoma may take fewer than 3 years. Genetically speaking, HNPCC syndromes are dominantly inherited with nearly 100% penetrance reported by one group of investigators (53) but only 70% to 80% (i.e., 20% to 30% of individuals with a predisposing mutation may never develop carcinoma) by others (60) Asymptomatic gene carriers can pass the causative mutation to their children (60). The disease is heterogeneous. All firstdegree relatives of a patient with HNPCC have a 50% risk of carrying one of the deleterious genes. Adenomas in patients carrying HNPCC gene mutations display microsatellite instability, suggesting that mismatch repair defects are important early events in colorectal carcinogenesis. Carcinoma formation requires inactivation of both copies of a given mismatch repair gene, one copy by germline mutation and the other by somatic (acquired) mutation (60). Individuals with the Lynch syndrome differ from patients with sporadic colorectal carcinoma in several ways (38). They show (1) an autosomal dominant mode of inheritance, (2) a predominance of proximal colonic carcinoma (72% of first colon carcinomas were located in the right colon, and only 25% were found in the sigmoid colon and rectum), (3) an excess of multiple primary colonic carcinomas (18%), (4) an early age of onset (mean, 44 years), (5) a significantly improved survival rate compared with family members with distal colonic and rectal carcinomas when compared with right-sided lesions in an American College of Surgeons series (53% vs. 35% 5-year survivals), and (6) 24% developed metachronous colon carcinoma with a risk for the development of metachronous lesions in 10 years of 40% based on life-table methods. In addition to the above features, Lynch type II syndrome is characterized by a high frequency of other adenocarcinomas and the occasional occurrence of cutaneous manifestations in the form of sebaceous adenomas and carcinomas, epitheliomas, or keratoacanthomas. In a study by Mecklin and Jarvinen (61), of 40 HNPCC families with 315 affected members, a total of 472 malignancies was identified. They included colorectal (63%), endometrial (8%), gastric (6%), biliopancreatic (4%), and uroepithelial (2%) carcinomas. The relative risk of these carcinomas ranges from 3 to 25 times that of the general population (13). The risk of colorectal carcinoma increases 1.6% per year from age 25 to 75 years in patients with these mutations (56). In a detailed pedigree analysis of 40 families with HNPCC Aarnio et al. (62) identified 414 patients affected with carcinoma. The risk of any metachronous carcinoma reached 90% after treatment of colorectal carcinoma and 70% after endometrial carcinoma; the second malignancy

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

was most often a new colorectal carcinoma or endometrial carcinoma (62). Other sites of carcinoma include breast, pancreas, and possible lymphoma and leukemia. The review by De Francisco and Grady (51) found the four most common extracolonic carcinomas include (in descending order) endometrial, ovarian, gastric, and transitional cell carcinoma of the uro-epithelial tract (bladder, kidney, ureter). Women with HNPCC are at a 10-fold increased risk of endometrial carcinoma, which is usually diagnosed between the ages of 40 and 60 years, that is, 15 years earlier than the general population. The estimated cumulative risk at age 70 years is 40% to 50%. With MSH2 and MLH1 mutations the risk of colorectal carcinoma or endometrial carcinoma before age 50 years is 20% to 25% compared with 0.2% for the general population (56). Ovarian carcinoma is less common (incidence approximately 9%). Gastric carcinoma occurs in 5% to 20% of HNPCC families. The relative risk of gastric carcinoma was 19.3 in MSH2 mutation carriers compared with the general population. Transitional cell carcinoma of the uro-epithelial tract occurs in 1% of HNPCC patients. Only the carriers of MSH2 mutations appear to have a significantly increased risk of carcinoma in the urinary tract (relative risk 75.3). Overall, the relative risk of gastric carcinoma, ovarian carcinoma and carcinoma of the urinary tract has been shown to be higher in patients with mutations in the MSH2 as compared with MLH1. Women with MSH6 mutations appear to be more likely to develop endometrial carcinoma. Other HNPCC-associated extracolonic neoplasms include carcinomas of the small bowel, pancreas, hepatobiliary tree, brain, and skin. Muir-Torre’s syndrome, first described in 1967 refers to patients with HNPCC who also develop benign or malignant sebaceous skin neoplasms (sebaceous adenomas, carcinomas, squamous or basal cell), and multiple kerato-acanthomas. Muir-Torre’s syndrome usually arises from mutations in MSH2. The development of glioblastoma multiforme, in association with HNPCC, is called Turcot’s syndrome. Muir-Torre’s syndrome is also used to describe central nervous system neoplasms occurring in FAP, although these are usually medulloblastomas instead of glioblastomas. Approximately a third of patients with Turcot’s syndrome have mutations in one of the mismatch repair genes. HNPCC is associated with a 50% risk of a second carcinoma within 15 years of the initial carcinoma diagnosis (56) compared with 5% in the general population. Information regarding the cumulative risk of developing various HNPCC-associated malignancies has been summarized from several reports and tabulated in Table 5 (13,51,56,62,63). Plaschke et al. (64) analyzed the involvement and phenotypic manifestations of MSH6 germline mutations in families suspected of HNPCC. Patients were preselected among 706 families by microsatellite instability, immunohistochemistry, and/or exclusion of MLH1 or MSH2 mutations and were subjected to MSH6 mutation analysis. Clinical and molecular data of MSH6 mutation families were compared with data from families with MLH1 and MSH2 mutations. They identified 27 families with 24 different pathogenic MSH6 germline mutations, representing 3.8% of the total of the families, and 14.7% of all families with DNA mismatch repair gene mutations. The median age of onset of colorectal carcinoma in putative mutation

63

TABLE 5 & Malignancies Associated with Hereditary

Nonpolyposis Colorectal Carcinoma

Malignancy Colorectal Endometrial Ovarian Gastric Urothelial (bladder, kidney, and ureter) Pancreaticobiliary Brain Small bowel

Median Age of Onset (Yr)

Lifetime Risk (%)

General Population (%)a

40–45 45 47 54 60

78–82 39–61 9–12 13–19 4–10

5 1.5 1 <1 <1

54 43 49

2–18 1–4 1–4

<1 <1 <1

a

From Ref. 56.

carriers was 10 years higher for MSH6 (54 years) compared with MLH1 and MSH2 (44 years). Relative to other malignancies, colorectal carcinoma was less frequent in MSH6 families compared with MLH1 and MSH2 families. In contrast, the frequency of non–HNPCC-associated neoplasms was increased. Later age of disease onset and lower incidence of colorectal carcinoma may contribute to a lower proportion of identified MSH6 mutations in families suspected of HNPCC. However, in approximately half of these families, at least one patient developed colorectal or endometrial carcinoma in the fourth decade of life. Therefore, a surveillance program as stringent as that for families with MLH1 or MSH2 mutations is recommended. Approximately 60% of families that meet the Amsterdam-I criteria (HNPCC) have a hereditary abnormality in a DNA mismatch repair gene. Carcinoma incidence in Amsterdam criteria-I families with mismatch repair gene mutations is reported to be very high, but carcinoma incidence for individuals in Amsterdam criteria-I families with no evidence of a mismatch repair defect is unknown. Lindor et al. (65) conducted a study to determine if carcinoma risks in Amsterdam criteria-I families with no apparent deficiency in DNA mismatch repair are different from carcinoma risks in Amsterdam criteria-I families with DNA mismatch repair abnormalities. Identification of 161 Amsterdam criteria-I pedigrees from multiple populationand clinic-based sources in North America and Germany, were grouped into families with (group A) or without (group B) mismatch repair deficiency by testing. A total of 3422 relatives were included in the analysis. Group A families from both population- and clinic-based series showed increased incidence of HNPCC-related carcinomas. Group B families showed increased incidence only for colorectal carcinoma (SIR, 2.3) and to a lesser extent than group A (SIR, 6.1). Families who have fulfilled Amsterdam criteria-I but who have no evidence of DNA mismatch repair defect do not share the same carcinoma incidence as families with HNPCC with mismatch-repair deficiency. Relatives in such families have a lower incidence of colorectal carcinoma than those in families with HNPCCLynch syndrome and incidence may not be increased for other carcinomas. These families should not be described or counseled as having HNPCC. To facilitate distinguishing these entities, the designation of ‘‘familial colorectal cancer type X’’ is suggested to describe this type of familial aggregation of colorectal carcinoma.

64

& PART I: COLORECTAL DISORDERS

Jass (66) dispelled any thoughts that the morphogenetic pathway involved flat adenoma or de novo carcinoma in these cases. In a review of 131 carcinomas, none was small and superficial. Residual adenoma (contiguous with carcinoma) was present in three of three (100%) in situ carcinomas, eight of nine (89%) carcinomas involving only submucosa, four of 14 (29%) carcinomas limited to the muscle coat, and 13 of 105 (12%), carcinomas extending beyond the muscle coat. Lynch type II syndrome colon carcinomas show a significant increase in the proportion of mucinous, poorly differentiated carcinomas, and Crohn’s-like lymphoid reaction around the lesion as well as a higher rate of synchronous and metachronous carcinomas (38,67). Other investigators have also studied HNPCC. Sankila et al. (68) compared the survival rates of 175 patients with hMLH1-associated HNPCC with those of 14,000 patients with sporadic colorectal carcinoma diagnosed at < 65 years. The overall 5-year cumulative relative survival rate was 65% for patients with HNPCC and 44% for patients with sporadic colorectal carcinoma. In a series of 1042 Japanese patients who underwent resection for colorectal carcinoma, 3.7% were found to have HNPCC (69). The characteristic early stage of onset, rightsided predominance, and favorable survival were confirmed. Metachronous colorectal carcinoma developed significantly more often in cases with HNPCC (12.8% vs. 1.8%), and metachronous extracolonic malignancies also developed more often (10.2% vs. 3.5%). In cases with HNPCC, the mean interval between initial operation and the diagnosis of the second malignancy was 61 months (range, 12 to 153 months). These findings stress the importance of long-term follow-up. Rodriguez-Bigas et al. (70) reported the experience from the HNPCC Registry at Roswell Park in Buffalo, New York, which included 301 people in 40 families. In 284 of 301 people, 363 carcinomas were identified. Colorectal carcinomas alone were identified in 64% and in conjunction with extracolonic malignancies in another 11%. Extracolonic malignancies alone were identified in 25%. The median age at diagnosis of colorectal carcinoma was 48 years. Right-sided malignancies predominated (55%), synchronous and metachronous lesions were noted in 33%, and synchronous or metachronous adenomas were documented in 51% of those studied. Generational anticipation was also noted. Hampel et al. (71) assessed the frequency of the mismatch repair genes MLH1, MSH2, MSH6, and PMS2, in patients with a new diagnosis of colorectal carcinoma to identify patients with the Lynch syndrome. Genotyping of the carcinoma for microsatellite instability was the primary screening method. Among patients whose screening results were positive for microsatellite instability, they searched for germline mutations in the MLH1, MSH2, MSH6, and PMS2 genes with the use of immunohistochemical staining for mismatch repair proteins, genomic sequencing, and deletion studies. Family members of carriers of the mutations were counseled and those found to be at risk were offered mutation testing. Of 1066 patients enrolled in this study, 19.5% had microsatellite instability, and 2.2% of these patients had a mutation causing the Lynch syndrome. Among the 23 probands with the Lynch syndrome, 10 were more than 50 years of age and 5 did not

meet the Amsterdam criteria or the Bethesda guidelines for the diagnosis of HNPCC. In the families of 21 of the probands, 117 persons at risk were tested, and of these 52 had Lynch syndrome mutations. They concluded, routine molecular screening of patients with colorectal adenocarcinoma for the Lynch syndrome identified mutations in patients and their family members that otherwise would not have been detected. In most centers it would not seem feasible to test all patients with colorectal carcinoma. The characteristics of HNPCC mandate the development of surveillance programs. Screening recommendations have varied, but the recommendation by Lynch, Smyrk, and Lynch (72) for gene carriers has been colonoscopy, beginning at age 20 to 25 years or at least 5 years earlier than the earliest age at which colon carcinoma was diagnosed in a particular kindred with the procedure repeated every other year until the patient is 30 years of age and then annually thereafter. The frequency of colonoscopic examination is justified by the finding that HNPCC adenomas have repair deficient cells with rapidly and relentlessly accumulating mutations that support the clinical concept of ‘‘aggressive adenomas’’ and accelerate the adenoma-carcinoma sequence. The authors strongly believe that germline carriers should be offered prophylactic subtotal colectomy. The importance of screening colonoscopy was underscored by the report of Sankila et al. (68) who compared the colorectal carcinoma death rates of two groups of HNPCC patients—those screened every 3 years and controls who refused screening. The authors’ data indicated that one carcinoma was prevented for every 2.8 polypectomies performed. This is in stark contrast to the National Polyp Study data for the general population, which gave a range of 41 to 119 polypectomies per carcinoma prevented. The importance of screening is further emphasized by the study that compared screened with unscreened (by choice)‘‘controls’’ evaluated over a 15-year period with 252 relatives at risk for HNPCC, 119 of whom declined screening. In the screened group, 8 of 133 (6%) developed colorectal carcinoma compared to 19 (16%) in the unscreened group (73). de Vos tot Nederveen Cappel and coworkers (74) examined the stage of the screening detected carcinomas in relation to the surveillance interval and to assess the risk of developing colorectal carcinoma while on the program in 114 families in the Dutch Family Registry. A total of 35 carcinomas were detected while on the program. With intervals between colorectal carcinoma and the preceding surveillance examination of two years or less, carcinomas were at Dukes’ stage A (n ¼ 4), B (n ¼ 11), and C (n ¼ 1). With intervals of more than two years, carcinomas were Dukes’ stage A (n ¼ 3), B (n ¼ 10), and C (n ¼ 6). The 10-year cumulative risk of developing colorectal carcinoma was 10.5% in proven mutation carriers, 15.7% after partial colectomy, and 3.4% after subtotal colectomy. There is a substantial risk of developing colorectal carcinoma while on the program. However, all carcinomas but one of subjects who underwent a surveillance examination two years or less before detection was at a local stage. They therefore recommend surveillance for HNPCC with an interval of two years or less. For women who are gene carriers, an annual endometrial aspiration curettage should begin at age 30 years. Those patients should be advised of the option of ovarian

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

carcinoma screening with trans-vaginal ovarian ultrasonography, Doppler color blood flow imagery, and CA-125 evaluation. They should be encouraged to have their children early so they can consider the option of undergoing prophylactic total abdominal hysterectomy and bilateral salpingo-oophorectomy between the ages of 35 and 40 years. Screening for breast and gastric carcinoma should be initiated earlier in life than might otherwise be considered. If at least one family member is affected, gastroscopy beginning at age 30 to 35 years and every one to two years is advised (56). Also, only if at least one family member is affected, ultrasonography of the urinary tract and urine cytology beginning at age 30 to 35 years and every one to two years is advised (56). Mecklin and Jarvinen (75) analyzed 22 Finnish HNPCC families followed up for 7 years. Metachronous neoplasms were diagnosed in 41% of patients treated by segmental resection and 24% of those treated by subtotal colectomy. Extracolonic carcinoma was diagnosed in 30%. The most common malignancy was biliopancreatic carcinoma, which accounted for all five deaths related to carcinoma. The authors concluded that subtotal colectomy is superior to hemicolectomy or segmental resection in patients with HNPCC and that regular follow-up is necessary for surveillance of remaining bowel and extracolonic malignancies. Itoh et al. (76) conducted a study of 130 HNPCC kindreds to determine the risk of death from malignancy in first-degree relatives. The authors found a sevenfold increase in the risk of colon carcinoma in both sexes. In women relatives, the risk of breast carcinoma was increased fivefold, and a lifetime risk of breast carcinoma was 1 in 3.7. From these results, the authors recommended a screening program. Because clinical premonitory signs such as multiple colonic polyps seen in cases of familial polyposis coli are lacking, the clinician must rely on the clinical findings and a family history typical of HNPCC. Disorders that have been considered in the differential diagnosis of HNPCC have included FAP, attenuated FAP, juvenile polyposis coli, and Peutz-Jeghers syndrome (38). Once the diagnosis has been made, the physician is faced with the responsibility of informing the patient that other family members are at risk because the genetic implications for carcinoma expression are profound. Identification of individuals at risk may be aided by inspection of the patient’s skin, which may provide clues to the existence of a carcinoma associated with genodermatosis. Criteria providing clues to the diagnosis of Lynch type II syndrome include (1) any patient presenting with an early onset of carcinoma of the colon (particularly of the proximal colon in the absence of multiple polyps), endometrium, or ovaries (particularly when the patient is younger than age 40 years); (2) any patient with multiple primary carcinomas in which the lesions are integral to Lynch type II syndrome (carcinoma of the colon, endometrium, or ovary and other adenocarcinomas); and (3) any patient who states that one or more first-degree relatives manifested early onset of the carcinomas integral to Lynch type II syndrome. It is important to determine whether an HNPCC syndrome is present because the operative procedure of choice for the index carcinoma is subtotal colectomy as opposed to a more limited resection (38,72). In the case of a woman who has completed her family, the resection may be

65

extended to a prophylactic hysterectomy and bilateral salpingo-oophorectomy because of the patient’s inordinately high risk of developement of carcinoma of the endometrium and ovaries. The lifetime risk of endometrial and ovarian carcinomas in HNPCC is up to 60% and 12%, respectively (77). Watson et al. (78) collected data on 80 ovarian carcinoma patients who are members of HNPCC families, including 31 known mutation carriers, 35 presentive carriers (by colorectal/endometrial carcinomas status), and 14 at-risk family members. Among frankly epithelial cases, most carcinomas were well or moderately differentiated. Ovarian carcinoma in HNPCC differs from ovarian in the general population in several clinically important respects. It occurs at a markedly earlier age (42.7 years). It is more likely to be epithelial (95.6%). If it is a frankly invasive epithelial carcinoma it is more likely to be well or moderately differentiated (85% were FIGO stage I or II at diagnosis). HNPCC patients with ovarian carcinoma are more likely to have a synchronous endometrial carcinoma (21.5%). Familial Colorectal Carcinoma Ten to fifteen percent of patients with colorectal carcinoma and/or colorectal adenomas have other affected family members but their family histories do not fit the criteria for either FAP or HNPCC and may not appear to follow a recognizable pattern of inheritance, such as autosomal dominant inheritance. Such families are categorized as having familial colorectal carcinoma. The presence of colorectal carcinoma in more than one family member may be due to genetic factors, shared environmental risk factors or even to chance (13). With a family history the risk of colorectal carcinoma is increased earlier in life than later such that at age 45 years the annual incidence is more than three times higher than average-risk people; whereas at age 70, risk is not significantly different. The incidence in a 35 to 40 year old is about the same as that of an average-risk person at age 50 years. A personal history of adenomatous polyps confers a 15% to 20% risk of subsequently developing polyps. A history of adenomatous polyps in a sibling or parent is also associated with increased risk of colorectal carcinoma. Expert recommendations on screening for such persons are similar to those with a positive family history of colorectal carcinoma. Most experts suggest that screening should begin at age 35 to 40 years when the magnitude of risk is comparable to that of a 50 year old. Because the risk increases with extent of family history there is room for clinical judgment in favor of even earlier screening, depending on the details of the family history. A common but unproven clinical practice is to initiate colorectal carcinoma screening 10 years before the age of the youngest colorectal carcinoma case in the family. Evidence demonstrates that patients in families with breast and colon carcinoma (hereditary breast and colon carcinoma) may have a carcinoma family syndrome caused by CHEK2 1100delC mutations that are present in a subset of the families (79). The CHECK2 mutation is incompletely penetrant.

Genetic Testing Genetic testing provides the ability to determine who is and who is not at risk for disease before the onset of symptoms. Such information is becoming essential for proper

66

& PART I: COLORECTAL DISORDERS

management of patients and their families. In individuals who inherit mutant genes, simple preventive measures often can reduce morbidity and mortality and allow more thoughtful planning for the future. The benefits of genetic testing are equally important for those families who are found not to carry the relevant mutation; these individuals are spared unnecessary medical procedures and tremendous anxiety. However, genetic testing is not without its problems. These problems can be broadly divided into psychological or technical in nature. From the societal view, issues related to insurance, employment, discrimination, and privacy have garnered much concern and attention. The technical challenges associated with genetic testing can be just as formidable and are often overlooked. Even when the mutated gene is known, routine genetic testing may fail to identify mutations. As a result of these uncertainties, genetic testing that fails to find the mutation is often inconclusive. Studies have shown that these inconclusive results may be misinterpreted by the patient and physicians and are a source of great anxiety. Because of the complex psychosocial and technical issues, it is clear that genetic testing should never be offered to patients without appropriate genetic counseling. Prior to embarking upon any genetic testing it is critically important that the individual at risk of developing carcinoma and the referring physician as well have an understanding of what the testing is designed to do. To this end, genetic counseling is of paramount importance for patients to fully understand the limitations of genetic testing and will aid in the management of patients who are susceptible to colorectal carcinoma. The management of individuals who may be at increased risk for an inherited colorectal carcinoma susceptibility is complex. Knowledge about genetic information can contribute to clinical management at several stages during patient care. Obtaining and interpreting genetic information can be a lengthy process but are essential to avoid harm and to heighten the benefits of genetic testing. Genetic counseling and interpreting genetic test results can be complex but without knowing the limitations of the methods used and the lifetime probability of developing carcinoma in individuals who carry a gene that predisposes to carcinoma, misinterpretation may lead to false assurance. Wong et al. (80) conducted a review of the literature and combined with their clinical and research experience suggested a guide to enable various health care providers to better counsel patients in their quest for advice on prevention, early detection, and surveillance for colorectal carcinoma. Much of the following information was extracted from their review. Interpretation of genetic tests for colorectal carcinoma is not a straightforward as interpreting a blood sugar test to ascertain whether it is elevated, normal or lowered. Genetic counseling is a dynamic communication process between the patient and the counselor who provides education and support within a multidisciplinary team. Counseling is targeted toward individuals who are interested in having a personal risk estimate. Those at greatest risk for developing malignancy may benefit by assessing current screening tests, whereas those at lower risk will be reassured that their risk for colorectal carcinoma is not as high as they thought.

Furthermore, survivors of malignancies may want to know their risk of recurrence or of a second primary neoplasm and the risk for their relatives, especially their children. Carcinoma is a genetic disease but is not necessarily inherited. Virtually all neoplasms and certainly all malignant ones result in part from genes that mutate somatically (during one’s lifetime). In contrast, a minority of neoplasms arises from an inherited gene that confers susceptibility in development. Genetic counseling is appropriate when the family history is suggestive of a heritable predisposition for malignancy. In general, the counseling process is composed of discrete elements: risk assessment, informative counseling, supportive counseling, and follow-up. Before initiating the process it is important for the counselor to gain insight into the action of these components from the perspective of the patient. This is initially obtained by contracting. Contracting allows patients to know what to expect from genetic counseling and ensures their needs will be elicited and addressed by the counselor. The presence or absence of a family history of benign and malignant neoplasms helps to determine whether a person has an increased probability of having an inherited susceptibility for malignancy. Other familial traits can be characteristic signs of a genetic syndrome and suggestive that an increased risk for carcinoma may be present in a family. Pedigree analysis is multigenerational and includes details from both sides of the proband’s family. Current age or age at death is ascertained for everyone, because early death may explain the absence of a positive family history. Genetic counseling is recommended for individuals with an absolute risk greater than 10% (Table 6) for developing carcinoma of the colon and rectum. Providing risk assessment is only one part of the overall education process in genetic counseling. Explaining what genes are, how they are passed to subsequent generations, and the natural history of the neoplastic disease TABLE 6 & Personal and Family Features that Confer a Higher than Average Risk for Developing Colorectal Carcinoma Absolute Risk to Age 70 (%) Population risk Polyps Family history Personal history < 1 cm in diameter > 1 cm in diameter Family history of colorectal carcinoma 1 FDR < Age 45 Age 45–55 > Age 70 1 FDR þ 1 SDR 2 FDR ICG-HNPCC (mutation status unknown) Mutation carrier in a gene associated with HNPCC FAP (not clinically screened and mutation status unknown) APC mutation carrier for FAP

3–6 4–7 3–6 9–18

15–30 6–25 3–6 10 15–30 35–45 70–90 40–45 80–90

Abbreviations: FDR, first-degree relative; SDR, second-degree relative; FAP, familial adenomatous polyposis; ICG-HNPCC, International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer. Source: From Ref. 80.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

is an integral part of genetic counseling. Other educational aims include the recommendation or provision of screening modalities for early detection and preventive options. Medical intervention such as prophylactic surgery or chemoprevention, and the possible limitations of these treatments are issues for members of hereditary carcinoma families. Obtaining knowledge may empower the individual and family by dispelling the myths about malignancy and by providing guidelines that help in decision making. It also establishes the foundation of informed consent for genetic testing. Indications Genetic testing is discussed when the patient’s family history of carcinoma falls in the autosomal dominant mode of inheritance. Appropriate situations include families who carry a mutation in APC, hMLH1, hMSH2, hPMS1, hPMS2, or hMSH6 or in families fulfilling the clinical FAP or ICG-HNPCC criteria. The decision to undergo genetic testing is a personal one based on informed consent. The first phase in determining whether genetic testing is appropriate is genetic counseling. Among FAP families, who are candidates for genetic testing? The diagnosis of FAP is made because of the presence of polyps on examination of the presenting patient (proband or index case). In autosomal dominant inheritance, on average, 50% of the patient’s first-degree relatives (parents, siblings, and offspring) will be at risk. As soon as the FAP diagnosis emerges, either before or after operation, the surgeon should recommend that all first-degree relatives be examined. If a mutation in APC is identified in the patient’s germline DNA, the immediate family and other branches of the family can be encouraged to come forward for genetic counseling and to know their risk status if so desired. If a mutation is not detected, further genetic testing for APC mutations in the family is contraindicated and the clinical decision is to monitor first-degree relatives by colonoscopy at recommended intervals. Depending on the expression of colorectal carcinoma in the proband and the pattern of colorectal carcinoma and other malignancies in the family, genetic testing for HNPCC may be advisable. In every case of FAP, a genetic test to pinpoint the heritable, germline APC mutation is justified, with the informed consent of the patient. As far as testing the patient’s family is concerned, a common exception (occurring in 30% of probands) is a patient with FAP whose parents are negative on endoscopic examination and who therefore has a ‘‘new mutation’’ in the APC gene. In this instance, the risk for colorectal carcinoma for siblings falls to that of the general population. Nonetheless, children of the patient will require testing (if a mutation was identified in the proband) if so desired, because they are at 50% risk for FAP. The only obvious example in which genetic testing is not recommended would be in an isolated patient with FAP with no living first-degree relatives. What patients with suspected HNPCC should be referred for genetic counseling and testing? Because HNPCC is not always obvious based on clinical findings, either preoperatively or during resection, the clinical management of a possible HNPCC patient is complex. A detailed family history, with sites of malignancy verified, is critical to determine whether the patient fits the Amsterdam criteria and other associated features of

67

HNPCC. Analysis of neoplastic material would provide assistance in determining whether to test the proband’s germline DNA (i.e., a blood sample) for HNPCC mutations (after genetic counseling) and refer the first-degree relatives should a mutation be detected. In the instance of a positive family history it is ideal for testing to be conducted despite microsatellite instablility (MSI) phenotype. The clinical algorithm would be much the same as for FAP, with the exception that the multiple genes might need to be analyzed unless tests on pathogenic samples narrow the search. It is important to point out that at present, detection of a confirmed pathogenic mutation occurs in less than 50% of HNPCC families. In the absence of an identified deleterious mutation, genetic testing is inconclusive, which must be explained to the patient by an individual skilled in genetic counseling. Family members can be given general advice for colorectal carcinoma screening of first-degree relatives of patients with this malignancy. Techniques The molecular techniques used in testing for germline mutations in hereditary colorectal carcinoma have been described in detail by Rabelo et al. (81) and the following is a synopsis of that work. Germline DNA can be tested in three ways: 1. With in vitro–synthesized protein assay (also called protein truncation test or in vitro truncation test), specific gene segments are amplified by polymerase chain reaction (PCR) or reversed transcription (RT) -PCR, transcribed and translated in vitro, and the protein product is analyzed by gel electrophoresis. If the amplified segment has a mutation that causes the production of a truncated protein, its smaller size will allow in it to be more mobile in the gel. 2. Single-stranded conformation analysis (SSCA) is also a mutation detection technique in which specific gene segments are amplified by PCR or RT-PCR, denatured to separate the stands of the DNA product and analyzed by gel electrophoresis. The mutant DNA often migrates differently under certain conditions, allowing the putative localization of mutation within the segment. 3. DNA sequencing is the criterion standard technique for mutation detection, with up to 99% accuracy. It allows the precise identification of the mutation in the DNA sequence by pinpointing any change in the number or identity of bases. Pathologic specimens are examined in two ways: 1. A clue to identify colorectal carcinoma caused by HNPCC is by an associated alteration in the length stability of repetitive tract DNA in the carcinoma of an affected individual. This is known as microsatellite instability also called replication error positive phenotype. The presence of MSIþ thereby increases the likelihood that direct testing for defects in HNPCC genes will be informative. There are three outcomes of MSI analysis (56). a. If at least two markers tested show instability, the result indicates an increased likelihood that the colorectal carcinoma arose due to an HNPCC-associated mutation (MSI-H high). b. If no marker shows instability in the absence of a compelling family history the result is usually interpreted as ruling out HNPCC (MSS stable).

68

& PART I: COLORECTAL DISORDERS

c. If instability is seen in only one of the markers the result does not indicate HNPCC. The clinical behavior of MSI-low carcinomas is identical to those with MSS, and does not have the mutational fingerprints of the clinical behavior of MSI-H carcinomas. MSI-L results occur as commonly as do MSI-H results. 2. Immunohistochemistry is a technique for diagnosing a carcinoma that lacks expression of a particular HNPCC gene. It examines either hMLH1 or hMSH2 protein expression in carcinomas by immunostaining using monoclonal antibodies. Formalin fixed, paraffin embedded adenocarcinoma tissue is the sample required for testing with a region of normal mucosa adjacent to the carcinoma for comparison. The absence of hMLH1 or hMSH2 immunostaining shows which gene is defective and should be analyzed for the presence of a germline mutation. Studies have shown that the lack of hMLH1 or hMLH2 immunostaining was associated with the presence of MSIþ. Genetic Testing for FAP Presymptomatic genetic diagnosis of FAP in at-risk individuals has been feasible with linkage and direct detection of APC mutations (13). If one were to use linkage analysis to identify gene carriers, ancillary family members, including more than one affected individual, would need to be studied. With direct detection, fewer family members’ blood samples are required than for linkage analysis, but the specific mutation must be identified in a least one affected person by DNA mutation analysis or sequencing. Because approximately 96% of the mutations in FAP lead to a truncated protein, it has become routine to use in vitro– synthesized protein (IVSP) assay (sometimes called protein truncation test) for mutation detection. When a truncated protein is identified, it is possible to localize the mutation to a specific segment of the gene and then use DNA sequencing to determine the mutated nucleotide(s). The use of IVSP as the sole genetic test in FAP misses approximately 20% of APC mutations. Another screening technique is based on analysis of electrophoretic migration of small segments of the wild-type mutant gene (SSCA). The sequential use of two molecular diagnostic tests has become a common practice: a simple and less expensive screening technique (of high sensitivity and moderate specificity) followed by a definitive test of high sensitivity, usually DNA sequencing. These mutation search methods, APC protein truncation testing (from lymphocyte RNA) considerably enhance the feasibility of testing at-risk individuals without requiring DNA from multiple affected family members (as linkage requires). In particular, it is useful for testing in small families or in patients with spontaneous or ‘‘de novo’’ mutations (the first occurrence of FAP in a kindred), which may account for as much as one third of incident cases (13). Only about 80% of APC mutations can be detected by this method. Therefore, the clinician cannot rule out FAP on the basis of this molecular test alone if other criteria support the FAP diagnosis. When the mutation in a family is known from one or more affected members, one test, usually IVSP, need be performed, as the expression pattern of the mutation is already established for that family. A positive result in such a test is considered a ‘‘mutation-positive’’

result, and the patient can be counseled as recommended. When the mutation in the family is not known, IVSP is performed. In the majority of cases, a mutation will be found in the APC gene that can be further characterized by DNA sequencing. Again, a ‘‘mutation-positive’’ result is obtained leading to the appropriate genetic counseling for the patient. Other options include DNA sequencing and linkage analysis. Even by combining two or more techniques, it is not possible to achieve 100% sensitivity, because the mutations may not be in the coding region of the gene or because a few FAP kindreds do not exhibit linkage to chromosome 5q. In such case, a ‘‘no mutation detected’’ result must not be interpreted as a ‘‘negative test’’ with the very important considerations for counseling the patient. For at-risk individuals who have been found to be definitively mutation-negative by genetic testing, there is no clear consensus on the need or frequency of colon screening, though it would seem prudent that at least one flexible sigmoidoscopy or colonoscopy examination should be performed in early adulthood (age 18–25 years) (13). Molecular Genetic Diagnosis of HNPCC The HNPCC syndrome is more complex in terms of genetic testing because at least six genes involved in DNA mismatch repair predispose to this syndrome. The two most commonly involved genes are hMSH2 (involved in approximately 45% of cases) and hMLH1 (involved in approximately 49% cases). Protein truncation is observed in > 80% of HNPCC cases caused by mutations in hMSH2 and to a lesser degree in hMHL1. Currently, IVSP or SSCA or both are used, followed by DNA sequencing for precise characterization. In the case of a mutation already known in an affected member of the kindred, the same technique used to detect that mutation can be used, and a ‘‘mutation-positive’’ result leads to genetic counseling. If no mutation has been previously described in the kindred, analysis of hMLH1 and hMSH2 in an affected individual by IVSP or SSCA followed by DNA sequencing for confirmation is indicated. If a truncating mutation is found, a causative role can be inferred for the mutation. This ‘‘mutation-positive’’ result allows counseling of the affected individual and further testing of the kindred. If a nontruncating mutation is found, tests such as linkage analysis and functional assays may be required to discriminate between an ‘‘inconclusive’’ and a ‘‘mutation-positive’’ result. It is estimated that approximately 30% of mutations will be missed if IVSP is used as the sole genetic test and that a similar or higher percentage of false negatives may also be expected if SSCA is used instead. Again, even by combining two or more techniques, it is not possible to achieve 100% sensitivity. Other genes may be responsible (locus heterogeneity). A ‘‘no mutation detected’’ result must not be interpreted as a ‘‘negative test’’ result. Summary A proposed algorithm for genetic testing of individuals in FAP and HNPCC kindreds is as follows. First, use IVSP to locate the mutation, then use DNA sequencing to characterize it precisely in the first affected family member. If the IVSP test is negative, DNA sequencing of the APC gene is warranted. When the mutation is defined for that kindred, the testing of other family members at risk may be accomplished

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

by using IVSP alone. If the mutation in the first affected individual was identified by DNA sequencing, it is logical to continue to use this method for genetic testing of at-risk individuals. Considering the weighty psychological, ethical, and legal implications of the results of genetic tests, the use of two different techniques, one to detect the mutation and the second to identify and confirm it, is recommended in all cases. DNA sequencing, the most sensitive method for mutation detection, should be one of the techniques used. For HNPCC, two techniques are appropriate for initial testing of mutations in these genes, IVSP and SSCA. Screening with IVSP is recommended but this preference may vary among laboratories. If a mutation is detected by any of these methods, DNA sequencing should be used to identify it precisely in the first affected member of a kindred. If these screening techniques are negative, the sequencing of both hMSH2 and hMLH1 may be warranted. If still negative, in view of the locus heterogeneity seen in HNPCC, the same screening protocol may have to be used for hPMS2, hPMS1, and hMSH6. Again, the use of two different tests in all cases is recommended and DNA sequencing is used and when the pathogenic mutation is defined for that kindred, the testing of other family members at risk may be accomplished with either IVSP or SSCA alone. When a mutation (a variant gene) is identified by any of the methods, it must be validated as a pathogenic mutation (true-positive result) instead of a polymorphism, a common variant in the protein that does not contribute to the disease phenotype (false positive). When a mutation is not identified (‘‘no mutation detected’’) the negative result may be a ‘‘true negative’’ (no predisposing mutations in any relevant genes), or a ‘‘false negative’’ (undetected mutation may exist in known or unknown genes). If the person is affected with colorectal carcinoma, this is either a false negative in a familial colorectal carcinoma syndrome or a sporadic case. This ‘‘no mutation detected’’ result is the one most likely to be misinterpreted, leading to inappropriate genetic counseling. The use of two genetic tests decreases the rate of false-negative results. DNA sequencing is the most sensitive technique for this purpose and its use is advocated whenever a negative result is obtained by IVSP or SSCP for affected members with a strong family history. Even if the APC gene or all five DNA mismatch repair genes are fully sequenced, there will still be a residue of false-negative results caused by mutations in the noncoding regions of those genes, locus heterogeneity, or in still-unidentified genes. A ‘‘negative test’’ in patients at risk for FAP or HNPCC will rule out these disorders only if a mutation has been identified in an affected family member. If tests to identify a mutation in an affected family member are negative, a ‘‘no mutation detected’’ result should be considered as noninformative or inconclusive (as if no test had been performed). The correct interpretation of a negative result in mutation analysis of hereditary colorectal carcinoma syndromes is mandatory to avoid adverse outcomes, because a false-negative result may lead to lack of appropriate endoscopic surveillance. Molecular genetic testing must be coupled with appropriate genetic counseling. Interpretation of Genetic Test Results FAP & A summary for interpreting APC genetic test results is as follows. The implication of a mutation-positive test

69

result is that APC mutation carriers need to be informed that prophylactic colectomy is necessary when adenomatous polyps become evident. Mutation carriers require surveillance by gastroscopy for extracolonic neoplasms in the upper gastrointestinal tract. Other variant manifestations in FAP include osteomas, cutaneous cysts, and congenital hypertrophy of retinal pigment epithelium (CHRPE). When a mutation has been identified in the family, direct gene testing of relatives who have not yet been clinically assessed will distinguish between those who carry the mutation and those who do not. About 30% of FAP cases are caused by new mutations in the APC gene. In these cases the parents will not carry this mutation and are not at risk of FAP; only descendants of the proband are at 50% risk. What are the implications of a ‘‘negative test’’ result versus ‘‘no mutation detected’’ result? A negative test result is given if the patient does not carry the mutation that is known to exist in their family. In this instance family members are not at increased risk for developing colorectal carcinoma compared with the general population and should follow guidelines for carcinoma surveillance for that group. If the patient is affected with FAP and complete coding sequence analysis of the APC gene fails to identify a mutation, this could mean that the APC gene is not responsible for the patient’s diagnosis. In this case, a ‘‘no mutation detected’’ result, APC gene testing has no predictive value for asymptomatic at-risk relatives. In these families firstdegree relatives should continue colorectal carcinoma surveillance annually between the ages of 12 and 25 years, every other year between the ages of 25 and 35 years, and every third year between 35 and 50 years. Family members who have not developed multiple adenomatous polyps on complete bowel examination by the age of 50 years are assumed to be unaffected by FAP. If a ‘‘no mutations detected’’ result occurs when an unaffected family member from a APC kindred is tested without testing an affected family member, it is important that the patient not be falsely reassured because other mechanisms may inactivate the APC gene or other genes may be involved. APC I1307K & A mutation in the APC gene termed I1307K

was found in about 6% of people of Ashkenazi Jewish descent and about 28% of Ashkenazim with a family history of colorectal carcinoma. It appears to be associated with an approximately twofold increase in colorectal carcinoma risk (13). This alteration is a transversion of a single base from thymidine to adenine in codon 1307 of the APC gene. Analysis for this mutation is done by allele specific oligonucleotide analysis (extremely sensitive and specific) and consequently delivers a conclusive positive or negative test result. Genetic testing for this alteration is possible but the clinical utility of such testing is uncertain. The mean age at which colorectal carcinoma occurs in people carrying the I1307K alteration is not known nor has the natural history of colorectal carcinoma in carriers of I1307K been assessed or compared to sporadic colorectal carcinomas. No screening outcomes have been assessed in carriers of I1307K. Therefore, it is not yet known whether the I1307K carrier state should guide decisions about the age at which screening is initiated, the optimal screening strategy or the optimal screening interval (13).

70

& PART I: COLORECTAL DISORDERS

HNPCC. A summary of interpreting HNPCC genetic testing results is as follows. The implication of a positive mutation result is that individuals carrying a mutation in a gene that predisposes to HNPCC should be educated on the risk of colorectal carcinoma as well as carcinoma in other sites: endometrium, stomach, small bowel, biliary tract, ovary, pancreas, renal pelvis, and ureter. The risk to age 70 years for developing any neoplasm may be as high as 90% for males and 70% for females (82). For mutation carriers who develop colorectal carcinoma, total abdominal colectomy with ileo-rectal anastomosis is recommended because the entire colonic mucosa is at risk for malignancy. Proctocolectomy may also be a consideration, because the distal rectum is also a possible site of carcinoma. Prophylactic resection may be an option for mutation carriers unwilling to undergo surveillance or when endoscopic polypectomy is difficult. It should be emphasized that there are no data on whether resection is effective in reducing overall mortality. All first-degree relatives of patients who carry a mutation in one of the mismatch repair genes should be counseled that they have a 50% chance of carrying the same mutation. What are the implications of a negative test result versus no mutation detected? A patient who tests negative for a known family mutation is not at risk for HNPCC, because it is a conclusive negative test result. Subsequent carcinoma surveillance should be as for the general population. If a mutation in a family with strong circumstantial evidence for HNPCC has not been identified, the interpretation of a ‘‘no mutation detected’’ test result in the HNPCC-associated gene or genes tested is complex. It may be necessary to perform more than one screening test to strengthen the suspicion that a mutation in a mismatch repair gene exists in the family. The difference between a negative test result and a ‘‘no mutation detected’’ result with its subsequent implications is dependent on whether a mutation has been previously identified in the family in HNPCC families. If the person is affected with HNPCC, and no mutation was detected in hMLH1 and hMSH2 with in vitro–synthesized protein assay and single-strand conformation polymorphism, other investigations may be offered to determine whether the patient carries a mutation in a mismatch repair gene. For example, testing for microsatellite instability (MSI) in neoplastic tissue may be a worthy investment before sequencing hMLH1 and hMSH2. If no abnormality in hMLH1 and hMSH2 is found by sequencing, analysis of the rarer HNPCC-causing genes is indicated. Caution should be exercised in providing genetic risk assessment on the basis of currently used germline mutation detection strategies. While screening for hMSH2 gene mutations in HNPCC kindreds, Xia et al. (83) observed that using RT-PCR and the protein truncation test, the hMSH2 exon 13 deletion variant was found in > 90% of individuals. This may lead to a misdiagnosis of HNPCC and has profound implications for counseling and genetic risk assessments of other family members. Impact of Genetic Testing Individuals need to know the implications and possible impact of genetic test results before testing. An ambiguous test result may be more distressing than a positive test result. Moreover, genetic testing often yields results that

are probabilistic. The psychological burden of knowing that one is at high risk for developing a neoplasm may outweigh the possible benefits from intervention. Results of a study by Keller et al. (84) suggest that expressed intention and attitude toward genetic testing do not reliably predict actual uptake of counseling or testing. Data on the actual uptake of genetic testing for HNPCC in a clinical sample of 140 patients fulfilling clinical criteria of HNPCC, was 26%. Some 60% of participants experienced pronounced distress related to their potential inheritance of the disorder compared to 35% among nonparticipants. Distress reached a clinically significant level in 28% of participants. Restricted communication within the family was observed frequently. Irrespective of groups, a positive attitude toward obtaining a gene test result predominated. Therefore, the benefits and risks of participating in genetic tests must be described before testing is begun. Undergoing early detection of colorectal carcinoma or prophylactic operation or both can save lives. Surveillance or operation can be targeted to specific sites that are more prone to neoplasia, depending on the identity of the gene. For instance, female members of HNPCC families have a 10-fold increased risk for developing endometrial carcinoma, with an age of onset 10 to 15 years earlier than the same disease in the general population. Options for intervention and early detection and provision of knowledge to other family members are common reasons why individuals seek testing for genes predisposing to carcinoma of the colon and rectum. Gritz et al. (85) examined the impact of HNPCC genetic test results on psychological outcomes among carcinomaaffected and -unaffected participants up to 1 year after results disclosure. A total of 155 persons completed study measures before HNPCC genetic testing, and 2 weeks and 6 and 12 months after disclosure of test results. Mean scores on all outcome measures remained stable and within normal limits for carcinoma-affected participants, regardless of mutation status. Among unaffected carriers of HNPCC-predisposing mutations, mean depression, state anxiety, and carcinoma worries scores increased from baseline to 2 weeks postdisclosure and decreased from 2 weeks to 6 months postdisclosure. Among unaffected noncarriers, mean depression and anxiety scores did not differ, but carcinoma worries scores decreased during the same time period. Affected and unaffected carriers had higher mean test-specific distress scores at two weeks postdisclosure compared with noncarriers in their respective groups; scores decreased for affected carriers and all unaffected participants from 2 weeks to 12 months postdisclosure. Classification of participants into high-versus low-distress clusters using mean scores on baseline psychological measures predicted significantly higher or lower follow-up scores, respectively, on depression, state anxiety, quality of life, and test-specific distress measures, regardless of mutation status. Although HNPCC genetic testing does not result in long-term adverse psychological outcomes, unaffected mutation carriers may experience increased distress during the immediate postdisclosure time period. Furthermore, those with higher levels of baseline mood disturbance, lower quality of life, and lower social support may be at risk for both short- and long-term increased distress. Examining a person’s DNA has the potential for farreaching consequences beyond the person being tested. Genetic testing can reveal information about relatives with

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

whom we share a common genetic legacy. Despite the possible benefits of testing for susceptibility for malignancy, the uncertainty and limitations of current efforts to lower mortality or morbidity contribute to the psychosocial and ethical complexities. Genetic information must remain confidential. However, breach of confidentiality has been considered permissible if all of the following conditions are met: reasonable effort to encourage disclosure has failed; harm is likely to occur if the information is withheld; the harm is serious and avoidable; and the disease is treatable or preventable (86). Conflict between issues of privacy and duty to forewarn family members arises because genetic information is, at once, individual and familial. Several key issues may be raised in genetic testing. First, at times genetic test results need to be interpreted in the context of the results of other family members. Second, family members should be independently and autonomously counseled. Third, the decision of each member regarding testing should be followed without coercion by other family members. Fourth, genetic test results have familial implications whether results are disclosed to other family members or not. To provide information useful in the education and counseling in individuals considering genetic testing, Lerman et al. (87) conducted structured interviews with 45 firstdegree relatives of colorectal carcinoma patients. Fifty-one percent of respondents indicated that they definitely would want to obtain a genetic test for colon carcinoma susceptibility when it is available. Motivation for genetic testing included the following: to know if more screening tests are needed, to learn if one’s children are at risk, and to be reassured. Barriers to testing included concerns about insurance, test accuracy, and how one’s family would react emotionally. Most participants anticipated that they would become depressed and anxious if they tested positive for a mutation; many would feel guilty and still worry if they tested negative. These preliminary results underscore the importance of the potential risks, benefits, and limitations of genetic testing with particular emphasis on the possibility of adverse psychological effects and implications for health insurance. Ethical issues of revealing or concealing genetic information will be debated long into the future because the potentially enormous consequences must be carefully considered. Hadley et al. (88) assessed the impact of genetic counseling and testing on the use of endoscopic screening procedures and adherence to recommended endoscopic screening guidelines in 56 symptomatic at-risk individuals in families known to carry an HNPCC mutation. They analyzed data on colonoscopy and flexible sigmoidoscopy screenings collected before genetic counseling and testing and 6 and 12 months postgenetic counseling and testing on 17 mutation-positive and 39 truenegative mutation individuals. Among mutation-negative individuals, use of colonoscopy and flexible sigmoidoscopy decreased significantly between pre- and postgenetic counseling and testing. Among mutation-positive individuals a nonsignificant increase in use was noted. Age was also associated with use of endoscopic screening after genetic counseling and testing. More mutation-negative individuals strictly adhered to guidelines than did mutation-positive individuals (87% vs. 65%). They concluded genetic counseling and testing for HNPCC significantly influences the use of colonic endoscopy and adherence to recommendations for colon carcinoma screening.

71

THREAT OF GENETIC DISCRIMINATION & Patients considering genetic testing need to be informed about the potential for genetic discrimination. This is of great concern for asymptomatic family members, because disclosure of a positive mutation result to third parties could be used to limit, raise rates for, or deny access to individual health insurance. However, nondisclosure could nullify a patient’s insurance contract. A survey of medical directors of United States life insurance companies indicated that familial colon or breast carcinoma constituted sufficient grounds to deny insurance (1 of the 27) or charge higher premiums (6 of the 27), despite not having actuarial data to support underwriting (calculating premiums) guidelines pertaining to these carcinomas. In the United States several states have safeguards against the potential misuse of genetic information. The potential for misuse of genetic information has led to federal initiatives in the United States to regulate the use of genetic information. In the context of insurance and employment it is unclear whether the Americans with Disabilities Act of 1990 (ADA) will provide adequate protection to carriers of carcinoma-susceptibility genes, because the Equal Employment Opportunity Commission does not view a person with genetic predisposition for a disease or a carrier of a gene for a late-onset disorder as having a disability. This implies that these persons are not protected under the ADA. Legislation has been introduced to extend the definition of disability to ‘‘genetic or medically identified potential of or predisposition toward a physical or mental impairment that substantially limits a major life activity.’’ For now, the ADA has been interpreted to offer protection only to those who develop symptoms. Health Insurance Portability and Accountability Act (Kennedy-Kassebaum bill) came into effect in the summer of 1997. It defined genetic information as part as an individual’s health status. This act was implemented to prohibit employers and insurers from excluding individuals in a group from coverage or charging higher premiums on the basis of health status. Also in 1997 the Genetic Information Nondiscrimination in Health Insurance Act (SlaughterSnowe bill) called for a ban of the use of ‘‘genetic information’’ in denying or setting rates for group health insurance. In this act genetic information was defined broadly to include genetic tests as well as information about inherited features. This bill called for a limit on the collection or disclosure of genetic information without the written consent of the individual. Fourteen states have enacted legislation to provide safeguards and to protect misuse of genetic information by insurers and employers. Other states have followed suit, and amendments are being made to those existing laws in some states, as reviewed by Offit (89). In summary, individuals undergoing genetic testing should understand the advantages and disadvantages of receiving and sharing genetic results. Attempting to protect the individual from genetic discrimination by disseminating results to only a few selected medical staff could impede care. So far, there has been no substantial use of genetic information by health insurers, and laws restricting its use have thus far maintained social fairness in the area of health insurance. When integrated with existing testing

72

& PART I: COLORECTAL DISORDERS

protocols for colorectal carcinoma and when applied with appropriate caveats particularly regarding interpretation of negative results, genetic testing can result in improved management of patients and families.

Alternate Pathway of Carcinogenesis An understanding of the mechanisms that explain the initiation and early evolution of colorectal carcinoma should facilitate the development of new approaches to effective prevention and intervention. Jass et al. (90) have proposed a model for colorectal neoplasia in which APC mutation is not placed at the point of initiation. Other genes implicated in the regulation of apoptosis and DNA repair may underlie the early development of colorectal carcinoma. Inactivation of these genes may occur not by mutation or loss but through silencing mediated by methylation of the gene’s promoter region. hMLH1 and MGMT are examples of DNA repair genes that are silenced by methylation. Loss of expression of hMLH1 and MGMT protein has been demonstrated immunohistochemically in serrated polyps. Multiple lines of evidence point to a ‘‘serrated’’ pathway of neoplasia that is driven by inhibition of apoptosis and the subsequent inactivation of DNA repair genes by promoter methylation. The earliest lesions in this pathway are aberrant crypt foci (ACF). These may develop into hyperplastic polyps or transform while still of microscopic size into admixed polyps, serrated adenomas, or traditional adenomas. Carcinomas developing from these lesions may show high- or low-level microsatellite instability (MSI-H and MSI-L, respectively) or may be microsatellite stable (MSS). The suggested clinical model for this alternative pathway is the condition hyperplastic polyposis. Hyperplastic polyposis presents a plausible model for the following reasons: 1. Polyps in this condition may show MSI and silencing of relevant DNA repair genes including hMLH1. 2. Methylation is demonstrated in DNA extracted from hyperplastic polyps in a subset of subjects with hyperplastic polyposis. The finding of methylation is concordant within multiple polyps in such cases, whereas discordant findings occur in subjects with multiple adenomas. 3. The requisite plasticity in methylator pathways is evident in hyperplastic polyposis in which all types of epithelial polyp may occur (hyperplastic, admixed, serrated adenoma, and traditional adenoma), and carcinomas may be MSI-H, MSI-L, or MSS (even within the same subject). 4. The condition hyperplastic polyposis may be familial. Jass et al. (90) believe that molecular and morphologic observations have stripped the hyperplastic polyp of its long-presumed innocence. However, the fact that the vast majority of hyperplastic polyps will never progress to carcinoma has not altered. It is impractical to advocate the removal of every minute hyperplastic lesion. On the other hand, more attention might be given to subjects with ‘‘high-risk’’ hyperplastic polyps. Highrisk features would include multiplicity (more than 20), size (greater than 10 mm), proximal location, associated polyps with dysplasia, and a family history of colorectal carcinoma. New diagnostic criteria and markers are required to distinguish innocent hyperplastic

polyps from their serrated counterparts with a malignant potential that belies their deceptively bland morphology.

& DIETARY FACTORS Dietary factors have been implicated in the etiology of colorectal carcinoma. Prime candidates are dietary fats and fiber-deficient diets. Studies thus far have been inconclusive with respect to which component is most important. Experimental, epidemiological, and clinical evidence show that diets consumed by Western populations have an important role in the modulation of this disease (91). The diet contains various mutagens and carcinogens that can be classified into three groups: (1) naturally occurring chemicals that include mycotoxins and plant alkaloids, (2) synthetic compounds exemplified by food additives and pesticides, and (3) compounds produced by cooking, which include polycyclic aromatic hydrocarbons and heterocyclic amines (92). Since heterocyclic amines are genotoxic compounds, a causal role in some stage of human colon carcinogenesis is plausible. Fat Populations with diets high in unsaturated fat and protein, especially if associated with low-fiber ingestion, are associated with a high incidence of colorectal carcinoma. In a Canadian study, patients with colon carcinoma reported a higher intake of total fat, saturated fat, and dietary cholesterol than controls, with the highest relative risk in saturated fat (93). In a prospective cohort study of diet and colon carcinoma, Willett et al. (94) gathered information on 98,464 nurses. Among the 150 nurses who developed colon carcinoma, the trend for risk with total fat intake (p < 0.05) and animal fat (p < 0.01) was significant. Animal fat from dairy sources (e.g., butter or ice cream) was not associated with risk. A study by Nigro et al. (95) found that animals fed 35% beef fat develop more colon carcinoma than those fed 5% beef fat. Reddy and Maruyana (96) found that a diet containg 20% of either corn or safflower oil increased the incidence of colon carcinoma in animals much more than in those fed 5% of the same fat. Not only is the amount of fat important, but the type may be equally important. Diets containing high levels of olive oil, coconut oil, or fish oil did not increase the incidence of colon carcinoma more than in animals fed 5% of the same fats. The mechanism of action is believed to be indirect, with the resulting high concentration of fecal bile acids and cholesterol stimulating cell proliferation and acting as promoters of carcinogenesis. However, this seemingly neat hypothesis is not universally accepted. It must be noted that the role of dietary fat is inextricably linked to excretion of bile acids and the ratio of anaerobic to aerobic bacteria. Meat and Fish Consumption of red and processed meat has been associated with colorectal carcinoma in many but not all epidemiological studies. Chao et al. (97) examined the relationship between recent and long-term meat consumption and the risk of incident colon and rectal carcinoma. A cohort of 148,610 adults aged 50 to 74 years (median, 63 years), residing in 21 states with population-based

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

registries, who provided information on meat consumption in 1982 and again in 1992/1993 were enrolled in the Cancer Prevention Study II Nutrition Cohort. Follow-up identified 1667 incident colorectal carcinomas. Participants contributed person-years at risk until death or a diagnosis of colon or rectal carcinoma. High intake of red and processed meat reported in 1992/1993 was associated with higher risk of distal colon carcinoma after adjusting for age and energy intake but not after further adjustment for body mass index, cigarette smoking, and other covariates. When long-term consumption was considered, persons in the highest tertile of consumption in both 1982 and 1992/ 1993 had higher risk of distal colon carcinoma associated with processed meat (RR, 1.50) and ratio of red meat to poultry and fish (RR, 1.53) relative to those persons in the lowest tertile at both time points. Long-term consumption of poultry and fish was inversely associated with risk of both proximal and distal colon carcinoma. High consumption of red meat reported in 1992/1993 was associated with a higher risk of rectal carcinoma (RR, 1.71) as was high consumption reported in both 1982 and 1992/1993 (RR, 1.43). Their results strengthen the evidence that prolonged high consumption of red and processed meat may increase the risk of carcinoma in the distal portion of the large intestine. Larsson et al. (98) prospectively examined whether the association of red meat consumption with carcinoma risk varies by subsite within the large bowel. They analyzed data from 61,433 women aged 40 to 75 years and free from diagnosed carcinoma at baseline. Diet was assessed at baseline using a self-administered food-frequency questionnaire. Over a mean follow-up of 13.9 years, they identified 234 proximal colon carcinomas, 155 distal colon carcinomas, and 230 rectal carcinomas. They observed a significant positive association between red meat consumption and risk of distal colon carcinoma but not of the proximal colon or rectum. The multivariate rate ratio for women who consumed 94 g/day or more of red meat compared to those who consumed less than 50 g/day was 2.22 for the distal colon, 1.03 for proximal colon, and 1.28 for rectum. Although there was no association between consumption of fish and risk of carcinoma at any subsite, poultry consumption was weakly inversely related to risk of total colorectal carcinoma. Norat et al. (99) prospectively followed 478,040 men and women from 10 European countries who were free of carcinoma at enrollment. After a mean follow-up of 4.8 years, 1329 incident colorectal carcinomas were documented. Colorectal carcinoma risk was positively associated with intake of red and processed meat [highest ( > 160 g/ day) and lowest ( < 20 g/day)] intake, HR ¼ 1.35 and inversely associated with intake of fish ( > 80 g/day vs. < 10 g/day, HR ¼ 0.69), but was not related to poultry intake. In this study population, the absolute risk of development of colorectal carcinoma within 10 years for a study subject aged 50 years was 1.71% for the highest category of red and processed meat intake and 1.28% for the lowest category of intake and was 1.86% for subjects in the lowest category of fish intake and 1.28% for subjects in the highest category of fish intake. Their data confirmed that colorectal carcinoma risk is positively associated with high consumption of red and processed meat and support an inverse association with fish intake.

73

Fiber Following his observation of the low incidence of colorectal carcinoma in African natives, Burkitt (100) promoted the idea that their high-fiber intake was responsible for this finding. He further deduced that the Western low-fiber diet along with high carbohydrate and animal fat intake was responsible for the higher incidence of maligiiancy of the large bowel. Burkitt also noted that when Africans abandon their customary diet, the incidence of carcinoma of the colorectum progressively increases. Diets with a high roughage content result in the production of soft frequent stools. A potential carcinogen in a patient with infrequent stools remains in contact with the colonic and rectal mucosa longer. Experimental work by Fleiszer et al. (101) has supported this thesis in that a high-fiber diet resulted in a diminished incidence of dimethylhydrazine (DMH) -induced carcinoma of the colon in rats. Yet the potential protective effect of fiber is controversial. A study by Nigro et al. (102) in which a group of rats was given a 30% beef fat and 10% fiber diet, demonstrated no protective effect from wheat bran or cellulose. Animals fed 5% fat and 20% or 30% fiber developed fewer carcinomas than fiber-free controls. This finding suggests that a large quantity of fat can overcome the protective effect of fiber. From an epidemiologic point of view, Greenwald, Lanza, and Eddy (103) analyzed 55 original reports and found evidence of an inverse association between a high fiber diet and the risk of colon carcinoma. The same group conducted a meta-analysis of 12 methodologically sound and descriptively complete case-control studies and showed protection with an odds ratio of 0.57 (95% CI, 0.50 to 0.64) (104). Those studies delineating vegetable fiber from total fiber suggested a stronger protective effect from vegetable fiber. Freudenheim et al. (105) conducted a case-control study of 850 pairs in which the fiber source was subdivided from grain, fruit, and vegetables and in which the results of consuming soluble vs. insoluble fiber were also compared. Fruit and vegetable fiber protected against rectal carcinoma in men and women and against colon carcinoma in men. Grain fiber protected against colon carcinoma only. Insoluble fractions of grain fiber were more protective in the colon, with both soluble and insoluble fractions from fruit and vegetables protective in the rectum. In a follow-up of 11 years of a German cohort of 1904 patients, the mortality rate for patients with colon carcinoma was reduced for individuals following a vegetarian lifestyle for over 20 years (106). A study with contrary findings was reported. Asano and McLeod (107) conducted a systematic review and meta-analysis to assess the effect of dietary fiber on the incidence or recurrence of colorectal adenomas, the incidence of colorectal carcinoma, and the development of adverse events. Five studies with 4349 subjects met the inclusion criteria. The interventions were wheat bran fiber, ispaghula husk, or a comprehensive dietary intervention with high fiber whole food sources alone or in combination. When the data were combined there was no difference between the intervention and control groups. The reviewers concluded there is currently no evidence from randomized clinical trials to suggest that increased dietary fiber intake will reduce the incidence or recurrence of adenomatous polyps within a two- to four-year period. Fuchs et al.

74

& PART I: COLORECTAL DISORDERS

(108) conducted a prospective study of 88,757 women who were 34 to 59 years old and had no history of carcinoma, inflammatory bowel disease, or familial polyposis. During a 16-year follow-up period, 787 cases of colorectal carcinoma were documented. In addition, 1012 patients with adenomas of the distal colon and rectum were found among 27,530 participants who underwent endoscopy during the follow-up period. After adjustment for age, established risk factors, and total energy intake, they found no association between the intake of dietary fiber and the risk of colorectal carcinoma or colorectal adenoma. The same investigators found no protective effect of fruit vegetable consumption (109). On the other hand, Terry et al. (110) found that total fruit and vegetable consumption was inversely associated with the risk of the development of colorectal carcinoma whereas they observed no association between colorectal carcinoma risk and consumption of cereal fiber. Michels et al. (109) prospectively investigated the association between fruit and vegetable consumption and the incidence of colon and rectal carcinoma in two large cohorts; the Nurses’ Health Study (88,764 women) and the Health Professionals’ Follow-up Study (47,325 men). With a follow-up including 1,743,645 person-years and 937 cases of colon carcinoma, they found little association of colon carcinoma incidence with fruit and vegetable consumption. Although fruit and vegetables may confer protection against some chronic diseases, their frequent consumption did not appear to confer protection from colon and rectal carcinoma in this study. Peters et al. (111) used a 137-item food-frequency questionnaire to assess the relation of fiber intake and frequency of colorectal adenoma. The study was done within the prostate, lung, colorectal, ovarian (PLCO) Cancer Screening Trial, a randomized controlled trial designed to investigate methods for early detection of carcinoma. In their analysis they compared fiber intake of 33,971 participants who were sigmoidoscopy-negative for polyps, with 3591 cases with at least one histologically verified adenoma in the distal large bowel (i.e., descending colon, sigmoid colon, or rectum). High intakes of dietary fiber were associated with a low risk of colorectal adenoma, after adjustment for potential dietary and nondietary risk factors. Participants in the highest quintile of dietary fiber intake had a 27% lower risk of adenoma than those in the lowest quintile. The inverse association was strongest for fiber from grains and cereals than from fruits. Risks were similar for advanced and nonadvanced adenoma. Risk of rectal adenoma was not significantly associated with fiber intake. Why these two studies reached different results is impossible to answer. The use of different types or sources and amounts of fiber may be one explanation. In a population-based case-control study, Meyer and White (112) found for both sexes that a higher dietary fiber intake was associated with lower relative risks for colon carcinoma. Howe et al. (113) examined the effects of fiber, vitamin C, and beta-carotene intakes on colorectal carcinoma risk in a combined analysis of data from 13 casescontrolled studies previously conducted in populations with differing colorectal carcinoma rates and dietary practices. Original data records for 5287 case subjects with colorectal carcinoma and 10,470 control subjects without disease were combined. Risk decreased as fiber intake

increased; relative risks were 0.79, 0.69, 0.63, and 0.53 for the four highest quintiles of intake compared with the lowest quintile. The inverse association with fiber is seen in 12 of the 13 studies and is similar in magnitude for left- and right-sided colon and rectal carcinomas, for men and for women, and for different age groups. In contrast, after adjustment for fiber intake, only weak inverse associations are seen for the intakes of vitamin C and beta-carotene. This analysis provided substantive evidence that intake of fiberrich foods is inversely related to risk of carcinomas of both the colon and rectum. If causality is assumed, they estimated that risk of colorectal carcinoma in the United States population could be reduced by 31% (55,000 cases annually) by an average increase in fiber intake from food sources of about 13 g/day, corresponding to an average increase of about 70%. Bingham et al. reached the same conclusion. In the biggest study ever published Bingham et al. (114) prospectively examined the association between dietary fiber and incidence of colorectal carcinoma in 519,978 individuals aged 25 to 70 years taking part in a study recruited from 10 European countries. Participants completed a dietary questionnaire in 1992–1998 and were followed up for the incidence of carcinoma. Follow-up consisted of 1,939,011 person-years, and data for 1065 reported cases of colorectal carcinoma were included in the analysis. Dietary fiber in foods was inversely related to the incidence of large bowel carcinoma (adjusted relative risk 0.75 for the highest vs. lowest quintile of intake), the protective effect being greatest for the left side of the colon, and least for the rectum. After calibration with more detailed dietary data, the adjusted risk for the highest versus lowest quintile of fiber from food intake was 0.58. No food source of fiber was significantly more protective than others, and nonfood supplement sources of fiber were not investigated. They concluded, in populations with low average intake of dietary fiber, an approximate doubling of total fiber intake from foods could reduce the risk of colorectal carcinoma by 40%.

Calcium Deficiency Slattery, Sorenson, and Ford (115) observed that the dietary intake of calcium decreased the risk of development of colon carcinoma. Calcium can bind intraluminally with bile acids and fatty acids, thus reducing their mitogenic effect (116). Calcium salts may have antiproliferative effects in the colon of patients who are predisposed to developing large bowel carcinoma (117). Dietary supplementation with calcium reduces colonic crypt cell production in both normal and hyperplastic mucosa. One of the main pathways used by extracellular calcium to exert its chemopreventive actions is through activation of a calcium-sensing receptor. This results in increased levels of intracellular calcium, inducing a wide range of biological effects, some of which restrain the growth and promote the differentiation of transformed colon cells (91). Calcium likely reduces lipid damage in the colon by complexing with fat to form mineral-fat complexes or soaps (118). It has been shown in an increasing number of animal experiments that calcium has the ability to inhibit colon carcinoma. In limited studies in man, the colonic hyperproliferation associated with increased risk of colon carcinoma has been reversed for short periods by the administration of supplemental

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

dietary calcium. In a population-based case-control study, Meyer and White (112) showed that calcium was associated with a decreased risk of colon carcinoma in women only. Baron et al. (119) conducted a randomized double-blind trial of the effect of supplementation with calcium carbonate on recurrence of colorectal adenomas. They randomly assigned 930 subjects (mean age, 61 years; 72% men) with a recent history of colorectal adenomas to receive either calcium carbonate [3 g (1200 mg of elemental calcium) daily] or placebo, with follow-up colonoscopies one and four years after the qualifying examination. Among the 913 subjects who underwent at least one study colonoscopy, the adjusted risk ratio for any recurrence of adenoma with calcium as compared with placebo was 0.85. At least one adenoma was diagnosed between the first and second follow-up endoscopies in 127 subjects in the calcium group (31%) and 159 subjects in the placebo group (38%). The effect of calcium was independent of initial dietary fat and calcium intake. They concluded calcium supplementation is associated with a significant—though moderate—reduction in the risk of recurrent adenomas. Wu et al. (120) examined the association between calcium intake and colon carcinoma risk in two prospective cohorts, the Nurses’ Health Study and the Health Professionals Follow-up Study. Their study population included 87,998 women in the former and 47,344 men in the latter. During the follow-up period, 15 years for the Nurses’ Health Study cohort, and 10 years for the Health Professionals Follow-up Study, 626 and 399 colon carcinomas were identified in women and men, respectively. In women and men considered together they found an inverse association between high total calcium intake ( > 1250 mg/day vs. 500 mg/ day) and distal colon carcinoma (women relative risk ratio 0.73, men relative risk ratio 0.58, and pooled relative ratio 0.65). No such association was found for proximal colon carcinoma (women relative risk ratio ¼ 1.28, men relative risk ratio ¼ 0.92, and pooled relative risk ratio 1.14). The incremental benefit of additional calcium intake beyond approximately 700 mg/day appeared to be minimal. Wallace et al. (121) examined the effect of calcium on the risk of different types of colorectal lesions. They used patients from the Calcium Polyp Prevention Study, a randomized double-blind, placebo-controlled chemoprevention trial among patients with a recent colorectal adenoma in which 930 patients were randomly assigned to calcium carbonate (1200 mg/day) or placebo. Followup colonoscopies were conducted approximately one and four years after the qualifying examination. The calcium risk ratio for hyperplastic polyps was 0.82, that for tubular adenomas was 0.89, and that for histologically advanced neoplasms was 0.65 compared with patients assigned to placebo. There were no statistically significantly differences between the risk ratio for tubular adenomas and that for other types of polyps. The effect of calcium supplementation on adenoma risk was most pronounced among individuals with high dietary intakes of calcium and fiber and with low intake of fat, but the interactions were not statistically significant. Their results suggest that calcium supplementation may have a more pronounced antineoplastic effect on advanced colorectal lesions than on other types of polyps. Taken together the available evidence suggests that increases in the daily intake of calcium in the diet

75

may provide a means of colorectal carcinoma control. Although a positive effect has not been proven, it might be prudent for the public to consume diets that contain adequate amounts of calcium.

Magnesium Larsson et al. (122) suggested that a high magnesium intake may reduce the occurrence of colorectal carcinoma in women. In a population-based prospective cohort of 61,433 women aged 40 to 75 years without previous diagnosis of carcinoma at baseline and a mean of 14.8 years (911,042 person-years) of follow-up, 805 incident colorectal carcinoma cases were diagnosed. Compared with women in the lowest quintile of magnesium intake, the multivariate rate ratio was 0.59 for those in the highest quintile. The inverse association was observed for both colon (RR, 0.66) and rectal carcinoma (RR, 0.45). Micronutrients and Chemical Inhibitors In geographic regions where the trace element selenium is found to be lacking, there is a higher incidence of colorectal carcinoma, while high selenium areas have low colorectal carcinoma rates (123). Significantly decreased selenium concentrations in blood samples from patients with colorectal carcinomas and various adenomas have been found when compared to normal controls (124). High selenium broccoli decreased the incidence of aberrant crypts in rats with chemically induced colon carcinomas by more than 50% compared with controls (125). In a case-controlled study, Nelson et al. (126) found that higher levels of selenium produced a protective effect against colon polyps or carcinomas. Jacobs et al. (127) conducted a combined analysis of data from three randomized trials—the Wheat Bran Fiber Trial, the Polyp Prevention Trial, and the Polyp Prevention Study—which tested the effects of various nutritional interventions for colorectal adenoma prevention among participants who recently had an adenoma removed during colonoscopy. Selenium concentrations were measured from blood specimens from a total of 1763 trial participants, and quartiles of baseline selenium were established from pooled data. Analyses of the pooled data showed that individuals whose blood selenium values were in the highest quartile (median ¼ 150 ng/mL) had statistically significantly lower odds of developing a new adenoma compared with those in the lowest quartile (OR ¼ 0.66). They concluded the inverse association between higher blood selenium concentration and adenoma risk supports previous findings indicating that higher selenium status may be related to decreased risk of colorectal carcinoma. A number of micronutrients and chemicals have been shown to have an inhibitory effect on the development of colorectal carcinoma: phenols, indoles, plant serols, selenium, calcium, vitamins A, C, and E, and carotenoids (128). They are present in small amounts in water and in whole grain cereals, fruits, and vegetables. The chemicals in foods that inhibit carcinoma development in laboratory animals have been summarized by Wargovich (129). They include plant phenols (in grapes, strawberries, and apples), dithiothiones and flavones (in cabbage, broccoli, brussels sprouts, and cauliflower), thioethers (in garlic, onions, and leeks), terpenes (in citrus fruits), and carotenoids (in carrots, yams, and watermelon). Some chemicals not

76

& PART I: COLORECTAL DISORDERS

present in the average diet affect the carcinogenic process in animals. Prostaglandin inhibitors and chemicals that influence cell proliferation and differentiation are examples (130). Some of these agents may be toxic, so the feasibility of administering a combination of agents was studied experimentally by Nigro et al. (131) By the addition of small nontoxic amounts of selenium, 13-cis-retinoic acid, and b-sitosterol, an additive inhibitory effect resulted in a significant diminution of intestinal carcinoma. Other combinations also have been found effective (128). Ascorbic acid, dialyl sulfide, and thioether found in garlic and onions also may aid in prevention (129,132). Most of the pleiotropic actions of vitamin D are mediated by binding to a nuclear receptor that interacts with specific consensus sites in promoters of specific genes, resulting in downregulation or upregulation of their expression. The actions of vitamin D involve cross-talk with growth factors/cytokines, inhibitory effects on the cell cycle and stimulation of apoptosis (91). Folate lies at the intersection of metabolic pathways involved in DNA methylation and biosynthesis. Three main mechanisms by which decreased levels of folate (and of other dietary one-carbon donors) might increase the risk of carcinoma are alteration of the normal DNAmethylation process; imbalance of the steady-state level of DNA precursors, leading to aberrant DNA synthesis and repair; and chromosome and chromatin change (91).

Alcohal Ingestion A relationship between alcohol ingestion and development of colorectal carcinoma has been reported (OR, 2.6) (112) specifically the development of rectal carcinoma in association with beer consumption (6). Daily alcohol drinkers experience a twofold increase in risk of colorectal carcinomas (133). The positive association had been accounted for primarily by an increased risk of carcinoma in men whose monthly consumption of beer was 15 L or more (134,135). Beer drinking increases the risk 1.3 to 2.4 times (135–138). In a study of 6230 Swedish brewery workers, the relative risk for rectal carcinoma was 1.7 while the risk of colon carcinoma was not significantly increased, supporting the hypothesis that high beer consumption is associated with an increased risk of rectal carcinoma (139). Newcomb, Storer, and Marcus (135) found that high levels of alcohol consumption in women (11 or more drinks per week) were associated with an increased risk of large intestinal carcinoma (RR ¼ 1.47). Maekawa (140) reported that the heavy cumulative intake of alcohol was associated with significantly higher risk of colorectal carcinoma than in nondrinkers (OR ¼ 6.8). The association of alcohol intake with the risk of colorectal carcinoma was not effected by the type of alcoholic beverage. Sharpe et al. (141) found the daily consumption of alcohol of any type was associated with increased risks of carcinoma of the distal colon (OR ¼ 2.3), and the rectum (OR ¼ 1.6) but not with an increased risk of a carcinoma of the proximal colon (OR ¼ 1.0). Smoking The association of smoking (odds ratio for > 40 pack years, 3.31) with adenomas (and by implication carcinoma) has been reported (142). Smoking for 20 years has a strong

relation to adenomas, but an induction period of at least 35 years is necessary for colorectal carcinoma (143,144). Chao (145) examined cigarette smoking in relation to colorectal carcinoma mortality, evaluating smoking duration and recency, and controlling for potential confounders in the Cancer Prevention Study II. This prospective nationwide mortality study of 1,184,657 adults (age  30 years) was begun by the American Cancer Society in 1982. After exclusions, their analytic cohort included 312,332 men and 469,019 women, among whom 4432 colon or rectal carcinoma deaths occurred between 1982 and 1996 among individuals who were carcinoma free in 1982. Multivariate-adjusted colorectal carcinoma mortality rates were highest among current smokers, were intermediate among former smokers, and were lowest in lifelong nonsmokers. The multivariate-adjusted relative rate for current compared with never smokers was 1.32 among men and 1.41 among women. Increased risk was evident after 20 or more years of smoking for men and women combined as compared with never smokers. Risk among current and former smokers increased with duration of smoking and average number of cigarettes smoked per day; risk in former smokers decreased significantly in years since quitting. If the multivariate-adjusted relative rate estimates in this study do, in fact, reflect causality, then approximately 12% of colorectal carcinoma deaths among both men and women in the general U.S. population in 1997 were attributable to smoking.

Clinical Dietary Studies A number of case control studies have been reported. Jain et al. (146) examined patients with large bowel malignancy and compared them with population and hospital controls. An increased risk was found in persons with an increased intake of saturated fat as well as calories, total protein, total fat, oleic acid, and cholesterol. The strongest effect was that of saturated fat. Potter and McMichael (147) found that dietary protein was the strongest predictor of colon carcinoma with a two- to threefold relative risk. In a review, Kritchevsky (148) reported that studies for specific vegetables have been carried out with carrots, broccoli, cabbage, lettuce, potatoes, and legumes. Of a total of 105 studies, 67% have shown no association. Prospective studies have examined the relationship between diet and subsequent development of colorectal carcinoma. Phillips and Snowdon (149) found a positive association for the risk of colon carcinoma with egg consumption, coffee intake, and weight greater than 125% of ideal weight. No association was noted for use of meat, cheese, milk, or green salad. Garland et al. (116) failed to note any association between dietary fat, animal or vegetable protein, ethanol, or energy intake and subsequent colorectal carcinoma. The authors did note a negative association between vitamin D and calcium intake and subsequent colorectal carcinoma. Stemmerman Nomura, and Heilbrun (150) studied the relationship between the intake of dietary fat and subsequent colorectal carcinoma during a 15-year follow-up in 7074 Japanese men and found a negative association between dietary total fat and saturated fat intake and colon carcinoma. The strongest effect was found in the right colon, whereas in contrast a weakly positive relationship was found in the rectum. Berry, Zimmerman,

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

and Ligumsky (151), in a study of patients undergoing colonoscopy, analyzed fatty acid and plasma lipid analogs and found that the quality of dietary fat did not influence the development of carcinoma or neoplastic polyps in their population. The fact that in most high-risk countries there is a positive association between fat or meat intake and risk of large bowel malignancy, but results are inconsistent in low-risk countries, implies that there is perhaps a threshold effect by which a minimal level of fat or meat is necessary to effect the development of colorectal carcinoma. There may be a confounding factor whereby dietary fat or protein increases the risk in persons eating low-fiber diets. In a case-control study of pathologically confirmed, single primary carcinomas of the rectum, the risk of rectal carcinoma increased with an increasing intake of kilocalories, fat, carbohydrate, and iron (152). The risk decreased with an increasing intake of carotenoids, vitamin C, and dietary fiber from vegetables. Fiber from grains, calcium, retinal, and vitamin E were not associated with risk. Associations of intake with risk were generally stronger for men than for women, except with vitamin C. The associations for carotenoids, vitamin C, and vegetable fiber persisted after stratification on intake of either kilocalories or fat. Giovannucci et al. (153) evaluated the relation between folate intake and incidence of colon carcinoma in a prospect of cohort study of 88,756 women from the Nurses’ Health Study. There were 442 with new cases of colon carcinoma. Higher folate intake was related to a lower risk of colon carcinoma (relative ratio 0.69) for intake > 400 mg/day compared with intake 200 mg/day after controlling for age; family history of colon carcinoma; aspirin use; smoking; body mass; physical activity; and intake of red meat, alcohol, methionine, and fiber. When intake of vitamins C, D, and E and intake of calcium were also controlled for, results were similar. Women who used multivitamins containing folic acid had no benefit with respect to colon carcinoma after four years of use. After 15 years of use, risk was markedly lower (relative ratio 0.25) representing 15 instead of 68 new cases of colon carcinoma per 10,000 women 55 to 69 years of age. Folate from dietary sources alone was related to a modest reduction in risk for colon carcinoma and the benefit of long-term multivitamin use was present across all levels of dietary intakes. In a large study designed to determine the relationship between fish consumption and risk of the development of carcinoma Fernandez et al. (154) noted a consistent pattern of protection against the risk of selected malignancies; colon OR 0.6 and rectum OR 0.5.

& IRRADIATION There have been sporadic reports of carcinoma of the colon and rectum that develops after radiation therapy for a variety of pelvic malignancies (155–160). The average interval between irradiation and the diagnosis of rectal carcinoma is 15.2 years with a range from 14 months to 33 years (158). There is controversy as to whether the relationship is one of cause and effect or purely coincidental. The characteristics are different from ordinary large bowel carcinoma in that there is a high incidence of mucin-producing carcinoma (53%) (159). Radiation injury was observed in 64% of cases (159). Radiation therapy for prostate carcinoma has been associated with an increased rate of pelvic malignancies,

77

particularly bladder carcinoma. Baxter et al. (161) conducted a retrospective cohort study using Surveillance, Epidemiology, and End Results (SEER) registry data. They focused on men with prostate carcinoma, but with no previous history of colorectal carcinoma, treated with either surgery or radiation who survived at least five years. They evaluated the effect of radiation on development of carcinoma for three sites: definitely irradiated sites (rectum), potentially irradiated sites (rectosigmoid, sigmoid, and cecum), and nonirradiated sites (the rest of the colon). A total of 30,552 men received radiation, and 55,263 underwent surgery only. Colorectal carcinomas developed in 1437 patients: 267 in irradiated sites, 686 in potentially irradiated sites, and 484 in nonirradiated sites. Radiation was independently associated with development of carcinoma over time in irradiated sites but not in the remainder of the colon. The adjusted hazards ratio for development of rectal carcinoma was 1.7 for the radiation group, compared with the surgery only group. Radiation had no effect on development of carcinoma in the remainder of the colon indicating that the effect is specific to directly irradiated tissue. Tami et al. (157) described some characteristics of radiation-associated rectal carcinoma. All four of their patients presented with chronic radiation colitis. Radiation-associated rectal carcinoma has a tendency to be diagnosed in the advanced stage and to have a poor prognosis. Since there are no reliable clinical or laboratory indicators of the presence of a curable colorectal carcinoma in the setting of chronic radiation proctocolitis, they recommend surveillance with a colonoscope should be done 10 years after irradiation in patients with previous pelvic radiotherapy.

& URETERIC IMPLANTATION Several reports have documented the development of neoplasia at or near the site of a ureterosigmoidostomy (162– 167). The risk of sigmoid carcinoma has been estimated to be anywhere from 8.5 to 10.5 and 80 to 550 times greater in patients with ureterosigmoidostomy than in the normal population (166,167). Adenomatous polyps at the level of the ureterosigmoidostomy may be precursors of a subsequent carcinoma. The interval between the implanation of ureters and the occurrence of colonic carcinoma varies from 5 to 41 years. Husmann and Spence (168) reviewed the literature to 1990 and found 94 patients who had colonic neoplasia after ureterosigmoidostomy for vesical exstrophy. The average patient age at diagnosis of neoplasia was 33 years with an average latency of 26 years. Blood in the stool and symptoms and signs of ureteral obstruction were cardinal warning signs. Treatment consisted of local extirpation of the lesion with reimplantation of one or both ureters back into the sigmoid in one third of patients. The preferred treatment is either endoscopic destruction of polyps after biopsy or resection of the involved segment with cutaneous loop diversions. If the polyp is located at the mouth of or immediately adjacent to the ureterocolic anastomosis, caution must be exercised to avoid obstruction of the orifice as a result of vigorous electrocautery. Death from carcinoma occurred in 30 of 49 patients. The current trend is away from this type of urinary diversion, but for those individuals who have already had this operation, periodic endoscopic surveillance appears to be in order.

78

& PART I: COLORECTAL DISORDERS

& CHOLECYSTECTOMY There is considerable epidemiologic evidence to suggest that bile acids play an important role in the development of colorectal malignancy, but their precise action in the initiation or promotion of the neoplastic process remains to be determined (169). A suggested explanation is that, although before cholecystectomy the bile acid pool circulates two or three times per meal, after cholecystectomy the pool circulates even during fasting. This enhanced circulation results in increased exposure of bile acids to the degrading action of intestinal bacteria, a step in the formation of known carcinogens. Hill (170) has noted a concentration of bile acids in populations of the United States and England that is seven times higher than that in Uganda and India. Populations with a high incidence of colorectal carcinoma have high fecal bile acid concentrations in comparison with those with a low incidence. A high proportion of anaerobic bacteria in these populations caused degradation of bile acids to form known carcinogens. Both deoxycholic and lithocholic acids have been shown to be promoters of colon carcinoma in animal models. These secondary bile acids are the products of bacterial dehydroxylation of the primary bile acids cholic and chenodeoxycholic acid, respectively. The primary bile acids are not promoters of carcinogenesis. Jorgensen and Rafaelsen (171) compared the prevalence of gallstone disease in 145 consecutive patients with colorectal carcinoma with gallstone prevalence in 4159 subjects randomly selected from a population. The group of patients had a significantly higher prevalence of gallstone disease than the population (odds ratio, 1.59), whereas cholecystectomies occurred with equal frequency in the two groups. There was a nonsignificantly trend toward more right-sided carcinomas in patients with gallstones than in patients without gallstones. Schernhammer et al. (172) conducted a prospective study of 85,184 women, 877 of whom developed colorectal carcinoma. They found a significant positive association between cholecystectomy and the risk of colorectal carcinoma (RR ¼ 1.21). The risk was highest for carcinomas of the proximal colon (RR ¼ 1.34) and the rectum (RR ¼ 1.58). Lagergren et al. (173) evaluated cholecystectomy and risk of bowel carcinoma with a slightly different result. Cholecystectomized patients identified through he Swedish Inpatient Register were followed up for subsequent carcinoma. In a total of 278,460 cholecystectomized patients, contributing 3,519,682 person-years, followed up for a maximum of 33 years. Cholecystectomized patients had an increased risk of proximal intestinal adenocarcinoma which gradually declined with increasing distance from the common bile duct. The risk was significantly increased for carcinoma (SIR 1.77) and carcinoids of the small bowel (SIR 1.71) and right-sided colon carcinoma (SIR 1.16). No association was found with more distal bowel carcinoma. These results, together with available literature, give substantial evidence for an association between gallstones and colorectal carcinoma, an association that is not due to cholecystectomy being a predisposing factor to colorectal carcinoma. Sporadic findings of an association between cholecystectomy and colorectal carcinoma can be explained by the above relationship. Wynder and Reddy (174) were impressed by the correlation between dietary fat intake and colon carcinoma. These authors reasoned that the dietary

fat content raised both the concentration of anaerobic bacteria and the amount of bile acid and cholesterol substrates in the gut, thus enhancing the production of bile acid and cholesterol metabolites, which may be the proximate carcinogens. Considerable attention has been directed to absence of the gallbladder and its possible relationship to the development of colorectal carcinoma. Many studies have been published offering arguments for and against such an association. In their comprehensive review of the controversy, Moore-head and McKelvey (175) found series that suggested an increased relative risk of developing colorectal carcinoma in the range of 1.59 to 2.27. Higher relative risks have been reported up to 3.5 for right-sided carcinoma in women, with the highest report being 4.5 for sigmoid lesions (176–178). In a total Icelandic population prospective study of 3425 individuals who underwent cholecystectomy and were followed 8 to 33 years, Nielsen et al. (179) found a relative risk of carcinoma in men of 2.73. McFarlane and Welch (180) found an overall odds ratio of 2.78, but it was 6.79 for right-sided lesions. However, this concept is not universally supported. Abrams, Anton, and Dreyfuss (181) found no relationship between cholecystectomy and the subsequent occurrence of proximal colonic carcinoma. Kune, Kune, and Watson (182) and Kaibara et al. (183) found no statistical association between previous cholecystectomy and the risk of colorectal carcinoma, either in general or in any subsite, age, or sex. Ekbom et al. (184), in a populationbased study of 62,615 patients who underwent cholecystectomy, found no overall excess risk of colorectal carcinoma but observed an increased risk among women for right-sided colon carcinoma 15 years or more after operation. In a casecontrol study conducted by Neugent et al. (185), no significant association was found between cholecystectomy and adenomatous polyps or carcinoma. Even if an association between large bowel neoplasia and biliary tract disease were confirmed, it is possible that there is no direct cause-and-effect relationship. It may be that the diet predisposing to one disease increases the risk of developing the other disease. Resolution of this controversy would be of considerable clinical significance by clearly exposing a readily identifiable at-risk group and offering more intensive screening opportunities to them.

& DIVERTICULAR DISEASE Most surgeons believe that because of the frequency with which both colon carcinoma and diverticular disease exist, it is not uncommon to see both conditions concomitantly without necessarily invoking a cause-and-effect relationship. A study by Boulos et al. (186) hinted that patients with diverticular disease might constitute a group at higher risk of developing neoplasia. Morini et al. (187) ascertained that adenomas and carcinomas were detected more often in patients with diverticular disease, with an overall odds ratio of 3.0. When examined separately, adenomas maintained their significantly higher frequency but no difference was observed for carcinomas. & ACTIVITY AND EXERCISE Persky and Andrianopoulos (188) reviewed studies examining the relation between exercise levels and risk of carcinoma. The authors found a relative risk increase of 1.3 to 2.0 for individuals with sedentary jobs. This association held true only for colon carcinoma, not for rectal

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

carcinoma. Thune and Lund (189) also found a reduced risk for colon carcinoma for men and women who engaged in physical activity at least 4 hours per week. The reduced risk was more marked in the proximal colon. No association between physical activity and rectal carcinoma was observed in men or women. In a case-control study of men with colorectal carcinoma in Japan, physical inactivity based on occupational category increased the risk from the lowest to the highest levels 1.32 to 1.92 times (rectum and proximal colon, respectively) (136). In a more recent report contrary results were found by Slattery et al. (190). They conducted a population-based study of 952 incident cases of carcinoma in the rectum and rectosigmoid junction with 1205 age- and sex-matched controls in Utah and Northern California at the Kaiser Permanente Medical Care Program (190). Vigorous physical activity was associated with reduced risk of rectal carcinoma in both men and women (OR ¼ 0.60, for men and 0.59 for women). Among men, moderate levels of physical activity were also associated with reduced risk of rectal carcinoma (OR ¼ 0.70). Participation in vigorous activity over the past 20 years conferred the greatest protection for both men and women (OR ¼ 0.55 for men and 0.44 for women). In another case-controlled study among men, increased physical activity (2 hr/wk) was associated with reduced risk for advanced adenomas (OR ¼ 0.4) and for nonadvanced adenomas (OR ¼ 0.8) (191). Colbert et al. (192) examined the association between occupational and leisure physical activity and colorectal carcinoma in a cohort of male smokers. Among the 29,133 men aged 50 to 69 years in the Alpha-Tocopherol, BetaCarotene Cancer Prevention study, 152 colon and 104 rectal carcinomas were documented during up to 12 years of follow-up. For colon carcinoma, compared with sedentary workers, men in light occupational activity had a relative risk of 0.60, whereas those in moderate/heavy activity had a relative risk of 0.45. For rectal carcinoma, there were risk reductions for those in light (relative risk 0.71) and moderate/heavy occupational activity (relative risk 0.50). These data provide evidence for a protective role of physical activity in colon and rectal carcinoma. Others support the concept of physical activity as being protective against the development of colon carcinoma (193,194). Various mechanisms for this protective effect have been extensively cited. Quadrilaters and Hoffman-Goetz (195) reviewed the published evidence of physical activity and the hypothesized mechanisms. These mechanisms included changes in gastrointestinal transit time, altered immune function and prostaglandin levels, as well as changes in insulin levels, insulin-like growth factors, bile acid secretion, serum cholesterol and gastrointestinal and pancreatic hormone profiles. There is currently little data to support any of the hypothesized biological mechanisms for the protective effect of exercise on colon carcinoma. It is likely that no one mechanism is responsible for the risk reduction observed in epidemiological and animal studies and therefore, the observed benefits of physical activity in colon carcinoma may be a combination of these and other factors.

& OTHER FACTORS A host of seemingly totally unrelated factors have been suggested to play a role in colorectal carcinoma genesis. Many of

79

these are cited below. The risk of colorectal carcinoma is increased following adenocarcinoma of the small bowel (196). A factor that has been studied but for which no definitive association has been documented is obesity. With regard to bowel function, a meta-analysis of 14 case-control studies revealed statistically significant risks for colorectal carcinoma associated with both constipation and the use of cathartics (pooled odds ratios, 1.48 and 1.46, respectively) (197). A meta-analysis of published data revealed that women with a history of breast, endometrial, and ovarian carcinomas have an increased relative risk of developing colorectal carcinoma of 1.1, 1.4, and 1.6, respectively (198). Large bowel malignancy correlates with the distribution of endocrine-dependent neoplasms (e.g., breast, endometrium, ovary, or prostate) and arteriosclerotic heart disease. It would be important to know whether there is a protective effect of hormone replacement therapy against colorectal carcinoma. Increased parity was reported to be associated with a decline in the risk of colon carcinoma (odds ratio, 0.44 for women with five children or more relative to nulliparous women), but not for rectal carcinoma (199). The association with colon carcinoma is restricted to women 50 years of age or older. A population based case-control study of postmenopausal women revealed that when compared with women who never used hormone replacement therapy, recent users had a relative risk of 0.54 for colon carcinoma and a relative risk of 0.91 for rectal carcinoma (200). Estrogen replacement therapy is associated with a substantial decreased risk in colon carcinoma (RR ¼ 0.71) (201). Users of 1 year or less have a relative risk of 0.81, while users of 11 years or more have a relative risk of 0.54. Other investigators found little overall association between colon carcinoma and oral contraceptive use, parity, age at first birth, hysterectomy, oophorectomy status, or age at menopause (202,203). In a meta-analysis of 18 epidemiologic studies of postmenopausal hormone therapy and colorectal carcinoma Grodstein et al. (204) found a 20% reduction in risk of colon carcinoma and a 19% decrease in the risk of rectal carcinoma for postmenopausal women who had ever taken hormone therapy compared with women who never used hormones. Much of the apparent reduction in colorectal carcinoma was limited to current hormone users (relative risk ¼ 0.66). From a case-control study among women, ever use of hormone replacement therapy was more strongly associated with reduced risk of advanced adenomas relative to polyp-free controls (OR ¼ 0.4) than with reduced risk of nonadvanced adenomas (OR ¼ 0.7) (191). Baris et al. (205) studied the patterns of carcinoma risk in Sweden and Denmark in 177 patients with acromegaly. Increased risks were found for colon (standardized incidence ratio ¼ 2.6) and rectum (standardized incidence ratio ¼ 2.5). Among other risks, the increased risk for several carcinoma sites among acromegaly patients may be due to the elevated proliferative and antiapoptotic activity associated with increased circulating levels of insulin-like growth factor-1. Bile acids are suspected from both clinical and experimental studies to have a role in colon carcinogenesis. The twofold increased mortality from colorectal carcinoma is apparent only 15 to 20 years after gastric surgery. It is suggested that the increased mortality from colorectal carcinoma after gastric surgery may be due to altered bile acid

80

& PART I: COLORECTAL DISORDERS

metabolism. To determine the relationship between bile acids and increased risk of colorectal neoplasms after truncal vagotomy, Mullen et al. (206) conducted a prospective screening study of 100 asymptomatic patients who had undergone truncal vagotomy at least 10 years previously. The patients were investigated by barium enema, colonoscopy, and gallbladder ultrasonography. Control data were obtained from forensic autopsy subjects. The incidence of neoplasms > 1 cm in the vagotomized group was 14% (11 adenomas and three carcinomas) and 3% in controls. The authors found increased proportions of chenodeoxycholic acid and lithocholic acid and decreased proportions of cholic acid in the duodenal bile of vagotomized patients and believe that these abnormalities in bile acid metabolism may help to explain the increased risk of colorectal neoplasia 10 years after truncal vagotomy. On the other hand, Fisher et al. (207), in their cohort study of 15,983 males, found no elevation of risk of large bowel carcinoma following gastric surgery for benign disease. Little et al. (208) tested the hypothesis relating bile acids, calcium, and pH to colorectal carcinoma in a large sample of asymptomatic subjects who had participated in fecal occult blood screening. Fecal samples were obtained from 45 cases of carcinoma, 129 subjects with adenoma, 167 fecal occult blood negative controls, and 155 fecal occult blood positive subjects in whom no-carcinoma or adenoma was found. No association between colorectal carcinoma and fecal bile acids or pH was observed. Although there was no overall association between colorectal adenomas and fecal bile acids or pH, villous adenomas were associated with increasing concentrations of major bile acids and decreasing concentration of minor bile acids, and there was a suggestion of an inverse association with an acid pH. High levels of fecal calcium were associated with a reduced risk of both colorectal carcinoma and adenoma, but this was not statistically significant. Their study does not support an association between colorectal carcinoma and fecal bile acids. However, there is evidence that increases in major bile acids are associated with villous adenomas. There is diverse evidence suggesting that intracolonic production of oxygen radicals may play a role in carcinogenesis (209). The relatively high concentrations of iron in feces, together with the ability of bile pigments to act as iron chelators that support Fenton chemistry, may very well permit efficient hydroxyl radicals generation from superoxide and hydrogen peroxide produced by bacterial metabolism. Such free radicals generation in feces could provide a missing link in our understanding of the etiology of colon carcinoma: the oxidation of procarcinogens either by fecal hydroxyl radicals or by secondary peroxyl radicals to form active carcinogens or mitogenic neoplastic promoters. Intracolonic free radical formation may explain the high incidence of carcinoma in the colon and rectum compared with other regions of the gastrointestinal tract, as well as the observed correlation of a higher incidence of colon carcinoma with red meat in the diet, which increases stool iron, and with excessive fat in the diet, which may increase the fecal content of procarcinogens and bile pigments. Epidemiologic studies and laboratory research have indicated an association between the metabolic activity of the intestinal microflora and carcinoma of the large bowel

(210). It has been suggested that activation of procarcinogens could be mediated enzymatically by intestinal bacteria. The levels of incriminated colonic bacterial enzymes are increased by dietary fat and inhibited by certain dietary fibers. Organic extracts of feces contain a mutagenic substance, presumably derived from bacterial metabolism in the large bowel. Whether this substance or some other organic chemical is the putative proximate carcinogen remains speculative, but the evidence continues to point to intestinal bacteria as the metabolic intermediary in colon carcinoma. High iron stores may increase the risk of colorectal carcinoma through their contribution to the production of free oxygen radicals. Knekt et al. (211) studied serum iron, total iron binding capacity, and transferrin saturation levels in a cohort of 41,276 subjects ranging in age from 20 to 74 years. The authors found a relative risk of 3.04 for colorectal carcinoma in subjects with transferrin levels exceeding 60%. Asthmatic patients have a reported elevated relative risk of colon (1.17) and rectum (1.28) carcinoma (212). The prevalence of colon carcinoma in patients with Barrett’s esophagus is 7.6% compared with 1.6% in a control population (213). Men with esophageal adenocarcinoma may be more likely to be diagnosed with colorectal carcinoma in their lifetime than expected. The opposite association may exist for women (214). A relative risk of 3.04 has been calculated for colorectal carcinoma as a result of anthranoid laxative abuse (215). Younes, Katikaneni, and Lechago (216) reported that 25% of patients who underwent appendectomy and were found to have mucosal hyperplasia were associated with adenocarcinoma of the colon. Plasma C-reactive protein concentrations are elevated among persons who subsequently develop colon carcinoma. These data support the hypothesis that inflammation is a risk factor for the development of colon carcinoma in average-risk individuals (217). Woolcott et al. (218) found colon carcinoma risk was inversely associated with coffee. Relative to those drinking fewer than one cup of coffee per day, the ORs for those drinking two cups was 0.9, for those drinking three to four cups 0.8, and for those drinking five or more cups 0.7. The reduced risk estimates were more pronounced with carcinoma of the proximal colon than the distal colon. Rectal carcinoma risk was not associated with either coffee or tea. In a study of premorbid and personality factors, aggressive hostility was the only variable found to be significant between colon carcinoma patients and controls (219).

& JUVENILE VS. ADULT CARCINOMA To add further confounding information to the matter, two different types of carcinoma may exist. Concerning the wide age range that colorectal carcinoma encompasses, Avni and Feuchtwanger (27) wrote a thought-provoking editorial in which they made a sharp differentiation between "juvenile" and "adult" carcinoma and suggested that a search be made for different etiologies. To support the concept that two different types of colon carcinoma may exist, the authors cited the following observations: 1. In the nonwhite population, the juvenile form occurs up to 16 times more frequently than in the white population, whereas the adult form occurs 10 times more frequently in the white population.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

2. Mucinous carcinomas comprise only 5% of all colonic carcinomas, although in the young age group they comprise 76%. 3. In adult patients with colon carcinoma, coexisting polyps can be found in 40% to 50% of patients, whereas in the juvenile group polyps are exceedingly rare. 4. Alleged nutritional factors in the genesis of colon carcinoma must be present for many years, which cannot be the case in the juvenile group. Moreover, diet in the nonwhite population in whom the juvenile type is more frequent differs from that of the white population. 5. In the adult population with colon carcinoma, a family history of disease is found in 20% to 30%, although this factor is almost nonexistent in the juvenile type. Only further investigation will resolve this issue.

& PROSPECTS FOR PREVENTION Epidemiologic evidence suggests that diet is the principal factor in the cause of colorectal carcinoma. Ingestion of excessive amounts of fat appears to be the major factor. Also noteworthy is the interrelationship between the varying relative amounts of fat and fiber in the diet. For example, in an experimental study on colorectal carcinogenesis by Galloway et al. (220), manipulation of diet resulted in significant differences in alteration of the surface architecture of the colonic mucosa. The high-fat, low-fiber diet was associated with the greatest risk for macroscopic malignant production, and the low-fat, highfiber diet, with the lowest risk. Furthermore, the sources of fat vary in the degree of their promotional effect (128). Fiber is considered to inhibit carcinoma, but only the type of fiber found in whole grain cereals, fruits, and vegetables, which contain large amounts of uronic acid, is effective. The exact nature of fat and fiber sources is yet to be delineated. An effective prevention strategy should be based on an understanding of the pathogenesis of carcinoma, but no such definitive understanding exists. From the available evidence it appears that a high intake of animal fat and protein promotes the formation of colon carcinoma and that certain cruciferous vegetables exert a protective influence. The two-step concept of carcinogenesis includes (i) initiation, about which little is known in human carcinoma, and (ii) promotion, which presumably takes a long time to complete. We may well be able to capitalize on the latter fact by the administration of inhibitors during the promotional phase, a concept realized experimentally by Nigro and Bull (128). Their studies suggested that an appropriate strategy for prevention of carcinoma of the large bowel would be a program aimed at a 10% reduction of fat consumption (i.e., from 40% to 30% of calories) and adoption of a more varied type of fat. The daily addition of 25 to 30 g of dietary fiber, especially grains and vegetables that contain cellulose and uronic acid, may prove effective. A third recommendation is to include in the diet chemical inhibitors such as selenium, retinoids, and plant steroids. Other studies have suggested that other factors, such as calcium and sulfur compounds present in garlic and onions, may be of value in the inhibition of colon carcinoma (129). Antioxidants such as b-carotene, vitamin C, vitamin E, and folic acid have been assessed (221). Indivi-

81

duals with a significant inheritance factor for colon carcinoma may require a supplement to enhance inhibition. Cassidy, Bingham, and Cummings (222) reported a strong inverse association and suggested a potentially important role for starch in the protection against colorectal carcinoma. This corresponds with the hypothesis that fermentation in the colon is the mechanism for preventing colorectal carcinoma. Experiments in animals and two epidemiologic studies in humans suggest that aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) may be protective against colon carcinoma. Thun, Namboodiri, and Heath (223) tested this hypothesis in a prospective mortality study of 662,424 adults who provided information in 1982 on the frequency and duration of their aspirin use. Death rates from colon carcinoma were measured through 1988. The possible influence of other risk factors for colon carcinoma was examined in multivariate analyses for 598 case patients and 3058 matched control subjects drawn from the cohort. Death rates from colon carcinoma decreased with more frequent aspirin use in both men and women. The relative risk among persons who used aspirin 16 or more times per month for at least 1 year was 0.60 in men and 0.58 in women. The risk estimates were unaffected when they excluded persons who reported at entry into the study that they had a malignancy, heart disease, stroke, or another condition that might influence both their aspirin use and their mortality. Adjustment for dietary factors, obesity, physical activity, and family history did not alter the findings significantly. No association was found between the use of acetaminophen and the risk of colon carcinoma. The authors concluded that regular aspirin use at low doses may reduce the risk of fatal colon carcinoma. Rosenberg et al. (224) assessed NSAID use in relation to risk of human large bowel carcinoma in a hospital-based casecontrol study of 1326 patients with colorectal carcinoma and 4891 control patients. For regular NSAID use that continued into the year before the interview, the multivariate relative risk estimate was 0.5. The inverse association was apparent for both colon and rectal carcinoma in men and women and in subjects younger and older than 60 years. Regular NSAID use that had been discontinued at least 1 year previously and nonregular use were not associated with risk. Almost all regular NSAID use was of aspirin-containing drugs. The present data suggest that the sustained use of NSAIDs reduces the incidence of human large bowel carcinoma. Smalley et al. (225) studied how patterns of use (duration, dose, and specific drug) of NSAIDs affected the incidence of colorectal carcinoma. The population-based retrospective cohort study of 104,217 persons aged 65 years or older had at least 5 years of enrollment. Incident histologically confirmed colorectal carcinoma was documented. Users of nonaspirin NSAIDs for at least 48 months of the previous 5 years had a relative risk of 0.49 for colon carcinoma when compared with those of no use of NSAIDs. Among those with more than 12 months of cumulative use, those using NSAIDs in the past year (recent users) had a relative ratio of 0.61 whereas those with no recent use had a relative ratio of 0.76. No specific NSAID offered a unique protective effect and low doses of NSAIDs appeared to be at least as effective as higher doses. Protection was most pronounced for rightsided lesions. The relative risk among recent users with

82

& PART I: COLORECTAL DISORDERS

more than 12 months of cumulative use was 0.81 for rectal carcinoma and 0.77 for left-sided carcinoma and 0.48 for right-sided colon carcinoma. In this elderly population, long-term use of nonaspirin NSAIDs nearly halved the risk of colon carcinoma. This study was consistent with previous studies that suggest that duration of use but not daily dose of NSAIDs is an important factor for chemoprevention. Their data also suggest that the protective effect is shared by most NSAIDs, and not confined to a small number of these drugs. Baron et al. (226) performed a randomized doubleblind trial of aspirin as a chemopreventive agent against colorectal adenomas. They randomly assigned 1121 patients with a recent history of histologically documented adenomas to receive placebo (372 patients), 81 mg of aspirin (377 patients), or 325 mg of aspirin (372 patients) daily. Followup colonoscopy was performed at least one year after randomization in 97% of patients. The incidence of one or more adenomas was 47% in the placebo group, 38% in the group given 81 mg of aspirin per day and 45% in the group given 325 mg of aspirin per day. Unadjusted relative risks of any adenomas (as compared with the placebo group) were 0.81 in the 81-mg group and 0.96 in the 325-mg group. For advanced neoplasms (adenomas measuring at least 1 cm in diameter or with tubulovillous or villous features, severe dysplasia, or invasive carcinoma), the respective relative risks were 0.59 and 0.83. They concluded low dose aspirin has a moderate chemopreventive effect on adenomas in the large bowel. Since adenoma development can be used as a surrogate marker for the development of carcinoma, the reduction in adenoma development by extrapolation would result in a decrease in the incidence of carcinoma. In contrast to most observational studies, the randomized Physician’s Health Study found no association between aspirin use and colorectal carcinoma after 5 years (227). In a randomized prospective cohort study 22,071 healthy male physicians who were 40 to 84 years of age in 1982 were given 325 mg of aspirin every other day. In 1988 the aspirin arm of the randomized trial was stopped early. Participants then chose to receive either aspirin or placebo for the rest of the study. Colorectal carcinoma was diagnosed in 341 patients during the study period. Over 12 years of follow-up, random assignment to aspirin was associated with a relative risk of colorectal carcinoma of 1.03. The relative risk for colorectal carcinoma in persons who used aspirin frequently after 1988 was 1.07. In the Physicians Health Study, both randomized and observational analyses indicate that there is no association between the use of aspirin and the incidence of colorectal carcinoma. In their review, Burke et al. (228) reported that in population-based observational studies, people had lower rates of colorectal carcinoma if they were taking various agents, including nonsteroidal anti-inflammatory drugs, calcium, and folate. In placebo-controlled trials in patients with familial adenomatous polyposis and in patients with sporadic colon adenomas, nonsteroidal anti-inflammatory drugs reduced the rates of adenomas, and there is a biologic rationale that they would be effective in reducing colorectal carcinoma as well. Large-scale chemoprevention trials sponsored by the National Cancer Institute are under way (229). Agents being evaluated include piroxicam, sulindac, aspirin, acarbose (a-glucosidase) inhibitors, and calcium carbonate.

Targeted populations include patients with previous adenoma, FAP, history of multiple polyposis, and subjects at risk for colon carcinoma.

& PATHOLOGY & MACROSCOPIC APPEARANCE On gross examination, adenocarcinoma of the colon appears as one of four fairly distinctive types— ulcerative, polypoid, annular, or diffusely infiltrating. The ulcerative carcinoma, the most common type, presents as a roughly circular mass with a raised, irregular, everted edge and a sloughing base. It is confined to one aspect of the bowel wall but may occupy a larger portion of the bowel circumference (Fig. 6). The polypoid, or cauliflower-type, carcinoma presents as a large fungating mass that projects into the lumen and is often of a low-grade malignancy. The ascending colon is a site of predilection (230) (Fig. 7). In approximately 10% of cases, the cut surface of the growth may have a gelatinous appearance, due to abundant mucin secretion, and this type has been referred to as a colloid carcinoma (230). The annular, or stenosing, carcinoma occupies the entire circumference of the bowel wall. The extent in the long axis is variable. The bowel lumen is usually considerably compromised (Fig. 8), and the proximal bowel may demonstrate varying degrees of dilation. Such carcinomas occur with greatest frequency in the transverse and descending colon (230). The diffusely infiltrating carcinoma produces a diffuse thickening of the intestinal wall and for the most part is covered with intact mucosa. It is extensively infiltrating, although it preserves the layers of the gastrointestinal wall. It occurs more often in the rectosigmoid, but any portion of the colon may be involved. This variety is similar to linitis plastica of the stomach. It is the type of carcinoma commonly associated with ulcerative colitis (Fig. 9). A review of the literature by Papp, Levine, and Thomas (231) revealed only 85 documented cases of primary linitis

FIGURE 6 & Macroscopic features of an ulcerated adenocarcinoma.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

FIGURE 7 & Macroscopic appearance of a polypoid adenocarcinoma.

plastica of the colon and rectum, characterized by presentation in younger patients and associated with metastatic disease, a higher mortality, and insidious growth, often making detection difficult.

& MICROSCOPIC APPEARANCE The histological appearance of carcinoma may vary considerably, with the major importance being related to prognosis. The lesion may be well differentiated (20%) (Fig. 10), moderately differentiated (60%) (Fig. 11), or poorly differentiated (20%) (Fig. 12) (231). The incidence of lymph node metastases is about 25%, 50%, and 80% in low-, average-, or high-grade malignancy, respectively. Furthermore, the histologic grade influences survival, with corrected 5-year survival rates of 77%, 61%, and 29%, for low-, average-, and high-grade rectal malignancies, respectively (230). Broders (232) popularized a method that divides adenocarcinoma into four grades. In grade 1, 75% to 100% of the cells are differentiated; in grade 2, 50% to 75%; in grade 3, 25% to 50%; and in grade 4, 0% to 25%. The principle of grading by differentiation is based on the biologic law that the higher the degree of differentiation, the less the power of reproduction; therefore it might be anticipated that well-differentiated carcinomas would proliferate at a slower pace than those that are comparatively undifferentiated. One difficulty in the application of histologic grading is the lack of uniformity in the degree of differentiation throughout the neoplasm. In

FIGURE 8 & Macroscopic features of an annular carcinoma. Arrows indicate associated adenomas.

83

FIGURE 9 & Macroscopic appearance of an infiltrating adenocarcinoma.

general, malignant cells are less differentiated at the invading margins than on the surface. Dukes (233) carefully differentiated his classification from that of Broders (232) by noting that the two methods answer different questions. Histologic grade is essentially an estimate of the pace of growth, whereas classification into Dukes’ A, B, and C cases is a measurement of the boundaries reached. Both methods permit grouping into cases with favorable and unfavorable outcome. Jass et al. (234) proposed a grading system that includes the parameters of tubule configuration, advancing margin, and lymphocytic infiltration. Contrary to what might be expected and what is generally believed, Gibbs (235) reported that rare undifferentiated carcinomas of the large intestine tend to spread circumferentially and do not readily give rise to lymphatic or hematogenous metastases. Affected patients may have a good prognosis. Colloid, or mucus-producing, lesions may present with varying degrees of differentiation and are said by some to have a poor prognosis (Fig. 13). Colloid adenocarcinoma can be classified as extracellular or intracellular, according to the predominant location of the mucin. Most colloid carcinomas are extracellular, with only approximately 2% of all carcinomas of the colon and rectum being the pure signet-ring cell variety. With the latter variety, the patient is unlikely to survive more than 2 years from the time of diagnosis (230). The largest number of mucinous

FIGURE 10 & Microscopic appearance of a well-differentiated adenocarcinoma with well-developed glands. Source: Courtesy of L.R. Be`gin, M.D., Montreal, Quebec.

84

& PART I: COLORECTAL DISORDERS

FIGURE 11 & Microscopic appearance of a moderately differentiated adenocarcinoma with gland formation that is less well defined. Source: Courtesy of L.R. Be`gin, M.D., Montreal, Quebec.

FIGURE 13 & Microscopic features of a mucin-producing adenocarcinoma with abundant extracellular mucin. Source: Courtesy of L.R. Be`gin, M.D., Montreal, Quebec.

carcinomas are found in the rectum, but the relative incidence of mucinous carcinoma is higher in the right colon (236). The primarily intracellular variety of mucinous carcinoma is classified as a signet-ring cell because the mucus pushes the nucleus to the periphery and thereby gives the cells their characteristic appearance. In a review of 426 patients with carcinoma of the rectum and rectosigmoid seen at Memorial Sloan-Kettering Cancer Center, Bonello, Sternberg, and Quan (237) found 4% of the carcinomas to be of the signet-ring cell type, and this accounted for 0.4% of all rectal carcinomas. Umpleby, Ranson, and Williamson (238) compared the clinical and pathologic features of 54 mucinous carcinomas of the large intestine with 576 nonmucinous carcinomas. Lesions were categorized as mucinous if they contained at least 60% mucin by volume. Those with moderate mucin content (60% to 80%) were indistinguishable in behavior from nonmucinous lesions. By contrast, those with a high mucin content ( >80%) showed several differences from nonmucinous carcinomas. They had a more proximal distribution through the large intestine, they comprised a greater fraction of carcinomas in the under-50 age group (24% vs. 7%), they were more likely to be Dukes’ stage D (58% vs. 31%), and local fixity was more common (70% vs. 37%). Consequently, the overall resection rate was reduced from 90% to 73%, the curative resection rate from 69% to 42%, and the 5-year

survival rate from 37% to 18%. Umpleby, Ranson, and Williamson (238) concluded that colorectal carcinomas of high mucin content require wide excision, tend to recur locally, and carry a poor prognosis. The dramatic difference in this report from others in the literature may be due to this study’s strict definition of a mucinous carcinoma as one with > 80% mucin content. In a review of 540 cases of colorectal carcinoma, Okuno et al. (239) found that mucinous carcinomas accounted for 6.4% of cases. Such carcinomas were more common in patients 39 years of age or younger and in women patients. They were most commonly located in the rectum, followed by occurrence in the right colon; however, the relative incidence was higher in the right colon (40.5% vs. 12.5%). These carcinomas were characterized by infiltration of the surrounding tissues (24.3% vs. 7.8%), positive lymph node involvement (75.7% vs. 48.6%), and peritoneal implant (21.6% vs. 4.1%). The cumulative 5and 10-year survival rates after resection of mucinous carcinoma were 45.5% and 39.8%, respectively; those after curative resection were 77.4% and 63.5%. The authors suggested the need for aggressive lymph node dissection and wide excision of the surrounding tissues for mucinous carcinoma, with special attention paid to local recurrence. This microscopic variety of carcinoma is associated with a higher incidence of metastases and a higher incidence of associated synchronous polyps and carcinoma than nonmucinous carcinomas. The clinical relevance of this association is that it is necessary to perform colonoscopy on patients with mucinous carcinoma (236). Anthony et al. (240) reported the largest series of signet cell carcinoma of the colon and rectum. There was equal distribution between the right and left colon. Synchronous carcinomas were present in 14%. Nodal or metastatic disease was present in 72% of patients at the time of diagnosis. The 5-year actuarial survival rate was 22%. Patient mortality was due to carcinomatosis in all 22 patients who died. Parenchymal liver involvement occurred in only two patients (9%). Ovarian metastases have been reported in 25% to 60% of patients at time of diagnosis and hence bilateral salpingo-oophorectomy should probably accompany the original resection. Primary linitis plastica of the colon and rectum is an uncommon entity with a poor prognosis. Papp, Levine,

FIGURE 12 & Microscopic features of a poorly differentiated adenocarcinoma with pleomorphic cells and little recognizable gland formation. Source: Courtesy of L.R. Be`gin, M.D., Montreal, Quebec.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

and Thomas (231) found 85 cases reported in the literature. Shirouza et al. (241) classified linitis plastica into two types according to histologic growth pattern—the more common scirrhous type and the lymphangiosis type. The scirrhous type is composed mainly of poorly differentiated or signet-ring cells and is accompanied by a severe desmoplastic reaction. The lymphangiosis type is composed mainly of moderately differentiated cells, frequently with glandular formation. The characteristic diffuse tubular thickening and rigidity are the result of fibrotic reaction around infiltrating malignant cells, such as in desmoplasia (see Fig. 9). The microscopic appearance is that of poorly differentiated pleomorphic cells (see Fig. 12). The pathologist can be instrumental in guiding the surgeon to the appropriate clinical management. Surgeons are often reluctant to recommend major operative procedures for patients with "minimally invasive" carcinoma. Hase et al. (242) conducted a study to determine the long-term outcome after curative resection of colorectal carcinomas that extend only into the submucosa. Seventy-nine patients who underwent curative resection were followed for at least 5 years. Formal operation followed attempted endoscopic removal in 25 patients. Lymph node metastases, found in 11 of 79 patients (13.9%), were associated with a worse outcome; 36.4% of node-positive patients developed recurrence vs. only 5.9% of node-negative patients. The cumulative survival rate was also worse in node-positive vs. node-negative patients; 72.7% vs. 91.1% at 5 years and 45.5% vs. 65.3% at 10 years. Five histopathologic characteristics were identified as risk factors for lymph node metastases: (i) small clusters of undifferentiated carcinoma cells ahead of the invasive front of the lesion (‘‘tumor budding’’), (ii) a poorly demarcated invasive front, (iii) moderately or poorly differentiated malignant cells in the invasive front, (iv) extension of the carcinoma to the middle or deep submucosal layer, and (v) malignant cells in lymphatics. Whereas patients with three or fewer risk factors had no nodal spread, the rate of lymph node involvement with four or more risk factors was 33.3% and 66.7%, respectively. Appropriate bowel resection with lymph node dissection is indicated if a lesion exhibits more than three histologic risk factors for metastasis. The coexistence of two or more cell types in colonic malignancy has been reported. Novello et al. (243) reported cases containing areas with clear adenocarcinomatous and squamous differentiation and morphologic as well as histochemical evidence of neuroendocrine differentiation.

& DEPRESSED CARCINOMA A unique macroscopic type of carcinoma rarely recorded in the Western world is the superficial depressed type, which represents de novo growth and is clearly different from that usually seen in the polyp-carcinoma sequence. This type of carcinoma, frequently described in the Japanese population, has a strong tendency to develop into invasive and advanced carcinoma. At our institution, Begin, Gordon, and Alpert (244) described a case of endophytic malignant transformation in a flat adenoma of the colon. The deep component was a well-differentiated adenocardnoma extending into the serosa and probably was an example of the depressed carcinomas described in the Japanese literature.

85

FIGURE 14 & Depressed type of colorectal carcinoma. Definition: Carcinoma whose surface is lower than that of neighboring normal mucosa. II c and II c þ II a types of carcinoma meet this definition. Source: From Ref. 246.

Kudo et al. (245) published an excellent description of the gross and microscopic features with the subtle nuances of making the diagnosis. The authors reported that the depressed type of invasive carcinoma represented 15.5% of all invasive carcinomas diagnosed on 30,311 endoscopic examinations. The definition of a depressed carcinoma is shown in Fig. 14. To detect the depressed type of carcinoma, it is important to pay special attention to slight changes in the mucosal color during endoscopy—slight redness or in some cases, pallor (Fig. 15). The detection rate was about 1 in 1000 endoscopic examinations. Indigo carmine spraying of suspicious areas reveal the underlying pathology of the depressed type of carcinoma (Fig. 16). Histologic confirmation of such a lesion is seen in Fig. 17. The degree of submucosal extension is depicted in Fig. 18. The management of patients with depressed carcinomas is outlined as an algorithm in Fig. 19. In sm 1 a and sm 1 b, extension without vessel invasion, strip biopsy is suitable. For sm 1 b with vessel invasion, resection is recommended because of the risk of lymph node metastases. For sm 2 and sm 3 (see Fig. 18), resection is the preferred treatment. Kubota et al. (246) examined 300 surgically resected specimens with a dissecting microscope and found 297 adenomas (240 polypoid, 32 flat, and 25 depressed) along with three nonpolypoid carcinomas. Nonpolypoid adenomas were mostly found in the transverse and descending colon. Almost all depressed adenomas were < 3 mm in

FIGURE 15 & Normal endoscopic image after indigo carmine spraying. Source: From Ref. 246.

86

& PART I: COLORECTAL DISORDERS

FIGURE 16 & Normal endoscopic image. Source: From Ref. 246.

dimension (96%), and almost all flat adenomas were < 3.5 mm in dimension (96.9%). The 3 minute carcinomas ranged in size from 2.4 to 2.9 mm. Minamoto et al. (247) suggested that superficial-type adenocarcinomas show rapid growth, aggressive behavior, and may not progress by the adenoma-carcinoma pathway but may rise from a very small superficial-type adenoma. Tada et al. (248) conducted a clinicopathologic study of 62 flat colorectal carcinomas and 80 polypoid colorectal carcinomas. The authors found the former to be smaller, more often in the proximal colon, and away from the rectosigmoid area, less frequently well differentiated, with fewer adenomatous remnants, and with more frequent deep invasion and lymphovascular permeation. Iishi et al. (249) reported on a series of 256 early colorectal carcinomas (Dukes’ A and B) of which 8% were superficial early carcinomas defined as < 3 mm in height. These lesions were found scattered throughout the large intestine and were often observed as reddish spots that were easily overlooked without careful observation. Histologically, 90% of these were well differentiated, 24% reached the submucosal layer, and 86% were not associated with adenoma. Flat lesions are being increasingly recognized with new colonoscopic techniques. Togashi et al. (250) reported 10,939 consecutive high-resolution video colonoscopies and indigocarmine spraying to detect flat lesions. All lesions suggesting neoplastic changes were removed by

FIGURE 17 & Carcinoma invading the submucosal layer (hematoxylin-eosin stain, original magnification X10). Source: From Ref. 246.

FIGURE 18 & Classification of submucosal invasion of early colorectal carcinoma: sm 1 ¼ invasion limited to the upper third of the submucosa; sm 1 a ¼ horizontal invasion limited to less than one quarter of the width of the carcinoma component in the mucosa; sm 1 b ¼ invasion limited to one quarter to half of the width of the carcinoma component in the mucosa; sm 1 c ¼ invasion extending to more than half of the width of the carcinoma component in the mucosa; sm 2 ¼ invasion limited to the middle third of the submucosa; sm 3 ¼ invasion of the lower third of the submucosa. Source: From Ref. 246.

polypectomy or operation. Carcinomas invading beyond the submucosal layer were excluded from this analysis. The gross appearance of flat-type lesions was classified as flat elevated type or flat depressed type based on the presence or absence of central depression. A total of 5408 neoplastic lesions were index lesions, including 5035 adenomas and 373 carcinomas (124 with submucosal invasion). The prevalence of flat depressed and flat elevated lesions were 2.8% and 18.1%, respectively. Submucosal invasion rates were 17.1% in the flat depressed, 0.8% in the flat elevated, 1.6% in the sessile, 4.0% in the pedunculated lesions, and 9% in the creeping lesions. The submucosal invasion rate in the flat depressed lesions was significantly higher than in any others, except for creeping lesions. The percentage of flat elevated and flat depressed carcinomas among all carcinomas invading the submucosa was 6.5% and 21.0%, respectively. They concluded that one-quarter of all colorectal carcinomas may be derived from flat lesions. Training in dye spray technique may result in a higher detection rate of flat colonic lesions. Nivatvongs (251) reviewed the subject of early colorectal carcinoma. Most such carcinomas can be treated by adequate local excision, such as colonoscopic polypectomy and peranal excision. If there are adverse risk factors, especially poorly differentiated carcinoma, lymphovascular invasion, or incomplete excision, a radical resection is indicated if there is no contraindication. In the case of a low rectal carcinoma, adjuvant chemoradiation should be considered. Recently, a new classification has been developed: sm 1 is invasion to the upper one-third of the submucosa, sm 2 is invasion to the middle one-third, and sm 3 is invasion to the lower one-third. Lesions of sm 1 and sm 2 have a low risk of local recurrence and lymph node metastasis; local excision is adequate. The sm 3 lesions and sm 2 flat and depressed types have a high risk of local recurrence and lymph node metastasis; further treatment is indicated.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

87

FIGURE 19 & Treatment algorithm for colorectal lesions, p ¼ pedunculated; ps ¼ subpedunculated; s ¼ sessile; LST ¼ laterally spreading type. Source: From Ref. 246.

& SENTINEL LYMPH NODE MAPPING Lymph node involvement with metastatic disease has long been recognized as a prognostic discriminant that decreases survival. Ultrastaging by serial sectioning combined with immunohistochemical techniques, improves detection of lymph node micrometastases. The introduction of sentinel lymph node mapping has provided the potential to facilitate ultrastaging. The impact of staging in the oncologic management of patients may have therapeutic implications. The challenge is to determine the biologic relevance and the prognostic implications. Mulsow et al. (252) reviewed the electronic literature (1996–1993) on sentinel node mapping in carcinoma of the colon and rectum. Lymphatic mapping appears to be readily applicable to colorectal carcinomas and identifies those nodes most likely to harbor metastases. Sentinel node mapping carries a false-negative rate of approximately 10% but will also potentially upstage a proportion of patients from nodenegative to node-positive following the detection of micrometastases. The prognostic significance of micrometastases in colorectal carcinoma is required before the staging benefits of sentinel lymph node mapping can be routinely adopted. Bilchik et al. (253) proposed a TNM classification for micrometastases and isolated malignant cells. They studied 120 patients who underwent lymph node mapping before resection of primary colorectal carcinoma. Sentinel nodes were identified using blue dye and/or radiotracer and were examined by hematoxylin-eosin staining, cytokeratin immunohistochemistry, and multilevel sectioning. The comparison group comprised 370 patients whose primary colorectal carcinomas were resected without lymph node mapping during the same period. Lymph node mapping was successfully performed in 96% of patients and correctly predicted the status of the nodal basin in 96% of patients. Nodal involvement was identified for 14.3%, 30%, 74.6%, and 83.3% of T1, T2, T3, and T4 carcinomas, respectively in the study group, and for 6.8%, 8.5%, 49.3%, and 41.8% of T1, T2, T3, and T4 carcinomas, respectively, in the comparison group. The study group had a higher percentage of nodal metastases (53% vs. 36%). They believe lymph node mapping and focused sentinel node analysis should be considered to better stage colorectal carcinoma.

Saha et al. (254) used the combination of isosulfan blue (Lymphazurin) 1% and 99mTc sulfur colloid, to test the feasibility and accuracy of lymphatic mapping for colorectal carcinoma. In 57 consecutive patients, mapping was successful in 100% of patients with isosulfan blue and in 89% with sulfur colloid. Lymphatic mapping was accurate in 93% of patients with isosulfan blue vs. 92% with sulfur colloid. The combined accuracy was 95%. A total of 709 lymph nodes were found (12.4 per patient): 553 nonsentinal lymph nodes (5.6% nodal positivity) versus 156 sentinel lymph nodes (16.7% nodal positivity). Isosulfan blue detected 152 sentinel lymph nodes, sulfur colloid detected 100, and both modalities detected 96. Of the sentinel lymph nodes detected by isosulfan blue only, 10.7% had nodal metastases, whereas 19.8% of sentinel lymph nodes detected with both modalities had nodal metastases. Nodal disease was detected in 41% of patients with invasive carcinoma. Metastases were detected only in the sentinel lymph nodes in 26% and only by micrometastases in 11% of these patients. The metastatic yield is significantly higher in sentinel lymph nodes identified by both modalities compared with isosulfan blue only. A less encouraging report was published by Bembenek et al. (255). They evaluated the feasibility and utility of lymphatic mapping in 48 patients with rectal carcinoma, 37 of whom had already undergone preoperative radiochemotherapy for locally advanced lesions. An endoscopic injection of sulfur colloid into the submucosa adjacent to the carcinoma was performed 15 to 17 hours before the operation. Ex vivo identification of the nuclide-enriched ‘‘sentinel lymph nodes’’ was performed using a hand-held gamma-probe. The selected sentinel lymph nodes were examined using serial sections and immunohistochemistry. One or more sentinel lymph nodes were found in 46 of the 48 patients. The sentinel lymph node detection rate was 96%. Lymph node metastases were present in 35%. A sensitivity of only 44%, and a false-negative rate of 56% were found. Further analysis showed that the method correctly predicted the nodal status only in the small subgroup of patients with early carcinoma without preoperative radiation. They concluded that although lymph node identification shows a relatively high detection rate, the sensitivity in patients with locally advanced irradiated rectal carcinoma is low.

88

& PART I: COLORECTAL DISORDERS

Further studies will be required to determine the ultimate utility of this modality but even this may not be necessary. The results of the Cancer and Leukemia Group B (CALGB) Protocol 80001, a prospective study conducted in 12 institutions, concluded that sentinel lymph nodes fail to predict nodal status in 52% of cases (256). Read et al. (257) reported that sentinel node mapping would have potentially benefited only 3% and failed to accurately identify nodal metastases in 24% of patients in their study. They concluded the fraction of patients benefiting from sentinel lymph node mapping and lymph node ultra processing techniques would be 2%. These studies might put the question of sentinel lymph node mapping for colorectal carcinoma to rest. The most recent review of the subject by Stojadinovic et al. (258), concluded that there are fundamental questions that remain unanswered, namely, (i) does sentinel lymph node mapping significantly upstage or increase staging accuracy? (ii) do patients with H and E negative nodes but nodal metastases have a significantly worse oncologic outcome than those without micrometastases?, and (iii) does treatment of nodal micrometastases with adjuvant chemotherapy translate into meaningful survival benefit? They believe that until these questions are answered, sentinel lymph node mapping for colorectal carcinoma will likely remain investigational. Cimmino et al. (259), reported that in 267 patients who underwent intraoperative lymphatic mapping with the use of both isosulfan 1% blue dye and radiocolloid injection, five adverse reactions to isosulfan blue were encountered—two cases of anaphylaxis and three cases of ‘‘blue hives.’’ The two patients with anaphylaxis experienced cardiovascular collapse, erythema, perioral edema, urticaria, and uvular edema. The blue hives in three patients

resolved and transformed to blue patches during the course of the procedures. The incidence of allergic reactions in their series was 2%. Should physicians expand the role of sentinel lymph node mapping, they should consider the use of histamine blockers as prophylaxis and have emergency treatment readily available to treat the life threatening complication of anaphylactic reaction.

& MODES OF SPREAD To produce metastases, malignant cells must succeed in invasion, embolization, survival in the circulation, arrest in a distant capillary bed, and extravasation into and multiplication in organ parenchyma (260) (Fig. 20). The outcome of this process depends on the interaction of metastatic cells with multiple host factors. Indeed the major obstacle to the effective treatment of colon carcinoma metastasis is the biologic heterogeneity of neoplasms. Another challenge to therapy is the finding that different organ environments can modify a metastatic malignant cell’s response to systemic therapy. Carcinoma of the colon may spread in one of the following ways: direct continuity, transperitoneal spread, lymphatic spread, hematogenous spread, and implantation. Direct Continuity Intramural spread of the carcinoma occurs more rapidly in the transverse than the longitudinal axis of the colon and has been estimated to proceed roughly at the rate of one quarter of the bowel circumference every 6 months. It is unusual for microscopic spread to occur more than 1 cm beyond the grossly visible disease. Radial extension

FIGURE 20 & Pathogenesis of metastases from colorectal carcinoma, including metastases from metastases.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

through the bowel wall may result in adherence to abdominal viscera such as the small or large gut, stomach, pelvic organs, or abdominal parietes. If the lesion is situated on the retroperitoneal aspect, infiltration of the posterior abdominal wall, duodenum, ureter, perirenal fascia, or iliacus or psoas muscles may occur. Knowledge of the degree of extension is necessary to effect a curative resection. As the carcinoma penetrates the bowel wall, neighboring structures are involved in 10% of patients (260). Probably one third to two thirds of such attached viscera are involved with the neoplastic process. An additional pattern of local spread is perineural invasion, which may reach as far as 10 cm from the primary lesion. Mechanisms involved in invasion of host tissues include (i) mechanical pressure produced by a rapidly proliferating neoplasm may force cords of malignant cells along tissue planes of least resistance, (ii) increased cell motility can contribute to malignant cell invasion, and (iii) malignant cells may secrete enzymes capable of degrading basement membranes, breaking down barriers between epithelial cells and the stoma (260).

Transperitoneal Spread Transmural extension of carcinoma ultimately penetrates the peritoneal surface, following which dissemination may occur in a transcelomic manner, with implants occurring anywhere on the peritoneal surface or omentum. Approximately 10% of patients with colon carcinoma develop peritoneal deposits (230). Lymphatic Spread The nature of extramural lymphatic spread is of paramount importance in planning the scope of an operation for carcinoma. Indeed, the extent of involvement of the lymphatic system is related to the prognosis of the patient. First metastases usually take place in the paracolic glands nearest the carcinoma and proceed in a stepwise way from gland to gland; however, exceptions to such an orderly progression do occur, and these glands may be missed with the first deposits detected in a more proximal location. Retrograde lymphatic metastases may occur when anterograde blockage is present. Hematogenous Spread Blood-borne metastases result in systemic spread of disease, with the liver being the organ most commonly involved. Circulating malignant cells have been found to occur in 28% of patients during induction of anesthesia and in 50% during laparotomy (261). Surprisingly, followup studies of patients with circulating malignant cells during operation have not shown an adverse effect on the ultimate prognosis. Using an experimental melanoma model it was found that < 1.0% of cells are viable 24 hours after entry into the circulation and < 0.1% of these cells eventually produce metastases (260). Other sites of hematogenous spread include the lung and, less commonly, bone. Weiss et al. (262) analyzed the sequence of events in hematogenous metastasis from colonic carcinoma using 1541 necropsy reports from 16 centers. The authors’ findings were consistent with the cascade hypothesis that metastases develop in discrete steps, first in the liver, next

89

in the lungs, and finally, in other sites. In only 216 of 1194 cases was there suggestive evidence that metastatic patterns (excluding lymph nodes) were causally related to lymph or nonhematogenous pathways. The concept that extrahepatic metastases arise from liver metastases is also supported by Taylor (263).

Implantation Several reports have been made concerning instances in which exfoliated malignant cells have implanted on raw surfaces, such as with hemorrhoidectomy wounds, fissuresin-ano, fistulas-in-ano, or along the suture line (264–269). Other forms of implantation may relate to an abdominal scar or at the mucocutaneous margin of a colostomy (260) (Fig. 21). & SITE OF SPREAD For every 100 patients with intestinal carcinoma, approximately one half will be cured by operation and five will die from lymphatic spread, 10 from local recurrence, and 35 from blood-borne metastases. The organs most frequently involved are the liver (77%), lungs (15%), bones (5%), and brain (5%). The spleen, kidneys, pancreas, adrenals (270), breast, thyroid, and skin are rarely involved (230). Even the trachea, tonsils, skeletal muscle, urethra, oral cavity, penis, and nail bed have become involved (230,271–275). & STAGING It is helpful for the treating physician to be aware of the extent of spread of the disease. By definition, the lesion must penetrate through the muscularis mucosa for it to be considered an invasive carcinoma. Cytologic malignant cells superficial to this layer are considered carcinoma-insitu. Dukes (233) originally proposed a classification based on the extent of direct extension along with the presence or absence of regional lymphatic metastases. Dukes’ A lesions are those in which growth is confined to the bowel wall, Dukes’ B lesions include those in which direct spread has progressed through the full thickness of the wall involving the serosa or fat, and Dukes’ C lesions are those in which

FIGURE 21 & Macroscopic appearance of anal implantation from a proximally located adenocarcinoma of the descending colon.

90

& PART I: COLORECTAL DISORDERS

regional lymph node metastases are present. Dukes’ C lesions are further divided into C1, in which lymph nodes are involved near the bowel wall, and C2, in which there is continuous involvement of nodes up to the point of ligature (276). Not unexpectedly, there is a correlation between histologic grade and the Dukes classification. By common usage, a fourth category, labeled D, has been added. In this group metastatic disease has advanced beyond the confines of surgical resection, as in the case of a patient with distant metastases or an unresectable local lesion (277). Wong et al. (278) conducted a study to determine the number of nodes that need to be examined to accurately reflect the histology of the regional lymphatics in colorectal carcinoma. Patients undergoing curative resection for T2 and T3 colorectal carcinomas were reviewed. The number of nodes examined ranged from 0 to 78 (mean, 17 nodes). Node-negative patients had fewer nodes examined (mean, 14 nodes) than node-positive patients (mean, 20 nodes). The entire sample had a node-positive rate of 38.8%. When at least 14 nodes were examined, the percent of patients with at least one positive node was 33.3%. They concluded the examination of at least 14 nodes after resection of T2 or T3 carcinomas of the colon and rectum will accurately stage the lymphatic basin. The staging of large bowel carcinoma remains a confused topic for most and has been discussed at length (279). In an effort to better define or categorize patients in order to achieve better prognostication, numerous eponymous modifications have been introduced, but to my mind these have served only to cloud the issue. Despite the intentions of well meaning physicians, all in the name of progress, few significant advances have been made. Some of the staging systems proposed are remarkably similar to the Dukes staging system, while others are so detailed that they are cumbersome and unmanageable, difficult to recall and apply, and, simply stated, not practical (277,279–289).

Further confusion has resulted from the fact that the Dukes staging system is frequently misquoted (287,289). An ideal staging system would be one that is simple and easy to remember and apply. It should include important prognostic discriminants without becoming too complex. The Australian Clinicopathological Staging (ACPS) system was proposed to incorporate clinical nformation in an effort to better define prognosis (283,285) (see Box 1). The tumor/node/metastasis (TNM) classification was proposed to incorporate findings at laparotomy and provide greater precision in the identification of prognostic subgroups, but its complexity has hindered general acceptance (281). The most current definitions of the TNM classification are depicted in Box 2. The completeness of resection designation is seen in Box 3 (280). The most commonly used classifications are depicted in Fig. 22. The value of any staging system is its potential application in the treatment of the patient. Since there is not yet an ideal method of definitively staging the patient’s disease preoperatively, a study of the resected specimen continues to have merit. Traditional pathologic variables considered important include histologic grading; depth of penetration; cell type; lymph node involvement; lymphatic, venous, or perineural invasion; distal margin of resection; and surrounding inflammatory response. In addition, it has been suggested that certain subsets of patients may benefit from adjuvant therapy. Being able to select appropriate therapy would be advantageous so that the remaining patients would not be subjected to potentially harmful treatments. Obtaining international agreement on a staging system will prove to be a Herculean task. Proponents of ‘‘new’’ classifications believe their system to be the best and their followers will continue to adhere to those systems. Only if and when a new system is proved to be beneficial from a prognostic or therapeutic viewpoint will

BOX 1 & Australian Clinicopathological Staging System When all known carcinoma has been removed by a ‘‘Curative resection,’’ the following categories are used for staging: ACPS O The carcinoma is confined to the mucosa in a patient who has undergone a bowel resection. ACPS A The carcinoma has spread into the bowel wall but not beyond the muscularis propria. There are no lymph node metastases, nor distant metastases.

ACPS B The carcinoma has spread beyond the muscularis propria into the adjacent tissues in continuity or into adjacent organs. There are no lymph node metastases, nor distant metastases. ACPS C The carcinoma may have spread varyingly into or through the bowel wall, but one or more lymph nodes contain cancer. There are no distant metastases.

ACPS D This category is used when there is clinical or microscopic evidence of any carcinoma remaining locally or at a distance, whether there has been a ‘‘palliative resection,’’ a ‘‘palliative operation,’’ a local excision or no operation because of the extent of the carcinoma. In the event of synchronous carcinomas being present in the resected specimen of bowel, the ACP stage allocated should be that of the most advanced carcinoma. ACPS X Category ‘‘X’’ is used where a local excision or other local procedure is done without lymphadenectomy. This may be subdivided into: XO—The carcinoma is confined to the mucosa. XA—The carcinoma does not spread beyond the muscularis propria. XB—The carcinoma spreads beyond the muscularis propria. Operations may be done with curative or palliative intent in this group. ACPS Y This category is used when the pathologic details are, for some reason, not known or are incomplete.

Source: From Davis NC, Newland RC. Terminology and classification of colorectal adenocarcinoma. The Australian Clinicopathological Staging System. Aust N Z J Surg 1983; 53:211–221.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

91

BOX 2 & TNM Definitions Primary tumor (T) TX T0 Tis T1 T2 T3 T4

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ: intraepithelial or invasion of the lamina propria
Tumor invades submucosa Tumor invades muscularis propria Tumor invades through the muscularis propria into the subserosa, or into nonperitonealized pericolic or perirectal tissues Tumor directly invades other organs or structures, and/or perforates the visceral peritoneum

,







Tis includes cancer cells confined within the glandular basement membrane (intraepithelial) or lamina propria (intramucosal) with no extension through the muscularis mucosae into the submucosa.





Direct invasion in T4 includes invasion of other segments of the colorectum by way of the serosa; for example, invasion of the sigmoid colon by a carcinoma of the cecum.






Tumor that is adherent to other organs or structures, macroscopically, is classified T4. However, if no tumor is present in the adhesion microscopically, the classification should be pT3. The V and L substaging should be used to identify the presence or absence of vascular or lymphatic invasion.

Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in 1 to 3 regional lymph nodes N2 Metastasis in 4 or more regional lymph nodes A tumor nodule in the pericolorectal adipose tissue of a primary carcinoma without histologic evidence of residual lymph node in the nodule is classified in the pN category as a regional lymph node metastasis if the nodule has the form and smooth contour of a lymph node. If the nodule has an irregular contour, it should be classified in the T category and also coded as V1 (microscopic venous invasion) or as V2 (if it was grossly evident), because there is a strong likelihood that is represents venous invasion. Distant metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis STAGE GROUPING Stage 0 I IIA IIB IIIA IIIB IIIC IV

T Tis T1 T2 T3 T4 T1-T2 T3-T4 Any T Any T

N N0 N0 N0 N0 N0 N1 N1 N2 Any N

M M0 M0 M0 M0 M0 M0 M0 M0 M1




Dukes
– A A B B C C C –

MAC
– A B1 B2 B3 C1 C2/C3 C1/C2/C3 D

Dukes B is a composite of better (T3 N0 M0) and worse (T4 N0 M0) prognostic groups, as is Dukes C (Any TN1 M0 and Any T N2 M0). MAC is the modified Astler-Coller classification. Note: The y prefix is to be used for those cancers that are classified after pretreatment, whereas the r prefix is to be used for those cancers that have recurred.

BOX 3 & Residual Tumor (R) R0 R1 R2

Complete resection, margins histologically negative, no residual tumor left after resection Incomplete resection, margins histologically involved, microscopic tumor remains after resection of gross disease Incomplete resection, margins involved or gross disease remains after resection

92

& PART I: COLORECTAL DISORDERS

FIGURE 22 & Colon wall with anatomic layers and depth of penetration for several proposed staging classifications of colorectal carcinoma.

it stand a chance of acceptance. The objectives of the staging system for carcinoma, succinctly enumerated by Davis and Newland (284), include: & & & & &

to aid the physician in planning treatment to give some indication of prognosis to assist in the evaluation of treatment results to facilitate the exchange of information between treatment centers and to contribute to the continuing investigation of human cancer.

It provides a method of conveying one group’s experience to others without ambiguity. To achieve these goals, a pilot project confirming their virtues would be necessary to convince the international community of the efficacy of the new system. As one example, from a practical point of view, demonstration of histopathologic penetration may be difficult, but Newland, Chapuis, and Smyth (290) have shown the distinction to be clinically relevant. In a prospective study of 1117 cases, they reported a 77% 5-year survival rate in node-negative patients with carcinoma limited to the bowel wall and a

54% 5-year survival rate when there was serosal penetration. One important advantage of a universal staging system would be to allow new or different modes of therapy or adjuvant therapy from different centers to be compared and a rational interpretation to be made. Indeed, such a system would facilitate or influence the choice of treatment. For example, it could extend the use of sphincter-saving operations, identify patients suitable for local treatment, and define those with considerable risk of local recurrence when combined operation and adjuvant treatment might be considered. Knowing the precise extent of the disease would be helpful for interpreting the influence of treatment on parameters such as disease-free survival so that we can identify small and useful improvements in prognosis. It is likely that much time will elapse before any such system will be accepted universally.

& BIOLOGY OF GROWTH The rate of growth of an individual carcinoma is almost certainly the most important prognostic discriminant. Colorectal carcinomas are relatively slow-growing neo-

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

plasms, and metastases occur relatively late. In his summary of the results of several series of studies, Spratt (291), using radiographic measurements, initially calculated a doubling time of 636.5 days in a follow-up of 7.5 years for an unusual case in which a patient was not treated by resection. Similar analysis of 19 other patients revealed a mean doubling time of 620 days, with a 95% range extending from 111 days to 3430 days. Spratt reported the doubling time of pulmonary metastases from colon and rectal carcinoma calculated radiographically to be 109 days, with a 95% range extending from 9 days to 1300 days. He believed that metastatic carcinomas increase their cellular complement six

93

times faster than do primary carcinomas. The doubling time of untreated hepatic metastases has varied between 50 and 95 days (292,293). It has been theorized that the absence of desquamation in the metastatic site accounts for this observed difference. Bolin, Nilsson, and Sjodahl (294) had the opportunity to measure radiographically lesions in 27 patients on two separate occasions, with a median interval of 11 months (4 to 91 months), and found median doubling time to be 130 days (53 to 1570 days). The authors suggested that the high growth rate observed in their investigation was due to the large size of the neoplasm at the time of initial examination. Burnett and Greenbaum (295) believe

94

& PART I: COLORECTAL DISORDERS

that although there is little doubt that many colonic carcinomas are slow growing, there is probably a subset of rapidly growing lesions. In a small group of patients, the authors observed doubling times as short as 53 to 150 days. The implication of their report is that a carcinoma seen on a barium enema study at a given time was not necessarily missed by a negative barium enema study reported, for example, 2 years earlier. Matsui et al. (296) estimated a statistical curve for growth from 31 patients with colorectal carcinoma in which initial lesions were diagnosed as mucosal carcinoma. These lesions were overlooked in the first or second investigations, but were detected later. Initial radiographic features were as follows; 4 pedunculated lesions, 1 semipedunculated lesion, 6 sessile lesions, 9 superficially elevated lesions, and 11 superficially depressed lesions. The diameters of the initial lesions were 12.1  6.1 mm. The final depths of invasion were 6 mucosal carcinomas, 12 submucosal carcinomas, 6 muscularis propria carcinomas, and 7 serosal carcinomas. The observation period between the initial and final examination was 41.5  25.8 months. Growth speed of early colorectal carcinoma was estimated through a statistically significant growth curve. Estimated doubling time of the volume of early colorectal carcinoma was 26 months.

& CLINICAL FEATURES & SYMPTOMS Patients with carcinoma of the large intestine may present in one of three characteristic ways: insidious onset of chronic symptoms, acute intestinal obstruction, or perforation with peritonitis. Aldridge et al. (297) reported that those conditions will occur in 77%, 16%, and 7% of patients, respectively. Mandava et al. (298) reported perforated carcinomas in 51 of 1551 patients (3.3%) with colorectal carcinoma. Localized perforation with abscess formation occurred in 31 patients (61%), and free perforation with generalized peritonitis occurred in 20 patients (39%). Runkel et al. (299), in a review of 923 patients, found the presentation with insidious onset, obstruction, and perforation to occur in 92.0%, 6.0%, and 2.0% of patients, respectively, with a combination of obstruction and perforation occurring in 0.5% of patients. In a report from Scotland of 750 consecutive patients, an unusually high number (33%) presented as an emergency (300). Depending on the location in the bowel, one or another of the following symptoms may predominate. Bleeding is probably the most common symptom of large bowel malignancy (301). All too often the bleeding is attributed by the patient, and regrettably by the physician as well, to hemorrhoids. Despite the fact that the most common cause of rectal bleeding is hemorrhoids, bleeding cannot be dismissed lightly, especially in a middle-aged or older individual. It has been estimated that visible blood per rectum occurs in 10% of the adult population over the age of 30 years (302). However, Beart et al. (301) found that in patients 40 years of age or older with a history of rectal bleeding and known hemorrhoids, 6% had rectal or colon carcinoma and 14% had colon polyps. This underscores the necessity of taking the symptom of bleeding seriously. Bleeding may be occult, as is often the case in right-sided

lesions, or overt, being either bright red, dark purple, or even black, depending on the location in the bowel. The second most common presenting symptom is probably a change in bowel habits, either constipation or diarrhea (301). The absolute frequency of defecation is not important, but a deviation from what is normal for a given individual may signal the presence of an intestinal neoplasm. Lesions in the proximal colon may not result in a change in bowel symptoms until they are very far advanced. Lesions in the distal colon are more likely to be manifest in symptoms since the stool is more formed in consistency and the bowel lumen is narrower. A progressive narrowing of the caliber of the stool may be reported in the case of a compromised lumen. Pain may occur with almost equal frequency as the preceding symptom. Abdominal pain may be vague and poorly localized, or it may result from a partially obstructing lesion of the colon. The latter type of pain is generally colicky in nature and may be associated with obstructive symptoms of bloating, nausea, and even vomiting. Rectal pain does not occur with carcinoma unless there is sacral nerve root or sciatic nerve involvement. However, tenesmus may occur with rectal carcinoma. Back pain is a late sign of penetration of retroperitoneal structures. Other symptoms include mucus discharge, which may coat the stool or be mixed with it. Investigation of weight loss as a solitary complaint only uncommonly yields a large bowel malignancy. However, when associated with a carcinoma, weight loss is usually a sign of advanced disease and bodes a poor prognosis. Nonspecific symptoms of general impairment of health, loss of strength, anemia, and sporadic fever also may be present. Rarely a carcinoma of the cecum obstructs the appendiceal orifice, and the patient presents with the signs and symptoms of acute appendicitis (303). Bladder involvement may result in urinary frequency, suprapubic pressure, and even pneumaturia if a sigmoidovesical fistula has developed. In the report of the Commission on Cancer (304), in which 16,527 patients were diagnosed with carcinoma of the colon, the presenting symptoms in order of frequency were abdominal pain (40.5%), change in bowel habits (33.2%), rectal bleeding (28.5%), occult bleeding (34.3%), malaise (16%), bowel obstruction (14.9%), pelvic pain (3.4%), emergency presentation (6.6%), and jaundice (1%). Iron deficiency anemia is a recognized complication of colorectal carcinoma especially with right-sided lesions and failure to investigate the anemia in older patients may lead to a delay in diagnosis. Acher et al. (305) conducted a study of all patients presenting with confirmed colorectal carcinoma in a catchment population of 280,000. The criteria for iron deficiency anemia was hemoglobin < 10.1 g. Of 440 patients with colorectal carcinoma, 38% had iron deficiency anemia at diagnosis and of the latter, 12% were known to have iron deficiency anemia for more than 6 months before diagnosis and 6% had iron deficiency anemia more than one year before diagnosis. Iron deficiency anemia was more common in right-sided carcinomas (65%) than those arising in the left side of the colon and rectum (26%). They concluded the investigation of iron deficiency anemia in older patients is important but in order to detect 26 patients with colorectal carcinoma a year earlier, the investigation of approximately 5000 patients would be required—a detection rate of less than 1%.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

Church and McGannon (306) performed a study to find out how often and how accurately a family history of colorectal carcinoma was recorded in the charts of 100 in patients on a colorectal surgical floor. The chart review was repeated 4 years later. In the initial review, they found that a family history was recorded in 45 of 100 charts. It was accurate for colorectal carcinoma in 36 charts. Four years later the rate of family history recording increased to 61 of 96, whereas the accuracy rate did not change. Despite improvement during a 4-year period, there is still room for further improvement.

& GENERAL AND ABDOMINAL EXAMINATIONS The general examination may be a guide to the patient’s state of nutrition. Any obvious excess weight loss may indicate advanced disease. Pallor may be a sign of anemia. The assessment may help in evaluating the patient’s fitness for operation. Usually the general abdominal examination fails to reveal significant abnormalities. Occasionally a mass may be present and may indicate the primary malignancy or possible metastatic disease. The liver may be enlarged, and if it is also umbilicated, this sign would be characteristic of metastatic disease. Ascites may be appreciated. Borborygmi may be present. Abdominal distention may suggest partial obstruction from a constricting lesion. Inguinal and left supraclavicular lymphadenopathy can occur rarely. Peritonitis from a perforated carcinoma may be difficult to differentiate from perforated diverticulitis. & DIGITAL RECTAL EXAMINATION Although a digital rectal examination will not identify the presence of a colon carcinoma, it should reveal the presence of a rectal lesion. The exam is described in detail in Chapter 24. Sometimes a sigmoid carcinoma that hangs down into the cul-de-sac may be palpable. & EXTRAINTESTINAL MANIFESTATIONS Rosato et al. (307) noted a number of cutaneous presentations associated with gastrointestinal malignancy. These included acanthosis nigricans, dermatomyositis, and pemphigoid (see Chapter 12). Halak et al. (308) reported on the incidence of synchronous colorectal and renal carcinomas and reviewed the literature on that issue. Among 103 patients who underwent colorectal surgery in their series, 5 cases of synchronous colorectal and renal carcinomas were detected (4.9%). A review of the literature suggests the incidence of simultaneous colorectal and renal carcinoma to be 0.04% to 0.5%. & SYNCHRONOUS CARCINOMAS Synchronous carcinomas are not uncommon, and in recent reviews the incidence was found to range from 2% to 8% in colon carcinoma patients (309–312). In view of this, it behooves the treating surgeon to evaluate the entire colon, if possible, to determine the presence of other neoplastic lesions. Colonoscopy performed preoperatively in patients undergoing elective resection is the optimal method for assessing the unresected colon for synchronous lesions. In a prospective study of 166 patients, Langevin and Nivatvongs (313) found synchronous carcinomas in eight patients, seven of whom required resection of more colon

95

than was necessary to treat the index carcinoma. In a study of 320 colorectal patients, Evers, Mullins, and Mathews (311) found synchronous carcinomas in different surgical segments in six of 21 patients (38%). Pinol et al. (312) conducted a study designed to identify individual and familial characteristics associated with the development of synchronous colorectal neoplasms in patients with colorectal carcinoma. During a 1-year period, 1522 patients with colorectal carcinoma attended in 25 Spanish hospitals were included. Synchronous colorectal neoplasms were documented in 505 patients (33.2%): adenoma (in 411 patients) (27%), carcinoma (in 27 patients) (1.8%), or both (n ¼ 67) (44%). Development of these lesions was associated with male gender (OR 1.94), personal history of colorectal adenoma (OR 3.39), proximal location of primary carcinoma (OR 1.4), TNM Stage II (OR 1.31), mucinous carcinoma (OR 1.89), and family history of gastric carcinoma (OR 2.03). Based on individual and familial characteristics associated with synchronous colorectal neoplasms, it has been possible to identify a subgroup of patients with colorectal carcinoma prone to multicentricity of neoplasm with potential implications on the delineation of preventive strategies. If technically and logistically possible, it would be ideal for all patients about to undergo elective resection carcinoma to have a preoperative colonoscopic examination. If not possible, it is suggested that patients should have postoperative colonoscopy.

& ASSOCIATED POLYPS Neoplastic polyps are frequently associated with carcinoma of the colon, and, indeed, on sigmoidoscopic examination, when a patient is found to have a neoplastic polyp, the entire colon should be assessed to rule out the presence of another associated polyp or an associated malignancy. Approximately 7% of patients with polyps will harbor an associated carcinoma of the colon or rectum (314). Slater, Fleshner, and Aufses (315) studied the relationship between the location of a colorectal carcinoma and the existence of adenomas in 591 patients. The overall incidence of adenomas was 29.7%, with resected specimens of right-sided carcinomas containing adenomas in 47% of cases and leftsided specimens having them in 22%. Thus the authors suggested that efforts be made to identify polyps preoperatively in patients with colorectal carcinoma. Also, since patients with carcinoma and associated adenomas are at increased risk of developing metachronous carcinoma, the group with right-sided carcinoma should be part of a particularly active surveillance program. Chu et al. (316) retrospectively studied the relationship of colorectal carcinoma with polyps in 1202 patients. Synchronous polyps were found in 36% of patients. Synchronous carcinoma was found in 4.4% of patients and metachronous carcinoma developed in 3.5% of patients. The incidence of synchronous carcinoma and a metachronous carcinoma increased with synchronous polyps, and varied according to number, size, and histologic features of the polyps. The adjusted 5-year survival rate was improved in patients with synchronous polyps compared with those without synchronous polyps. The pattern of relapse was the same for the synchronous polyp and nonsynchronous polyp groups. The authors even went so

96

& PART I: COLORECTAL DISORDERS

far as to recommend subtotal colectomy for colorectal carcinoma and synchronous polyps in good-risk patients. There is a general trend for the incidence of multiple carcinomas to rise as the number of adenomas increases whereas in patients in whom there is evidence of only one adenoma being present, the incidence of multiple carcinomas is less than 2%, the figure for those with more than five adenomas rises to 30% (231).

& OTHER ASSOCIATED MALIGNANCIES In a review of 9329 cases of colorectal carcinoma from the literature, Lee et al. (317) found that the incidence of primary malignant neoplasms at sites other than the colon ranged from 3.8% to 7.8%. In a series of 14,235 patients with colon and rectal carcinomas followed for an average of 3.6 years, Tanaka et al. (318) observed an elevated risk of second primary malignancies as follows: rectum (observed/expected [O/E] 2.0, men; O/E 4.3, women), corpus uteri (O/E, 8.2), ovary (O/E, 4.3), and thyroid gland (O/E 4.7, women). These observations were more notable in right-sided colon carcinoma patients than left-sided colon carcinoma and in those younger than 50 years. Schoen, Weissfeld, and Kuller (319) also found that women with a history of breast, endometrial, or ovarian carcinoma were at a statistically significant increased risk for subsequent colorectal carcinoma (1.1, 1.4, and 1.6, respectively).

& COMPLICATIONS A number of well-recognized complications of carcinoma of the large bowel may alter its clinical presentation.

& OBSTRUCTION Depending on the macroscopic features and location of the malignancy, interference with the passage of intestinal contents may occur. Carcinoma is the most common cause of large bowel obstruction, contributing to 60% of cases in the elderly (320). Ohman (321) summarized the results of 26 series on colorectal carcinoma reported in the literature. The combined total comprised 23,434 patients of whom 15% presented with obstruction. Incidences ranged from 7% to 29%. In the right colon, carcinomas are usually polypoid, and because of the liquid nature of the intestinal contents, obstruction is unlikely unless it involves the critical ileocecal valve, in which case obstruction may supervene. In the left colon, where the intestinal contents are solid and the nature of the malignancy is more inclined to be annular, occlusion is more likely. Complete obstruction of the colon from carcinoma is often entirely insidious in its onset. The patient often has had progressive difficulty in moving his or her bowels and has taken increasing doses of laxatives until the abdomen has become more distended with pain and eventual obstipation. Nausea and vomiting may supervene. Alternatively, the patient may present with sudden, severe, colicky abdominal pain that persists, and investigation may reveal a complete obstruction. The offending lesion may be surprisingly small as seen in Fig. 23. On examination, the patient’s general condition usually is found to be good, because dehydration and electrolyte depletion are often late

FIGURE 23 & Small adenocarcinoma of splenic flexure in patient who presented with acute colonic obstruction.

phenomena. Examination of the abdomen reveals that it is distended and tympanitic but not tender. Hyperactive peristalsis may be present. It is unlikely that an abdominal mass will be felt in the presence of a distended abdomen. Digital examination may reveal a balloon-type rectum and, in the exceptional case, a palpable carcinoma may be present. A mass in the cul-de-sac may be appreciated, representing either a sigmoid loop that is hanging down or a cul-de-sac implant. Sigmoidoscopy may reveal the lower edge of a constricting lesion. The diagnosis is usually suggested from a history of intestinal symptoms and physical findings. Plain x-ray films of the abdomen will reveal the presence of obstruction and indicate its level. The amount of small bowel distention will depend on the competence of the ileocecal valve. The presence of an obstructing carcinoma can be confirmed with an emergency barium enema study. A nonspecific type of colitis may develop proximal to an obstructing carcinoma of the colon (322,323). It is reported to occur in 2.0% to 7.5% of cases (324,325). In most cases, there is a short segment of normal mucosa immediately proximal to the carcinoma, above which the colitic process appears. The etiology of these changes in the colonic mucosa proximal to the obstructed lesion is obscure, although the gross pathology and microscopic features are consistent with those of ischemic colitis in the subacute or chronic phase (324). The extent of the colitic process varies. It may be unsuspected and discovered only at the time of operation. The extent of resection may be dictated by the length of inflamed bowel. Seow-Choen, Chua, and Goh (325) reported that four of 204 new cases of colorectal carcinoma seen in a 24-month period were proven to have concomitant proximal ischemic colitis. Ischemic colitis associated with obstructing carcinoma of the colorectum may present dramatically with gangrene or colonic perforation if the acute vascular insufficiency is severe. Less severe degrees of ischemic insult must be recognized intraoperatively because the incorporation of ischemic colon in a colonic anastomosis may result in an anastomotic leak. The National Surgical Adjuvant Breast and Bowel Project (NSABP) trials have suggested that obstructing carcinomas in the right colon carry a more significant risk of recurrence and carcinoma-related mortality than obstructing carcinomas in the left colon (326). The diminished survival rate in patients presenting with obstructing malignancies

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

does not appear to correlate significantly with annularity of the lesion or the presence of lymph node metastases.

& PERFORATION The incidence of perforation associated with carcinoma of the colon is in the 6% to 12% range of all hospital admissions for colon carcinoma (323). A perforation may result in peritonitis, abscess formation, adherence to a neighboring structure, or fistulous communication into a viscus. Perforation is found in conjunction with obstruction in approximately 1% of patients with colorectal carcinoma. In patients with obstruction, concurrent perforation is found in 12% to 19% of patients (323,327). If acute obstruction supervenes in the middle or distal colon, the cecum may perforate. However, the most common form of perforation is associated with the carcinoma itself. Such perforation may develop suddenly with diffuse peritonitis or more gradually with a localized peritonitis; when it occurs in the cecum, it may resemble appendicitis. Patients with obstruction and a proximal perforation present as desperately ill individuals with generalized peritonitis, dehydration, and electrolyte depletion. Immediate exploratory laparotomy is demanded. Patients without obstruction but with perforation of the carcinoma are also gravely ill and require immediate laparotomy after some correction of dehydration and electrolyte depletion. Patients without obstruction but with perforation of a carcinoma may present with a localized peritonitis. Under these circumstances it may be confused with diverticulitis if the sigmoid colon is involved, or with appendicitis or Crohn’s disease if the right colon is involved. & BLEEDING Bleeding is a common symptom of colorectal carcinoma, but massive bleeding is an uncommon presentation. & UNUSUAL INFECTIONS ASSOCIATED WITH COLORECTAL CARCINOMA Unusual infections associated with colorectal carcinoma may, in some instances, be the sole clue to the presence of a malignancy. The infections are either related to invasion of tissues or organs in close proximity to the neoplasm or secondary to distant seeding by transient bacteremia arising from necrotic carcinomas. Panwalker (328) identified a series of patients whose clinical presentations included endocarditis (Streptococcus bovis bacteremia), meningitis (S. bovis bacteremia), nontraumatic gas gangrene (Escherichia coli), empyema (E. coli, Bacteroides fragilis), hepatic abscesses (Clostridium septicum), retroperitoneal abscesses (E. coli, B. fragilis), clostridial sepsis, and colovesical fistulas with urosepsis (E. coli). Panwalker also reviewed the English language literature and identified other infections associated with colon carcinoma, including nontraumatic crepitant cellulitis, suppurative thyroiditis, pericarditis, appendicitis, pulmonary microabscesses, septic arthritis, and fever of unknown origin. Lam, Greenberg, and Bank (329) reported an unusual presentation of colon carcinoma with a purulent pericarditis and cardiac tamponade caused by B. fragilis. With or without endocarditis, septicemia caused by S. bovis may be associated with an occult colonic malignancy (330,331). The septicemia also has been described with a variety of gastrointestinal pathologic conditions, such as

97

colonic adenomas, inflammatory bowel disease, and carcinoma of the esophagus. Patients may be entirely asymptomatic, but all patients with endocarditis caused by S. bovis should be evaluated for concomitant colon carcinoma. In a literature review, Panwalker (328) found S. bovis bacteremia reported in 467 adults with endocarditis present in 62%, absent in 13%, and of unclear status in the remaining patients. Other organisms associated with endocarditis are S. salivarius and enterococcus. An asymptomatic colonic malignancy has been reported in a patient with meningitis caused by S. salivarius (332). S. bovis meningitis was reported in 10 adults (328). Silent colon carcinoma has been reported to present as hepatic abscesses (333,334), and therefore anaerobic hepatic abscesses might alert the physician to the possibility of a malignancy of the large intestine. Panwalker (328) found 55 patients with the dramatic clinical presentation of gas gangrene associated with colorectal carcinoma (16 of which were cecal). The gas gangrene was metastatic in 10 patients. Sites included the neck, chest wall, upper extremities, shoulders, and axilla. Kudsk (335) subsequently reported five cases of painful, rapidly spreading gas-producing infection of the lower extremity (three cases), upper extremity (one), and pelvis (one), which represented metastatic C. septicum infections in diabetic patients. All had occult carcinomas of the right colon. More recently, Lorimer and Eldus (336) reported three cases of invasive C. septicum infection associated with colorectal carcinoma. The authors cited a previous review of 162 cases of nontraumatic C. septicum infection identified-malignant disease in 81%, approximately half of which were colorectal. The carcinoma is typically right sided and always ulcerated, which can occur in three circumstances: occult carcinoma (80%), anastomotic recurrence, or carcinoma that is unresectable or has been bypassed. In summary, gas gangrene associated with colorectal carcinoma is a catastrophic illness that appears to be clostridial, affects diabetic patients disproportionately, and, in almost 50% of cases, is the result of an otherwise silent cecal carcinoma.

& DIAGNOSIS In a case of carcinoma of the colon the patient’s history may not be helpful. Consequently, early diagnosis may depend on screening, which may be directed at the identification of high-risk groups, the use of screening tests, and the investigation of patients with positive screening test results. Early detection is described in Chapter 25. Suffice it to say that a combination of occult blood testing and flexible fiberoptic sigmoidoscopy is the minimum current recommendation (337). Colonoscopy is ideal and preferable. Atypical dyspepsia and vague abdominal symptoms should be investigated since malignancy may be the cause of otherwise un-explained ill health and anemia. Of course, the symptoms suggestive of the disease, as previously described, should be pursued with the appropriate modalities as described in the discussion of investigations. For differential diagnosis, a host of conditions may be considered, depending on the predominant symptom complex with which the patient presents. These include inflammatory bowel disease, either of the nonspecific type, such

98

& PART I: COLORECTAL DISORDERS

as ulcerative colitis or Crohn’s disease, or of the specific type, such as amebiasis, actinomycosis, or tuberculosis. For patients presenting with narrowed bowel, ischemic strictures may be included in the differential diagnosis. Acute abdomen may be confused with conditions such as diverticulitis or appendicitis with abscess, Crohn’s disease, foreign body perforation, or even an infarcted appendix epiploica. If marked bleeding is the prominent feature, vascular ectasis, diverticulosis, or acute ischemia are possibilities. With acute obstruction, volvulus, diverticulitis, and Crohn’s disease might be considered. Extrinsic pressure from a metastatic carcinoma, endometriosis, or even pancreatitis are possibilities in appropriate circumstances. Uncommon conditions (e.g., colitis cystica profunda) also can be considered.

& INVESTIGATIONS & OCCULT BLOOD TESTING Occult blood determinations are of value in the screening setting; however, for patients who have symptoms that are suggestive of large bowel disease, occult blood testing is inadequate (337). Certainly patients who relate a history of rectal bleeding do not need occult blood testing to confirm its presence.

ment of the distance of a carcinoma from the anal verge and this judgement cannot be made satisfactorily or accurately with a flexible instrument.

Colonoscopy The role of colonoscopy also has assumed greater importance in the evaluation of colon disease (Chapter 3). In particular, colonoscopy plays a major role in screening for colorectal carcinoma, especially in high-risk patients, as described in Chapter 25. With specific reference to its value in the assessment of patients with large bowel malignancy, colonoscopy has been recommended as a preoperative examination to detect the presence of synchronous neoplastic polyps or carcinomas. Its necessity in the preoperative or at least perioperative setting has been debated, but growing evidence suggests that it plays an important role (310,341,342). The rationale is that synchronous carcinomas exist in 2% to 7% of cases (313,343,344). It has been suggested that preoperative colonoscopy alters the operative procedure in one third of patients (342). Another group of surgeons, concerned with the potential for implantation of malignant cells, exfoliated by preoperative colonoscopy, has opted for intraoperative palpation to detect synchronous carcinomas and postoperative colonoscopy to clear the colon of polyps (345).

& ENDOSCOPY

& RADIOLOGY

Anoscopy and Sigmoidoscopy Use of an anoscope to determine the presence of any significant internal hemorrhoids is of value, especially for patients who present with bright-red rectal bleeding. Sigmoidoscopic examination is an indispensable diagnostic tool in the assessment of rectal carcinoma (338). The appearance of rectal carcinoma is usually quite distinctive. A protruding mass into the lumen may be seen, but more characteristically a raised everted edge with a central, sometimes necrotic, sloughing base will be noted. The distance from the lower edge of the lesion to the anal verge should be carefully determined because it may be crucial in deciding whether intestinal continuity can be restored. Two points also should be noted: which wall the lesion is located on and whether the lesion is annular. In addition, information can be obtained regarding the mobility, of the lesion by placing the end of the sigmoidoscope against the lower margin of the lesion and exerting gentle pressure along the long axis. Mobile carcinomas can be moved and present a quite different sensation from the rigid feel of a fixed carcinoma. The size of the lesion should be recorded, and finally a biopsy should be performed to confirm the diagnosis.

Barium Enema The method by which the largest number of carcinomas of the colon are diagnosed is the barium enema examination, although in some hands colonoscopy is the preferred investigative modality. Various radiologic features may be demonstrated, such as an annular, or ‘‘napkin ring,’’ appearance, as often seen in the left colon (Fig. 24). Features distinguishing it from spasm are the irregular, jagged outline and destruction of the mucosa with a typical "apple

Flexible Fiberoptic Sigmoidoscopy Flexible fiberoptic sigmoidoscopy has assumed an increasing role in the diagnosis of colon disease (339,340). Because of its greater length and flexibility than the rigid sigmoidoscope, the flexible sigmoidoscope allows for the detection of a larger number of neoplastic lesions. In addition, equivocal lesions seen on a contrast enema can be elucidated if they are within the reach of the scope. Some surgeons have replaced rigid sigmoidoscopy with flexible sigmoidoscopy, but others, including myself, believe that the rigid instrument is still superior for the crucial assess-

FIGURE 24 & Barium enema study appearance of an annular adenocarcinoma of the colon. Source: Courtesy of M. Rosenbloom, M.D., Montreal, Quebec.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

99

FIGURE 25 & Barium enema study appearance of a polypoid adenocarcinoma of the colon. Source: Courtesy of M. Rosenbloom, M.D., Montreal, Quebec.

core" appearance and overhanging edges. A large filling defect representing a bulky neoplasm projecting into the lumen is seen more often in the right colon (Fig. 25). A polypoid sessile lesion occupying only one wall of the bowel is demonstrated in Fig. 26. Sometimes a polypoid lesion on a pedicle may be entirely malignant. Occasionally a complete retrograde obstruction may be present (Fig. 27), but this radiologic appearance is not necessarily indicative of a clinically anterograde obstruction. In an audit of 557 barium enema studies in patients with known carcinomas, a malignant lesion was recorded in 85% (346). Reviewing cases in which there was failure to perceive a demonstrated lesion or failure to analyze a perceived lesion indicated that 94% of carcinomas should have been reported. The air-contrast barium enema has been considered superior to the full-column barium enema for detection of small polyps. For detection of large constricting lesions, the single-contrast enema is superior to the doublecontrast enema.

FIGURE 26 & Barium enema study appearance of a sessile adenocarcinoma of the colon. Source: Courtesy of M. Rosenbloom, M.D., Montreal, Quebec.

FIGURE 27 & Barium enema study appearance of a complete retrograde obstruction due to adenocarcinoma of the colon. Source: Courtesy of M. Rosenbloom, M.D., Montreal, Quebec.

In the event of a negative barium enema report, subsequent management will depend on clinical assessment. Should there be a strong degree of clinical suspicion of a carcinoma, colonoscopic evaluation would be in order. If no strong suspicion exists and the barium enema was ordered because of minimal symptoms of constipation, no further investigation may be deemed necessary. If rectal bleeding was the prominent feature and the patient exhibited prominent internal hemorrhoids, hemorrhoid treatment should be performed to eliminate confusion with bleeding of colonic origin. Should bleeding persist after such treatment, colonoscopic examination would be indicated.

Intravenous Pyelography There is controversy as to whether an intravenous pyelogram is necessary preoperatively. Supporters suggest that it is helpful in advance to know whether the ureters or bladder are involved with the neoplastic process, whether two kidneys are present, or whether additional ureters may be present. Furthermore, in the follow-up of patients who develop urinary tract problems postoperatively, it provides a baseline for differentiating whether these problems are due to a surgical complication or are secondary to preexisting urologic disease (347,348). In one study, abnormalities were found in as many as 26% of patients (348). Adversaries suggest that intravenous pyelography is not cost-effective and is not necessarily reliable in demonstrating the absence of disease involvement (349,350). For the most part, a CT scan will provide the information previously sought by the IVP. Ultrasonography Sonographic evaluation of colorectal carcinoma can be performed via trans-cutaneous or intracorporeal approaches. The transcutaneous approach evaluates the liver for metastases and identifies ascites, adenopathy, and an omental cake. Intracorporeal imaging can be performed endoluminally, intraoperatively, or laparoscopically.

100

& PART I: COLORECTAL DISORDERS

Preoperative assessment of the liver by ultrasonography may provide valuable information to consider when recommending the appropriate management of patients with colorectal carcinoma. In a preoperative ultrasonographic study of 195 patients, Grace et al. (351) found a false negative rate of 7.2%. Both rigid and flexible echoendoscopes are available. Because of the lack of input on the management of the patient with a carcinoma above the peritoneal reflection, endosonography currently finds its role in staging rectal carcinoma. The echo-colonoscope, although available, has not been evaluated extensively. With the advent of laparoscopic colectomy, preoperative endosonography may have an increased role in the future. Intraoperative liver ultrasonography is currently being touted as the most accurate method for detecting colorectal metastases. It has served to supplement inspection and palpation at the time of laparotomy. The liver can be surveyed by a radiologist experienced in intraoperative sonography in 5 to 10 minutes. Although peripheral lesions are easily palpable by the surgeon, deeper lesions may be missed by manual examination. Additional findings may very well alter the course of management, either extending or eliminating the planned procedure. Rafaelsen et al. (352) compared diagnostic accuracies of measuring liver enzymes, preoperative ultrasonography, manual palpation, and intraoperative ultrasonography in the detection of liver metastases in 295 consecutive patients with colorectal carcinoma. The presence of metastases was further assessed by ultrasonography 3 months postoperatively. The sensitivity of intraoperative ultrasonography (62 of 64) was significantly superior to that of manual palpation (54 of 64) and preoperative ultrasonography (45 of 64). The lowest sensitivity was presented by the measurement of liver enzymes. The authors concluded that intraoperative ultrasonography reduced the number of patients with liver metastases from being subjected to superfluous or even harmful liver surgery and may increase the number in whom liver surgery will prolong life. They cite six other studies that suggest a higher sensitivity of intraoperative ultrasonography than preoperative ultrasonography in the detection of liver metastases. Meijer et al. (353) studied a series of 122 patients who underwent elective resection—34 with suspected liver lesions seen on preoperative computed tomography (CT) and ultrasonography and 88 without suspected lesions. Of the 34 patients with suspected lesions, the diagnosis was confirmed with intraoperative ultrasonography in 21 and, in the remaining 13 patients, the suspected lesion was shown to be benign. Of the 88 patients with normal preoperative imaging, lesions were detected in five patients. During an 18-month follow-up, 6.5% of patients developed liver metastases not recognized during operation. Machi et al. (354) evaluated 189 patients with colorectal carcinoma at the time of operation and revealed that the sensitivity of intraoperative ultrasonography (93.3%) was significantly higher than that of preoperative ultrasonography (41.3%), conventional CT (47.1%), and manual palpation (66.3%). Metastatic liver lesions were detected solely by intraoperative ultrasonography in 9.5% of patients. During the postoperative follow-up period of 18 months or more, liver metastases that were unrecognized during operation appeared in 6.9% of patients. Reevaluation based on these follow-up results indicated that the sensitivity

of intraoperative ultrasonography decreased to 82.3%, which was still significantly better than that of other methods. Takeuchi et al. (355) reported on a consecutive series of 119 colorectal carcinoma patients who were studied by preoperative extracorporeal ultrasonography, inspection and palpation of the liver at laparotomy, and intraoperative liver ultrasonography. Hepatic metastases were diagnosed in 19 patients—eight by extracorporeal ultrasonography, seven by palpation, and the last four only after intraoperative ultrasonography. Follow-up for a median of 38 months revealed another eight patients with liver metastases detected at a mean time from operation of 14.7 months. The authors concluded that although intraoperative sonography is a sensitive and useful method in detecting liver metastases, some occult hepatic metastases will remain undetected. Laparoscopic ultrasonography has been used to image the liver and may lead to a less invasive manner of staging the patient and providing pathologic confirmation.

Computed Tomography When there is suspicion on ultrasonography of the presence of a hepatic metastatic lesion or if there is clinical suspicion that the primary malignancy has invaded adjacent viscera or anterior or posterior abdominal walls, CT scanning of the abdomen may help delineate the extent of the disease. This may help in the preoperative planning of the extent of the operation. CT scanning of the abdomen is useful in the preoperative investigation of patients to detect occult metastases. Kerner et al. (356) used CT of the abdomen and pelvis to augment the preoperative evaluation of 158 consecutive patients with primary, colorectal carcinoma. In 88 patients findings present on CT were otherwise unknown. Of those, 35% were clinically significant in that they allowed the surgeon to alter the propsed operative procedure or added additional technical information for consideration preoperatively. Findings included liver metastases (26), atrophic kidney (three), and abdominal wall or contigous organ invasion (11). In addition, two other solid organ carcinomas were detected. The authors concluded that CT eliminates the need for a preoperative intravenous pyelogram, improves the preoperative staging of metastatic disease, and provides a baseline for comparison during the postoperative follow-up if recurrence is suspected or adjuvant therapy is planned. CT has a low accuracy for identifying early stages of primary colorectal carcinoma, but is useful in examining patients suspected of having extensive disease or recurrent disease with an extrinsic component at an anastomosis (357). Mauchley et al. (358) assessed the clinical utility of the practice of routine preoperative CT scanning and determined its cost-effectiveness in 130 colon carcinoma patients. CT scans provided information that was used in treatment planning in 33% of patients and definitively altered the mode of treatment in 16% of patients. The practice saved the institution US$24,018 over 6 years. They concluded routine preoperative CT scanning-definitively alters treatment in a small number of cases and is cost-effective. Magnetic Resonance Imaging Magnetic resonance imaging (MRI) is a technique that creates images by evaluating nuclei for the absorption or emission of electromagnetic energy in the presence of a stable magnetic

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

field (359). It has a greater tissue contrast resolution than CT, is multiplanar, and involves no ionizing radiation. The imaging times are longer, the spatial resolution is less than CT, calcifications and bone detail are not as obvious, and the study is more expensive. Certain patients cannot be safely imaged and these include those with cardiac pacemakers, implanted drug infusion pumps, ferromagnetic vascular aneurysm clips, and ocular foreign bodies. MRI has limitations in assessing depth of penetration of the bowel wall but this limitation can be partly overcome by the use of an endorectal coil with up to 90% accuracy. The sensitivity for detecting lymph node metastases ranges from 13% to 40%. Initially MRI was believed to be better than CT in distinguishing between fibrotic changes and recurrences, but this is probably not so. MRI can be used to assess hepatic metastases. Metastatic lesions to the lungs are better defined by CT. For metastatic lesions involving osseous structures, especially the spine and central nervous system, MRI should be the imaging modality of choice. Zerhouni et al. (360) evaluated the accuracy of CT and MRI in staging 478 patients with colorectal carcinoma. CT was more accurate than MRI in the definition of penetration of the muscularis propria by rectal carcinoma (74% vs. 58%). Accuracies were equivalent in depiction of transmural extent of colonic carcinoma. CT and MRI exhibited accuracies of 62% and 64%, respectively, in assessment of lymph node involvement with sensitivities of 48% and 22%, respectively. The accuracy of MRI and CT in the evaluation of liver metastases was equivalent (85%).

Positron Emission Tomography Positron emission tomography (PET) is a method of imaging using a positron-emitting isotope-labeled compound that is incorporated into the biochemical process occurring in organs and tissues of the body. The anatomic and morphologic characteristics are not as well delineated as with other imaging modalities such as CT and MRI, but PET images provide useful information about the nature and physiology of the cellular function of the tissue and have been used to evaluate neoplasms, including colorectal carcinoma. The most widely used isotope is 2-deoxy-2(18F) fluoro-D-glucose, or fluorodeoxyglucose (FDG). Imaging relies on the premise that there is an enhanced rate of glucose in malignant tissue. PET scanning is more accurate than CT scanning in identifying malignancy. This is especially true in the postoperative follow-up in differentiating recurrent carcinoma from fibrosis. Tempero et al. (359) reviewed three studies that compared the results of PET to CT, MRI, and radioimmunoguided scintigraphy (RIGS). In each case PET sensitivity and specificity were very high and better than the other modalities. The major problems are to clarify the role of PET in clinical management and determine who will benefit from this additional expensive, time consuming study. PET scans may have a role in both the diagnosis of and response to therapy of hepatic metastases. Several studies have been performed comparing computed tomography scan with positron emission tomography scan in clinical decision making. Unfortunately, therapeutic decisions are being made based on positron emission tomography scan data without a clear understanding of how well the diagnostic findings correlate with the clinical find-

101

ings. Johnson et al. (361) conducted a retrospective review of 41 patients with metastatic colorectal carcinoma. All patients had both a computed tomography scan and a positron emission tomography scan before surgical exploration. PET scan was found to be more sensitive than CT scan when compared with actual operative findings in the liver (100% vs. 69%), extrahepatic region (90% vs. 52%), and abdomen as a whole (87% vs. 61%). Sensitivities to PET scan and CT scan were not significantly different in the pelvic region (87% vs. 61%). They concluded PET scanning is more sensitive than CT scanning and more likely to give the correct result when actual metastatic disease is present.

& RADIOIMMUNODETECTION Radioimmunodetection of carcinoma is usually accomplished by obtaining whole-body gamma scans of patients who have been injected intravenously with an antibody labeled or conjugated with a gamma-emitting radionuclide (359). In colorectal carcinoma, a variety of antigens have been targeted (e.g., TAG-72 and carcinoembryonic antigen [CEA]). A variety of radionuclides used include iodine-131 (131I), technetium-11 (11Tc), indium-111 (111In), and iodine-125 (125I). Combined results using a variety of antibodies and radionuclides suggests that the sensitivity and detection are high in selected patients; the reports range from a sensitivity of 60% to more than 90% (359). A number of benefits are attributed to radioimmunodetection. Although it can localize metastases, it is not yet clear whether any changes in management translate into long-term patient benefits. Another goal in monitoring recurrent colorectal carcinoma is to distinguish colorectal carcinoma from benign conditions (e.g., postoperative or radiation changes in the pelvis). False positive rates of antibody imaging may be as high as 13%, and it is thus advisable to pursue positive scans with biopsy confirmation (359). Liver metastases may be diagnosed but the identification of extrahepatic sites might spare the patient an unnecessary liver resection. Bertsch et al. (362) reported on 32 patients with primary colorectal carcinoma who underwent RIGS after being injected with anti-TAG-72 murine monoclonal antibody CC49 labeled with iodine-125. Sixteen patients had gross disease and RIGS-positive tissue removed (five with en bloc resection and 11 with extraregional tissues resected, two liver resections, and 25 lymphadenectomies— 10 in the gastrohepatic ligament, 5 celiac axis, 6 retroperitoneal, and 4 iliac) and 16 had only traditional extirpation of disease because RIGS-positive tissue was too diffuse. With a median follow-up of 37 months, survival in the former group was 100%, and 14 of 16 patients had no evidence of disease. In the latter group, 14 of 16 patients died and two were alive with disease. The same authors reported that for patients with recurrent disease, the addition of RIGS increased the detection rate at operation from 116 sites without RICS to 184 with RIGS (a 57% increase) (363). Using murine antibody B72.3 labeled with indium-111 (111In-CYT-103, Cytogen), Dominguez et al. (364) studied 15 patients with recurrent colorectal carcinoma. It was more accurate than a CT scan but when the value of the scan was examined with respect to the potential

102

& PART I: COLORECTAL DISORDERS

contribution to patient management, it was beneficial in only 13% of patients. Moffat et al. (365) assessed the clinical impact of an anti-CEA Fab’ antibody fragment labeled with technetium-99m pertechnetate in 210 patients with advanced or metastatic colorectal carcinoma. When compared with conventional diagnostic modalities, the CEA scan was superior in the extra-hepatic abdomen (55% vs. 32%) and pelvis (69% vs. 48%). Potential clinical benefit was demonstrated in 89 of 210 patients. Corman et al. (366) evaluated the role of immunoscintigraphy with 111Insatumomab pendetide in 103 patients with colorectal carcinoma. In the 84 patients for whom histopathologic information was available, the sensitivity was 73% and the specificity was 100%, with an overall accuracy of 85% for determining the presence and extent of malignant disease. Investigators judged that the antibody imaging mode was a beneficial contribution in 44% and a negative effect in 2% of case studies. Treatment plans were altered in 17 patients. The exact role of immunoscintigraphy has not yet been defined. Galandiuk (367) suggested three potentially major roles of immunoscintigraphy in the management of patients who have undergone curative resection: (1) the detection of recurrent disease in patients with an elevated CEA level and either a negative investigation or equivocal findings on CT scan, (2) the exclusion of extra-abdominal disease, prior to planned resection of a presumably isolated recurrence, and (3) earlier detection of recurrence in the follow-up of high-risk patients. Immunoscintigraphy may allow recurrent disease to be detected at a time when curative resection may still be feasible and result in improved survival and/or effective palliation.

Miscellaneous Investigations The results of hydrocolonic sonography, a technique of trans-abdominal sonography following retrograde installation of water into the colon were compared with conventional transabdominal sonography by Limberg (368). In a study of 29 patients with carcinoma, the diagnosis was correctly made in 97%, compared with only 31% by conventional sonography. Further studies will be required to determine the utility of this technique. Deranged liver blood flow patterns have been detected in patients with hepatic metastases (369). The sugar moiety detected from rectal mucus by the galactose oxidase-Schiff (Sham’s test) has been studied in screening for colorectal carcinoma. Dian-Yuan et al. (370) found a sensitivity of 85.7% for colorectal carcinoma and 47.1% for adenomas in a study of 6480 subjects older than 40 years of age. This compared favorably to 90.5% and 41.2% for fecal occult blood testing in the same group of individuals. In a study of 330 asymptomatic individuals, Sakamoto et al. (371) found an overall specificity of Sham’s test of 92.2%. & CYTOLOGY The value of cytology in establishing the diagnosis of carcinoma for the most part has been appreciated only in investigational studies. To date, clinical application has been limited. Brush cytology via the colonoscope has been reported to be 86% accurate in establishing the diagnosis preoperatively, a% identical to that of biopsy (372). A special

circumstance in which cell brushings may be of value is when a stricture prevents the colonoscope from reaching the lesion, a situation in which a biopsy cannot be performed. In a study of 33 patients, 15 of whom had proven carcinoma, colonic cytology was positive in 93% of cases, whereas the cytology of control patients was negative (373).

& BLOOD MARKERS Liver Function Tests To determine the presence or absence of anemia, a complete blood count is indicated. Liver function tests often will point to metastatic disease, but a normal liver profile does not rule out hepatic metastases. Carcinoembryonic Antigen The initial report of a tumor-specific antigen in human colonic carcinomas by Gold and Freedman (374) heralded a new era in the assessment of the status of patients with colorectal carcinoma. However, CEA evaluation has not fulfilled its promise as a simple blood test that would afford an early diagnosis of carcinoma of the colon. In a most comprehensive treatise on the subject, Gold (375) described his life’s work on CEA. The following information is drawn from that most comprehensive publication. The CEA molecule is considered an oncodevelopmental human marker of neoplasia initially found in adenocarcinomas of the human digestive system. The molecule has a nominal molecular mass of 180 kDa. The CEA gene family comprises 29 gene-like sequences in two defined clusters on chromosome 19. It is of interest that human colonic carcinoma develops in mucosal tissue that has already undergone multiple steps of genetic change. It has been postulated that these multiple steps create a field effect that is characterized by morphologically normal, but biologically altered, epithelial cells. CEA has been used as a phenotypic marker of this field effect by examining the immunohistochemical expression of CEA on morphologically normal mucosa adjacent to colonic adenocarcinomas. It has been shown very clearly that CEA expression occurs in ’normal’ mucosa adjacent to a carcinoma and that there is a gradient of CEA expression, falling off at increasing distances from the carcinoma. These data are relevant to both the biology of human colorectal carcinoma and, more practically, to the optimal location of surgical resection. Kyzer et al. (376), using statin as a marker, similarly demonstrated that the proliferative rate of mucosa adjacent to a colonic carcinoma is elevated and returns to normal at 5 cm from the carcinoma, even though it is morphologically indistinguishable from its normal counterparts. The role of CEA in clinical medicine first became a consideration with the development of a radioimmunoassay for circulating CEA. The first series of data, derived from patients with established colonic carcinoma, were most exciting, but more extensive studies revealed the clearly expected false negative assays, particularly in early-stage bowel carcinoma, and false positive results in patients with nonenteric carcinoma as well as in others with nonmalignant conditions. Over the years, suggestions

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

for the use of the CEA assay have included detection, diagnosis, monitoring, staging and classification (prognosis), pathology, localization, and therapy. Normal concentrations of CEA are 2.5 to 5.0 ng/ml, depending on the assay used. CEA concentrations are in general more often elevated in smokers than in nonsmokers, more frequently elevated in men than in women, and more frequently elevated in older subjects than in younger individuals. Racial differences in the frequency of serum elevations of CEA have been suggested but not established. Elevated CEA levels have been described in advanced breast carcinoma, pancreatic carcinoma, lung carcinoma, and other noncolonic adenocarcinomas, but they do not detect early stages of these diseases. Although 80% or more of patients with advanced colonic adenocarcinoma have circulating CEA, the CEA assay should not be used as the sole diagnostic test for suspected carcinoma. CEA levels are not presently useful for distinguishing locally invasive polyps from benign lesions. Preoperative serum CEA levels in diagnosed colorectal carcinoma are elevated in 40% to 70% of patients. Preoperative serum CEA concentrations correlate inversely with the grade of the carcinoma and directly with the pathologic stage. The CEA is elevated in 95% of patients with well-differentiated lesions, while it is elevated in as few as 30% of those with poorly differentiated adenocarcinomas. The higher the preoperative CEA level, the greater the likelihood of a postoperative recurrence. A significant negative correlation between preoperative elevated plasma CEA levels and patient survival has been observed. Despite disagreement between various groups that have explored the relationship between preoperative CEA levels and prognoses, most studies report that a high preoperative CEA level is indicative of a poor prognosis. This association is often as discriminating as pathologic staging and grading. It is still uncertain which absolute preoperative CEA value reliably discriminates high-risk from low-risk cases for postoperative recurrence (see discussion of Prognostic Discriminants on p. 590). An increase in the blood CEA concentration in a patient after apparently successful surgical treatment for carcinoma has repeatedly been shown to signal a recurrence of the carcinoma. After apparently complete surgical resection of colorectal carcinoma, the blood CEA concentration, if elevated before operation, decreases to the normal range in nearly all patients. The decrease usually occurs within 1 month but sometimes takes up to 4 months. If levels fail to decrease to the normal range, it is likely that the resection has been incomplete or that the carcinoma has already metastasized. A sustained and progressive rise is strong evidence for recurrence at the primary area or at distant sites. Serial CEA monitoring is currently considered by some authors as the best noninvasive technique for detecting recurrent colorectal carcinoma. The debate concerning the merit of postoperative determinations of CEA values in monitoring patients with resected colon carcinoma continues to rage. At one extreme are those who feel that the carcinoma cures attributable to CEA monitoring are too infrequent to justify the substantial costs and physical and emotional stress that this intervention may cause for patients. Others believe that intensive follow-up using CEA assays can identify treatable recurrences at a relatively

103

early stage. It has been suggested that when CEA increases more rapidly than an average of 12.6% per month, recurrence should be strongly suspected. Because the overall prognosis for patients with recurrent disease after surgical resection is dismal, serum CEA determination may offer the only chance of a cure for a select group of individuals. Immunohistochemically, CEA has been identified in carcinomas of the colorectum, breast, lung, uterine cervix, gallbladder, stomach, pancreas, liver, prostate, urinary bladder, and uterus and in neuroendocrine neoplasms associated with the larynx, lung, and thyroid (377,378). There are good reasons to regard ulcerative colitis and certain colonic adenomas as premalignant lesions. Immunoperoxidase staining for CEA supports the concept of a polyp-adenoma-carcinoma sequence. Both chronic inflammatory bowel disease and colorectal adenomas show higher tissue CEA concentrations than normal colonic mucosa, suggesting that these situations can be regarded biochemically as premalignant conditions. Results concerning the correlation between positive CEA test results in the carcinoma immunohistochemistry and histologic grade, lymph node involvement, locoregional recurrence, disease-free interval, and patient survival remain controversial. Initial difficulties have been overcome and several, virtually instant, and easy-to-use radiolabeling kits for antiCEA antibodies are available. The reported sensitivities for the detection of liver metastases have ranged from 0% to 94% in different studies, indicating differences largely due to technical ability and knowledge in avoiding pitfalls. The most complete clinical report so far was conducted in Italy. F(ab)2 fragments of the monoclonal antibodies antiCEA FO23C5, determined to be more suitable than intact immunoglobulin (IgG) or Fab fragments for immunoscintigraphy, were labeled with either iodine-131 or indium-111. The variation in results reported by various groups reflects the gamut of potential variables, including the radiolabel used, the route of administration of the conjugate, the size and location of the lesion, the vascularity of the lesion, the patient population studied, the imaging technology used, the unavoidable parameter of subjective interpretation of scans, even in blinded situations, and the type of antibody preparation used. On the basis of the foregoing information, it may, therefore, be concluded that primary lesions with a high CEA content have the best anti-CEA antibody uptake and are most easily imaged and that large fungating carcinomas accumulate a high proportion of the injected labeled antibody in contrast to ulcerating carcinomas with poor vascularity. Hepatic metastases have a high CEA content and a high antibody uptake but may not image well because of the relatively high background uptake of the conjugate by the normal liver. To date, adjuvant chemotherapy has not been very successful in those patients whose carcinomas have seen resected but who are at high risk for recurrence (i.e., Dukes’ B and C lesions). The ability to detect carcinomas by radioimmunolocalization raises the possibility of treating such lesions by targeting with the same technology. An even more intriguing approach to the treatment of CEA-producing carcinomas has been developed by incorporating the recombinant CEA gene, in whole or in part, into an appropriate vector, such as the vaccinia virus. This construct, with or without other gene products to enhance immunity, has been

104

& PART I: COLORECTAL DISORDERS

shown to enhance cell-mediated anti-CEA immunity in both animal and human studies. Based on the results to date, a clinical trial of this form of therapy has been initiated.

Other Markers A host of less well-described markers have been reported to have potential screening value. For example, ornithine decarboxylase activity has been found to be significantly lower patients with adenomas than in controls, and even lower in patients with carcinoma (379). Opposite findings were reported by Narisawa et al. (380), who found increased levels in patients with carcinoma. Compared with its activity in control mucosa, urokinase activity has been found to be significantly elevated in adenomas and carcinomas, with levels significantly higher in carcinomas (381). Tissue plasminogen activator activity is reduced in adenomas and carcinomas, with levels significantly lower in carcinomas.

& TREATMENT The evolution of operative treatment for carcinoma of the colon has involved several stages because of the original enormous risk of sepsis. These stages have included initially nothing more than a diverting colostomy, subsequent recommendations for exteriorization and double-barrel colostomy, efforts at reestablishing intestinal continuity with various internal stents, resection and anastomosis with proximal diversion, and ultimately resection and primary anastomosis. The acceptance of a one-stage procedure was achieved through a better understanding of the mechanical and antimicrobial bowel preparation.

& CURATIVE RESECTION Preoperative Evaluation The general assessment of the patient is discussed in Chapter 4. Fazio et al. (382) described a dedicated prognostic index for quantifying operative risk in colorectal carcinoma surgery from data collected from 5034 consecutive patients undergoing major surgery. Primary end point was 30-day operative mortality. The patients’ median age was 66 years. Operative mortality was 2.3% with no significant variability between surgeons or through time. Multivariate analysis identified the following independent risk factors: age (OR ¼ 1.5 per 10-year increase), American Society of Anesthesiologists grade (OR for ASA II, III, IV-V vs. I ¼ 2.6, 4.3, 6.8), TNM staging (OR for Stage IV vs. I– III ¼ 2.6), mode of surgery (OR for urgent vs. nonurgent ¼ 2.1), no-carcinoma resection versus carcinoma resection (OR ¼ 4.5), and hematocrit level. The model has implications in every day practice, because it may be used as an adjunct in the process of informed consent and for monitoring surgical performance through time. Bowel Preparation Preoperative bowel preparation has been the subject of considerable controversy. It has long been believed that adequate mechanical preparation and antibiotic preparation are necessary. This has recently been brought into question (see Chapter 4). How each of these is accomplished has also been debated. Mechanical cleansing

may be accomplished by the use of vigorous laxatives along with repeated enemas until clearing. For a number of years, an oral lavage with a polyethylene glycol hypertonic electrolyte solution, such as GoLytely, was used extensively and still is by some colorectal surgeons. Oral Phospho-soda preparations have become increasingly popular but certain precautions regarding their use are discussed in Chapter 4. In the area of antibiotics, controversy exists as to whether the patient should receive oral or systemic antibiotics or possibly both. Further, there is the question of which antibiotic is the most appropriate. Clearly, whichever antibiotics are chosen, they should be selected on the basis of gram-positive and gram-negative aerobic and anaerobic coverage. There is no question that there are a number of acceptable combinations. Further argument centers around the timing of antibiotic administration, but it is certain that antibiotics should be started preoperatively. The duration of antibiotic administration is also controversial, but the antibiotic is probably not necessary after the day of operation. Our current antibiotic regimen consists of one preoperative dose and two postoperative doses of systemic ticarcillin (Timentin 3.1 gm) (383).

Exploration of Abdomen Before operation a Foley catheter is routinely inserted into the bladder. A nasogastric tube is not necessary in the vast majority of cases, but for those cases in which it becomes necessary, the tube can be inserted during the postoperative period. For access to the abdominal cavity, incisions should be made in a manner that provides maximum exposure for the planned resection. All operations can be performed through a midline incision and this is the access preferred by most surgeons. For a patient about to undergo a right hemicolectomy, an oblique right-sided abdominal incision is usually most adequate; if necessary, the incision can be extended. For transverse colon lesions, a supraumbilical transverse incision places access immediately in the area of the planned operation, and the incision can be extended in either direction if there is any difficulty encountered in taking down the flexures. For splenic flexure lesions, Rubin et al. (384) advocate the use of a left subcostal transverse incision combined with the right lateral position. For leftsided colonic lesions, a subumbilical transverse incision can be used. This permits adequate exposure for even a low anterior resection. For descending colon lesions, an oblique incision may prove very convenient. The use of paramedian incisions appears quite antiquated. For emergency operations, a midline incision seems the access of choice. When the abdomen has been opened, attention should be directed to ruling out the presence of metastatic disease, with special attention given to the liver and the pelvis. A relatively new technique advocated for the detection of occult hepatic metastases is intraoperative contact ultrasonography (385). Lesions > 1 cm in diameter can be detected in 95% of cases, and those between 0.5 cm and 1.0 cm, in 66% of cases. After assessment of the abdomen, attention is focused on the primary lesion to determine its resectability. Principles of Resection Dogma abounds with respect to the technical aspects of operation for colorectal carcinoma. The general principles

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

FIGURE 28 & Extent of resection for carcinoma in the cecum or ascending colon.

advocated for all operations for carcinoma include removal of the primary lesion with adequate margin, including the areas of lymphatic drainage. The definition of an adequate margin especially for rectal carcinoma remains controversial. Approximately one half of the patients seeking operative treatment already have metastatic disease spread to the regional lymph nodes. Controversy exists as to the

FIGURE 29 & Extent of resection for carcinoma in the hepatic flexure.

105

FIGURE 30 & Extent of resection for carcinoma in the transverse colon.

appropriate extent of lymphatic dissection. Is a segmental resection adequate therapy? Or with a sigmoid carcinoma, for example, should a formal left hemicolectomy be performed? For the most part the literature suggests that no survival advantage can be attributed to extended lymph node dissections for left colon and rectal carcinoma (386–389). An exception is the isolated report by Enker, Laffer, and Black (390). Without doubt this operation does result in increased morbidity, with patients often suffering from impotence, bladder difficulties, and potential vascular problems. Any marginal improvement is outweighed by the considerable morbidity. The principle of en bloc resection of involved structures is firmly established. Continued controversy surrounds radical lymph node dissection, luminal ligation, oophorectomy, and the ‘‘no-touch technique.’’ What is becoming increasingly evident is that differences in outcome among different surgeons suggest that technique is important. Whether a properly performed lymphadenectomy may produce a therapeutic benefit or whether it is simply a more accurate staging procedure is unknown. For lesions located in the cecum or the ascending colon, a right hemicolectomy to encompass the bowel served by the ileocolic, right colic, and right branch of the midcolic vessels is recommended (Fig. 28). For lesions involving the hepatic flexure, a more extended resection of the transverse colon is indicated (Fig. 29). For lesions in the transverse colon, depending on the portion involved, a segment of bowel is removed as shown in (Fig. 30). Splenic flexure lesions require removal of the distal half of the transverse colon and the descending colon (Fig. 31). Sigmoid lesions are appropriately treated by excision of the sigmoid colon (Fig. 32). Some surgeons prefer more radical excisions, but there is no convincing evidence to suggest that prolonged survival or decreased local

106

& PART I: COLORECTAL DISORDERS

FIGURE 31 & Extent of resection for carcinoma in the splenic flexure.

recurrence will result (Fig. 33). Indeed, operative mortality and postoperative complications are reportedly higher (386). For patients who have synchronous carcinomas in different portions of the colon, a subtotal colectomy seems appropriate (Fig. 34). Other suggested indications for subtotal colectomy include associated polyps (not removed by colonoscopy), acute or subacute obstruction, associated sigmoid diverticulosis (symptomatic), prior transverse colostomy for obstruction, young patient age (< 50 years)

FIGURE 32 & Extent of resection for carcinoma in the sigmoid colon.

FIGURE 33 & Extended resection preferred by some surgeons for carcinoma in the sigmoid colon.

with a positive family history, and adherence of the sigmoid colon to a cecal carcinoma (391). The techniques described in the following section pertain to good-risk patients. For poor-risk patients or patients undergoing palliative resection, segmental resections are more appropriate. Certain intraoperative precautions have been proposed to eliminate or at least minimize the dissemination of malignant cells. Concern has been expressed that manipulation of the carcinoma results in blood-borne metastases. There is also the risk of exfoliated malignant cells adjacent to the primary lesion becoming implanted at the suture line, in the peritoneal cavity, or in the wound. It has been postulated that handling of the primary lesion early in the operation promotes such dissemination. This thesis was supported by the demonstration of malignant cell in the circulation (392,393), a finding that led Turnbull et al. (394) to popularize the no-touch technique in which lymphovascular channels were ligated prior to any manipulation of the primary lesion. After using this maneuver, they reported improved survival rates for Dukes’ C lesions. However, this was not a controlled trial, and Turnbull’s results have not been duplicated. Therefore the technique has not been adopted by most surgeons as standard therapy. In an effort to avoid implantation of malignant cells shed from the primary carcinoma, Cole, Packard, and Souffiwic (393) recommended encirclement of the bowel lumen proximal and distal to the primary lesion. This is a simple addition to the operation and can usually be performed easily. Wound edges can be covered to prevent malignant cell implantation in the wound. In an effort to diminish the risk of implantation of malignant cells distal to the lesion, a host of cytotoxic agents (e.g., Dakin’s

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

FIGURE 34 & Extent of resection for synchronous carcinomas in different portions of the colon.

solution or bichloride of mercury) have been used to irrigate the distal bowel; distilled water has also been used for this purpose. Since none of these cytotoxic agents has been used in a clinical trial setting, their value is in question. As a means of diminishing suture line implantation, iodized catgut was popular for a brief time (395). Another effort is the intraluminal installation of diluted formalin, in which case local recurrence was reportedly reduced to 2.6% from 14.3% for untreated patients (396). None of the techniques tried thus far, including irrigation of the peritoneal cavity or the use of iodized catgut, has been proven of value. However, proximal and distal ligation of the bowel would appear to be a harmless practice. Adjuvant chemotherapy administered directly into the bowel lumen at the time of operation has been used but not in a trial setting. These empirical maneuvers were the forerunners of the clinical trials currently in progress.

107

FIGURE 35 & Incision of parietal peritoneum.

and horizontally along the upper border of the transverse colon, with division of any adhesions to the gallbladder. As dissection continues, the lesser sac is opened by dividing the gastrocolic ligament. Mobilization is continued for as far as the resection is planned. During this stage of the operation, the second and third portions of the duodenum are

Technique RIGHT HEMICOLECTOMY & With the appropriate retractor

in place and the small bowel packed toward the left side of the abdomen, the procedure is begun by incising the parietal peritoneum from just below the terminal ileum toward the hepatic flexure (Fig. 35). This can be done with Metzenbaum scissors or preferably by use of diathermy. If feasible, the colon is encircled above and below the carcinoma with umbilical tapes. This procedure is performed as soon as the lesion is appropriately mobile. The right colon is elevated from the retroperitoneum, with care taken not to injure the ureter, gonadal vessels, or inferior vena cava (Fig. 36). As dissection is carried toward the hepatic flexure, attention is given to avoiding any injury to the duodenum. Peritoneal division is continued around the hepatic flexure

FIGURE 36 & Mobilization of the right colon. Exposure of the duodenum with care taken to avoid injury to retroperitoneal structures.

108

& PART I: COLORECTAL DISORDERS

exposed, and caution should be exercised to prevent injury to this structure. Next, the greater omentum is transected vertically (Fig. 37). The medial aspect of the peritoneum then is incised along the planned area of resection. Mesenteric attachments are divided until the vascular anatomy is clear. The vessels are now displayed, and their trunks are triply clamped and divided, and the remaining end is doubly ligated (Fig. 38). The ileocolic, right colic, and right branches of the middle colic artery are divided in turn. The small vessels adjacent to the small bowel and transverse colon at the level of the proposed transection are divided between clamps, and hemostasis is secured. The two ends of bowel are now ready for division (Fig. 39). The technique the surgeon adopts—stapling or suturing—will-direct how the bowel is handled. Surgeons, who advocate the no-touch isolation technique ligate the lymphovascular structures as the initial maneuver of the operation. An incision is made in the root of the mesentery, and the trunks of the vessels are identified, divided, and ligated prior to any mobilization. The major concern with this method is the potential need to deal with the ureter, gonadal vessels, and duodenum without the benefit of adequate exposure, which may keep these structures out of harm’s way. In terms of long-term survival advantage, the efficacy of this method has not been supported in surgical trials against standard operative procedures. In a multicenter prospective randomized controlled trial, Wiggers, Jeekel, arid Arends (397) compared the no-touch isolation technique to a conventional technique. Both overall and corrected survival data did not differ between the two groups, although there was a tendency toward reduction in the number of occurrences

FIGURE 37 & Transection of the greater omentum.

FIGURE 38 & Ligation and division of vessels.

and the length of time to the development of liver metastases with the no-touch technique. For the reestablishment of intestinal continuity, my preference is to use staplers. A functional end-to-end anastomosis is created in the following way (398). Once the site of transection has been selected, a small area (less than that necessary for hand-sutured anastomoses) is cleared. Enough fat is cleared from the edge so that when the bowel opening is closed, there is room for the linear stapler to be applied without inclusion of mesenteric fat or appendices epiploicae. Obesity per se is not a contraindication to the use of staplers. In fact, since less clearing is necessary, staplers may have an advantage in such circumstances. The linear anastomosing instrument is applied in the mesenteric-antimesenteric plane (Fig. 40A). If the bowel diameter is too large to fit within the jaws of the instrument, such as with the transverse colon, the instrument is placed so that

FIGURE 39 & Application of stapler to divide the bowel.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

109

FIGURE 40 & (A) After complete bowel mobilization, the segment of bowel to be resected is transected with the anastomosing stapling instrument. Meseriteric edges are approximated. (B) Alignment of proximal and distal resected ends with closure of the mesentery completed. (C) Excision of the antimesenteric corner to accommodate the anastomosing stapler. (D) Insertion of each limb of the instrument into the bowel ends to be anastomosed. (E) With the bowel ends snugly fitted to the neck of the stapler to-provide maximum length of anastomosis, the instrument is activated and the anastomosis is accomplished. Suture is placed through antimesenteric border of bowel just beyond the bowel anastomosis. (F) Staple line is carefully inspected for proper completion and possible bleeding from the line. (G) Approximation of the bowel edges performed so that the previous staple lines are staggered and the side-to-side anastomosis created is in the open shape of a V. Application of Allis clamps to include the full thickness of the bowel wall and. the complete circumference of the bowel. (H) Application of the linear stapler with excision of excess tissue. Alternatively, bowel opening can be closed with the anastomosing instrument. (I) Completed anastomosis with the staple lines clearly demonstrated.

110

& PART I: COLORECTAL DISORDERS

the tips of the instrument are on the antimesenteric border. Then if the bowel is not completely transected and stapled closed, the open area will occur where an opening would have been made to create the functional end-to-end anastomosis. The advantage of a stapled transection is that little or no devascularization of the bowel is necessary. With manual anastomoses too little clearing may make anastomotic suturing insecure, and too much may jeopardize the viability of the bowel ends. Caution, however, must be exercised in patients with diverticular disease. It is often easier to close the mesentery before the anastomosis is constructed, especially with a right hernicolectomy and with patients who are overweight. Completing this step prior to constructing the anastomosis also will diminish the possibility of the error of rotation of the ileum (Fig. 40B). The antimesenteric borders are aligned, and at each corner an amount of tissue just adequate to insert the fork of the anastomosing instrument is excised (Fig. 40C). The instrument is inserted to its full length to create a large anastomosis (Fig. 40D). The halves of the instrument are joined and the bowel is drawn up, which ensures an anastomosis that is the full length that the instrument is capable of creating (Fig. 40E). The anastomosis should be checked to see that it is being created through the antimesenteric border and that no fat, omentum, mesentery, sponge, or viscus is trapped. It might be best to insert the ‘‘sleeping suture of Steichen’’ before the instrument is removed. The suture is placed at the end of the anastomosis on the antimesenteric border of the bowel to diminish any possible tension on the anastomosis. It should be noted that the use of this suture is omitted in patients with Crohn’s disease for fear of possible development of a fistula. Alternatively, the suture may be placed prior to insertion of the stapler. The halves of the instrument should be gently separated because excessive force may result in disruption of the anastomosis. Some bleeding at the suture line is not uncommon and may be controlled by sponge compression, light cautery, or suture ligature. Excessive cautery may result in a weakened anastomosis and may predispose to leakage. Heavy bleeding should be controlled with fine sutures. The anastomotic suture lines are held apart in preparation for the application of the linear stapler (Fig. 40F). Welter, Charlier, and Psalmon (399) showed that with the functional end-to-end anastomosis the area of the anastomosis can be increased by up to one third of the original bowel lumen if the linear stapler closing staple line is applied so as to hold the anastomosing staple lines in an open V position. There is also the very remote possibility

that if the suture lines remain in apposition, unwanted healing may occur from one to the other. The staple lines should be staggered when the linear stapler is applied (Fig. 40G). This modification was suggested by Chassin, Rifkind, and Turner (400) to avoid too many intersecting staple lines, which may create an ischemic point with potential for a leak. Allis clamps are applied to the tissues being approximated to prevent a portion of the bowel circumference from slipping back as the jaws of the instrument compress the tissue (Fig. 40H). The instrument is fired, and the excess tissue is cut away prior to release of the instrument to avoid injury to the anastomosis. A fine ooze of blood is reassuring of a good blood supply to the anastomosis, but more brisk bleeding should be controlled with gentle cautery or a fine suture. Alternatively, the bowel opening can be closed with the application of the anastomosing instrument. This technical variation has been suggested to reduce the cost of the anastomosis. If adopted, care should be taken to ensure that the instrument application does not compromise the size of the anastomosis. The anastomosis is checked to ensure that it is complete and no leak is present (Fig. 40I). To determine the results of our experience with the use of staples for construction of anastomoses following colonic resection, we reviewed a series of 223 anastomoses performed in 205 patients (401). Indications for operation included malignancy, benign neoplasms, inflammatory bowel disease, and several miscellaneous entities. A functional end-to-end anastomoses using the standard GIA cartridge and the TA 55 instruments was performed. The operative mortality rate was 1.5%, with none of the deaths related to the anastomosis. Intraoperative complications encountered included bleeding (21), leak (one), tissue fracture (one), instrument failure (four), and technical error (three). Early postoperative complications related to or potentially related to the anastomosis included bleeding (five), pelvic abscess (one), fistula (one), peritonitis (two), and ischemia of the anastomosis (one). Late complications included five patients with small bowel obstruction, two of whom required operation. Anastomotic recurrences developed in 5.9% of patients. Our experience with stapling instruments has shown them to be a reliable method for performing anastomoses in the colon in a safe and expeditious manner. Complications after functional end-to-end anastomoses reported by other authors are depicted in Table 7. A variety of commercial instruments are available for the construction of stapled anastomoses (Fig. 41).

TABLE 7 & Complications After Functional End-to-End Anastomosis

Author(s) Chassin et al. (402) (1978) Fortin, Poulin, and Leclerc (403) (1979) Brodman and Brodman (404) (1981) Reuter (405) (1982) Scher et al. (406) (1982) Steichen and Ravitch (407) (1984) Tuchmann et al. (408) (1985) Kyzer and Gordon (401) (1992) Kracht et al. (409) (1993)


Not addressed.

No. of Cases 181 118 88 69 35 264 51 223 106

Bleeding (%) 0 0 0 0


0.4 2.0 2.2


Fistula or Leak (%)

Intraperitoneal Abscess (%)

Obstruction or Stenosis (%)

Operative Mortality (%)

1.1 5.0 0 9.0 2.9 3.4 6.0 0.9 2.8

1.7 0 2.3

1.1 0.8













0.7 2.5 0 2.9 8.6




0.8










0.4 1.9

0

0.4 1.5 1.9







CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

111

FIGURE 41 & Instruments for constructing stapled anastomoses. (A) Proximate linear cutter. (B) Proximate linear stapler. (C) Flexible proximate access stapler. (D) GIA staplers. (E) TA stapler. (F) Roticulator. (G) Computer-powered linear surgical stapling products: (a) Reusuable 60 mm linear cutter, (b) right-angle linear cutters: 45 mm, 30 mm, and 30 mm vascular. Source: (A, B, C): courtesy of Ethicon Inc., Sommerville, NJ; (D, E, F): courtesy of U.S. Surgical Corp., Norwalk, CT; (G): courtesy of Power Medical Interventions, Langhorne, PA.

112

& PART I: COLORECTAL DISORDERS

FIGURE 42 & (A) Posterior placement of outer layer of seromuscular interrupted sutures. (B) Placement of continuous inner layer of absorbable sutures. (C) Continuation of inner layer of sutures anteriorly. (D) Completion of anastomosis by anterior placement of layer of interrupted seromuscular sutures.

When a hand-sutured anastomosis is elected, the bowel edges are transected obliquely to ensure adequate blood supply at the bowel edges to be anastomosed. Today most surgeons prefer end-to-end anastomoses. Even where there is a disparity in size of the bowel lumina, as occurs in anastomoses of the ileum to the transverse colon, the discrepancy can be readily overcome by division of the antimesenteric border of the ileum. Considerable controversy has been engendered about whether to use a onelayer or two-layer anastomosis and the type of suture material to be used. The two-layer technique, which I have used successfully in the past, consists of a posterior row of 4–0 silk or vicryl placed on a fine atraumatic needle into the seromuscular layer. An inner layer of 4–0 chromic catgut or vicryl is placed through the full thickness of the bowel wall, begun on one edge, and continued on the posterior wall in a simple running over-and-over suture but with a change to the Connell suture on the anterior half. The anterior seromuscular layer is then completed with 4–0 silk or vicryl sutures (Fig. 42). A growing number of surgeons have favored a singlelayer inverting interrupted technique. A posterior interrupted single layer of an absorbable suture such as 3–0 or 4–0 Vicryl or Dexon has been used. The suture is then continued on the anterior wall (Fig. 43). Care must be taken not to invert excessive amounts of tissue, thereby causing narrowing of the lumen, but this is true for any type of hand-sutured anastomosis. Some surgeons prefer to use a Gambee suture for the single-layer anastomosis (Fig. 44). Other suture materials such as Prolene and wire have been used. Copious saline irrigation of the abdominal cavity is performed to remove blood, bacteria, and debris. Drains are not necessary. Wounds are closed with continuous absorbable sutures for the peritoneum and fascia, with staples or subcuticular continuous absorbable material used for the skin. Resection of Transverse Colon The appropriate operation for a carcinoma of the transverse colon has been a controversial matter. The reason is the

desire to fulfill the criteria for resection of the regional lymphatic drainage. Depending on portion of the transverse colon that is involved, drainage may occur through the middle and/or right colic branches and possibly the left colic branches. For a lesion that is located in the midtransverse colon, a transverse colectomy would be in order. The procedure might begin with division of the greater omentum from the greater curvature of the stomach, either above or below the gastroepiploic arterial arcade, with care taken not to injure the wall of the stomach

FIGURE 43 & (A) When disparity in bowel ends exists, smaller end may be fishmouthed. Alignment of bowel ends for anastomosis. (B) Placement of single layer of interrupted sutures in a posterior row. (C) Completed anastomosis by placement of anterior row of interrupted sutures. Source: From Ref. 410.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

FIGURE 44 & Placement of the Gambee suture is begun with the fullthickness bowel and continued with the mucosa and submucosa on the same side. Suturing is continued on the opposite side with the submucosa and mucosa and then the full-thickness wall. Inverted anastomosis is thus created.

(Fig. 45). In the event of a very redundant transverse colon, the omentum may be divided vertically on either side at the proposed proximal and distal lines of resection of the colon. For a short transverse colon, the entire omentum may be included in the resected specimen. To avoid tension on the anastomosis, either one or both of the hepatic and sple-

FIGURE 45 & Division of the greater omentum from the greater curvature of the stomach.

113

nic flexures will require mobilization. It is often technically easier to resect the right and transverse colon rather than attempt to mobilize both flexures. The technique of mobilization of the hepatic flexure has been described in the discussion of right hemicolectomy. Mobilization of the splenic flexure is facilitated by incising the lateral peritoneal attachment along the descending colon (Fig. 46). As the splenic flexure is approached, great care must be exercised to avoid injury to the spleen. The lienocolic ligament can be accentuated by passage of a finger along the colonic wall from the descending colon side toward the splenic flexure. The ligament then can be clamped and divided, or, alternatively, it can be divided with the use of cautery. Great caution should be exercised in this maneuver since there are frequently numerous adhesions to the splenic capsule. The peritoneum is incised on the mesocolon, and in the process the splenic flexure is mobilized downward and to the right, exposing the retroperitoneum. If the greater omentum becomes a limiting factor, division of the omentum is begun. Varying other posterior attachments may require division, with care taken not to incite bleeding in this location. The trunk of the middle colic vessel and smaller vessels are secured (Fig. 47). It should be noted that the origin of the middle colic vessels is quite proximal on the superior mesenteric vessels and must be pursued with extreme caution to prevent injury to these structures. The bowel is divided and an anastomosis is created as previously described. In reconstituting the mesenteric defect between the ileum and the descending colon, care

FIGURE 46 & Mobilization of the splenic flexure by division of the lienocolic ligament.

114

& PART I: COLORECTAL DISORDERS

between the distal transverse and proximal sigmoid colon. Some surgeons advocate a more formal left hemicolectomy.

FIGURE 47 & Ligation and division of the middle colic and adjacent smaller vessels.

must be exercised to avoid narrowing of the duodenojejunal junction. For lesions at or near the hepatic flexure or the ascending colon, a right hemicolectomy is performed. For lesions near the splenic flexure, a partial left colectomy with anastomosis of the transverse colon to the proximal sigmoid is performed (Fig. 48). Resection of this type may necessitate division of the left branch of the middle colic and left colic vessels. Resection of Descending Colon For lesions of the descending colon, the left branch of the middle colic artery remains intact, but the left colic artery and, depending on the level of the lesion, the first sigmoidal vessels are ligated. The anastomosis is performed

Sigmoid Resection Some controversy exists as to the most appropriate procedure for removal of a sigmoid carcinoma. One school of thought supports the necessity for a radical left hemicolectomy with anastomosis of the transverse colon to the rectum. However, there is a growing number of surgeons who realize that the extended resection has not resulted in increased survival rates. When patients who have lymphatics involved to the root of the inferior mesenteric artery have these resected, no increased survival rate is noted in comparison with patients who have a less radical procedure (388). It would, therefore, seem that the extra mobilization, with its potential risks and prolonged operating time, is not justified. The extent of the resection depends on the portion of the sigmoid colon involved. Lesions of the proximal sigmoid would require an anastomosis performed between the descending colon and the distal sigmoid, those of the distal sigmoid would involve an anastomosis between the proximal sigmoid and the upper rectum, and those of the mid-portion of the sigmoid, depending on the redundancy of the colon, would require an anastomosis between the sigmoid-descending junction and the rectosigmoid. The splenic flexure is not routinely mobilized, but, depending on the location of the lesion and the redundancy of the colon, it may require mobilization to avoid tension on the anastomosis. The patient may be placed in the supine position, but for more distal lesions it is preferable to have the patient in the modified lithotomy position so that simultaneous access can be obtained through the abdomen and the rectum. This access is necessary to allow use of the circular stapling device or inspection of the anastomosis by proctosigmoidoscopy.

FIGURE 48 & (A) Extent of resection for carcinoma near the splenic flexure. (B) Result after resection.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

115

FIGURE 51 & Mobilization of the sigmoid colon with care taken not to injure retroperitoneal structures. FIGURE 49 & Incision of peritoneum along the white line of Toldt.

The procedure is initiated by incising the peritoneum along the white line of Toldt in the left paracolic gutter, freeing the distal descending colon and the sigmoid from their developmental attachments from the splenic flexure to the pelvic brim (Fig. 49). In the midportion of the sigmoid meso-colon is the intersigmoid fossa, a small depression in the peritoneum that acts as a guide to the underlying ureter (Fig. 50). As the sigmoid mesentery is further mobilized, care is taken to displace the mesosigmoid from the left ureter, which is seen coursing over the iliac vessels (Fig. 51). The gonadal vessels should be protected in a similar way because injury will result in troublesome bleeding. After lateral mobilization and determination of the proximal line of resection, the peritoneum over the medial aspect of the mesosigmoid is incised toward the root of the inferior mesenteric artery to the level of the proposed ligation and then downward toward the pelvis. The inferior mesenteric artery, with its left colic and sigmoidal branches, will be identified (Fig. 52). The inferior mesenteric artery distal to

FIGURE 50 & Intersigmoid fossa.

the left colic branch is then divided and ligated. Smaller vessels leading toward the planned lines of resection are secured, and the bowel is transected proximally and distally. As with the technique for right hemicolectomy, some surgeons advocate ligation and division of the blood supply prior to any other manipulation, but the same general principles pertain. In such a situation, depending on the extent of resection, the inferior mesenteric artery at its origin from the aorta (or distal to the left colic branch) and the inferior mesenteric vein at the level of the duodenum (or more distally for a lesser resection) require ligation and division. Abcarian and Pearl (411) have described a simple

FIGURE 52 & Identification of the inferior mesenteric artery and its left colic and sigmoidal branches.

116

& PART I: COLORECTAL DISORDERS

FIGURE 53 & (A) High ligation of the inferior mesenteric vein lateral to the fourth portion of the duodenum through a paraduodenal peritoneal incision. (B) High ligation of the inferior mesenteric artery at its origin through a diagonal extension of the paraduodenal incision medially to expose the infrarenal aorta.

technique for high ligation of the inferior mesenteric artery and vein. After completion of abdominal exploration, the small bowel is packed away toward the right side of the abdominal cavity to expose the duodenojejunal flexure. The peritoneum overlying the lateral border of the fourth portion of the duodenum is incised, exposing the inferior mesenteric vein. The vein is mobilized for 2 to 3 cm, ligated in continuity with non-absorbable suture, and divided. This incision is extended diagonally 5 to 6 cm medially to expose the infrarenal aorta proximal to its bifurcation. The inferior mesenteric artery is easily identified, doubly ligated in continuity, and divided at its origin. The lymph nodes surrounding the takeoff of the inferior mesenteric artery are dissected sharply in a proximal-to-distal manner to allow for their complete excision (Fig. 53). Heald (412) recommends division of the inferior mesenteric artery approximately 2 cm from the aorta to preserve the autonomic nerves, which split around its origin. The anastomosis is then created according to the surgeon’s method of choice. However, if the anastomosis is low, my preference is to use the circular stapler as described in detail in Chapter 24. For surgeons who deem it necessary to perform a radical left hemicolectomy, the operation is conducted in a similar way by combining the mobilization of the sigmoid, the splenic flexure, and the distal transverse colon. The notable difference is the level at which the vessels are secured. To accomplish the radical left hemicolectomy, the posterior parietal peritoneum is incised to expose the inferior mesenteric vessels. The artery is tied flush with the aorta, and the vein is ligated separately at the level of the duodenum (Fig. 54). Bilateral Salpingo-Oophorectomy In a review of their experience and the surgical literature, Birnkrant, Sampson, and Sugarbaker (413) found the

incidence of ovarian metastases from colorectal carcinoma to be approximately 6% with a range of 1.5% to 13.6%. In a prospective controlled study, Graffner, Aim, and Oscarson (414) detected ovarian metastases in 10.3% of patients undergoing operations on all segments of the large bowel. Since bilateral involvement occurs between 50% and 70% of the time, a bilateral oophorectomy is recommended, especially for postmenopausal women. However,

FIGURE 54 & High ligation of the inferior mesenteric artery and vein.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

controversy exists about the role of prophylactic oophorectomy dunrig resecticin of primary colorectal carcinoma. Sielezneff et al. (415) attempted to prospectively assess the prognostic impact of simultaneous bilateral oophorectomy in postmenopausal women undergoing curative resection for colorectal carcinoma. Ovarian metastases were detected in 2.4% of the operative specimens. Local recurrence or liver metastases rates were not affected by oophorectomy. Five-year actuarial survival rates were not significantly different whether patients had oophorectomy (81.6%) or not (87.9%). Their results suggested that microscopic synchronous ovarian metastases is rare at the time of curative resection of a colorectal carcinoma in postmenopausal women and does not modify prognosis. Young-Fadok et al. (416) conducted a prospective randomized trial of 152 patients to evaluate the influence of oophorectomy on recurrence and survival in patients with Dukes’ B and C stage colorectal carcinoma. In 76 patients randomized to oophorectomy, no incidence of gross or microscopic metastatic disease to the ovary was found. Preliminary survival curves suggest a survival benefit for oophorectomy of 2 to 3 years after operation, but this benefit does not appear to persist at 5 years (Fig. 55). There has been no incidence of colorectal carcinoma metastatic to the ovaries in this series of Dukes’ B and C stage carcinoma, unlike other nonrandomized studies of all stages, which have reported a 4% to 10% incidence. These authors concluded that the possibility of a survival advantage emphasizes the need to continue this preliminary work. Concomitant oophorectomy is controversial, and its efficacy in prolonging survival has been questioned since few patients survive 5 years after operation. However, it generally adds little to the operation and will prevent the subsequent development of ovarian carcinoma (417). Removal of the ovaries at the time of bowel resection will eliminate the need for repeat laparotomy to resect an ovarian mass in approximately 2% of women with large bowel carcinoma (413). Oophorectomy is often recommended more strongly for patients who have carcinoma of the rectum, but no site in the colon results in a greater proportion of ovarian metastases. The apparent greater proportion from the left colon is probably due to the fact that this portion of the large bowel harbors the largest number of malignancies.

FIGURE 55 & Survival curves for prophylactic oophorectomy during resection for primary colorectal carcinoma. Source: From Ref. 416.

117

Oophorectomy should be performed in premenopausal women if any gross abnormality of the ovary is detected (413). Indeed, ovarian metastases of colorectal origin have been reported to occur more commonly in premenopausal women, with rates ranging from 3.8% to 28% (413,418). This finding might support the recommendation for prophylactic oophorectomy regardless of patient age (419). Certainly, contiguous involvement necessitates en bloc resection. Ovarian involvement carries with it a poor prognosis (420).

Postoperative Care The postoperative care of the patient is discussed in Chapter 4. It should be noted that it is not necessary to use nasogastric suctioning on a routine basis. & ADJUVANT THERAPY In the last half century, there has been very little improvement in the ultimate 5-year survival rate of patients operated on for colorectal carcinoma, and physicians are constantly trying methods that will improve the current results. Five general principles underlie adjuvant therapy (421): 1. There may be occult, viable malignant cells in circulation (intravascular, intralymphatic, or intraperitoneal) and/ or established, microscopic foci of malignant cells locally, at distant sites, or both. 2. Therapy is most effective when the burden of malignancy is minimal and cell kinetics are optimal. 3. Agents with reported effectiveness against the carcinoma are available. 4. Cytotoxic therapy shows a dose-response relationship and therefore must be administered in maximally tolerated doses, and the duration of therapy must be sufficient to eradicate all malignant cells. 5. The risk-to-benefit ratio for therapy must be favorable to individuals who may remain asymptomatic for their natural life expectancy after resection of their malignancy.

Radiotherapy Although radiotherapy has been used extensively in various settings for the treatment of rectal carcinoma, little use has been made of it in the treatment of colon carcinoma. Exceptions relate to the presence of a carcinoma in a portion of the colorectum fixed to the retroperitoneum (cecum, ascending colon, and descending colon) or pelvis (rectosigmoid or rectum). Indications that have been considered appropriate for postoperative radiotherapy include (i) involvement of lymph nodes, (ii) known inadequate margins of resection, (iii) adherence to the retroperitoneum, sacrum, or pelvic side walls, (iv) transmural penetration to a macroscopic degree, and (v) extensive microscopic penetration with the presence of positive lymph nodes (422). Radiotherapy has been used in both the adjuvant setting and in situations of incomplete removal (423–425). In a non-randomized prospective trial, Dutterhaver et al. (423) documented an improved 5-year survival rate in patients with B3, C2, and C3 colon carcinomas (Gunderson and Sosin staging modification) treated with postoperative radiation compared with historical controls. Similarly,

118

& PART I: COLORECTAL DISORDERS

Kopelson (426,427) found postoperative adjuvant pelvic irradiation to be beneficial with respect to disease control in patients with transmural carcinoma of the cecum, ascending colon, descending colon, or sigmoid colon as well as improved survival for these cases. Minsky (428) reviewed the literature on the use of whole abdominal radiation therapy for resectable colon carcinoma. In general, patients received 20 to 30 Gy to the whole abdomen with or without a local boost. Additional 5-fluorouracil (5-FU) was delivered in a variety of doses and schedules. Combined results revealed an in-field failure rate of 12% to 50%. Significant toxicity varied from 5% to 38%. With a median follow-up of 5 years, the 5-year disease-free and overall survival rates were 58% and 67%, respectively.

Chemotherapy The use of adjuvant chemotherapy is attractive because it may offer the possibility of identifying patients who are likely to have occult, residual, or disseminated disease at the time of operation. Chemotherapy is most effective when the burden of carcinoma is smallest and the fraction of malignant cells in growth phase is the highest (429). Many chemotherapeutic agents have been used singly or in combination in an effort to improve survival rates. A large number of studies have directed enormous energy to the subject of intravenous or oral administration of chemotherapeutic agents beginning 4 to 6 weeks after operation. For the most part, none of the early studies directly indicated that adjuvant chemotherapy was of overall benefit to the survival of patients with colon carcinoma (430–436). Results were often equivocal. In 1988 a meta-analysis of randomized adjuvant studies using systemic 5-FU for up to 1 year demonstrated a marginal overall reduction in the odds ratio of death of approximately 17% and an absolute 5-year survival benefit of 3.4% (436). Promise was offered by the combination of leucoyorin and 5-FU or levamisole and 5-FU (437). An encouraging publication by Laurie et al. (438) reported that the use of a combination of 5-FU and levamisole administered to patients undergoing curative resection for Dukes’ C carcinoma of the colon resulted in a 5-year survival rate of 49% compared with 37% in patients receiving no postoperative treatment. The North Central Cancer Treatment Group (NCCTG) —Mayo Clinic Trial (Intergroup Study) randomized 1296 patients with Dukes’ B and C carcinoma to treatment for 1 year with levamisole and 5-FU or to operation alone (439). Patients with Dukes’ C could be assigned to levamisole alone. With a median follow-up of 3 years, Dukes’ C patients treated with levamisole and 5-FU had a reduced risk of carcinoma recurrence of 41% (chiefly within the liver) and overall death rate reduced by 33%. This initial improvement corresponded with an absolute survival benefit of approximately 5%, which was maintained at 5 years. Treatment with levamisole alone had no detectable effect. The survival benefit for patients with Dukes’ B carcinoma did not reach conventional levels of significance, and there was no reduction in the incidence of liver metastases (440). In April 1990 the National Institutes of Health (NIH) sponsored a consensus development conference on adjuvant chemotherapy for colorectal carcinoma (421). It was concluded that based on the results of the NCCTG (434)

and a subsequent trial by the Intergroup consortium (439), there was compelling evidence to recommend the combination of 5-FU and levamisole as ‘‘standard’’ adjuvant treatment for patients with stage III colon carcinoma (regional lymph node involvement) if they are unable to enter a clinical trial. However, an update of the Intergroup Trial INT-0089 in which there were four arms [leucovorin and 5-FU vs. leucovorin and 5-FU (different dose and schedule) vs. levamisole plus 5-FU vs. levamisole and leucovorin plus 5-FU] were compared revealed a similar survival (441). Intergroup Trial 0153 (Southwest Oncology Group trial 9415) was developed to compare the efficacy of continuousinfusion FU plus levamisole to FU/LV plus leucovorin plus levamisole in the adjuvant treatment of high-risk Dukes’ B2 and C1 or C2 colon carcinomas (442). A total of 1135 patients were registered. At least one grade 4 toxicity occurred in 39% of patients receiving FU/LV and 5% of patients receiving continuous-infusion 5-FU. However, almost twice as many patients receiving continuousinfusion 5-FU discontinued therapy early compared with those receiving FU/leucovorin. The 5-year overall survival is 70% for FU leucovorin and 69% for continuous-infusion FU. The corresponding five-year disease-free survival is 61% and 63%, respectively. For all patients, 5-year overall survival is 83%, 74%, and 55%; five-year disease-free survival is 78%, 67%, and 47% for N0, N1, and N2-3, respectively. They concluded, continuous infusion of 5-FU has less severe toxicity but did not improve disease-free survival or overall survival in comparison with bolus FU/LV. Thirion et al. (443) presented an update of the metaanalysis with a longer follow-up and the inclusion of 3300 patients randomized in 19 trials on an intent-to-treat basis. Their overall analysis showed a twofold increase in response rates (21% for FU leucovorin vs. 11% for FU alone; OR ¼ 0.53) and a small but statistically significant overall survival benefit for FU leucovorin over FU alone (median survival 11.7 vs. 10.5 months, respectively; hazards ratio 0.9) which were primarily seen in the first year. They observed a significant interaction between treatment benefit and dose of FU with response and overall survival advantage of FU leucovorin over FU alone being restricted to trials in which similar dose of FU was prescribed in both arms. In what they termed their final report, Moertel et al. (444) reported their experience with 5-FU plus levamisole as adjuvant therapy for patients undergoing resection of stage III colon carcinoma. Patients were assigned to observation only, to levamisole alone (50 mg orally, 3 times a day for 3 days, repeated every 2 weeks for 1 year), or to this regimen of levamisole plus 5-FU (450 mg/m2 body surface area intravenously daily for 5 days and then beginning at 28 days weekly for 48 weeks). In all 929 patients followed for 5 years or more (median, 6.5 years) 5-FU plus levamisole reduced the recurrence rate by 40% and the death rate by 33%. Levamisole reduced the recurrence rate by only 2% and the death rate by only 6%. With few exceptions, toxicity was mild, and patient compliance was excellent. No evidence of late side effects was seen. The authors concluded that 5-FU plus levamisole should be considered the standard treatment for all patients with stage III colon carcinoma not entered into clinical trials.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

In the NSABP Protocol C-01 (445), 1166 patients with Dukes’ B and C carcinoma of the colon were randomized to one of three therapeutic categories: (i) no further treatment following curative resection, (ii) postoperative chemotherapy consisting of 5-FU, semustine, and vincristine, or (iii) postoperative bacille Calmette-Gue`rin (BCG). With an average study time of 77.3 months, patients in the chemotherapy arm had an overall improvement in disease-free survival (p ¼ 0.02) and survival (p ¼ 0.50) and 1.31 times the likelihood of dying compared with patients treated by operation alone. Comparison of the BCG-treated group with opertion alone revealed no statistical significance in disease-free survival (p ¼ 0.90), but there was a survival advantage in favor of the BCG-treated group (p ¼ 0.03). At 5 years the risk of dying in the operation-only group was 1.28 times the risk of the BCG-treated patients. Further analysis disclosed that the survival advantage was the result of a diminution in deaths not related to carcinoma; when these deaths were eliminated, the survival difference was not statistically significant. Findings from the C-01 trial demonstrated a significant disease-free survival and survival benefit that lasted 5 years after colon resection, but at 8 years, the disease-free survival and survival differences were no longer significant. In a subsequent NSABP protocol C-03, Wolmark et al. (446) reported data from 1081 patients with Dukes’ B and C carcinoma of the colon. Patients were randomly assigned to receive either lomustine (MeCCNU), vincristine, and 5-FU (MOF), or leucovorin-modulated 5-FU (leucovorin plus 5-FU). The mean time, of the study was 47.6 months. Comparison: between the two groups indicates a disease-free survival advantage for patients treated with leucovorin and 5-FU. The 3-year disease-free survival rate for patients in this group was 73% compared with 64% for patients receiving MOF. The coresponding percentage of patients surviving was 84% for those randomized to receive leucovorin plus 5-FU and 77% for the MOF-treated cohort. At 3 years of follow-up, patients treated with postoperative leucovorin plus 5-FU had a 30% reduction in the risk of developing a treatment failure and a 32% reduction in the mortality risk compared with similar patients treated with MOF. The disease-free survival and survival rates with 5FU and leucovorin in the NSABP protocol C-03 appear comparable to those with 5-FU and levamisole. A meta-analysis was performed on nine randomized clinical trials that compared 5-FU with 5-FU plus intravenous leucovorin for the treatment of advanced colorectal carcinoma (447). Therapy with 5-FU plus leucovorin administered either as a weekly or monthly regimen showed a highly significant benefit over single-agent 5-FU in terms of the response rate of the carcinoma (23% vs. 11%). This increase in response did not, however, result in a discernible improvement of overall survival. In 1989 the NSABP initiated a clinical trial C-04 in which 2151 patients with Dukes’ B and C colon carcinoma were randomized to receive 5-FU plus leucovorin, 5-FU plus levamisole, or 5-FU plus leucovorin and levamisole. With an average time on study of 63.4 months, there was no significant difference in disease-free survival or survival rates in the three arms. The authors concluded that when compared with 5-FU plus levamisole, 5-FU plus leucovorin is at least as effective and possibly produces a disease-free

119

survival and survival advantage in patients with Dukes’ B and C colon carcinoma (440). Wolmark et al. (448) reported data presented for 2176 patients with Dukes’ stage B or C carcinoma entered into NSABP protocol C-05. Patients were randomly assigned to receive either 5-FU þ Leucovorin or 5-FU þ Leucovorin and Interferon. The mean time on the study was 54 months. There was no statistically significant difference in either disease-free survival (69% vs. 70%) or overall survival (80% vs. 81%) at four years of follow-up. Toxic effects of grade 3 or higher were observed in 61.8% of subjects in the group treated with 5-FU þ LV and in 72.1% of subjects in the group treated with 5-FU þ LV þ IFN. The addition of Interferon to 5-FU þ LV adjuvant therapy conferred no statistically significant benefit but did increase toxicity in this study. Although the benefit from adjuvant chemotherapy has been established for Dukes’ C colon carcinoma patients, many still question the worth of such therapy in patients with Dukes’ B disease. Mamounas et al. (449) examined the relative efficacy of treatment observed in four NSABP studies and reported that in patients with Dukes’ B disease, the relative reduction of recurrence and mortality was comparable to that in patients with Dukes’ C disease. A report from a pooled analyses of 1016 patients from five separate other trials found no significant difference in event-free survival or overall survival. They felt their data did not support the use of chemotherapy in Dukes’ B colon carcinoma (450). Three cooperative groups evaluated 5-FU and leucovorin individually, but because their protocols were so similar, their data were pooled to allow a combined analysis (451). Patients with Dukes’ B and C were randomized to operation alone or to operation followed by 5-FU and leucovorin administered for 5 days every 28 days for 6 months. The crude rate of recurrence in the liver was twice as high in the control group compared with the treatment group. The 3-year overall survival rate for patients with Dukes’ B disease was not significantly different between the control and chemotherapy groups. There was a statistically significant improvement in overall survival only in patients with Dukes’ C lesions (hazard ratio for overall survival, 0.7). Accordingly there is now considerable evidence of a survival benefit produced by systemic chemotherapy in patients with Dukes’ C disease, but some uncertainty of benefit in those with Dukes B lesions. In another review of the efficacy of adjuvant chemotherapy, Dube´, Heyen, and Jenicek (452) conducted a qualitative and quantitative meta-analysis of 39 randomized clinical trials. Patients with colon carcinoma who received chemotherapy could expect an increase of 5% in the survival rate and those with rectal carcinoma a 9% increase. Given the high incidence of colorectal carcinoma, even a small benefit is far from negligible. In a recent report Sakamoto et al. (453) performed a meta-analysis on individual data from three randomized trials conducted by the Japanese Foundation for Multidisciplinary Treatment for Carcinoma involving a total of 5233 patients with stages I to III colorectal carcinoma to assess the survival and disease-free survival benefits of treating patients after resection of a primary carcinoma with oral fluoropyrimidines for 1 year. Compared to a control group of untreated patients oral therapy reduced the risk of death

120

& PART I: COLORECTAL DISORDERS

by 11% and the risk of recurrence or death by 15%. There was no significant difference in benefit of oral therapy with respect to stage (I, II, or III), site (rectum or colon), patient age, or patient sex. In a recent study, Andre et al. (454) randomly assigned 2246 patients who had undergone curative resection for stage II or III colon carcinoma to receive 5-FU and leucovorin alone or with oxaliplatin for 6 months. After a mean followup of 37.9 months, 21% in the group given 5-FU and leucovorin þ oxaliplatin had had a carcinoma-related event, as compared with 26.1% in the 5-FU leucovorin group. The rate of disease-free survival at 3 years was 78.2% in the group with 5-FU leucovorin þ oxaliplatin and 72.9% in the 5-FU leucovorin group. In the group with 5-FU leucovorin þ oxaliplatin, the incidence of febrile neutropenia was 1.8%, the incidence of gastrointestinal adverse effects was low, and the incidence of grade 3 sensory neuropathy was 12.4% during treatment, decreasing to 1.1% at 1 year of follow-up. Six patients in each group died during treatment (death 0.5%). They concluded, adding oxaliplatin to a regimen of 5-FU and leucovorin improved the adjuvant treatment of colon carcinoma. Martenson et al. (455) conducted a trial to determine whether radiation therapy added to an adjuvant chemotherapy regimen improves outcome in high-risk patients. Patients with resected colon carcinoma with adherence or invasion of surrounding structures or with T3 N1 or T3 N2 carcinomas of the ascending and descending colon were randomly assigned to receive fluorouracil and levamisole therapy with or without radiation therapy (45–50.4 Gy in 25–28 fractions beginning 28 days after starting chemotherapy). Overall five-year survival was 62% for chemotherapy patients and 58% for chemoradiotherapy patients; five-year disease-free survival was 51% for both groups. Toxicity ( grade 3) occurred in 42% of chemotherapy patients and 54% of chemoradiotherapy patients. They concluded the overall survival and disease-free survival were similar but toxicity was higher among chemoradiotherapy patients. Based on the recognition that the liver is usually involved when large bowel malignancy disseminates, present in 25% to 30% of patients on initial diagnosis, and also that the liver is the most frequent site of subsequent metastatic disease (40% to 50% of patients) (456), Taylor et al. (457) proposed the strategy of immediate postoperative portal vein perfusion of chemotherapy. In the update of their randomized trial, in which 127 control patients were compared with 117 patients who received intraportal 5-FU and heparin via an umbilical vein catheter for 7 days after operation, benefit was demonstrated in treated patients, with 5-year survival rates of 70% vs. 45%. When individual groups were analyzed, only patients with Dukes’ B carcinoma had a significant improvement in overall survival. Taylor reported that there have been 10 published series of adjuvant cytotoxic portal vein infusion in patients with primary colorectal carcinoma (263). Portal vein cytotoxin infusion is usually begun at operation or in the immediate postoperative period and given continuously for the first 7 days after operation. This short perioperative therapy is in marked contrast to long-term chemotherapy in which 5-FU is given systemieally for 6 months. Toxicity

is reduced as, indeed, is cost. A Swiss group (458) reported a 25% incidence of relapse confined to liver in the control arm compared with an incidence of 12% in the portal vein infusion group in node-positive patients. Overall, portal vein infusion reduced the recurrence by 21% (hazard ratio, 0.90) and the risk of death by 26% (0.74). The major risk reduction was in patients with Dukes’ C disease as, indeed, was the overall absolute survival improvement. This differed from the original report of Taylor et al. (457), in which only patients with Dukes’ B carcinoma obtained a survival benefit. In the NSABP protocol C-02, 1158 patients with Dukes’ A, B, and C carcinoma of the colon were randomized to either no further treatment following curative resection or to postoperative 5-FU and heparin administered via the portal vein (459). Therapy began on day of operation and consisted of constant infusion for 7 successive days. The average time on study was 41.8 months. A comparison of the two groups of patients indicated both an improvement in disease-free survival (74% vs. 64%) at 4 years, and a survival advantage (81% vs. 73%) at 4 years in favor of the chemotherapy-treated group. When compared with the treated group, patients who received no further treatment had 1.26 times the risk of developing a treatment failure and 1.25 times the likelihood of dying after 4 years. Particularly significant was the failure to demonstrate an advantage produced by 5-FU in decreasing the incidence of hepatic metastases. The liver was the first site of treatment failure in 32.9% of 82 patients with documented recurrences in the control group and in 46.3% of 67 patients who received additional treatment. Therapy is administered via a regional route to affect the incidence of recurrence within the perfused anatomic boundary. Since, in this study, adjuvant portal-vein 5-FU infusion failed to reduce the incidence of hepatic metastases, it may be concluded that the disease-free survival and survival advantages are a result of the systemic effects of 5FU. This report differs from those of Taylor et al. (457) and Wereldsma et al. (460), who reported a decreased incidence of liver metastases. A meta-analysis of all 10 published trials of portal vein cytotoxic infusion, including a total of 3499 patients, has provided confirmatory evidence (461). Overall, portal vein infusion was associated with a mean 18% reduction in the annual risk of death from any cause, with a 20% reduction in death from colorectal carcinoma. In this meta-analysis, overall survival at 5 years was 64% in the portal vein infusion group vs. 59% in the operation-only group. The benefit was significantly greater for patients with Dukes’ C disease (54.2% vs. 46.6%) than for those with Dukes’ B lesions (76.3% vs. 72.1%). The published results of both adjuvant systemic chemotherapy for 6 to 12 months or portal vein infusion for 1 week are roughly comparable, with an approximate 5% improvement in absolute survival at 3 years (263). The major benefits from both treatments occur in patients with lymph node-positive disease. This is entirely consistent with the hypothesis that chemotherapy inhibits occult hepatic micrometastases rather than producing systemic effects, outside the liver. Subsequent relapse in extra-hepatic organs is also inhibited, since extrahepatic metastases spread from liver metastases.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

The adjuvant x-ray and 5-FU infusion study (AXIS) is the largest trial yet reported to determine the efficacy of postoperative portal vein infusion of 5-FU. Consenting patients with colorectal carcinoma were randomized to operation with or without seven days of portal vein infusion (1 g 5-FU plus 5000 units heparin in 1 L per 5% dextrose infused over each 24-hour period). In addition, patients with rectal carcinoma could be randomized to radiotherapy or no radiotherapy to be given either before or after operation. A total of 3583 patients were randomized with respect to portal vein infusion. No survival benefit was seen in the 761 patients randomized with respect to radiotherapy and although not statistically significant, the impact on local recurrence rates was similar to that reported in the literature. No overall benefit of portal vein infusion was established in the AXIS study when colonic and rectal carcinomas were considered together but the evidence suggesting a differential treatment effect according to site of carcinoma in AXIS was strongly supported by a meta-analysis incorporating the previous trials. Combining the data gave hazard ratios of 0.82 and 1.00 for colonic and rectal carcinoma respectively, equating to an absolute survival benefit for patients with colonic carcinoma of 5.8%, a level close to that seen for prolonged systemic therapy. There is now sufficient evidence to recommend that patients with Dukes’ C (stage III) disease who are otherwise sufficiently fit should receive adjuvant chemotherapy (usually a combination o 5-FU plus leucovorin or 5-FU plus levamisole). A recent review noted bolus scheduling of 5FU þ leucovorin is favored in North America while infusion of 5-FU leucovorin is favored in Europe (462). They noted infused 5-FU leucovorin may be safer with new drugs such as oxaliplatin and irinotecan. A review by Haydon (463) reported that in stage III disease chemotherapy appears to be equally effective whether it is given daily for 5 days per month or on a weekly scheduled. The overall effect is a relative reduction in recurrence of 25% or an absolute improvement in survival of 10%. However, doubt remains as to the role of adjuvant chemotherapy in stage II colon carcinoma. Most of the randomized trials have demonstrated a relative reduction in recurrence but have not shown any significant impact on survival. Even for patients with lymph node involvement the indications may not be uniformly obvious. Chapuis et al. (464) conducted a study to identify patient and carcinoma characteristics that might assist in patient selection for chemotherapy after resection of clinicopathological stage III colonic carcinoma. From an initial 2980 patients, after exclusions, 378 remained who had a potentially curative operation for colonic carcinoma with nodal metastases and did not receive adjuvant chemotherapy. Both overall and colonic carcinoma-specific survival rates were negatively associated with serosal surface involvement, apical node metastasis, high histological grade, and venous invasion. The survival of patients with stage III disease who had none of these adverse features was not significantly different from that of patients with stage II lesions. However, survival diminished significantly when one or more of the adverse features were present. Patients with stage III disease but none of the identified adverse features experience relatively good survival and are unlikely to

121

benefit from adjuvant chemotherapy. In this series such patients accounted for 40.5% of patients with stage III disease. There is still some uncertainty, however, with regard to the benefits for patients with Dukes’ B lesions. Figueredo et al. (465) conducted a systematic review to address whether patients with stage II colon carcinoma should receive adjuvant therapy. Thirty-seven trials and 11 metaanalysis were included. The evidence for stage II colon carcinoma comes primarily from a trial of fluorouracil plus levamisole and a meta-analysis of 1016 patients comparing fluorouracil with folinic acid versus observation. Neither detected an improvement in disease-free or overall survival for adjuvant therapy. A recent pooled analysis of data from seven trials observed a benefit for adjuvant therapy in a multivariate analysis for both disease-free and overall survival. The disease-free survival benefits appeared to extend to stage II patients. A meta-analysis of chemotherapy by portal vein infusion has also shown a benefit in diseasefree and overall survival for stage II patients. A meta-analysis was conducted using data on stage II patients where data were available (n ¼ 4187). The mortality risk ratio was 0.87. They concluded there is preliminary evidence indicating that adjuvant therapy is associated with a disease-free survival benefit for patients with stage II colon carcinoma but these benefits were small and not necessarily associated with improved overall survival. An American Society of Clinical Oncology panel in collaboration with the Cancer Care Ontario Practice Guideline Initiative, conducted a literature based meta-analysis to determine whether medically fit patients with curatively resected stage II colon carcinoma should be offered adjuvant chemotherapy as part of routine clinical practice (466). They found no evidence of a statistically significant survival benefit of adjuvant chemotherapy. However, there are populations of patients with stage II disease that could be considered for adjuvant therapy, including patients with inadequately sampled lymph nodes, T4 lesions, perforation, or poorly differentiated histology. The ultimate clinical decision should be based on discussions with the patient about the nature of the evidence supporting treatment, the anticipated morbidity of treatment, the presence of high-risk prognostic features on individual prognosis, and patient preferences. A refinement of the selection of patients for adjuvant chemotherapy might be devised from the study by Ribic et al. (467) who investigated the usefulness of microsatellite instability status as a predictor of the benefit of adjuvant chemotherapy with fluorouracil in stage II and III colon carcinoma. Specimens were collected from patients with colon carcinoma who were enrolled in randomized trials. Among 287 patients who did not receive adjuvant therapy, those with carcinomas displaying high-frequency microsatellite instability had a better 5-year rate of overall survival than patients with carcinomas exhibiting microsatellite stability or low-frequency instability (hazard ratio for death, 0.31). Among patients who received adjuvant chemotherapy, high-frequency microsatellite instability was not correlated with increased overall survival (hazard ratio for death, 1.07). Adjuvant chemotherapy improved overall survival among patients with microsatellite-stable and low-frequency microsatellite instability carcinomas.

122

& PART I: COLORECTAL DISORDERS

Consideration must be given to the definite toxicities. Treatment is often suspended or modified because of gastrointestinal or marrow complications. These observations emphasize the need to evaluate the potential benefits of chemotherapy in a trial setting. In addition, consideration should be given to absolute survival as well as to the quality of life during and following treatment. The complication side of chemotherapy was poignantly highlighted in the report by Culakova et al. (468) who reviewed the hospitalization of colorectal carcinoma patients with febrile neutropenia at 115 academic medical centers. Over a 6-year period, 957 patients with colorectal carcinoma experienced 1046 hospitalizations with febrile neutropenia. The mean (median) length of stay was 9.0 days while the average (median) cost per hospitalization was $12,484 ($7153). The 30% of patients hospitalized 10 days accounted for 62% of all hospital days and 66% of all costs. Clinical infection was reported in 57% of patients, with microbiological documented infection reported in 31% including gram-negative sepsis in 8.5%, gram-positive in 2.7%, and other bacterial infections in 17.5%. Pneumonia was reported in 5% while 5% presented with hypotension and 34% with evidence of hypovolemia. Death occurred in 8.5% of hospitalizations. Mortality rates were greatest in those with gram-negative sepsis (35%), gram-positive sepsis (23%), hypotension (23%), and pneumonia (32%). One or more major comorbidities were reported in 57% of patients. Mortality rates increased with the number of comorbidities. Significant independent risk factors for mortality in multivariate analysis were gram-negative sepsis (OR ¼ 6.1), gram-positive sepsis (OR ¼ 3.8), hypovolemia (OR ¼ 2.5), pneumonia (OR ¼ 4.9), lung disease (OR ¼ 4.6), liver disease (OR ¼ 4.4), heart disease (OR ¼ 3.9), and other vascular disease (OR ¼ 13.3). One-third of hospitalized patients with febrile neutropenia experienced complicated hospitalizations with significant morbidity and mortality. Arora and Potter (469) reviewed the literature to determine whether adjuvant chemotherapy was safe and effective in patients > 70 years. Of 3347 patients from 7 randomized trials, 43% had stage II and 57% had stage III disease. The probability of death without recurrence of carcinoma was strongly associated with age. The oldest patients had a higher probability of dying without evidence of recurrence (13%) than the youngest patients (2%). In addition, 32% of deaths of the oldest patients but only 5% of deaths of the youngest patients were due to causes other than the carcinoma. Approximately 30% of the patients in each group died with recurrence of carcinoma over the 8-year follow-up period. The 5-year overall survival was 71% for those who received adjuvant therapy, compared with 64% of those untreated. The 5-year recurrence-free rate was 69% in the treated patients compared with 58% in untreated patients. After five years, patients aged 70 years and older had decreased overall survival because of death from other causes. Patients treated with fluorouracil plus levamisole had significantly more leukopenia and nausea or vomiting, whereas those treated with fluorouracil plus leucovorin had significantly more stomatitis and diarrhea. Leukopenia was significant in patients with levamisole and fluorouracil treatment (31% of subjects 70 years vs. 17% of subjects < 70 years) and borderline significantly higher in patients with fluorouracil and leucovorin treatment (8% of subjects 70 years vs. 4% of subjects

< 70 years). This analysis supports the benefit of adjuvant chemotherapy in patients with resected stage II and III colon carcinoma. The patients enrolled in these trials may not be representative of all older patients with colon carcinoma. Only 0.7% of patients in the trial were aged 80 years or older so caution is advised in extrapolating these findings to the oldest patients. The decision to treat older people who have functional limitations and comorbid conditions should be individualized. Rothenberg et al. (470) conducted a review to assign attribution for the causes of early deaths on two National Cancer Institute-sponsored cooperative group studies involving irinotecan and bolus fluorouracil and leucovorin. The inpatient, outpatient, and research records of patients treated on Cancer and Leukemia Group B protocol C89803 and on North Central Cancer Treatment Group protocol N9741 were reviewed by a panel of five medical oncologists not directly involved with either study. Each death was categorized as treatment-induced, treatmentexacerbated, or treatment-unrelated. Patients treated with irinotecan plus bolus 5-FU/leucovorin had a threefold higher rate of treatment-induced or treatment-exacerbated death than patients treated on the other arm(s) of the respective studies. For C89803, these rates were 2.5% for irinotecan/fluorouracil leucovorin group versus 0.8% for bolus weekly 5-FU and leucovorin. For N9741, these rates were 3.5% for irinotecan/fluorouracil leucovorin, 1.1% for oxaliplatin plus bolus and infusional 5-FU and leucovorin, and 1.1% for oxaliplatin plus irinotecan. The majority of deaths in both studies were attributed to thromboembolic events or gastrointestinal toxicities. The colorectal carcinoma chemotherapy study group of Japan (471) reported on the efficacy of postoperative treatment with mitomycin C and oral 5-FU for colorectal carcinoma. They designed a second trial to evaluate the effectiveness of additional preoperative chemotherapy to postoperative treatment with mitomycin C and oral 5-fluorouracil for curatively resected colorectal carcinoma patients. There were 1355 patients (colon 755 and rectum 600) enrolled in this study. The pre- and postoperative chemotherapy group were treated preoperatively with 5-FU (320 mg/m2/day) by continuous intravenous infusion for 5 days beginning on day 6 before operation and postoperatively with mitomycin C (6 mg/m2 on days 7 and 14 and in months 2, 4, and 6 by bolus injection), and oral 5-FU (200 mg/day) for 6 months. The postoperative chemotherapy group received postoperative chemotherapy only. In an intent-to-treat analysis, the 5-year survival rate in the pre- and postoperative chemotherapy group and the postoperative chemotherapy group was 77.3% and 75.7% for colon carcinoma and 67.2% and 69% for rectal carcinoma, respectively. In a per-protocol analysis, the five-year disease-free survival rate in the pre- and postoperative group and the postoperative group was 76% and 80.7% for colon carcinoma and 60.5% and 63% for rectal carcinoma, respectively, indicating no significant difference between the two groups. Douillard et al. (472) compared oral uracil/tegafur and oral leucovorin to conventional intravenous fluorouracil and leucovorin in previously untreated metastatic colorectal carcinoma. Eight hundred sixteen patients were randomized to receive either uracil/tegafur (300 mg/m2/ day) and leucovorin (75 or 90 mg/day for 28 days every

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

35 days) or IV bolus 5-FU (425 mg/m2/day) and leucovorin (20 mg/m2/day) for 5 days every 28 days. Comparable survival was noted in both groups with a median of 12.4 months for the oral group and 13.4 months in the other group. The overall response rate did not differ between treatment arms. Oral uracil provided a safer, more convenient alternative to a standard bolus of intravenous 5-FU leucovorin. Diarrhea, nausea and vomiting and stomatitis and mucositis were significantly less frequent with the oral regimen as was myelosuppression. Patients treated with the oral medication had fewer episodes of febrile neutropenia and documented infections. Bevacizumab Bevacizumab (Avastin; Genentech Inc., South San Francisco, California, U.S.A.) is a recombinant humanized antivascular endothelial growth factor monoclonal antibody that inhibits neoplastic angiogenesis, and has demonstrated survival benefit in patients with previously untreated metastatic colorectal carcinoma when combined with irinotecan/fluorouracil/leucovorin (IFL). Kabbinavar et al. (473) combined analysis of data from three randomized clinical studies evaluating bevacizumab in combination with FU/LV alone. The median duration of survival was 17.9 months in 5-FU/LV bevacizumab group compared with 14.6 months in the combined control group, corresponding to a hazard ratio for death of 0.74. The median duration of progression-free survival was 8.8 months in the FU/LV bevacizumab group, compared with 5.6 months in the combined control group, corresponding to a hazard ratio for disease progression of 0.63. The addition of bevacizumab also improved the response rate (34.1% vs. 24.5%). In a phase III trial, combining bevacizumab with irinotecan, bolus fluorouracil, and leucovorin (IFL) increased survival compared with IFL alone in first-line treatment of patients with metastatic colorectal carcinoma. Hurwitz et al. (474) described the efficacy and safety results of the patient cohort who received bevacizumab combined with fluorouracil leucovorin and compared them with results of concurrently enrolled patients who received IFL. Median overall survivals were 18.3 and 15.1 months with fluorouracil leucovorin bevacizumab (n ¼ 110) and IFL/placebo (n ¼ 100), respectively. Median progression-free survivals were 8.8 and 6.8 months, respectively. Overall response rates were 40% and 37% and median response durations were 8.5 and 7.2 months, respectively. Adverse events consistent for those expected from the fluorouracil leucovorin or IFL-based regimens were seen, as were modest increases in hypertension and bleeding in the bevacizumab arm, which were generally easily managed. They concluded the fluorouracil leucovorin bevacizumab regimen seems as effective as IFL and has an acceptable safety profile. They further concluded that fluorouracil leucovorin bevacizumab is an active alternative treatment regimen for patients with previously untreated metastatic colorectal carcinoma.

Immunotherapy Immunotherapy was believed to have some effect on colon carcinoma, but there is no conclusive evidence to indicate significant improvement in survival (445). A review of prospective randomized trials by Lise et al. (475), which included an immunotherapy arm, failed to demonstrate any benefit. A report of a controlled randomized trial con-

123

sisting of a 2-year program of vaccination with BCG and neuraminidase-treated autologous carcinoma cells at 5-year follow-up failed to alter either the disease-free interval or the survival of patients (476). A controlled clinical trial of interferon-a as postoperative surgical adjuvant therapy for patients with colon carcinoma demonstrated significant enhancement of nonspecific immune function but no significant difference in patient survival (477). A study in which 189 patients with Dukes’ C colorectal carcinoma underwent resection for cure were randomized to observation or postoperative treatment with 17-1A antibody. After a median follow-up of 5 years, antibody treatment was reported to have reduced the overall death rate by 30% (478). In the future, genetic engineering techniques may allow generation of substances during the immune response, and these may have therapeutic value by modifying the biologic response to malignancy (479).

& COMPLICATED CARCINOMAS Previous studies have reported that emergency presentation of colorectal carcinoma is associated with poor outcome. McArdle and Hole (480) conducted a study aimed to establish, after adjusting for case mix, the magnitude of the differences in postoperative mortality and survival between patients undergoing elective surgery and those presenting as an emergency. Of 3200 patients who underwent surgery for colorectal carcinoma, 72.4% of 2214 elective patients had a potentially curative resection compared with 64.1% of 986 patients who presented as an emergency. Following curative resection, the postoperative mortality rate was 2.8% after elective and 8.2% after emergency operation. Overall survival at five years was 57.5% after elective and 39.1% after emergency curative operation; carcinoma-specific survival at 5 years was 70.9% and 52.9%, respectively. The adjusted hazard ratio for overall survival after emergency relative to elective surgery was 1.68 and that for carcinoma-specific survival was 1.90. Jestin et al. (481) identified risk factors in emergency surgery for colonic carcinoma in a large population of 3259 patients; 806 had an emergency and 2453 an elective procedure. Patients who had emergency surgery had more advanced carcinomas and a lower survival rate than those who had an elective procedure (5-year survival rate 29.8% vs. 52.4%). There was a stage-specific difference in survival with poorer survival both for patients with stage I and II carcinomas and for those with stage III carcinomas after emergency compared with elective surgery. Emergency surgery was associated with a longer hospital stay (mean 18 days vs. 10 days) and higher costs (relative cost 1.5) compared with elective surgery. The duration of hospital stay was the strongest determinant of cost. Obstruction When complete obstruction of the colon arises as a result of a carcinoma, the recommended treatment depends on the level of the colon that is obstructed as well as the beliefs and experience of the treating surgeon (482,483). In their review of 115 obstructing carcinomas, Sjodahl, Franzen, and Nystrom (484) found that 37% were right sided (proximal to splenic flexure) and 63% were left sided. Only 4% were Dukes’ A while 15% already had distant metastases.

124

& PART I: COLORECTAL DISORDERS

Interestingly, a study by Nozoe et al. (485) found the mean size of the obstructing carcinoma was 3.7 cm which was significantly smaller than that of nonobstructing carcinomas (5.4 cm). The proportion of lymph node metastases in obstructing carcinomas was 66.9% which was significantly higher than that in nonobstructing carcinomas (42.4%). The proportion of carcinomas classified into Dukes’ C or D in obstructing carcinomas was 84.6% and was significantly higher than that in nonobstructing carcinomas (52.5%). If the patient’s condition can be stabilized and there is evidence of resolution of the occlusion, bowel preparation and elective resection is the ideal solution. This clinical course is unusual, and therefore decisions on how to proceed must be made. For right-sided colonic obstructions, it is generally accepted that the treatment of choice is a resection and primary anastomosis with removal of the right and proximal transverse colon (486). Even though the bowel is not prepared, the resection usually can be readily accomplished. When the obstruction is located in the distal transverse colon, the matter of how to proceed is controversial. Some surgeons believe that the patient should have a proximal diversion followed by a definitive resection. However, under these circumstances a growing number of surgeons have adopted the procedure of an extended right hemicolectomy followed by a primary ileodescending colon anastomosis. Lee et al. (487) compared the operative results of 243 patients who had emergency operations for right-sided and left-sided obstructions from primary colorectal carcinomas. One hundred seven patients had obstruction at/or proximal to the splenic flexure (right-sided lesions), and 136 had lesions distal to the splenic flexure (left-sided lesions). The primary resection rate was 91.8%. Of the 223 patients with primary resection, primary anastomosis was possible in 88% of patients. Among the 101 primary anastomosis in patients with left-sided obstruction, segmental resection with on-table lavage was performed in 75 patients and subtotal colectomy was performed in 26 patients. The overall operative mortality rate was 9.4%, although that of the patients with primary resection and anastomosis was 8.1%. The anastomotic leakage rate for those with primary resection and anastomosis was 6.1%. There were no differences in the mortality or leakage rates between patients with right-sided and left-sided lesions (mortality 7.3% vs. 8.9% and leakage 5.3% vs. 6.9%). Colocolonic anastomosis did not show a significant difference in leakage rate when compared with ileo-colonic anastomosis (6.1% vs. 6%). Three-Stage Procedure For patients with an obstruction of the left colon, greater controversy exists and a larger number of options are available. Traditionally, these patients have undergone a threestage operation, with the first-stage being a transverse colostomy or possibly a cecostomy, followed by resection and anastomosis, and finally by closure of the colostomy. In a review of the subject, Deans, Krukowski, and Irwin (488) reported that between 70% and 80% of patients having a transverse colostomy undergo resection of their carcinoma during the first hospitalization, with a hospital stay of 30 to 55 days. Overall, 25% of patients do not undergo closure of their colostomy because they are unfit or unwilling to undergo an additional operation. Overall mortality rates range from 2% to 15%, mostly in the 10% range, with morbidity rates ranging from 20% to 37%, often

related to stoma complications, ranging from 6% to 14%. Although many reports show that the combined mortality rate of the three-stage procedure is similar to that of primary resection with delayed anastomosis, there is the suggestion that long-term survival is decreased in the three-stage operation (488). Sjodahl, Franzen, and Nystrom (484) found a modest increase in 5-year survival rate of 38% for immediate resection compared with a rate of 29% for a staged resection. Although proximal decompression is still promoted as a simple, safe initial option, the cumulative morbidity and mortality rates, survival disadvantage, prolonged hospital stay, and necessity of repeated operations make the three-stage procedure most unfavored. Hartmann’s Procedure Some surgeons have advocated an immediate resection without anastomosis (i.e., a proximal colostomy and mucous fistula or closed rectal stump, Hartmann’s procedure). The perceived advantages include immediate removal of the carcinoma, avoidance of an anastomosis in less than ideal circumstances, and more rapid convalescence and shorter hospital stay. In the event it proves to be permanent, a left-sided colostomy is much less of a burden than a transverse colostomy. The overall operative mortality rate has ranged from 6% to 12%, mostly in the 10% range (488), with hospital stay ranging from 17 to 30 days. Rates of colostomy closure of 60% or more are common. It must be remembered that significant morbidity can be associated with colostomy closure. In their report on 130 stomas and their subsequent closure, Porter et al. (489) experienced a complication rate of 44%. Nevertheless, Hartmann’s procedure combines primary resection and relief of the obstruction with acceptable morbidity and mortality rates. It is particularly appropriate for a patient with perforation of the left colon and for the elderly unfit patient. Subtotal Colectomy More recently, some surgeons have recommended a subtotal colectomy with primary ileosigmoid anastomosis or even ileorectal anastomosis. Advantages offered by this operation include (i) no stoma problems, (ii) a one-stage, procedure with a single hospitalization, (iii) a shorter hospital stay with financial savings, and (iv) removal of synchronous proximal neoplasms and reduced risk of metachronous lesions. Wong et al. (490) reported on 35 patients who presented with left-sided obstructing carcinoma. Unsuspected synchronous proximal lesions occurred in 12 patients (32%)—three carcinomas, eight adenomas, and one with another synchronous carcinoma and polyp. Initial reports stressed the technical demands of this operation but, with care, good results can be obtained. Operative mortality rates of 3% to 11% have been reported and morbidity rates are low, with a leakage rate of 4% and a hospital stay of 15 to 20 days (488). Subtotal colectomy carries a risk of diarrhea and/or fecal incontinence, particularly in elderly patients. However, most reported experience has not rated this a significant problem. The overall morbidity rate (6% vs. 44%) and length of hospital stay (17 vs. 34 days) are significantly less than after combined procedures (491). Perez et al. (492) evaluated the results of emergency subtotal colectomy in 35 patients with obstructing carcinoma of the left colon. The postoperative mortality rate was 6%, and complications were significant:

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

wound infection, 28%; ileus, 17%; evisceration, 8%; intestinal obstruction, 8%; and anastomotic leak, 11%. In a series of 35 patients, Lau, Lo, and Law (493) reported a complication rate of 31%, which included an anastomotic leak rate of 3%. Their review of the literature revealed leak rates that ranged from 0% to 4.5% for subtotal colectomy and 0% to 14% for colonic lavage methods. Chrysos et al. (494) reported four patients with obstructing carcinoma of the rectosigmoid junction and upper rectum who underwent a total colectomy followed by construction of a 10 cm ileal-J pouch that was subsequently anastomosed to the distal rectal stump. One year postoperatively, all patients experienced one to three normal bowel motions daily and no episodes of incontinence. They believe total colectomy with ileal-J-pouch-rectal anastomosis is a reasonable operative alternative in cases with obstructing carcinomas of the rectosigmoid junction which necessitate removal of the upper rectum. On-Table Lavage Still others have recommended resection of the primary disease combined with on-table lavage and primary anastomosis. A major perceived disadvantage of on-table lavage is that it is time-consuming. The operative technique consists of the mobilization of the appropriate segment of colon according to oncologic principles. In most circumstances, both the hepatic and splenic flexures require mobilization. The bowel at an appropriate distance distal to the carcinoma is divided, as is the proximal bowel 5 to 10 cm distal to the proximal site of the anastomosis, thus removing the carcinoma-bearing portion of colon. A No. 22 or No. 24 Foley catheter is inserted into the cecum through the freshly amputated appendicular stump or through the terminal ileum if the patient has had an appendectomy (Fig. 56). The catheter balloon is inflated and held

FIGURE 56 & On-table colonic lavage.

125

in place by a pursestring suture. A noncrushing clamp is placed across the terminal ileum to prevent reflux of the irrigation fluid. A standard intravenous infusion set is connected to the Foley catheter. The bowel, having been adequately mobilized, permits the distal portion to be placed in a kidney basin and hard fecal scybala can be ‘‘milked’’ into the kidney basin. A sterile corrugated anesthetic scavenger tube, 22 mm in diameter, is then inserted into the distal bowel and secured in place with strong tapes. The free end of this tube is draped over the side of the patient and secured in an appropriate collecting system. The colon is then lavaged with warm isotonic saline solution until the effluent in the corrugated tube is clear. The volume of lavage solution required is determined by the extent of fecal loading but can usually be accomplished with 3 to 6 L. Lavage time may vary from 20 to 45 minutes. Once the effluent is clear, the Foley catheter is removed, and the appendiceal opening or ileum is closed. A short segment of bowel containing the irrigated tube is resected and an anastomosis created according to the surgeon’s preference. Most series quote an operative mortality rate of approximately 10% (488). Anastomotic leakage rates following primary anastomosis are low. Tan and Nambiar (495) performed 36 primary resections and anastomoses following intraoperative antegrade colonic lavage for left-sided obstructing carcinoma. There were two deaths (one from anastomotic leak). Other complications included chest infection in 11% and wound infection in 19.4%. Others support this form of management (492,496,497). Leakage rates are approximately 4% (488). Wound infection rates remain a problem, with rates of 25% to 60%, and hospital stays around 20 days (488). The Scotia study group (498) conducted the first multicenter prospective randomized trial comparing subtotal colectomy with segmental resection and primary anastomosis following intraoperative irrigation for the management of malignant left-sided colonic obstruction. Of the 91 eligible patients recruited by 12 centers, 47 were randomized to subtotal colectomy and 44 to on-table irrigation and segmental colectomy. Hospital mortality and complication rates did not differ significantly, but 4 months after operation, increased bowel frequency (three or more bowel movements per day) was significantly more common in the subtotal colectomy group (14 of 35 vs. 4 of 35). More patients in the subtotal colectomy group reported that they had consulted their general practitioner with bowel problems than those in the segmental resection group (15 of 37 vs. 3 of 35). The authors believe that segmental resection following intraoperative irrigation is the preferred option except when there is cecal perforation or if synchronous neoplasms are present in the colon, when subtotal colectomy is more appropriate. Chiappa et al. (499) reported 39 patients who were treated with intraoperative decompression, on-table lavage, resection, and primary anastomosis. The primary anastomosis was intraperitoneal in 74% and below the peritoneal reflection of the rectum in 26% of patients. Operative mortality was 3% and anastomotic leakage was observed in 6% of patients. Complications included intra-abdominal abscess (3%), and wound infections (8%). Ohman (321) also found a higher operative mortality rate for primary resection (14%) compared with staged resection (5%) and, although there was an early apparent

126

& PART I: COLORECTAL DISORDERS

superior survival rate with the staged procedures, it did not persist into the fourth and fifth years: Umpleby and Williamson (327) reported a better 5-year survival rate following resection and anastomosis (48%) than after staged procedures (18%). Primary Resection Rather boldly, some surgeons have performed a resection with primary anastomosis in the absence of bowel preparation (500). An intracolonic bypass has been suggested as treatment (501). Still others have suggested a primary resection with anastomosis and proximal diversion. In an effort to shed light on the issue, Kronborg (502) conducted a randomized trial in which he compared the results of traditional staged procedures with an initial transverse colostomy, followed by curative resection, and subsequent colostomy closure with immediate resection and end colostomy and mucous fistula with subsequent reanastomosis. He found no difference in mortality or carcinoma-specific survival rates between the two treatments. From this constellation of choices, it becomes difficult to select the best one. Ultimately, the selection depends on the surgeon’s experience and preference. An informed decision rests on the recognition of the comparable morbidity and mortality rates for the single procedure compared with the combined morbidity and mortality rates of the multiple operations of the staged procedures. Fielding, StewartBrown, and Blesovsky (503) recorded an operative mortality rate of 25% for primary resection and 34% for staged resection. This prospective study compared the outcome of primary staged resection in colonic obstruction and failed to show any difference in mortality rates between these options. My own preference is to extend the primary resection for, lesions as far as the sigmoid colon. It appears worthwhile to cleanse the bowel distal to the obstruction, and a primary anastomosis then can be effected between the terminal ileum and the sigmoid colon. The morbidity and mortality rates are lower than those found with the staged approach, and the length of hospitalization is shorter. By eliminating a second or third hospitalization and a temporary colostomy, palliation is better for those patients who ultimately die from recurrent disease. Furthermore, those patients who undergo resection for cure may have increased rates of long-term survival. If the lesion is so distal that there would be little remaining reservoir by resecting all the obstructed colon proximal to the carcinoma, a reasonable alternative would be to cleanse the bowel distal to the carcinoma, perform a primary resection, and use on-table lavage, with a primary anastomosis. In the very debilitated patient, consideration should be given to a right transverse colostomy. In the exceptional case in which the obstructing lesion is deemed unresectable, a bypass should be performed when possible. For right-sided lesions, an ileotransversostomy can be performed. In other circumstances a colocolostomy might be deemed appropriate; in any event, this choice would be preferable to a permanent stoma, which would be the last option. There have been reports on the role of preresectional laser recanalization for obstructive carcinomas of the colon and rectum. Eckhauser and Mansour (504) reported on use of the neodymium: yttrium-aluminum-garnet

(Nd:YAG) laser to successfully accomplish decompression and allow for a formal bowel preparation and a definitive one-stage operation. The authors’ experience with 29 patients did not involve compromise of patient safety. In several other series, the success rate for recanalization was 80%, with a 2% to 50% procedure-related morbidity and mortality (488).

Stenting Intestinal stenting is a procedure that is rapidly coming into more widespread use. It was first introduced by Doharto in 1991 (505) as definitive palliative treatment for patients with obstructive disease where resection for cure was not appropriate due to very advanced local disease, metastatic disease, or because of an unacceptably high operative risk. In 1994 Tejero et al. (506) proposed stent placement as a ‘‘bridge-to-surgery’’ for emergency relief of colonic obstruction with an aim to subsequent elective resection. The technique can be applied in patients who refused operative treatment. Colostomy can be avoided with an improved quality of life especially in the palliative setting. Suitable lesions for endoluminal colorectal stenting include both obstructing primary left-sided colorectal carcinomas as well as extracolonic malignancies such as prostate, bladder, ovarian, or pancreatic. It is not required for right-sided obstructing carcinomas and it is also not appropriate for lesions less than 5 cm from the anal verge. The actual length of the lesion is not a theoretical limitation. It is contraindicated in the presence of colonic perforation with peritonitis and would not prove effective with multiple sites of obstruction. Although it is not mandatory, it is probably best that stents be placed under endoscopic guidance with the aid of fluoroscopy. The administration of prophylactic antibiotics is probably wise. The procedure is conducted under conscious sedation. A catheter over a guide wire is advanced through the lesion. Contrast is injected into the proximal lumen. Once deployed, the stents expand and become incorporated into the surrounding tissue by pressure necrosis thus anchoring the stent. Post procedure plain abdominal x-rays are obtained for 3 or 4 days. Dauphine et al. (507) reviewed their experience with 26 self-expanding metal stents as the initial interventional approach in the management of acute malignant large bowel obstruction. In 14 patients, the stents were placed for palliation, whereas in 12, they were placed as a bridge-to-surgery. In 85%, stent placement was successful on the first occasion. In the remaining four individuals, one was successfully stented at the second occasion, and three required emergency operation. Nine of the 12 patients (75%) in the bridge to surgery group underwent elective colon resection. In the palliative group, 29% had reobstruction of the stents and in 9% the stent migrated. In the remaining 64%, the stent was patent until the patient died or until the time of last follow-up. Colonic stents achieved immediate nonoperative decompression and proved to be both safe and effective. Since first described, there have been numerous publications on the subject. Khot et al. (508) conducted a systematic review of the published data on stenting for the treatment of colorectal obstruction. A total of 58 publications was found of which 29 case series were included in the analysis. Technical and clinical success, complications,

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

and reobstruction, both in palliation and as a ‘‘bridge to surgery’’ were assessed. Pooled results showed that stent insertion was attempted in 598 instances. Technical success was achieved in 92% and clinical success in 88%. Palliation was achieved in 90% of 336 cases while 85% of 262 insertions succeeded as a ‘‘bridge-to-surgery’’ (95% had a one-stage operative procedure with a mean time to the operating room 8.9 days). Technical reasons for failure included inability to place the guide wire, malposition or perforation. Clinical failures included perforation, persistent obstructive symptoms, or adhesion of colonic wall to the stent. There were three deaths (1%). Perforation occurred 22 times (4%); 1% in balloon dilatation versus 2% in nonballoon dilatation. Stent migration was reported in 10% of 551 technically successful cases. Management included stent removal, stent reinsertion, operation, and no immediate intervention but proceed to planned operation. The rate of stent reobstruction was 10% of the 525, mainly in the palliative group. Reason for obstruction included ingrowth of malignancy, stent migration, and fecal impaction. Bleeding occurred in 5% of patients, the majority requiring no treatment but three patients received transfusions. Another 5% of patients experienced pain, either abdominal or rectal, and this was controlled with oral analgesics. They concluded, that the evidence suggests that colorectal stents offer good palliation, and are safe and effective as a ‘‘bridge-to-surgery.’’ Stent usage can avoid the need for a stoma and is associated with low rates of mortality and morbidity. Dilatation of malignant strictures at the time of stent placement appears to be dangerous and should be avoided. Law et al. (509) evaluated the outcomes of selfexpanding metallic stents as a palliative treatment for malignant obstruction of the colon and rectum. The insertion of self-expanding metallic stents was attempted for palliation in 52 patients. Successful insertion of the stent was achieved in 50 patients. The median survival of patients was 88 (range, 3–450) days. Complications occurred in 13 patients (25%). These included perforation of the colon (n ¼ 1), migration or dislodgement of the stents (n ¼ 8), severe tenesmus (n ¼ 1), colovesical fistula (n ¼ 1), and ingrowth of malignancy (n ¼ 2). Insertion of a second stent was required in eight patients. Subsequent operations were performed in nine patients, and stoma creation was required in seven patients. Saida et al. (510) evaluated the long-term prognosis of expandable metallic stent insertion compared with emergency operation without expandable metallic stent. Forty emergency operations and 44 expandable metallic stent insertions were retrospectively compared. Postoperative complications were significantly less frequent in the expandable metallic stent group: wound infection was 14% vs. 2%; leakage following anastomosis was 11% versus 3%, 3-year overall survival rate was 50% vs. 48%; 5-year survival rate was 44% vs. 40% in the emergency operation and expandable metallic stent groups, respectively. They concluded that because preoperative expandable metallic stent insertion for obstructive colorectal carcinoma had good postoperative results and no disadvantages in longterm prognosis, this procedure should be used in preoperative treatments of obstructive colorectal carcinoma. Martinez-Santos et al. (511) evaluated primary anastomosis and morbidity rates obtained with self-expandable

127

stents in comparison with the results of emergency surgical treatment. Patients with left-sided malignant colorectal obstruction were enrolled. Forty-three patients were assigned to preoperative stent and elective operation or palliative stent (emergency surgical treatment). In the stent group, the obstruction was relieved in 95% after the stent placement. Of 26 patients who underwent operative treatment, a primary anastomosis was possible in 84.6% vs. 41.4% in the immediate operative group, with lower need for a colostomy (15.4% vs. 58.6%) in the immediate operative group. The anastomotic failure rate was similar and the reintervention rate was lower (0% vs. 17%). The total stay (14.2 days vs. 18.5 days), the intensive care unit stay (0.3 days vs. 2.9 days), and the number of patients with severe complications (11.6% vs. 41.2%) were significantly lower in the stent group. Johnson et al. (512) studied 36 patients of whom 18 had obstructing left-sided colon carcinomas relieved by placement of endolumnal stents. These were compared with 18 historical controls with similar clinicopathological features that were treated more conventionally with palliative stoma formation. Both groups of patients gained relief of obstructive symptoms. There were no differences in survival or inhospital mortality. The median length of palliation was 92 days for stenting and 121 days for palliative stoma formation. Formation of a stoma required a significantly longer stay in the intensive care unit but hospital stay was similar. They concluded as an alternative to palliative operation, selected patients benefit from colonic endoluminal stenting with relief of obstructive symptoms and no adverse effect on survival. Patients may be spared the potential problems associated with palliative stoma formation and the morbidity of operation. Stenting can be offered to the very frail patient who would otherwise be managed conservatively. Meisner et al. (513) reported on 104 procedures with self-expanding metal stents performed in 96 patients. The goals of the procedure were either postponement of emergency operation or definitive palliative treatment. Technical success was achieved in 92% and clinical success in 82%. Procedure-related complications included perforation in three patients during stenting and in one instance 6 to 7 hours after. Other technical problems could mainly be overcome by introducing an additional stent. They believe complications seen in the group treated with self-expanding metal stents and subsequent resection (mortality 18% and anastomotic leakage 18%), do not differ from the number of complications usually seen in patients who undergo colorectal resection. Suzuki et al. (514) reviewed 36 patients with malignant obstruction, and 6 patients with benign obstructive disease who underwent placement of self-expandable stents using a combined endoscopic and fluoroscopic technique. Stent placement was successful in 86%. Complication occurred in 44%, migration (n ¼ 7), reobstruction (n ¼ 5), perforation (n ¼ 2), fistula formation (n ¼ 1), and stent fracture (n ¼ 1). Stent placement was successful in 100% of patients with benign strictures but poststent migration was frequent (2/6). Tomiki et al. (515) compared the clinical outcome of 18 patients who had stent placement and 17 patients who underwent only colostomy. The postoperative hospital stay was 22.3 days for stent placement compared with 47.4 days

128

& PART I: COLORECTAL DISORDERS

for colostomy. The duration to readmission was 129.2 days for stent placement and 188.4 days for colostomy. The estimated duration of primary stent patency was 106 days. Mean survival period was 134 days in patients with stent placement and 191 days in patients with colostomy. They concluded, stent placement increases the option of palliative treatment and is an effective treatment contributing to improving quality of life. Sebastian et al. (516) systematically reviewed the efficacy and safety of self-expanding metal stents in the setting of malignant colorectal obstruction. Fifty-four studies reported the use of stents in a total of 1198 patients. The median technical and clinical success rates were 94% and 91%, respectively. The clinical success when used as a bridge to surgery was 71.7%. Major complications related to stent placement included perforation (3.8%), stent migration (11.8%), and reobstruction (7.3%). Stent-related mortality was 0.58%. Carne et al. (517) compared the use of expandable metallic stents as a palliative measure to traditional open surgical management. Patients with left-sided (splenic flexure and distal), colorectal carcinoma and nonresectable metastatic disease (stage IV) were treated with expandable metal stents or open resection or stoma. Twenty-two of 25 patients had colonic stents successfully inserted and 19 patients underwent open operation. The malignancies were primary in 22 stent procedures and 18 open operations. The open operations were laparotomy only (n ¼ 2), bypass (n ¼ 1), stoma (n ¼ 7), resection with anastomosis (n ¼ 4), resection without anastomosis (n ¼ 5). The complications after open operation were urinary (n ¼ 2), stroke (n ¼ 1), cardiac (n ¼ 2), respiratory (n ¼ 2), deep venous thrombosis (n ¼ 1), anastomotic leak (n ¼ 1). There were no stent-related complications. The mean length of stay was significantly shorter in the stent group (4 vs. 10.4 days). There was no difference in survival between the two groups (median survival: stent group 7.5 months; open operation, 3.9 months). They concluded, patients treated with stents are discharged earlier than after open operation. Stents do not affect survival. Although stents are expensive, the procedure appears to be cost-effective since emergency operation can be avoided with acute bowel obstruction, and in those with advanced disease no resection of the colon is necessary.

Perforation Perforation has been reported to occur in 3% to 9% of patients with colorectal carcinomas (299). Patients who develop a free perforation of the colon associated with a carcinoma present with signs and symptoms of generalized peritonitis. The carcinoma itself may be perforated, or there may be a left-sided carcinoma associated with a right-sided perforation. Each situation is handled differently: In the clinical setting, for treatment of a perforated carcinoma, older reports recommended that the perforation be managed by diversion, with a proximal colostomy or cecostomy performed in association with repair of the perforation. However, this treatment does not relieve the septic process, and the aim of therapy should be to remove the diseased segment. Otherwise, contamination will continue from the level of the stoma to the level of the perforation. On completion of the resection, the question arises as to

how to handle the bowel ends. If the patient already has generalized peritonitis, it seems inappropriate to perform a primary anastomosis. In this event, the proximal bowel is brought out as a stoma, and the distal bowel is drawn out as a mucous fistula or closed as a Hartmann pouch. For a right-sided perforation, a similar procedure can be performed. Another option is to resect the perforated diseased bowel and perform a primary anastomosis with a proximal diversionary stoma, either a proximal colostomy or a loop ileostomy. If technically feasible, the two ends of bowel should be brought out adjacent to each other as described for the end loop stoma. The advantage of this technique is that bowel continuity can be established at a later date without the need for a formal laparotomy. When there is an obstructing lesion of the left colon and a perforation of the right colon, a viable option is a subtotal colectomy encompassing removal of the perforated colon and the malignancy in one operation. Saegesser and Sandblom (323) stressed the fact that simple suture repair of an ischemic colon will not hold and that a temporary colostomy placed in an ischemic or inflamed bowel will pull through. The authors believe that the practice of closure of the perforation and relief of obstruction by colostomy or by exteriorization of the perforated cecum is illogical and inadequate. The surgeon should proceed with resection of the carcinoma and the entire distended part of the ischemic and perforated colon. A subtotal colectomy might even be considered if only a left-sided perforation is present, since this operation would fulfill the criteria of removing the diseased and unprepared bowel. Another option for management of the patient with a perforation remote from the diseased segment is to bring out the perforated segment as a stoma, either by colostomy or cecostomy. For the patient who presents with localized peritonitis on the right side, the diagnosis may be confused with that of appendicitis. If the diagnosis is definite at the time of laparotomy, it is reasonable to proceed with a right hemicolectomy and primary anastomosis. If the localized peritonitis occurs on the left side, the differential diagnosis will include diverticulitis. Resection of the diseased segment is indicated, and management of the ends involves the same considerations as with the obstructed unprepared bowel.

Bleeding Massive bleeding from a carcinoma is an unusual complication, but when it arises, it offers the built-in advantage of being a colonic cathartic. Therefore, if bleeding is so profuse that urgent operation is required, a mechanical cleansing is automatically present, and the affected portion of bowel can be resected with a primary anastomosis. Obstructive Colitis Obstructive colitis is an ulceroinflammatory condition that occurs in a dilated segment of the colon proximal to an obstructing or partially obstructing lesion. The entity is rarely reported in the literature and the following information was drawn from the review by Tsai et al. (518). Obstructive colitis is only encountered in 0.3% to 3.1% of all colorectal carcinomas and affects both men and women over 50 years of age. Minor degrees of obstructive colitis may be overlooked and its prevalence may be as high as

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

7% when specifically sought. The left side of the colon, especially the sigmoid colon, is usually involved in obstructive colitis. Patients with obstructive colitis usually complain of bleeding per rectum and abdominal pain as well as nausea and vomiting, all of which are indistinguishable from the symptoms of colorectal carcinoma. Regardless of severity and distribution pattern, a diagnostic feature of obstructive colitis is the presence of an intact mucosal segment of about 2- to 6-cm long between the carcinoma and the colitis. The area of colitis is usually a single confluent area, often with regular geographic margins, which is well demarcated from the surrounding normal mucosa. Microscopically, focal areas of colitis associated with obstructive colitis show replacement of mucosa by active granulation tissue. Acute and chronic inflammatory cells are moderate in amount and seldom extend beyond sites of granulation. Pseudopolyps of granular tissue or edematous mucosa may occur, and crypt abscesses may involve the mucosa at the ulcer margin. The mucosa in the intervening segment and distal to the obstructing lesion is usually normal. It is differentiated from ulcerative colitis which is characterized histologically by an intense inflammation of the mucosa and submucosa in addition to the presence of multiple crypt abscesses. The rectum is always involved and the disease extends proximally for varying distances but always with continuity of involvement to the proximal extent of the disease process. One suggested pathogenic mechanism of obstructive colitis is that of secondary ischemia caused by hypoperfusion. Additional factors, such as preexisting atheroma, anemia, or a past history of pelvic irradiation may play a role in precipitating the colitis. Obstructive colitis can cause both diagnostic and therapeutic problems. The signs and symptoms arising from obstructive colitis may be attributed to the primary obstructive lesion, which is usually most obvious on radiological and endoscopic studies. Areas of colitis may be a source of septicemia or may perforate and lead to peritonitis. Anastomoses in the unrecognized area of colitis may break down. Up to 25% of cases of obstructive colitis have been associated with anastomotic complications. Their frequently normal appearance at operation may lead to involved segments of colon being used for anastomoses with consequent complications. Because perforation through a colonic carcinoma is a grim prognostic event with negligible 5-year survival, it is important to distinguish this from perforation through the obstructive colitis, which may have a different prognosis. Awareness of the features and incidence of obstructive colitis should help surgeons avoid these diagnostic and therapeutic problems.

Invasion of Adjacent Viscera Occasionally a carcinoma becomes attached to the abdominal wall or the adjacent viscera, such as the small bowel, urinary bladder, uterus, stomach, spleen, ureter, or duodenum. It is estimated that such attachment occurs in approximately 10% of all patients with colon carcinomas with a reported range of 3.1% to 16.7% (519). The philosophy of treatment to be followed in these circumstances might best be expressed by the quote attributed to Hippocrates: ‘‘To extreme diseases, extreme remedies.’’ In order to perform

129

FIGURE 57 & (A) Carcinoma of the transverse colon attached to the spleen and greater curvature of the stomach resected enbloc. (B) Carcinoma of the transverse colon invading the sigmoid colon.

an adequate curative operation, it sometimes becomes necessary to excise en bloc all or part of the attached viscus (Fig. 57). Often these adhesions are inflammatory in nature and not caused by malignant infiltration, so the prognosis frequently is better than might have been anticipated originally (Table 8). With this knowledge, the surgeon should not hesitate to resect attached structures. An exception to these recommendations might be when the duodenum or bladder base is extensively involved, in which case the primary lesion is removed and the structures at risk are marked with metal clips. Under these circumstances, the morbidity and mortality rates of the radical operation involving an anterior exenteration or Whipple’s procedure would probably exceed any possible benefit derived from a very radical operation. However, Curley, Evans, and Ames (530) reported on 12 patients in whom the carcinoma involved the duodenum or pancreatic head and who underwent an en bloc extended right hemicolectomy and pancreaticoduodenectomy. There were no operative deaths, and malignant invasion was confirmed in all patients. At time of report, eight of the 12 patients were alive at a median 42 months. Similarly Koea et al. (531) reported their experience with 8 patients with bulky primary carcinomas of the right colon infiltrating the duodenum (n ¼ 4) or pancreatic head (n ¼ 4) surgically managed at Memorial Sloan-Kettering Cancer Center. Six patients presented with anemia, and one

130

& PART I: COLORECTAL DISORDERS

TABLE 8 & Nature of Adhesions Between Colon and Adjacent Viscus

Inflammatory

5-Year Survival (%)

Operative Mortality (%)

49 55 39 57 84

51 45 61 43 16

70 52 61 52 0–76a

– 12 0 0 2

71 70 54 69 38 53

29 30 46 31 62 47

30 54 44 mo (mean) 34–78b 51 77

0 4 1 4 – 0

Adhesions (%) Author(s) Glass et al. (520) (1986) Gall, Tonak, and Altendorf (521) (1987) Hunter, Ryan, and Schultz (522) (1987) Orkin et al. (523) (1989) Eisenberg, Kraybill, and Lopez (524) (1990) Montesani et al. (525) (1991) Curley et al. (526) (1992) Izbicki et al. (527) (1995) Rowe et al. (519) (1997) Carne et al. (528) (2004) Nakafusa et al. (529) (2004)

No. of Patients

Carcinomatous

69 121 28 65 58 35 101 83 118 53 53

a

0% survival for those with lymph node metastases; 76% for those without lymph node metastases. 5 year survival: 78% for inflammatory adhesions plus negative lymph nodes; 58% for inflammatory adhesions plus positive lymph nodes; 34% for invasive adhesions plus positive lymph nodes; 64% for invasive adhesions plus negative lymph nodes; 71% for negative lymph nodes; 47% for positive lymph nodes.

b

patient each with epigastric pain and an abdominal mass. All patients had T4 lesions, whereas five had lymph node metastases at presentation. All patients were resected with clear pathologic margins either by right colectomy and en bloc duodenectomy (n ¼ 4) or en bloc pancreaticoduodenectomy (n ¼ 4). The 30-day mortality rate was 0. Six patients remained alive and free of disease at a median follow-up of 26 months, and there was one long-term survivor who was alive and free of disease at 84 months after resection. Talamonti et al. (532) reviewed 70 patients who underwent resection of a carcinoma of the colon and rectum with en bloc total cystectomy (36 patients) or partial cystectomy (34 patients). There were three postoperative deaths in the total cystectomy group but none, after partial cystectomy. The 5-year actuarial survival rate for the 64 patients with negative resection margins was 52%. In exceptionally good-risk patients, consideration may be given to a total pelvic exenteration. In a review of 1918 patients who underwent curative resection for colorectal carcinoma, Gall, Tonak, and Altendorf (521) noted that 121 patients had multivisceral organ involvement. Extended multivisceral radical resections resulted in a postoperative mortality rate of 12% (compared to 6% without such resection), with a 5-year survival rate of 54% for patients with inflammatory adherence and 49% for patients with malignant infiltration. In this series the most frequently used extension of resection was total hysterectomy (39%), small bowel (21%), urinary bladder (16%), and abdominal wall (4%). It is worthy of note that when carcinoma was inadvertently torn or transected during resection, the 5-year survival rate dramatically dropped to 17%. Despite the increased operative mortality with extended resection, the authors of this review believe that the benefit outweighs the disadvantages. Hunter, Ryan, and Schultz (522) reviewed their results of colorectal carcinoma in three treatment groups: standard colectomy, en bloc resection, and colectomy with separation of adherent organs. The 5-year survival rates were 55%, 61%, and 23%, respectively. No operative mortality occurred with en bloc resection. The 5-year survival rate, recurrence rate, and local recurrence rate for standard colectomy were 55%, 33%, and 11%, respectively; for en bloc resection, 61%, 36%, and

18%; and for separation of organs, 23%, 77%, and 69%. The authors concluded that colorectal carcinoma adherent to adjacent organs must be treated by en bloc resection because separation of organs results in unacceptably high local recurrence and poor 5-year survival rates. On the other hand, the results of en bloc resection were comparable to those of standard colectomy for nonadherent carcinomas. Natafusa et al. (529) evaluated the short-term or longterm outcome of multivisceral resection relative to that of the standard operation. Of 323 patients, 16.4% received multivisceral resection because of adhesion to other organs. Overall, morbidity rates were 49.1% for multivisceral resection versus 17.8% for the standard operation and postoperative mortality was 0% in both groups. Only multivisceral resection (odds ratio, 2.7) was an independent factor for overall postoperative complications. The survival of patients with multivisceral resection was similar to that after the standard operation (5-year rate 76.6% vs. 79.5%). Lymph node metastases (hazard ratio 2.5) and blood transfusion (hazard ratio 2.4) were independently associated with patient survival. Kroneman, Castelein, and Jeekel (533) evaluated the results of 33 patients who underwent curative en bloc resection. Adherent organs excised included small bowel, urinary bladder, abdominal wall, uterus, duodenum, pancreas, stomach, and kidney. The postoperative morbidity rate was 6%, the mortality was 3%, and the 4-year survival rate was 33%. Poeze et al. (534) reported on 1346 patients with colorectal carcinoma, 144 (11%) of whom underwent multivisceral resections for invasion of adjacent organs. In patients who had disease-free margins, there was no compromise of long-term survival (i.e., local invasion to adjacent organs with or without lymph node involvement was not related to survival). The overall operative mortality rate was 5%. Izbicki et al. (527) reported on 83 patients who underwent en bloc resection. Mean survival was 44 months after extended resection. The postoperative mortality, morbidity, and survival rates were comparable to those in patients who underwent nonextended resections. Landercasper et al. (535) reported on 54 of 1284 patients (4%) who underwent potentially curative resections of right colon lesions found to be adherent to adjacent organs, abdominal wall, or retroperitoneum. Postoperative

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

complications developed in 24% of patients. The mortality rate was 1.9% and the 5-year survival rate was 31%. Only one of nine patients with pancreatic or duodenal adherence treated with limited resection remained disease free. The authors recommend radical en bloc resection if no distant metastases are present. Adjuvant radiation therapy or chemotherapy did not improve survival. To determine the perioperative mortality and morbidity and the long-term prognosis of patients undergoing extended pelvic resections for localized advanced primary adenocarcinoma of the rectum, Orkin et al. (523) reviewed their experience with 65 patients. Local invasion without distant metastases was present in all patients at operation and en bloc resection of all involved organs was performed with intent of cure. Average age at operation was 61 years; 23% were men and 77% were women. Operations included abdominoperineal resection in 57%, low anterior resection in 31% and Hartmann procedure in 12%. Additionally, women (81%) with intact uteri underwent en bloc hysterectomy, 77% women with intact ovaries had oophorectomy, and 50% of women had partial vaginal resection. Twenty-six percent of the 65 patients had a cystectomy, and two patients had a portion of small intestine resected in continuity with their carcinoma. Pathologic examination revealed lymph node involvement in 45% and histologic confirmation of adjacent organ extension in 57%. There were no perioperative deaths. Overall, 5-year survival was 52% with 65% of deaths attributable to either recurrent carcinoma or a new primary lesion. The cumulative probability of recurrence at five years was 39%. Carne et al. (528) reported on multicenter experiences of en bloc bladder resection for colorectal carcinoma adhering to the urinary bladder. Fifty-three patients were identified of which 45 had en bloc partial cystectomy performed, 4 en bloc total cystectomy and four had the adhesions disrupted and no bladder resection. All patients who did not have a bloc resection developed local recurrence and died from their disease. Mean follow-up was 62 months. The extent of bladder resection did not seem important in determining local recurrence. The decision to perform total rather than partial cystectomy should be based on the anatomic location of the carcinoma. Rowe et al. (519) determined the therapeutic benefit of multivisceral resection in patients with locally advanced colorectal carcinomas. The study population was composed of 118 patients whose resection of the primary lesion included one or more adhesed adjacent secondary organs or structures. Their survival is reported in Table 8 but clinical relevance is that there was no statistically significant difference in the 5-year survival rates when multiple adjacent secondary organs or structures were resected and therefore they believe an aggressive operative approach is warranted. Yamada et al. (536) reported 64 patients with locally advanced primary or recurrent rectal carcinoma who underwent abdominoperineal resection with sacral resection in 9 patients, anterior pelvic exenteration in 8 patients, total pelvic exenteration in 27 patients, and total pelvic exenteration with sacral resection in 20 patients. Rates of morbidity, reoperation, and mortality were 50%, 4.5%, and 0% in 22 patients with primary carcinoma, and 60%, 2.4%, and 2.4% in 42 patients with recurrent disease, respectively. Major complications, such as

131

sepsis, intra-abdominal abscess, and enteric fistula caused one hospital death and reoperation in two patients. In 21 patients who underwent curative resection for primary carcinoma, the overall 5-year survival rates were 74.1% for Dukes’ B and 47.4% for Dukes’ C although the difference was not statistically significant. Thirty patients with recurrent carcinoma who underwent curative resection, had significantly improved survival with a 5-year survival rate of 22.9% compared with 12 patients who underwent palliative resection resulting in a survival rate of 0%. In an excellent clinical review of the role of extended resection in the initial treatment of locally advanced colorectal carcinoma, Lopez and Monafo (537) collated information on the results of multivisceral resection for colorectal carcinoma. In 11 publications in which 609 patients underwent extended resection for colorectal carcinoma, the operative morbidity rate was 27%, the operative mortality rate was 6%, and lymph node metastases occurred in 39%. The 5-year survival rate was 68% for node-negative status and 23% for node-positive status. If adherence to adjacent viscera was benign, the 5-year survival rate was 68% but declined to 40% if the attachment was malignant. Survival in locally advanced colorectal carcinoma is more dependent on lymph node status than on the extent of local invasion (524). In 23 publications in which 248 patients underwent total pelvic exenteration for rectosigmoid carcinoma, the operative morbidity rate was 60% and the operative mortality rate was 12%. The 5-year survival rate was 64% for node-negative status and 32% for node-positive status. In the unique situation in which there is isolated invasion of the prostate by a rectal carcinoma, Campbell et al. (538) described the use of radical retropubic prostatectomy in conjunction with restorative proctosigmoidectomy for en bloc excision. This novel technique offers an alternative to total pelvic exenteration, thereby obviating the need for urinary and fecal diversion. The expected 5-year survival of patients subjected to en bloc resection ranged from 30% to 79% (see Table 8) and thus justifies an aggressive approach.

Urinary Tract Involvement by Colorectal Carcinoma McNamara et al. (539) recently reviewed the literature on urinary tract involvement by colorectal carcinoma with the aim of highlighting technical and oncologic issues that should be considered when dealing with this complex problem. From the relevant literature, they identified three distinct clinical scenarios in which the urinary tract may be affected by colorectal carcinoma: involvement by primary colorectal carcinoma, involvement of recurrent carcinoma, and unexpected intraoperative findings of urinary tract involvement. The following information and guidelines draw heavily from their dissertation. Primary Involvement of the Urinary Tract Involvement of the urinary tract system occurs in 5% of patients with primary colorectal carcinoma. Any level of the urinary tract can be affected by direct invasion or be involved with an associated inflammatory mass. Three sites are most commonly affected: the dome of the bladder, the lower ureter, and the base of the bladder. Adherence to or

132

& PART I: COLORECTAL DISORDERS

invasion of the dome of the bladder is the most common presentation and most frequently occurs in rectosigmoid malignancies. Locally advanced disease with direct invasion of adjacent organs may result in fistula formation, but half of such patients have no symptoms at presentation. Involvement of the trigone may compromise the intramural ureter. Lower third lesions of the rectum may involve the prostate gland and prostatic urethra. A CT is usually performed as part of the standard investigation of patients with sigmoid or rectal carcinoma but is mandatory in patients with urinary symptoms. In addition to staging, CT allows localization of the ureters and confirms bilateral renal function, although it tends to overestimate the need for urinary organ resection. CT is more likely to produce a false-positive diagnosis of pelvic floor or piriform muscle invasion than magnetic resonance imaging and is less likely to identify sacral bone invasion when it is present. Modern high-resolution MRI (sensitivity 97% and specificity 98%) is superior to CT (sensitivity 70% and specificity 85%) in staging locally advanced primary or recurrent rectal carcinomas, with better detection of penetration of the fascia propria and involvement of the potential circumferential resection margin. Cystoscopy diagnoses the cause of genitourinary symptoms in 79% to 87% of patients with rectal carcinoma. Only 57% of patients with a mucosal abnormality at cystoscopy have bladder invasion at final pathology, yet locating the vesical opening of a malignant rectovesical fistula improves identification of patients who require pelvic exenteration for adequate resection. Bladder Involvement If involvement of the dome of the bladder is suspected, en bloc resection of the carcinoma and all adherent bladder should be performed, because of the well documented difficulty in distinguishing between adherence and invasion macroscopically and the greatly diminished survival experienced by patients in whom the carcinoma is breached during resection. This policy carries the risk that the adjacent organ in the resected specimen may show no evidence of malignant invasion but is justified because no increase in morbidity is reported following multivisceral resection, especially partial cystectomy. No adverse effect on local recurrence or survival has been demonstrated when partial cystectomy is performed instead of total cystectomy for localized malignant involvement, provided the resection is R0. Involvement of the trigone is less straightforward, and curative resection requires total pelvic exenteration. Total pelvic exenteration is appropriate for direct invasion of the trigone, vesicoureteric junction, or intramural ureter in the absence of distant metastases and has been used in both primary and locally recurrent disease. Total pelvic exenteration may be combined with sacral resection, especially in patients with local recurrence extending into the presacral space. Bladder reconstruction requires construction of a urinary conduit, of which an ileal conduit is the most common, although cecal or colonic conduits are sometimes used. Supralevator exenteration with doublepouch reconstruction using a colonic J-pouch and a Mainz pouch with sphincter-preserving urethral anastomosis has been described, but long-term results are not available and recurrence in this setting may result in catastrophic

complications. Early urologic complications of urinary diversion include ileoureteral anastomotic dehiscence and early hydronephrosis. Late urologic complications include ureteral stenosis and late hydronephrosis. Unsuccessful endoscopic and radiologic management of these complications may lead to the necessity for nephrectomy. Operative mortality rates following total pelvic exenteration ranging between 5% and 33% have been quoted. There is a trend toward increased morbidity in patients who receive preoperative radiotherapy. Review of the literature reveals 3-year survival figures ranging from 30% to 64.5% and 5-year survival figures ranging from 9% to 61%. Some surgeons routinely include intrapelvic dissection of the internal iliac and obturator nodes in their approach to total pelvic exenteration but no convincing survival advantage has been demonstrated. Total pelvic exenteration has been reported to have a sixfold greater mortality than, lesser, exenterative procedures. Ureteric Involvement Bilateral involvement of the ureters may occur because of compression from extensive nodal disease at the pelvic brim or by invasion of the trigone by the primary carcinoma but both scenarios usually require total pelvic exenteration if curative resection is desired. In contrast, unilateral ureteric invasion may be approached by en bloc resection of the affected segment followed by appropriate reconstruction. Ipsilateral ureteroureterostomy over a double-J stent is the simplest form of anastomosis but even when combined with use of a vesicopsoas hitch is suitable only for short resections of the distal ureter. Reconstruction following resection of a longer segment may require use of a Boari flap in which a well-vascularized flap of bladder is constructed into a tube to which the proximal ureter may be anastomosed. Cystoureterectomy and ureteric crossover is recommended for unilateral involvement of the ureterovesical junction and may be performed without significantly increasing postoperative morbidity and mortality. Ileal interposition has satisfactory oncologic results and allows resection of a long ureteric segment but may result in renal damage because of transmission of high intravesical pressures and should only be performed in carefully selected patients. Rarely, nephrectomy may be an acceptable option. Fistula Rectourinary fistulation is an uncommon event that rarely occurs in females because of the protective effect afforded by the interposition of the female genital tract. The classic triad of pneumaturia, fecaluria, and recurrent urinary tract infection is unusual, and patients more commonly present with fever, a pelvic mass, or cystitis. Most patients have a urinary tract infection, but pneumaturia is reported by only 10% of patients. Only 21% of fistulas associated with a rectal carcinoma contain malignant cells. The remaining 79% result from interventions (including operation, radiotherapy and chemotherapy) for rectal carcinoma. The success rate for initial and reoperative surgery was 21% and 88%, respectively, when malignant cells were identified in the fistula tract as compared with success rates of 44% and 100% for treatment-related fistulas. The decision to administer neoadjuvant chemoradiotherapy must balance the possibility of improved survival and less radical

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

operation against the reported increase in preoperative fistulization and perioperative morbidity and mortality. Hydronephrosis In a patient with primary colorectal carcinoma, the most common cause of hydronephrosis is regional nodal disease from a sigmoid or rectal carcinoma at the pelvic brim but direct extension of a primary carcinoma, local inflammation, and isolated ureteric metastases are all possible. Malignant hydronephrosis detected at the time of first diagnosis of colorectal carcinoma is a worrying finding because less than half of such patients have resectable disease. Radiotherapy The role of preoperative radiotherapy in rectal carcinoma involving the urinary tract is not yet clear. Downstaging may improve resectability by reducing the extent of operation necessary to obtain negative margins and rendering some inoperable carcinomas resectable. Unexpected Intraoperative Involvement A particular difficulty arises when unexpected local extensive disease is identified at operation. Discovery of a rectosigmoid carcinoma adherent to the bladder for which one can envisage a relatively straightforward en bloc resection with primary closure of the bladder clearly differs from a carcinoma likely to require complex reconstruction. Important are issues relating to the quality of the preoperative informed consent, particularly if the proposed resection requires a procedure with the potential for considerably greater morbidity and mortality than anticipated or an unexpected impact on postoperative quality of life such as necessity to create a stoma. In some circumstances, the correct decision is to defer resectional operation in favor of radiotherapy or a subsequent more aggressive one-stage procedure. Recurrent Colorectal Carcinoma The finding of hydronephrosis after a previous colorectal resection usually indicates pelvic sidewall disease that precludes resection. It is associated with concomitant metastatic disease in 50% of patients and predicts poor survival, even after salvage operation. Investigation of a patient with suspected recurrence involving the urinary tract should be vigorous to avoid the morbidity and mortality of salvage operation in patients unlikely to benefit. Inoperable metastatic disease should be excluded with spiral CT and MRI or PET. Rarely, urinary and/or fecal diversion may be justified in the presence of metastatic disease in patients who are symptomatic but cannot be successfully palliated with less invasive radiologic or endourologic techniques. Abnormal Renal Function Patients with abnormal preoperative renal function require optimization of their condition before operation. An elevated preoperative urea level is independently predictive of increased 30-day mortality, while patients who develop acute renal failure postoperatively have a 30-day mortality in excess of 50%. Patients may require preoperative urinary

133

decompression. Early urinary decompression is a priority to prevent or minimize irreversible renal damage. This may take the form of initial retrograde double-J stenting or percutaneous nephrostomy with subsequent endourologic stent insertion. Palliation Treatment of unresectable carcinoma involving the urinary tract or potentially resectable local disease in the presence of unresectable metastases should maximize survival without adverse effects on quality of life. Most malignant strictures of the ureter can be treated by an endourologic approach with minimal morbidity, allowing normal micturition without external drainage and with durable results. Pelvic arterial infusion of 5-FU and mitomycin C may reduce pain and improve hydronephrosis while systemic steroid therapy has also been suggested for the temporary improvement of hydronephrosis in the palliative setting.

Unresectable Carcinoma In the unusual circumstance in which a lesion is totally unresectable, it usually can be bypassed satisfactorily. Palliative Resection One of the most unsatisfying situations facing any surgeon who operates on patients with colon and rectal carcinoma is that of recommending a major abdominal procedure, with its potential complications, to a patient who has definite evidence of incurable disease. The decision regarding operative intervention is usually reached with some trepidation, since many of these patients are in poor physical condition and have a limited life expectancy. However, even for patients with metastatic carcinoma of the large bowel, resection performed to eliminate the symptoms of local disease has been advocated as a worthwhile procedure for avoiding the potential complications of obstruction and massive bleeding and the effects of local invasion of the primary lesion. In general, resections relieve patients of their symptoms and sometimes may even prolong life expectancy (540). The most common symptoms are pain and bleeding (541,542). It has been estimated that 10% to 20% of patients who are seen with primary operable colorectal carcinoma already have associated liver metastases (540). Unfortunately, not all patients will benefit from resection, and, in fact, some patients will be caused additional morbidity. This morbidity, together with the mortality of the operative procedure, may exceed the benefit of any temporary symptomatic relief. Thus the role of palliative resection for malignant neoplasms has been questioned from time to time. This is especially true for the decision to perform a palliative abdominoperineal resection, an operation that entails not only an operative procedure of considerable magnitude but also the establishment of a permanent colostomy in a patient with only a chance of limited survival. In the presence of metastatic disease, survival will depend on the nature and extent of the metastases. Indeed, some metastatic lesions should be resected in addition to the primary lesion. Survival depends on the pattern of metastatic disease. For example, Joffe and Gordon (541) noted survival with unilobar liver metastases to be 16.9 months,

134

& PART I: COLORECTAL DISORDERS

while with bilobar metastases, survival was only 8.5 months. Cady, Monson, and Swinton (540) noted a survival of 13 months; Takaki, Ujiki, and Shields (543), 12 months; and Goslin et al. (544) a similar length of survival. Under such circumstances, the recommendation for resection should be tempered by a consideration of factors such as extensive hepatic replacement or jaundice, marked ascites, or massive peritoneal seeding, in which case life expectancy is very short and no benefit could be accrued from a resection. The prognosis is poor for patients with extensive liver metastases, patients older than 75, and patients with a previous history of cardiovascular disease (541). Makela et al. (545) reviewed 96 patients who underwent palliative operations with an 8% postoperative mortality rate (5% for resections and 17% nonresection procedures) and a 24% postoperative morbidity rate. Median survival was 10 months (15 months for resections and 7 months for nonresection procedures) and 5% of patients survived longer than 5 years. The median relief of symptoms related to the malignancy was 4 months (4 months after resection and 1 month after nonresection procedures). Twenty-five patients underwent a second palliative operation. Liu et al. (542) studied 68 patients with incurable colon carcinoma to try to identify objective criteria that might help surgeons decide which patients will benefit from palliative operations. The postoperative mortality rate was 10% and the complication rate was 10%. The mean survival after palliative resection was 10.6 months, after bypass was 3.4 months, and after diagnosis in patients not operated on was 2.0 months. Of the variables studied, the only factors affecting survival were poorly differentiated lesions and > 50% replacement of liver. The authors concluded that although resection carries with it a relatively high postoperative mortality rate, it is worthwhile as long as hepatic metastases occupy < 50% of liver volume. The macroscopic features of the primary disease must be taken into consideration because of the ever-present concern of obstruction. However, the advent of endoscopic, laser therapy appears entirely suited to maintaining an adequate caliber of lumen to prevent obstruction. Endoluminal stenting is another way of dealing with obstructive symptoms.

Synchronous Carcinomas Recommendations for the appropriate treatment of synchronous carcinomas of the colon are at least in part based on the magnitude of the risk of development of metachronous adenomas and carcinomas after conventional resections. The incidence of synchronous carcinomas has been reported to be 1.5% to 7.6% (546). In a series of 2586 patients, an incidence of 1.8% was reported (546). Bussey, Wallace, and Morson (547) reported on 3381 patients who survived conventional resections for carcinoma of the colon and rectum at St. Mark’s Hospital in London and found an overall incidence of metachronous carcinoma of 1.5%. The incidence rose to 3% in those cases followed up for at least 20 years. For those patients in whom an associated adenomatous polyp was found in the original operative specimen, the level rose to 5%. In a more recent study, synchronous carcinomas were found in 4.4% of patients (548). Passman, Pommier, and Vetto (549) reported on an

18-year multi-institutional data base of 4878 patients with colon carcinoma. There were 160 patients (3.3%) with 339 synchronous carcinomas. Eight percent of these patients had more than two lesions at the time of diagnosis. Based on highest stage lesion, 1% of patients were at stage 0, 28% at stage I, 33% at stage II, 25% at stage III, and 11% at stage IV. The disease-specific 5-year survival rate by highest stage was 87% for stage 0 or I, 69% for stage II, 50% for stage III, and 14% for stage IV. These ‘‘highest stage’’ survival rates for patients with synchronous carcinomas were not significantly different from survival of patients with same-stage solitary carcinomas in their data base. In light of this, it seems reasonable that if synchronous carcinomas are located in the same anatomic region, a conventional resection should be performed. When the carcinomas are widely separated, a subtotal colectomy is the operation of choice.

Synchronous Polyps and Carcinoma Recommendation for the treatment of patients with colon carcinoma and associated polyps involves the same considerations as for synchronous carcinomas. However, it also depends on the number, location, and size of these polyps. For example, if the polyps were confined to the region of the index carcinoma, the conventional operation for that portion of the bowel would be indicated. With the availability of colonoscopy, assessment and possible therapy of associated polyps can be accomplished. If the remaining bowel contains only occasional polyps that can be easily excised with the colonoscope, it would appear reasonable to have these polyps excised and to proceed with a conventional resection of the carcinoma. If one of the excised polyps should contain a carcinoma or if the polyps were of a size deemed in excess of colonoscopic polypectomy, a subtotal colectomy would be appropriate (550). Subtotal colectomy even has been recommended for colon carcinoma and synchronous polyp in good-risk patients (548). An individual who has exhibited the propensity for growth of many polyps in the colon, although not in adequate numbers to be considered familial adenomatous polyposis, would still qualify for a subtotal colectomy and ileorectal, or at least ileosigmoid, anastomosis. Metachronous Carcinoma Gervaz et al. (551) assessed the incidence of metachronous colorectal carcinomas in a population-based study of 500,000 residents. Of this total, 5006 patients had sporadic carcinoma of the colon or rectum with 34% being located proximal to the splenic flexure. Occurrence of a second primary colorectal carcinoma was observed in 2.4% of this population. The risk for developing a second incidence of primary colorectal carcinoma was higher in patients whose initial carcinoma was located in the proximal colon (3.4% vs. 1.8%, odds ratio 1.9). The risk for each segment of large bowel was as follows: cecum, 3.4%; right colon, 3%; transverse colon, 3.8%; left colon, 2.8%; sigmoid colon, 1.7%; and rectum 1.8%. By contrast, the risk for developing a second extracolonic carcinoma did not differ between patients with proximal and distal carcinomas (13.7% vs. 13.4%).

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

Shitoh et al. (552) reported that microsatellite instability can be regarded as an independent marker for predicting the development of metachronous colorectal carcinoma after operation. In a study of 328 colorectal carcinoma patients surveyed by periodic colonoscopy for at least three years after operation, 17 metachronous colorectal carcinomas were detected during the follow-up period. The% of microsatellite instability-positive cases was 26.4%. Incidences of metachronous colorectal carcinomas in microsatellite instability-positive and microsatellite instability-negative cases were 15.3% and 3%, respectively. The cumulative 5-year incidence of metachronous colorectal carcinomas was significantly higher in microsatellite instability positive cases than in microsatellite instability-negative cases (12.5% vs. 2.5%).

Treatment of Metastatic Disease Liver Metastases to the liver from carcinoma of the colon or rectum are frequent occurances. Indeed, the liver is the dominant site of treatment failure and the major cause of death in patients with colorectal carcinoma. Studies have demonstrated that up to 30% of patients undergoing apparently curative operation already have hepatic metastases that are not evident to the surgeon at the time of laparotomy (369,553,554). Furthermore, another 50% have recurrent disease develop within the liver (555). Some 90% of patients who die from colorectal carcinoma have liver metastases (264). In a study of doubling times, Finlay et al. (556) determined that the mean doubling time for overt metastases was 155  34 days ( SEM) compared with 86  12 days for occult metastases. The mean age of the metastases at the time of operation was estimated by extrapolation of the observed growth curve, assuming Gompertzian kinetics, to be 3.7  0.9 years ( SEM) for overt metastases and 2.3  0.4 years for occult metastases. There is a perception that streamline flow of blood in the portal vein may influence the anatomic distribution of liver metastases, depending on the site of the primary lesion. It has previously been reported that carcinomas arising in the right colon are distributed to the right lobe of the liver 10 times more commonly than to the left lobe, whereas liver metastases from carcinomas arising from the left colon and rectum are believed to be distributed homogenously. Wigmore et al. (557) collected data prospectively on the anatomic site of hepatic metastases in 207 patients with colorectal metastases. This study could not find any evidence to support a differential pattern of metastasis within the liver dependent on the location of the primary colorectal carcinoma. In an effort to accurately detect liver metastases, Van Ooijen et al. (558) prospectively compared continuous CT angiography to preoperative ultrasonography and conventional CT in 60 patients with primary or secondary colorectal carcinoma. The standard references were palpation of the liver and intraoperative ultrasonography. Continuous CT angiography had a high sensitivity of 94% in contrast to ultrasonography (48%) and conventional CT (52%). However, there was a higher false positive rate because of variations in the perfusion of normal liver parenchyma. Overall, continuous CT angiography had the highest accuracy (74%) compared with ultrasonography (57%) and CT (57%). The low specificity will hamper its routine application.

135

Strasberg et al. (559) reviewed 43 patients with metastatic colorectal carcinoma referred for hepatic resection after conventional staging with CT. PET scanning was performed on all patients. PET identified additional carcinoma not seen on CT in 10 patients. Operation was contraindicated in six of these patients because of the findings on PET. Laparotomy was performed in 37 patients. In all but two, liver resection was performed. The Kaplan-Meier estimate of overall survival at three years was 77%. This figure is higher than 3-year estimate of survival found in previously published series. They concluded preoperative PET scan lessens the recurrence rate in patients undergoing hepatic resection for colorectal metastases to the liver by detection of disease not found on conventional imaging. Some patients who will not benefit from operation can thus be spared a laparotomy and major resection. Liver surgeons usually recommend against biopsy of colorectal liver metastases because of the risk of local dissemination. Rodgers et al. (560) conducted a multicenter retrospective review of cases of colorectal liver metastases presenting for operation that had undergone a preoperative biopsy. Of 231 cases of colorectal metastases, 18.6% had undergone a preoperative biopsy. Evidence of dissemination related to the biopsy was 16%. Within the operative period (median 21 months), three of the seven cases with evidence of dissemination and 11 of the 35 without dissemination were alive without disease. They concluded there is a significant risk of local dissemination with biopsy of colorectal liver metastases. The value of intraoperative hepatic ultrasonography has been discussed in detail on p. 619. Fuhrman et al. (561) reported on the use of intraoperative ultrasonography in the assessment of porta hepatis lymph nodes and the evaluation of resection margins to determine whether this modality would improve the selection of patients likely to benefit from operation. Of 151 patients undergoing exploration, 30 patients were considered unresectable, 14 (9.2%) demonstrated by intraoperative ultrasonography. The authors concluded that intraoperative ultrasonography did, indeed, improve the selection process. The question of what to do for patients with these metastases has been a matter of controversy. At one point any suggestion of an operative approach to metastatic disease was deemed foolish by some. The natural history of untreated hepatic metastases confirms a median survival of 6 to 12 months and of 4.5 months if metastases are synchronous (562–564). If not resected, 3-year survival rates ranged from 3% to 7% and only 1% to 2% of patients will survive for 5 years (555). Six studies of the natural history of such metastases in a total of 1151 patients described a 5-year survival rate of 3%. In a study of 484 untreated patients, six independent determinants of survival were identified in the following order: (1)% liver volume replaced by carcinoma, (2) grade of malignancy of the primary lesion, (3) presence of extrahepatic disease, (4) mesenteric lymph node involvement, (5) serum CEA, and (6) patient age (565). The prognosis is closely related to the extent of liver replacement. A variety of chemotherapeutic regimens, including systemic chemotherapy and direct intraportal and intraarterial modes of administration, have been attempted, all with limited and short (if any) benefit, but with the cost

136

& PART I: COLORECTAL DISORDERS

TABLE 9 & Survival Following Resection of Hepatic Metastases Survival (%) Author(s) Hughes, Scheele, and Sugarbaker (567) (1989)a Schlag, Hohenberger, and Herforth (568) (1990) Petrelli et al. (569) (1991) Doci et al. (570) (1991) Rosen et al. (571) (1992) Nakamura et al. (572) (1992) Van Ooijen et al. (573) (1992) Grayowski et al. (574) (1994) Scheele et al. (575) (1995) Fuhrman et al. (561) (1995)b Hananel, Garzon, and Gordon (576) (1995) Rougier et al. (577) (1995) Wade et al. (578) (1996) Wanebo et al. (579) (1996) Ohlsson et al. (580) (1998) Fong et al. (581) (1999) Buell et al. (582) (2000) Elias et al. (583) (2003) Kato et al. (584) (2003)c Teh and Ooi (585) (2003) Weber et al. (586) (2003)

No. of Patients 800 122 62 100 280 31 118 204 434 107 26 123 133 74 111 1001 110 111 585 96 62

3-Year 40

47 45 43 45

35

57 54 38 71 45

5-Year 32 30 26 30 25 45 21 32 33 44 31 21 26 24 25 37 40 20 33 22

Operative Mortality (%)

Complication Rate (%)

4 8 5 4 3 8 1 4 3 0

34 30 41 16 35 22 66

4 7 4 3 2 4

31 21 28

0 0

7 36

35

a

Tumor registry of 24 institutions (24%, 5-year disease-free survival). Ultimate patient selection with intraoperative ultrasonography. c Post operative hepatic artery chemotherapy in 33% but no difference in survival noted in those with or without chemotherapy. b

of considerable toxicity and anxiety. Systemic chemotherapy has resulted in response rates ranges 18% to 28% (566), and the median survival rate ranges of 8 to 14 months (555). Other efforts have been directed at hepatic artery embolization, hepatic artery ligation, and even irradiation, all without significant worthwhile benefit. The lack of effective therapeutic alternatives has made hepatic resection the primary treatment consideration. Indeed, worthwhile survival rates in selected patients have been reported (Table 9). The timing and extent of operation varies. In the patient who presents with a synchronous lesion, which is amenable to operation, it appears appropriate to excise the lesion at the time of operation. If the lesion requires a major hepatic resection, the combination of partial hepatectomy and colectomy appears to be too great a task for one operation. After the colonic resection, if there is no other evidence of metastases and if a thorough evaluation, including a CT scan, has demonstrated removable disease, proceeding with resection is the treatment of choice. If the patient presents with metastatic disease at a later date, evaluation is necessary to ensure that the metastatic disease is confined; at the same time evaluation should be performed to rule out the presence of recurrent disease at the area of the index carcinoma. The resection of hepatic metastases in patients with intra-abdominal extrahepatic disease is of no proven benefit (587). Even with preoperative staging, as many as 26% of patients will have intra-abdominal extrahepatic metastases, most commonly in portal and celiac lymph nodes (587). It is necessary to rule out evidence of other metastatic disease. Unfortunately, investigation of patients rarely unveils a solitary lesion. Only approximately 10% of patients develop metastases suitable for operation. In their study of the natural history of hepatic metastases from colorectal

carcinoma, Wagner et al. (588) found that the median survival rate for unresected solitary and multiple unilobar metastases was 21 and 15 months, respectively. Earlier series reported untreated patients to have a median survival of 6 to 12 months. It is understandable why hepatic resection became an attractive option. Surgical resection of primary colorectal carcinoma in patients with stage IV disease at initial presentation remains controversial. Although bowel resection to manage symptoms such as bleeding, perforation, or obstruction has been advocated, management of asymptomatic patients has not been well defined. Patient-dependent factors (performance status, comorbid disease) and extent of distant metastases are among the considerations that have an impact on the decision to proceed with operative management in asymptomatic stage IV colorectal carcinoma. To ascertain the natural history of a group of untreated patients and to evaluate simultaneously in another group whether or not the administration of systemic chemotherapy modifies this natural history, Luna-Perez et al. (589) followed 77 patients with liver metastases from colorectal carcinoma. Untreated patients consisted of 45 patients; 41 developed extrahepatic metastatic disease and their median survival rate was 13 months. The group who received chemotherapy included 32 patients; 29 developed extrahepatic metastatic disease and their median survival was 15 months. There were no differences in overall survival in both groups. The administration of systemic chemotherapy did not modify the natural course of the disease. Dismal results of this nature mandate a better form of therapy, namely, operative. Ruo et al. (590) reviewed 127 patients who underwent elective resection of their asymptomatic primary colorectal carcinoma. Over the same time period, 103 stage IV patients who did not undergo resection were

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

identified. The resected group could be easily distinguished from the nonresected group by a higher frequency of right colon carcinomas and metastatic disease restricted to the liver or one other site apart from the primary carcinoma. Resected patients had prolonged median (16 months vs. 9 months) and two years (25% vs. 6%) survival compared with patients never resected. Univariate analysis identified three significant prognostic variables (number of distant sites involved, metastases to liver only, and volume of hepatic replacement by malignancy) in the resected group. Volume of hepatic replacement was also a significant predictor of survival. Subsequent to resection of asymptomatic primary colorectal carcinoma, 20% developed postoperative complications. Median hospital stay was six days. Two patients (1.6%) died within 30 days of operation. They concluded stage IV patients selected for elective palliative resection of asymptomatic primary colorectal carcinomas had substantial postoperative survival that was significantly better than those never having resection. A review by Blumgart and Fong (555) revealed an operative mortality of < 5% in most series, but complications arose in excess of 20% in most series. Myocardial complications were seen in 1%, pleural effusion requiring thoracostomy in 5% to 10%, pneumonia in 5% to 22%, and pulmonary embolism in 1%. Complications specifically related to liver resection included liver failure (3% to 8%), bile leak and biliary fistula (4%), perihepatic abscess (2% to 10%), and significant hemorrhage (1% to 3%). The most common sites for failure were the liver and lung with the liver involved as a site of recurrence in 45% to 75% of patients having liver resection. In light of this, adjuvant systemic chemotherapy seems to be an attractive option, but to date its role is unproven. Because the liver is the most common site of recurrence and may be the sole site in up to 40% of patients, regional hepatic chemotherapy is theoretically attractive, but studies in this arena have also failed to prove the benefit of that therapy. However, some studies are encouraging. The prognosis of metastatic carcinoma is grave. Kuo et al. (591) collected data from 74 patients with stage IV colorectal carcinoma to identify prognostic factors for predicting selection criteria for operative treatment in patients with metastatic disease. Overall survival time was 16.1 months. Survival in the curative resection group was significantly longer than in the noncurative groups (31.9 months vs. 12.7 months). The operative mortality and morbidity rates were 5.6% and 21%, respectively. The two most common complications were leakage at the site of anastomosis and urinary tract infection. Based on these results, they concluded that patients older than 65 years with metastases at multiple sites, intestinal obstruction, preoperative carcinoembryonic antigen level 500 ng/ mL, lactate dehydrogenase 350 units/L, hemoglobin < 10 mg/dL, or hepatic parenchymal replacement by metastatic disease > 25% have poor prognosis for operative intervention. They noted the more aggressively they performed radical resection and metastasectomy in selected patients the more survival benefits the patients obtained. SIMULTANEOUS COLORECTAL AND HEPATIC RESECTION &

Weber et al. (586) reported that in selected patients, simultaneous resection of the colorectal primary carcinoma and liver metastases does not increase mortality

137

or morbidity rates compared with delayed resection, even if a left colectomy and/or a major hepatic resection are required. De Santibanes et al. (592) reviewed the results of liver resection performed simultaneously with colorectal resection in 71 cases. The median hospital stay was eight days. Morbidity was 21% and included nine pleural effusions, seven wound abscesses, four instances of hepatic failure, three systemic infections, three intraabdominal abscesses, and one colonic anastomotic leakage. Operative mortality was 0%. Recurrence rate was 57.7% and progression of disease was detected in 33.8%. Overall and disease-free survivals at one, three, and five years were 88%, 45%, and 38% and 67%, 17%, and 9%, respectively. Prognostic factors with notable influence on patient outcomes were nodal stage as per TNM classification, number of liver metastases, diameter (smaller or larger than 5 cm), liver resection specimen weight (lighter or heavier than 90 g) and liver resection margin (smaller or larger than 1 cm). Chua et al. (593) retrospectively analyzed 96 consecutive patients with synchronously recognized primary carcinoma and hepatic metastases who underwent concurrent (64 patients) or staged (32 patients) colonic and hepatic resections. No significant differences were observed between concurrent and staged in type of colon resection, or hepatic resection, overall operative duration, blood loss, volume of blood products transfused, perioperative morbidity (53% vs. 41%), disease-free survival from date of hepatectomy (median 13 months vs. 13 months) or overall survival from date of hepatectomy (median 27 months vs. 34 months). There was no operative mortality. Overall duration of hospitalization was significantly shorter for concurrent than for staged resection (mean 11 days vs. 22 days). They concluded, concurrent colectomy and hepatectomy is safe and more efficient than staged resection and should be the procedure of choice for selected patients in medical centers with appropriate capacity and experience. Tocchi et al. (594) reviewed the results of 78 patients who underwent resection of primary colorectal carcinoma and hepatic metastases with curative intent. Adverse predictors of the long-term outcome included the number of metastases ( > 3), preoperative CEA value > 100 ng/mL, resection margin < 10 mm, and portal nodal status. Tanaka et al. (595) reported on 39 consecutive patients with synchronous colorectal carcinoma metastases to the liver who underwent curative simultaneous ‘‘one-stage’’ hepatectomy and resection of the colorectal primary. Only the volume of the resected liver was selected as a risk factor for postoperative complications (350 g mean resected liver volume in patients with postoperative complications vs. 150 g in those without complications). Patient age of 70 years or older and poorly differentiated mucinous adenocarcinoma as the primary lesion predicted decreased overall survival. They concluded, a one-stage procedure appears desirable for synchronous colorectal hepatic metastases except for patients requiring resection of more than one hepatic section, patients aged 70 years or older, and those with poorly differentiated or mucinous adenocarcinomas as primary lesions. Currently there is no consensus as to which factors are important in selecting patients for operation and which factors are important in determining the patient’s prognosis.

138

& PART I: COLORECTAL DISORDERS

For example, Attiyeh and Wichern (596) found no significant difference in the survival rates of patients with a solitary metastasis and in those with multiple lesions, nor was survival influenced by the size of the metastasis. The survival rate was better in patients whose primary colorectal lesion was Dukes’ B compared with those whose lesion was Dukes’ C. Adson (597) listed several determinants for a favorable prognosis, including (1) primary colorectal carcinoma of limited locoregional extent (Dukes’ A or B), (2) presence of fewer than four liver lesions, (3) metastases that appear a long time after the primary lesion was removed, (4) lesions that can be removed with wide margins, and (5) lack of extra-hepatic metastases. Combining their own experience with reports in the literature, Bozzetti et al. (598) found that sites of failure after liver resection were hepatic in 16%, extrahepatic in 15%, and both in 14%. Patterns of recurrence in our patients were hepatic in 31%, hepatic and an extrahepatic site in 15%, and lung in 15% (576). Nagorney (563) reported that the only characteristic associated with prolonged survival was the stage of the primary lesion, with Dukes’ B patient survival being greater than Dukes’ C. In Nagorney’s review, site of origin and degree of differentiation of the primary carcinoma did not correlate with survival rate. Characteristics of metastatic disease that influenced survival included the number of hepatic metastases (one to three are better than four or more), the interval between diagnosis of the primary lesion and hepatic metastases, the resection margin (a margin > 1 cm is better), and the presence of extrahepatic disease. The size and distribution of lesions within the liver had no associated with survival. We were also interested in variables related to survival and reviewed 26 selected patients with liver colorectal metastases who underwent hepatic resection (576). The patient’s age, sex, site of primary lesion, histologic grade, lymph node involvement, location, size, and number of hepatic metastases, type of hepatic resection, and preoperative CEA blood levels were documented. Complete removal with histologically negative resection margins were accomplished in 24 patients. The extent of resection performed was hepatic lobectomy in 12 patients, segmentectomy in eight patients, and wedge resection in four patients. The 5-year survival rate was 30.5%. Patients with metachronous metastases had a better survival rate than those with synchronous lesions (46.6% vs. 13.6%, respectively). None of the other factors studied showed a significant effect on survival. During a median follow-up of 30.9 months, 20 patients developed recurrence of their disease (60% in the liver). There was no perioperative mortality. Morbidity arose in 66% of patients, with a majority of the complications minor. Wanebo et al. (579) reported a significant relationship with survival and the number of metastases (three or fewer vs. four or more), the presence of bilobar vs. unilobar metastases, and the extent of liver resection (wedge and segmental vs. lobectomy and trisegmentsctomy). They believe that resection of bilobar disease or extended resection should generally be avoided, especially in medically compromised patients. Nakamura et al. (572) adopted a very aggressive approach for patients suffering from liver metastases. Of 31 patients, 22 underwent lymph node dissection of the hepatic hilus, in the minds of most surgeons, a current contraindication to hepatic resection.

Six of the 22 patients who underwent lymph node dissection had nodes positive for carcinoma. Ten patients underwent removal of recurrent lesions in the liver, lung, adrenal glands, and brain after initial hepatic resection. Based on an overall 5-year survival rate of 45%, the authors concluded that repeat hepatectomy and dissection of hilar lymph nodes improves prognosis in selected patients with hepatic metastases of colorectal carcinoma. In an analysis of risk factors, Grayowski et al. (574) found that gender, Dukes’ classification, site of primary colorectal carcinoma, histologic differentiation, size of metastatic lesion, and intraoperative transfusion requirement were not statistically significant prognostic factors. In patients 60 years of age or older, an interval of 24 months or less between colorectal and hepatic resection, four or more metastatic lesions, bilobar involvement, positive resection margins, lymph node involvement, and the direct invasion of adjacent organs were significant poor prognostic factors. Hughes, Scheele, and Sugarbaker (567) collated information from a registry of 24 participating institutions. Factors that they found to affect prognosis detrimentally were (1) more than four metastaltic lesions, (2) a short disease-free interval from initial resection to appearance of metastases ( < 1 year), (3) a pathologic margin of < 1 cm on the liver specimen, and (4) the presence of lymph node metastases at the time of initial resection. Using a multivariate regression analysis, Scheele et al. (575) found that survival was dependent on the presence of satellite metastases, grade of the primary carcinoma, time of the diagnosis of metastases (synchronous vs. metachronous), diameter of the largest metastases ( > 5 cm), anatomic vs. nonanatomic approach, year of resection, and mesenteric lymph node involvement. Rougier et al. (577) studied 544 patients with resected hepatic metastases from colorectal carcinoma to determine prognostic factors. Among the 20 variables assessed, eight items were singled out. In decreasing order of relative risk, they included performance status (2 to 4 vs. 0 to 1), alkaline phosphatase level (greater than normal vs. normal), number of involved segments (4 vs. 3), chemotherapy (no vs. yes), extrahepatic metastases (yes vs. no), primary location (right vs. other), prothrombin time ( < 75% vs. > 75%), and resection of the primary carcinoma (no vs. yes). Specific criteria for the selection process are constantly evolving. Adson (597) has offered a thoughtful set of guidelines. Patients whose primary colorectal lesions are well confined, who have one to three evident unilobar hepatic metastases that likely can be removed with wide margins, and who have no evidence of extrahepatic metastases should undergo resection. Patients with extrahepatic metastases, numerous hepatic metastases involving more than one half of the liver, large lesions that encroach on major hepatic veins, or contralateral hilar ducts or veins or lesions sited so as to preclude resection with free margins have an unfavorable prognosis and should not undergo resection. Unfortunately, many patients do not fall neatly into one of these categories, and the surgeon must exercise considerable judgment in making a definitive recommendation. The role of neoadjuvant chemotherapy for patients with multiple (five or more) bilobar hepatic metastases

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

irrespective of initial resectability is still under scrutiny. Tanaka et al. (599) compared the outcome of hepatectomy alone with that of hepatectomy after neoadjuvant chemotherapy for multiple bilobar hepatic metastases from colorectal carcinoma. The outcome of 48 patients treated with neoadjuvant chemotherapy followed by hepatectomy was compared with that of 23 patients treated by hepatectomy alone. Patients who received neoadjuvant chemotherapy had better three and five year survival rates from the time of diagnosis than those who did not (67.0% and 38.9% vs. 51.8% and 20.7%, respectively) and required few extended hepatectomies (four segments or more) (39 of 48 vs. 23 of 23). In patients with bilateral multiple colorectal liver metastases, neoadjuvant chemotherapy before hepatectomy was associated with improved survival and enabled complete resection with fewer extended hepatectomies. Allen et al. (600) compared the treatment and outcome in patients referred for staged resection of synchronous colorectal liver metastases between patients who did not receive neoadjuvant chemotherapy and had exploratory operations after recovery from colon resection and patients who did receive chemotherapy before liver resection. Neoadjuvant chemotherapy was given to 52 patients; in 29 of them the disease did not progress but in 17 the disease progressed while they were receiving treatment. Median follow-up was 30 months. Five-year survival was statistically similar between patients who did and those who did not receive neoadjuvant therapy (43% vs. 35%). Patients within the neoadjuvant group whose disease did not progress while they were receiving chemotherapy experienced significantly improved survival as compared to patients who did not receive chemotherapy (85% vs. 35%). In the setting of synchronous colorectal metastases the response to neoadjuvant chemotherapy may be a prognostic indicator of survival and may assist in the selection of patients for conventional or experimental adjuvant therapies. Fong et al. (581) reported on 1001 consecutive patients undergoing liver resection for metastatic colorectal carcinoma. These resections included 237 trisegmentectomies, 394 lobectomies, and 370 resections encompassing less than a lobe. The operative mortality rate was 2.8%. The 5-year survival rate was 37% and the 10-year survival rate was 22%. Seven factors were found to be significant and independent predictors of poor long-term outcome: positive margin, extrahepatic disease, node-positive primary, disease-free interval from primary to metastases < 12 months, number of hepatic lesions > 1 cm, largest hepatic lesion > 5cm, and carcinoembryonic antigen level > 200 ng/mL. When the last five of these criteria were used in a preoperative scoring system, assigning one point for each criterion, the total score was highly predictive of outcome. The fiveyear actuarial survival for patients with 0 points was 60%, 1 point was 44%, 2 points was 40%, 3 points was 20%, 4 points was 25%, and 5 points was 14%. In fact, no patient with 5 points survived five years. Patients with up to two criteria can have a favorable outcome. Patients with three, four, or five criteria should be considered for experimental adjuvant trials. Iwatsuki et al. (601) examined various clinical and pathologic risk factors in 305 consecutive patients who underwent primary hepatic resection for metastatic

139

colorectal carcinoma. Preliminary multivariate analysis revealed that independently significant negative prognosticators were: (1) positive surgical margins, (2) extrahepatic carcinoma involvement including the lymph nodes, (3) three or more metastatic lesions, (4) bilobar metastases, and (5) time from treatment of the carcinoma to hepatic recurrence of 30 months or less. Because the survival rates of the 62 patients with positive margins or extrahepatic metastases were uniformly very poor, multivariate analysis was repeated in the remaining 243 patients who did not have these lethal risk factors. The reanalysis revealed that independently significantly poor prognosticators were: (1) three or more metastases, (2) metastases size greater than 8 cm, (3) time to hepatic recurrence of 30 months or less, and (4) bilobar metastases. Risk scores (R) for recurrence were divided into five groups: grade 1, no risk factors; grade 2, one risk factor; grade 3, two risk factors; grade 4, three risk factors; and grade 5, four risk factors. Grade 6 consisted of the 62 culled patients with positive margins or extrahepatic metastases. Estimated five-year survival rates of grade 1 to 6 patients were 48.3%, 36.6%, 19.9%, 11.9%, 0%, and 0%, respectively. The proposed risk-score grading predicted the survival differences. Smith et al. (602) found that in patients who are undergoing curative resection of hepatic colorectal metastases, an elevated expression of the biomarkers hTERT and Ki-67 are better predictors of poor long-term survival than is a score based on clinical features. Kato et al. (584) reported on 585 patients who underwent hepatectomy at 18 institutions. The 5-year survival rate for those treated by hepatectomy was significantly higher (32.9%) than for those not undergoing hepatectomy (3.4%). After hepatectomy for hepatic metastases, the most prevalent form of recurrence was in the remnant liver (41.4%), followed by recurrence of pulmonary metastases (19.2%), and other (7.2%). Factors of the primary carcinoma adversely affect prognosis included poorly differentiated adenocarcinoma or mucinous carcinoma, depth of invasion, lymph node metastases of Stage n3 and n4 by the Japanese classification of colorectal carcinoma, number of metastatic lymph nodes of more than four, and Dukes’ stage D. Factors at the time of hepatectomy adversely affecting prognosis after operation for hepatic metastases included residual carcinoma, extrahepatic metastases, hepatic metastases of degree H3 stipulated by the Japanese classification of colorectal carcinoma, number of metastases of four or more, pathology of hepatic metastases of poorly differentiated carcinoma, resection margin of < 10 mm, and carcinoembryonic antigen value higher than normal preoperative and one month postoperative. Indications for hepatectomy in patients with four or more hepatic colorectal metastases remain controversial. Imamura et al. (603) reviewed data from 131 patients who underwent a total of 198 hepatectomies. Patients were grouped according to the number of metastases. The 5-year survival rate of patients with 1 to 3, 4 to 9, and 10 or more metastases were 51%, 46%, and 25%, respectively. They concluded hepatic resection for patients with four to nine metastases clearly is warranted. On the other hand, for patients with 10 or more nodules operation cannot be insured absolutely to be contraindicated in high volume centers at which the operative mortality rate is nearly zero.

140

& PART I: COLORECTAL DISORDERS

In the review by Jaeck (604) the 5-year survival rate for resection of colorectal liver metastases ranged from 20% to 54%. However, the resectability rate of colorectal liver metastases is reported to be less than 20%. This limitation is mainly due to insufficient remnant liver and to extrahepatic disease. Among extrahepatic locations, lymph node metastases are often considered indications of a very poor prognosis and a contraindication to resection. He found that the presence of hepatic pedicle lymph node metastases ranged from 10% to 20%. When located near the hilum and along the hepatic pedicle, they should not be considered an absolute contraindication to resection and extended lymphadenectomy should be performed. However, when they reach the celiac trunk, there is no survival benefit after resection of colorectal liver metastases. Elias et al. (605) reported the long-term outcome and prognostic factors of 75 patients who underwent a complete R0 resection of extrahepatic disease simultaneously with hepatectomy for colorectal liver metastases. Extrahepatic disease localization included peritoneal carcinomatosis (limited), hilar lymph nodes, local recurrences, retroperitoneal nodes, lung, ovary, and abdominal wall. The mortality rate was 2.7% and morbidity was 25%. After a median follow-up of 4.9 years, the overall 3and 5-year survival rates were 45% and 28%, respectively. They concluded extrahepatic disease in colorectal carcinoma patients with liver metastases should no longer be considered as a contraindication to hepatectomy. However, there must be an intended R0 resection and it is inappropriate for patients with multiple extrahepatic disease sites or more than five liver metastases. The optimal operative strategy for the treatment of synchronous resectable colorectal liver metastases has not been defined. Martin et al. (606) reviewed their experience with 240 patients who were treated surgically for primary adenocarcinoma of the large bowel and synchronous hepatic metastasis. One hundred thirty-four patients underwent simultaneous resection of a colorectal primary and hepatic metastases in a single operation (group 1), and 106 patients underwent staged operations (group 2). Simultaneous resections tend to be performed for right colon primaries, smaller, and fewer liver metastases, and less extensive liver resection. Complications were less common in the simultaneous resection group, with 49% sustaining 142 complications compared with 67% sustaining 197 complications for both hospitalizations in the staged resection group. Patients having simultaneous resection required fewer days in hospital (median 10 days vs. 18 days). Perioperative mortality was similar (simultaneous n ¼ 3 staged, n ¼ 3). They believe simultaneous resection should be considered a safe option in selected patients with resectable synchronous colorectal metastases. Nelson and Freels (607) assessed the effect of posthepatic resection, hepatic artery chemotherapy on overall survival. Trials were sought in Medline, the Cochrane Controlled Trial Register, the Cochrane Hepatobiliary Group Trials Register, and through contact of trial authors and reference lists using key words. Overall survival at five years in the hepatic artery group was 45% and 40% in the control group. No significant advantage was found in the meta-analysis for hepatic artery and chemotherapy measuring overall survival. Adverse events related to hepatic artery therapy were common including five therapy-related deaths. They concluded

that this added intervention for the treatment of metastatic colorectal carcinoma cannot be recommended at this time. More recently, Clancy et al. (608) conducted a metaanalysis of prospective clinical trials to determine if adjuvant hepatic arterial infusion confers a survival benefit to treat residual microscopic disease after curative hepatic resection for colorectal carcinoma metastases. Prospective clinical trials comparing hepatic arterial chemotherapy after curative hepatic resection for colorectal carcinoma metastases against a control arm were included. The outcome measure was survival difference at one and two years after operation. Seven studies met the inclusion criteria, and all except one were randomized trials. The survival difference in months was not statistically significant at two years. Based on these findings, they concluded routine adjuvant hepatic artery infusion after curative resection for colorectal carcinoma of the liver cannot be recommended. Bines et al. (609) reported on a review of 131 patients who underwent hepatic resection for metastatic colorectal carcinoma. There were 31 recurrences and, of those, 13 underwent re-resection with a morbidity rate of 23%, a mortality rate of 8%, and a 5-year survival rate of 23%. The authors concluded that in properly selected patients, repeat resection yields results similar to those after initial resection. Wanebo et al. (579) reported that 12% of their patients had repeated resection of metastases, with an overall 5-year survival rate of 43% after the first resection and 22% after the second resection. In their review of 10 reports, Blumgart and Fong (555) noted that between 15% and 40% of patients who undergo resection for hepatic metastases have the liver as the sole site of recurrence, and approximately one third will be candidates for further resection. In the 146 patients collated, the operative mortality rate was 3%, and the complications encountered were similar to those that developed after initial resection. These results were in highly selected patients. The median survival was > 30 months when calculated from the time of second liver resection and > 47 months when calculated from the time of the first resection. However, there were only four 5-year survivors. Although resection is feasible, only approximately 5% of all patients undergoing further resection will come to a second resection (555). Wanebo et al. (610) reported recurrence rates in from 65% to 85% of patients after initial hepatectomy for metastases for colorectal carcinoma. Approximately one half of these have liver metastases and in 20% to 30%, only the liver is involved. The opportunity for resection is frequently limited because of diffuse liver disease or extrahepatic extension, and only approximately 10% to 25% of these patients have conditions amenable to resection. The authors’ comprehensive review of the 28 series showed that the mean interval between the first and second liver resections varied from 9 to 33 months and was approximately 17.5 months in the two largest series. The median survival in the series reporting 10 or more patients was 19 months (mean, 24 months), which is comparable to data in the single resection series. In the large French Association series containing 1626 patients with single resections and 144 patients with two resections, the 5-year survival rates were 25% and 16%, respectively. The recurrence rate after repeat resection was high ( > 60%), and half of the recurrences were in the liver. The prognostic factors favoring repeat resection are variable,

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

FIGURE 58 & Survival curve for 134 patients undergoing second hepatic operations for colorectal metastasis (collected from 15 reports in the literature). Source: From Ref. 611.

but they include absence of an extrahepatic extension of carcinoma and a complete resection of liver metastases. The authors concluded that repeat hepatic liver resection for metastatic colorectal carcinoma in carefully selected patients appears warranted. From their own experience and review of the literature, Pinson et al. (611) came to the same conclusion. For patients collected from the literature, the authors constructed a survival curve (Fig. 58). Fernandez-Trigo, Shamsa, and Sugarbaker (612) were also encouraged to perform repeat hepatic resections for colorectal metastases because it remains the only curative treatment. Others concur with this course of management (613). Takahashi et al. (614) reviewed clinical data of patients undergoing repeat hepatectomy for metastatic colorectal carcinoma compared with those of initial hepatectomy to determine criteria for repeat hepatectomy. For 22 patients who underwent repeat hepatectomy, no mortality and an 18% morbidity rate were observed. The 3-year survival rate after hepatectomy was 49%. The only poor prognostic factor after repeat hepatectomy was a serum carcinoembryonic antigen level greater than 50 ng/mL before initial hepatectomy. Suzuki et al. (615) assessed the risks and clinical benefits of repeat hepatectomy for those patients who underwent hepatectomy for colorectal metastases. There was no operative mortality after repeat hepatectomy in 26 patients. Operative bleeding was significantly increased in the second hepatectomy but operating time, duration of hospital stay, and performance status after the second hepatectomy were comparable with those of the initial hepatectomy. The median survival time from the second hepatectomy was 31 months and the three 3- and 5-year survival rates were 62% and 32%, respectively. A short disease-free interval (six months or less) between the initial hepatectomy and diagnose of hepatic recurrence in the remnant liver was significantly associated with poor survival after the second hepatectomy. Oshawo et al. (616) conducted a study aimed to compare outcome in patients with solitary colorectal liver metastases treated by operation (n ¼ 20) or by radiofrequency ablation. Most patients in both groups also received systemic chemotherapy. Median survival after liver resec-

141

tion was 41 months with a 3-year survival of 55.4%. There was one postoperative death and morbidity was minimal. Median survival after radiofrequency ablation was 37 months with a 3-year survival of 52.6%. In this study, survival after resection radiofrequency ablation of solitary colorectal liver metastases was comparable. The latter is less invasive and requires either an overnight stay or day-case facilities only. Berber et al. (617) determined the predictors of survival at the time of laparoscopic radiofrequency thermal ablation in 135 patients with colorectal liver metastases who were not candidates for resection. The median survival for all patients was 28.9 months. Patients with a CEA less than 200 ng/mL had improved survival compared with those of a CEA more than 200 ng/mL (34 months vs. 16 months). Patients with the dominant lesion less than 3 cm in diameter had a median survival of 38 months versus 34 months for lesions 3 to 5 cm, and 21 months for lesions greater than 5 cm. Survival approached significance for patients with one to three metastases versus more than three metastases (29 months vs. 22 months). The presence of extrahepatic disease did not affect survival. Only the largest liver metastases more than 5 cm was found to be a significant predictor of mortality with a 2.5-fold increased risk of death versus the largest liver metastases less than 3 cm in size. Ueno et al. (618) collected data from 68 patients who underwent resection of colorectal liver metastases who might benefit from prophylactic regional chemotherapy. The extrahepatic recurrence rate at three years after hepatectomy was 57.8%. Three variables were independently associated with extrahepatic recurrence including raised serum level of carcinoembryonic antigen after hepatectomy (relative risk 5.4), venous invasion of the primary carcinoma (relative risk 4.0) and high-grade budding of the primary carcinoma (relative risk 3.1). Patients with none of these risk factors had a 3-year extrahepatic recurrence rate of 7.1% compared with 61.6% for those with one risk factor and 100% for those with two or three risk factors. This system might be used on an individual basis to select patients with colorectal liver metastases for regional chemotherapy or systemic chemotherapy after operative intervention. The value of postoperative chemotherapy following resection of hepatic metastases was reviewed by Cohen and Kemeny (619). Two studies compared hepatic artery infusion with no treatment and no overall survival benefit was reported. In one study there was a modest two-year survival improvement from 72% to 86%. Drugs used included FUDR, 5-FU/ leucovorin and in light of newer drugs used for systemic chemotherapy, this improvement may not be relevant. For patients with unresectable liver metastases, Gray et al. (620) reported on the use of embolization of yttrium-90-containing microspheres into the liver via a catheter inserted into the hepatic artery at laparotomy. In 29 patients, the CEA levels fell in the 26 patients in whom this therapy was tested, and there was CT evidence of reduction in 48% of the 22 patients re-examined. Some patients also received continuous chemotherapy infusion to potentiate the radiation effect. Although this is an important first step, there is no evidence that reduction of disease is translated to either improved survival or quality of life. Stubbs et al. (621) treated 50 patients with advanced nonresectable colorectal liver metastases with selective internal

142

& PART I: COLORECTAL DISORDERS

radiation therapy. Estimated liver involvement was less than 25% in 30 patients, 25% to 50% in 13 patients, and greater than 50% in 7 patients. A single dose of between 2.0 and 3.0 GBq of 90 yttrium microspheres was injected into the hepatic artery via a subcutaneous port and followed at 4-week intervals by regional chemotheraphy with 5-FU. Treatment-related morbidity did occur including a 12% incidence of duodenal ulceration. Median survival for patients with extrahepatic disease was 6.9 months. For patients with no extrahepatic disease median survival was 17.5 months. Substantial destruction of liver metastases can be achieved in more than 90% of patients with a single treatment. Lang and Brown (622) recommended the selective embolization of doxorubicin and ethiodized oil for unresectable hepatic metastases. In his review, Stuart (623) found that chemoembolization for patients with metastatic colorectal carcinoma appears to be a reasonable alternative for many who are not operative candidates. Response rates of approximately 50% have been reported, with survival longer than would be expected in studies of systemic therapy among patients who had failed standard chemotherapy. Survival may be especially enhanced in treated patients who have no extrahepatic metastases. Cryoablation has also been used in this clinical setting but has not yet been proven to improve outcome (555). For patients with unresectable hepatic metastases, Weaver, Atkinson, and Zemel (624) reported the use of hepatic cryosurgery with or without resection in 140 patients, 119 of whom had carcinomas that were colorectal in origin. The median number of lesions treated was three. The operative mortality rate was 4%, and complications included coagulopathy, hypothermia, myoglobinuria, pleural effusion, acute tubular necrosis and infection. The median survival rate was 27 months. Ruers et al. (625) reported on the long-term efficacy of cryosurgery as an adjunct to hepatic resection in patients with colorectal liver metastases not amenable to resection alone. Thirty patients met the following inclusion criteria: metastases confined to the liver and judged irresectable, 10 or fewer metastases, cryosurgery alone or in combination with hepatic resection allowed disease clearance. Median follow-up was 26 months. Overall, 1- and 2-year survival rates were 76% and 71%, respectively. Median survival was 32 months. Diseasefree survival at one year was 35%, two years 7%. Six patients developed recurrence at the site of cryosurgery; given that the total number of cryosurgery-treated lesions was 69 the local recurrence rate was 9%. Lung Pulmonary metastases occur in approximately 10% of all patients with adenocarcinoma of the colon and rectum, and the majority of these are only of one facet of a generalized spread of disease. About 10% of these patients, that is, l% of the total, will develop solitary pulmonary metastases. The criteria for determining resectability of pulmonary metastases are similar to those applied to resection of hepatic metastases:

1. Ideally the pulmonary metastasis should be solitary. If more than one is involved, the lesions should be confined to one lung; if bilateral, the lesion in each lung must be solitary.

2. The primary colorectal carcinoma should be controlled locally. 3. There should be no other evidence of metastases. 4. The patient’s medical condition should allow for thoracotomy and pulmonary resection. A number of prognostic discriminants, including disease-free interval, number of metastases, grade, stage, and location of the primary carcinoma, age and sex of the patient, location of the pulmonary metastases, and type of pulmonary resection, have been examined. None is uniformly reliable in the management of a particular patient, and there is no agreement on the importance of each individual factor. The predictive value of the route of venous drainage on prognosis was investigated in a consecutive series of 44 patients who underwent curative resection of pulmonary metastases from colorectal carcinoma (626). The primary lesion was located in the colon in 14 patients and in the upper third of the rectum in 11 patients, thus indicating blood drainage directed toward the portal vein (group I). In 10 and nine cases, respectively, the initial growth was in the middle and lower third of the rectum with the venous outflow at least partially directed into the vena cava (group II). There was no obvious difference in the two groups regarding the initial site of carcinoma recurrence. The liver was involved in four of 15 patients failing in group I as opposed to four of 13 patients with hematogenous relapse in group II. Median survival and disease-free survival times were significantly longer in patients in group I (58.4 and 50.2 months, respectively) than in patients in group II (30.9 and 16.8 months, respectively), and, even more pronounced, in colon carcinoma patients (75.4 and 60.2 months, respectively) when compared with rectal carcinoma patients (31.0 and 17.9 months, respectively). In contrast, survival curves did not differ significantly when the two groups with different routes of drainage (5-year survival rate, 53% vs. 38%; 5-year disease-free survival rate, 43% vs. 37%), or carcinomas of the colon and rectum (5-year survival rate, 67% vs. 38%; 5-year disease-free survival rate, 60% vs. 32%) were compared using the log-rank test. The primary carcinoma site does therefore not become a major criterion in selecting patients for surgical resection. Saclarides et al. (627) reported on 23 patients who underwent 35 thoracotomies for metastatic colorectal carcinoma. The pulmonary disease was diagnosed within an interval of 0 to 105 months (average, 33.4 months) after colon resection. Fifteen patients underwent a single thoracotomy (12 patients had solitary lesions and three patients had multiple nodules). Eight patients underwent multiple thoracotomies. The median survival following thoracotomy was 28 months, the 3-year survival rate was 45%, and the 5-year survival rate was 16%. Factors that had no significant bearing on survival included the origin and stage of the primary carcinoma and the patient’s age and sex. An interval before thoracotomy of 3 years had an impact on survival. Patients who underwent multiple thoracotomies had a significantly prolonged survival. Patients who underwent a single thoracotomy for a solitary lesion had a significantly prolonged survival compared with patients who underwent a single thoracotomy for multiple metastases. After thoracotomy, 14 patients eventually developed

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

recurrent disease, which was confined to the lung in only four patients. Of these 14 patients, 11 subsequently died of carcinoma. The authors concluded that thoracotomy for metastatic disease should be considered when the primary carcinoma is controlled, the lungs are the only site of metastatic disease, and there is adequate lung reserve to withstand surgery. In a review of 12 series of patients, including their own, Brister et al. (628) summarized the results of 335 patients who underwent pulmonary resection. The authors found overall 2-year and 5-year survival rates of 70% and 30%, respectively. In their series the only factor significant in determining survival was a long disease-free interval. It seems logical that a longer disease-free interval reflects a slower growing malignancy and would in turn be associated with a longer postthoracotomy survival. In a series of 76 pulmonary resections, Wade et al. (578) found a projected 5-year survival rate of 36%, a mean survival rate of 38 months, and a 3% operative mortality rate. Kanemitsu et al. (629) studied factors that might be helpful in predicting survival in 313 patients (the largest number) with pulmonary metastases from colorectal carcinoma who were candidates for thoracotomy. Pulmonary resections included lobectomy 137, partial resection 132, segmentectomy 38, and pneumonectomy 6. Overall survival rates were 90.4% in one year, 53% at three years, and 38.3% at five years. The 1-, 3- and 5-year survival rates of patients with pulmonary metastases from colorectal carcinoma who did not undergo thoracotomy were 58.6%, 8.5%, and 1.9%, respectively. They identified five variables as independent predictors of 3-year survival: prethoracotomy carcinoembryonic antigen level, number of pulmonary lesions, presence of hilar or mediastinal infiltrated lymph nodes, histology of the primary carcinoma and presence of extrathoracic disease. Their model has moderate predictive ability to discriminate between patients who are likely to survive after thoracotomy for pulmonary metastases from colorectal carcinoma. Watanabe et al. (630) also reviewed 49 patients to identify prognostic factors for overall survival and risk factors for further intrapulmonary recurrence after resection of pulmonary metastases from colorectal carcinoma. Survival after resection of pulmonary metastases was 78% at three years and 56% at five years. Solitary pulmonary metastases were significantly correlated with survival. The pathological features of the primary colorectal carcinoma had no impact on survival. Histologically incomplete resection of pulmonary metastasis significantly correlated with pulmonary re-recurrence. Negri et al. (631) reported on the development of a preoperative chemotherapy strategy for patients selected to undergo pulmonary metastasectomy from colorectal carcinoma in 31 patients. The median age at operation was 61 years. Twenty (65%) proceeded directly to operation and five of these patients received postoperative chemotherapy. Eleven (35%) received preoperative chemotherapy which consisted of fluoropyrimidine in combination with either Oxaliplatin, or Mitomycin C, except for one patient who received single agent CPT-11; 82% had a partial response and 18% had stable disease. In total, 39 thoracic operations (six bilateral and one incomplete) were undertaken. There were no postoperative deaths. Twenty percent who had initial operation had postoperative complications

143

compared to 18% of the preoperative chemotherapy group. Overall 3- and 5-year survival rates after the first thoracic operation were 65.2% and 26.1%, respectively. Disease-free interval, number of pulmonary metastases, previous resection of hepatic metastases, prethoracotomy carcinoembryonic antigen (CEA), and preoperative chemotherapy were not found to be significant prognostic factors for survival. They concluded resection of lung metastases has a low morbidity and mortality and results in long-term survival for 20% to 30% of patients. Furthermore, in this clinical setting preoperative chemotherapy produced a high response rate with no patients experiencing disease progression prior to operation. King et al. (632) assessed the safety and efficacy of imaging-guided percutaneous radiofrequency ablation for local control of lung metastases from colorectal carcinoma. Forty-four metastatic lesions in 19 patients were treated successfully at 25 treatment sessions. Five of 19 patients were retreated for new lesions. There were 13 pneumothoraces following the 25 treatments, and six patients required drainage. Six months after treatment CT demonstrated that three lesions had progressed, 25 metastases were stable or smaller, and 11 were no longer visible. At 12 months five metastases had progressed, 11 were smaller or stable and 9 were not visible. Retreatment for recurrence or new metastases is feasible and occurred in five patients in this series. Some patients also received concomitant chemotherapy. Improving long-term survival is not the only goal of treatment. Relief of symptoms such as cough, hemoptysis, and pain is also beneficial. Furthermore, survival benefit has been shown for repeat pulmonary metastasectomy. Ike et al. (633) evaluated results of their strategy for intensive follow-up after resection of colorectal carcinoma and aggressive resection of lung metastases. The followup program for lung metastases includes a serum CEA assay every two months and chest x-ray every six months. Operative resection of lung metastases was performed if the primary and any nonpulmonary metastases had been controlled, lung metastases numbered four or fewer, and pulmonary functional reserve was adequate. Standard operation for lung metastases was lobectomy and lymph node dissection was added in cases where the metastases were over 3 cm in size. Forty-two patients underwent 50 lung resections for metastatic colorectal carcinoma. Overall five-year survival rate after resection of lung metastases from colorectal carcinoma was 63.7%. Patients with well differentiated primary carcinoma, a solitary metastatic nodule, and disease-free interval of at least two years after initial operation are likely to be long-term survivors. Ishikawa et al. (634) reviewed retrospectively the clinical course of 37 patients who underwent operative resection of primary colorectal carcinoma and metastatic lung disease. Multivariate analysis indicated that the existence of an extranodal malignant deposit in the primary lesion (hazard ratio ¼ 4.55) and three or more lung metastases (hazard ratio ¼ 2.9) were significant indicators for poor prognosis. They divided the patients into two groups; Goup A (n ¼ 12) had neither of these two parameters, and Group B (n ¼ 25) comprised all other patients. Survival rate at three and five years were 90.9% and 90.9% in Group A and 16.1% and 8.1% in Group B, respectively; and disease-free survival after

144

& PART I: COLORECTAL DISORDERS

thoracotomy (3-year and 5-year disease-free survival rate, 52.9% and 39.7% in Group A and 5.3% and 5.3% in Group B, respectively). They identified an extra nodal malignant deposit at the primary carcinoma site as a new significant prognostic factor after resection of pulmonary metastases from colorectal carcinoma. Resection of pulmonary metastases is expected to be very useful for patients without extra nodal deposits and fewer than three pulmonary metastases. Zink et al. (635) reviewed medical records of 110 patients operated on pulmonary metastases of colorectal origin. The median time interval between diagnosis of the primary carcinoma and thoracotomy was 35 months. After resection of the pulmonary metastases, the 3- and 5-year post-thoracotomy survival measured 57% and 32.6%, respectively. The overall survival was significantly correlated with the disease-free interval and the number of intrapulmonary metastases. Treatment, stage, and grade of the primary carcinoma, occurrence of liver metastases and local recurrences, mode of treatment of metastases, and postoperative residual stage had no significant correlation with either total or post-thoracotomy survival. Irshad et al. (636) reviewed 49 patients treated operatively for pulmonary metastases from colorectal carcinoma. The perioperative death rate was 4%. Overall 5- and 10-year survival rates were 55% and 40%, respectively. The mean interval between the initial colonic resection and resection of pulmonary metastases was 36 months. Variables that carried a poor prognosis included more than one pulmonary lesion, a disease-free interval less than two years, and moderately or poorly differentiated colorectal carcinoma. The 16 patients who received chemotherapy after their thoracotomy had a 5-year survival rate of 51% compared with 54% for the 33 patients who did not receive chemotherapy, demonstrating postoperative chemotherapy has no survival benefit. Vogelsang et al. (637) evaluated clinically relevant prognostic factors to define a subgroup of patients who would most benefit from resection of lung metastases from colorectal carcinoma. There were 75 patients with pulmonary metastases from colorectal carcinoma who underwent 104 R0 lung resections. Patients who had no evidence of recurrent extrathoracic disease, no more than three metastases on either side, lobectomy as the maximum operative procedure, and adequate cardiorespiratory function were eligible for operation. Overall median survival was 33 months with 3- and 5-year survival rates of 47% and 27%, respectively. Prognostic groups included patients with a maximum metastasis size of 3.75 cm or less with a diseasefree interval of more than 10 months and patients with larger metastases and a shorter disease-free interval. Median survival and 5-year survival were 45 months and 39% in the former group and 24 months and < 11% in the latter. When a solitary pulmonary shadow occurs synchronously with a large bowel carcinoma, the dual problems of the nature of the pulmonary lesion and the priority of management arise. Based on the principles of management of metastatic lesions, this clinical situation should not be a dilemma. The large bowel malignancy should be handled without regard for the pulmonary lesion. If the surgeon achieves a curative resection, thus fulfilling the criterion of control of local disease, and there is no other evidence

of metastatic disease, attention then can be directed toward the pulmonary lesion, usually 2 or 3 months later. The pulmonary lesion may, in fact, prove to be a primary lung carcinoma. Assessment of the contribution of surgery to the treatment of pulmonary metastases from carcinoma of the colorectum demonstrated that it is a valid treatment option with survival benefit. An aggressive operative policy directed toward metastatic carcinoma confined to the lung has resulted in a rewarding rate of disease-free survivals and appreciable palliative benefit for appropriately selected patients. Even multiple metastases may be successfully managed as long as the cardinal principles of patient selection are observed. Pulmonary resection is attended with little risk (operative mortality rates ranging from 0% to 4% and major complication rates from 0% to 12%), the survival results justify an aggressive operative approach (627). Accordingly, patient follow-up should include regular chest x-ray examinations to detect subsequent metastases amenable to operative therapy. Liver and Lung Although simple lung or liver metastasectomy from colorectal carcinoma has proved effective in selected patients, the subject of simultaneous bi-organ metastasectomies is seldom addressed. Mineo et al. (638) reported on 29 patients who presented simultaneous (n ¼ 12) or sequential liver before lung (n ¼ 10) and lung before liver (n ¼ 7) metastases. All metastases were successfully resected in a total of 56 separate procedures. In 35 thoracic procedures, 45 metastases were removed by wedge resection (n ¼ 36) or lobectomy (n ¼ 9). In addition, 47 liver metastases were resected with wedge (n ¼ 24), segmentectomy (n ¼ 13), or lobectomy (n ¼ 10). There were no perioperative deaths and the morbidity rate was 10.7%. All patients were followed for a minimum of three years. Median survival from the second metastasectomy was 41 months, with a 5-year survival rate of 51.3%. Risk-factor distribution among the three metastatic pattern groups was insignificant. Premetastasectomy elevated levels of both CEA and CA19–9, and mediastinal or celiac lymph node status were significantly associated with survival, although number of metastasectomies, disease-free interval, and simultaneous versus sequential diagnosis were not. In the multivariate analysis, only elevated CEA plus CA19–9 was significantly associated with survival. They concluded that either simultaneous or sequential lung and liver metastasectomy can be successfully treated by operation. Ike et al. (639) retrospectively analyzed 48 patients who underwent pulmonary resection for metastatic colorectal carcinoma, 27 of whom had lung metastases alone and 15 had previous partial hepatectomy, and 6 had previous resection of local or lymph node recurrence. Five-year survival rates after resection of lung metastases were 73% in patients without preceding recurrence, 50% following previous partial hepatectomy, and 0% after resection of previous local recurrence. There was no significant difference in survival after lung resection between patients who had sequential liver and lung resection versus those who had lung resection alone. Nagakura et al. (640) analyzed retrospectively a total of 136 patients who underwent resection of hepatic or pulmonary metastases of colorectal origin. Eighty-four

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

patients underwent hepatectomy alone, 25 underwent pulmonary resection alone, and 27 underwent both hepatic and pulmonary resection. The 27 patients undergoing hepatic and pulmonary resection were divided into two groups; 17 patients with sequentially detected hepatic and pulmonary metastases and 10 patients with simultaneous detected metastases. Patient survival after hepatic and pulmonary resection was comparable with that after hepatic alone and that after pulmonary resection alone. Among the 27 patients undergoing hepatic and pulmonary resection, the outcomes after resection were significantly better in patients with sequentially detected metastases (cumulative 5-year survival of 44%) than in those with simultaneously detected ones (cumulative 5-year survival of 0%). They concluded patients with sequentially detected hepatic and pulmonary metastases from a colorectal primary are good candidates for aggressive metastasectomy but simultaneous detection of these metastases does not warrant resection. Ovary The mechanism of spread of large bowel carcinoma to ovary is not clear. Postulated methods include implantation from intraperitoneal spread, hematogenous spread, and lymphatic dissemination. Immunostaining for cytokeratin 7 (CK7) (positive in ovarian carcinoma) and cytokeratin 20 (positive in colorectal carcinoma) is a useful technique for making the distinction between the organs of origin (231). The development of ovarian metastases was discussed in the section on treatment. Colon carcinoma may present as metastatic disease to the ovaries in a Krukenberg-like pattern (641). Treatment consists of bilateral salpingooophorectomy (Fig. 59). In a study of a series of patients who presented with what appeared to be primary ovarian neoplasms but actually were ovarian metastases from a colonic origin, Herrera-Ornelas et al. (642) found that survival was similar to that of patients who were primarily diagnosed as having large bowel carcinoma and subsequently developed ovarian metastases. Average life expectancy after diagnosis was 16.5 months. In their review of 63 patients with metachronous ovarian metastases, Morrow and Enker (643) found that such disease was part of diffuse intraabdominal disease in 55% of patients. The mean survival rate following operation was 16.6 months. Ability to remove all

FIGURE 59 & Example of bilateral ovarian metastases in a patient presenting with acute colonic obstruction.

145

gross disease at the time of oophorectomy was the major determinant of survival. Patients rendered disease free had a mean survival of 48 months compared with 9.6 months in patients with unresectable disease. Morrow and Enker believe that bilateral oophorectomy is warranted as part of the palliative treatment of women with metastatic disease to prevent the development of large symptomatic metastases that require further therapy. Huang et al. (644) reviewed the impact of elective and therapeutic oophorectomy on the natural history of colorectal carcinoma. A total of 155 patients were studied. Synchronous ovarian metastases occurred in 90 patients (58.1%); metachronous ovarian metastases occurred in 41.9%. Estimated 5-year survival for patients with synchronous ovarian metastases was 9% versus 20% for metachronous ovarian metastases. Resection of metastatic disease was associated with an improved 5-year survival for synchronous ovarian metastases (15% vs. 0%) and metachronous ovarian metastases (24% vs. 0%) if patients were disease-free postoperatively. Other clinical characteristics including age, menopausal status, stage, location of primary carcinoma, had no significant impact on survival. Thus, ovarian metastases from colorectal carcinoma are associated with a poor outcome. Although there is no survival advantage associated with resection of occult microscopic disease, long-term survival is possible if patients are rendered surgically disease free. Bone Metastases to the bone are usually associated with widely disseminated disease. Besbeas and Stearns (645). reported osseous involvement in 6.9% of patients, 5.1% as part of widespread metastases and 1.8% with skeletal metastases only. Sites of metastatic disease included the skull, scapula, clavicle, ribs, vertebrae, pelvic bones, humerus, and femur. The interval from initial diagnosis to manifestation of osseous metastases ranged from 10 months to 6 years and 11 months in that report. Bonnheim et al. (646) reported a 4% incidence of osseous metastases. Scuderi et al. (647) reported a case of sternal metastases as the initial presentation of an unknown rectal carcinoma. They stated osseous metastases occur in 3.8% to 10.5% of cases of rectal carcinoma. Isolated bony metastases are very rare and usually represent a late manifestation, being part of diffuse metastatic disease. Operation is an option in cases of solitary sternal localization but must be reserved for patients in good general condition. Treatment is directed toward pain control, and this is often achieved through radiotherapy. The mean period from onset of osseous metastases to death was 10 to 13.2 months (645,646). Brain Wong and Berkenblit (648) recently reviewed therapeutic options and expected outcomes for patients with brain metastases. They reported that the median survival with no treatment was 1 to 2 months; with steroids 2 to 3 months; with whole-brain radiotherapy 3 to 6 months; with operation and whole-brain radiotherapy 10 to 16 months; with radiosurgery and whole-brain radiotherapy 6 to 15 months; and with chemotherapy 8 to 12 months. Because the majority of cytotoxic agents seem to be unable to penetrate the blood–brain barrier, the role of chemotherapy in the

146

& PART I: COLORECTAL DISORDERS

treatment of brain metastases remains controversial. A few of the newly developed cytotoxic agents can cross the blood-brain barrier and may have a role in the treatment of patients with brain metastases. They noted that recent studies have demonstrated the antineoplastic activity of topotecan against brain metastases, with objective response rates ranging from 33% to 63% in patients with various solid malignancies mostly of lung origin. This result may be explained by the lack of exposure of brain metastases to previous cytotoxic agents, suggesting a role for topotecan in patients with brain metastases. Early studies have also suggested that topotecan, an apparent radiosensitizer, may be particularly effective in combination with radiotherapy, the current standard of care for patients with brain metastases. Alden, Gianino, and Saclarides (649) reviewed their experience with brain metastases from colorectal carcinoma. The authors identified 19 of their own patients and collected information from other reports. Fifty-eight percent of the patients had disseminated disease at initial diagnosis. The mean interval between treatment of the primary lesion and the diagnosis of brain metastases was 32.1 months. The brain was the sole site of metastatic disease in 21% of patients. Lesions were solitary in 63%, exclusively cerebral in 53%, cerebellar in 32%, or both in 15%. Presenting complaints ranged from ataxia (63%), headaches (21%), dizziness (26%), and weakness (32%) to seizures (16%), dysphasia (21%), and mental status changes (21%). Diagnosis is established by CT scanning (Fig. 60) or MRI. Treatment consists of steroids to decrease intracranial swelling and radiation or craniotomy in special circumstances. Survival is dismal with no 1-year survivors in the series of 19 patients reviewed by Alden, Gianino, and Saclarides (649). The median survival rate following craniotomy was 4.9 months and following radiation was 2.6 months. Survival was not affected by the number or location of metastatic lesions or whether the brain was the sole site of metastatic disease. Because of the dismal survival rate, the authors believe that craniotomy is rarely indicated, except in the rare patient who has minimal

FIGURE 60 & CT scan demonstrating metastasis to brain. The hyperdense thick-walled lesion is surrounded by a ‘‘halo’’—-vasogenic (white matter) edema.

neurologic impairment, a long disease-free interval, a solitary metastasis, and no extracranial disease. The authors believe that for most patients, radiation is the treatment of choice. Hammond et al. (650) reported on 100 patients with brain metastases secondary to colorectal carcinoma. Of these patients, 36 underwent operation, 57 underwent radiotherapy alone, and the remaining seven received steroids. The median interval between the diagnosis of the primary carcinoma and the diagnosis of brain metastasis was 26 months. The median survival time was 1 month for patients who received only steroids, 3 months for those who received radiotherapy, and 9 months for those who underwent operation. The early onset of brain metastases was associated with a poor prognosis. Farnell et al. (651) reported that brain metastases occur in 25% to 35% of all patients with malignancies, with colorectal carcinoma accounting for approximately 8% of these. Of 150 patients with brain metastases, 82% had concomitant extracerebral metastases, especially in the lungs. Only 16% of patients survived > 1 year. The median survival rates for all patients receiving operation and radiotherapy (39), operation alone (11), radiotherapy alone (79), and supportive care (17) were 42, 45, 16, and 8 weeks, respectively. Of the patients treated with radiotherapy, 30% showed regression and three had complete regression. Given the similar results in patients treated with operation plus radiotherapy and those treated with operation alone, the authors believe that consideration should be given to withholding radiotherapy to obviate its side effects. Peritoneal Carcinomatosis Intraperitoneal carcinomatosis accounts for 25% to 35% of recurrences of colorectal carcinoma (652). Studies demonstrate that peritoneal carcinomatosis is not necessarily a terminal condition with no options for treatment for cure. The combination of aggressive, cytoreductive surgery and intra-abdominal hyperthermia chemotherapy improves long-term overall survival in selected patients but is a time-consuming procedure (approximately 12 hours) and entails high mortality (5%) and morbidity (35%) (652). Most commonly used drugs are Mitomycin C and platinum compounds, which have synergistic toxic effects on malignant cells when hyperthermia is applied. The three principal studies dedicated to the natural history of peritoneal carcinomatosis from colorectal carcinoma consistently showed median survival ranging between six and eight months. Glehen et al. (653) conducted a retrospective multicenter study to evaluate the international experience with cytoreductive surgery and perioperative intraperitoneal chemotherapy and to identify the principal prognostic indicators. The study included 506 patients from 28 institutions. The median age was 51 years. The median follow-up was 53 months. The morbidity and mortality rates were 22.9% and 4%, respectively. The overall median survival was 19.2 months. Patients in whom cytoreductive surgery was complete had a median survival of 32.4 months, compared with 8.4 months for patients in whom complete cytoreductive surgery was not possible. Positive independent prognostic indicators by multivariate analysis were complete cytoreduction, treatment by a second procedure, limited extent of peritoneal carcinomatosis, age less than 65 years, and use of adjuvant chemotherapy. The use of neoadjuvant

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

chemotherapy, lymph node involvement, presence of liver metastases, and poor histologic differentiation were negative independent prognostic indicators. Culliford et al. (654) reported aggressive treatment of peritoneal metastases from colon carcinoma by surgical cytoreduction and infusional intraperitoneal chemotherapy may benefit selected patients. There were 64 patients having surgical debulking and intraperitoneal (FUDR) plus leucovorin for peritoneal metastases. Primary carcinoma sites were 47 in the colon and 17 in the appendix. Peritoneal metastases were synchronous in 48 patients and metachronous in 16 patients. Patients received intraperitoneal FUDR (1000 mg/m2 daily for three days) and intraperitoneal leucovorin (240 mg/m2) with a median cycle number of 4. The median number of complications was 1 with no treatment related mortality. Only 9% required termination of intraperitoneal chemotherapy because of complications. The median follow-up was 17 months. The median survival was 34 months; 5-year survival was 28%. The 5-year survival was 54% for complete and 16% for incomplete resection. Occasionally a few discrete nodules on the peritoneum are present at the time of colonic resection and it would seem appropriate to excise these. Verwaal et al. (655) conducted a randomized trial of cytoreduction and hyperthermia intraperitoneal chemotherapy versus systemic chemotherapy and palliative surgery in patients with peritoneal carcinomatosis of colorectal carcinoma. Of the 105 patients randomly assigned with a median follow-up of 21.6 months, the median 5-year survival was 12.6 months in the standard therapy arm and 22.3 months in the experimental therapy arm. The treatment-related mortality in the aggressive therapy group was 8%. In their review of the literature of 11 other reports on similar therapy, median reported survival ranged from 6 to 39 months mostly in the 15 month rage but the best series reported a 30% 5-year survival. Verwaal et al. (656) reported updated data on 117 patients treated by cytoreduction and hyperthermic intraperitoneal chemotherapy. The median survival was 21.8 months. The 1-, 3-, and 5-year survival rates were 75%, 28%, and 19%, respectively. In 59 patients a complete cytoreduction was achieved, and in 41 patients there was minimal residual disease. The median survival of these patient groups was 42.9 and 17.4 months, respectively. When gross macroscopic disease was left behind, as was the case in 17 patients, the median survival was five months. Involvement of the small bowel before cytoreduction was associated with poor outcome. In patients with widespread peritoneal deposits, such an approach seems excessively aggressive. Improved outcomes for selected patients with peritoneal spread has been reported. Shen et al. (657) reviewed their experience of cytoreductive surgery and intraperitoneal hyperthermic chemotherapy with mitomycin C in 77 patients. Peritoneal carcinomatosis was synchronous and metachronous in 27% and 73% patients, respectively. Seventy-five percent of patients had received chemotherapy prior to intraperitoneal hyperthermic chemotherapy. Complete resection of all gross disease was accomplished in 48% of patients. Overall survival at one, three, and five years was 56%, 25%, and 17%, respectively. With a median follow-up of 15 months, the median overall survival was 16 months. Perioperative morbidity and mortality were 30% and 12%, respectively.

147

Hematologic toxicity occurred in 19%. Poor performance status, bowel obstruction, malignant ascites, and incomplete resection of gross disease, were independent predictors of decreased survival. Patients with complete resection of all gross disease had a 5-year overall survival of 34% with a median overall survival of 28 months. Elias et al. (658) conducted a two-center prospective randomized trial comparing postoperative peritoneal chemotherapy plus systemic chemotherapy alone, both after complete cytoreduction surgery of colorectal peritoneal carcinomatosis. Analysis of 35 patients showed that complete resection of peritoneal carcinomatosis resulted in a two-year survival of 60%, far above the classic 10% survival rate among patients with colorectal peritoneal carcinomatosis treated with systemic chemotherapy and symptomatic surgery. In this small series, postoperative intraperitoneal chemotherapy did not demonstrate any advantage for survival. Other Metastatic Disease Metastatic carcinoma that involves the spleen is usually a manifestation of widely disseminated disease, but solitary splenic metastases have been reported (659). Cutaneous metastasis of rectal adenocarcinoma is a rare event occurring in fewer than 4% of all patients with rectal carcinoma (660,661). When present, it typically signifies disseminated disease with a poor prognosis (662). Metastatic colonic carcinoma may present in an old operative scar. Metastases to the glans penis (663), pancreas (664), and vagina (not contiguous disease) (665) have been reported. Metastatic colon carcinoma has even presented as a testicular hydrocele (666) or to the testis (667).

Carcinoma in Young Patients It has been reported that carcinoma of the colon occurring under the age of 40 carries with it a poor prognosis. It has been suggested that this is due to the fact that patients present at a later stage in their development because the diagnosis had not been suspected. (See page 518.) In a depressing report by Radhakrishan and Bruce (668), eight children with primary carcinoma of the colon presented with the common symptom being right iliac fossa pain. All children had poorly differentiated, highly aggressive lesions. In spite of operation and adjuvant therapy, all the children died within one year of presentation.

& POSTOPERATIVE COMPLICATIONS The complications that may be encountered following colonic surgery are discussed in detail in chapter 36. Nevertheless, one very detailed and extraordinarily carefully studied series of cases should be cited. Killingback et al. (669) reviewed 1418 elective resections with anastomoses by a single colorectal surgeon. Postoperative mortality was 1.6%. Significant adverse events which were potentially avoidable occurred in 45.5% of the patients who died. The morbidity rate was (41.6%). Clinical anastomotic leaks occurred more frequently in extraperitoneal anastomoses (4.7%) than in intraperitoneal anastomoses (0.2%). Anastomotic leak caused the death of two

148

& PART I: COLORECTAL DISORDERS

TABLE 10 & Results of Curative Operation for Colon Carcinoma 5-Year Survival (%) Author(s)

No. of Patients

Corman, Veidenheimer, and Coller (7) (1979) Pihl et al. (671) (1980) Stefanini, Castrini, and Pappalardo (672) (1981) Zhou, Yu, and Shen (673) (1983) Umpleby et al. (674) (1984) Isbister and Fraser (675) (1985) Wied et al. (676) (1985) Glass et al. (520) (1986) Davis, Evans, and Cohen (677) (1987) Moreaux and Catala (678) (1987) Brown, Walsh, and Sykes (679) (1988) Enblad et al. (680) (1988) Jatzko, Lisborg, and Wette (681) (1992) Clemmensen and Sprechler (682) (1994) Singh et al. (199) (1995) Carraro et al. (683) (2001) Read et al. (684) (2002) Morin et al. (685) (2006)

1,008 434 436 302 439 1,505 442 413 405 646 550 38,166 223 212 304 256 316 310

Resectability Rates (%)

Operative Mortality (%)

95

4 7 3 2 13

81 71

Crude

76 73 27 43 47

3 3 1 7 3 2

85 98 99 —

Corrected

59

82 52

38 78

46 81 47

3 4 2 1

59 (10 yr) 60 84 72

ably had a curative operation. Some authors present survival data for subsets of patients according to the Dukes classification, while others state that they are using the Dukes classification but, in fact, are using some variation of it, and therefore survival data are not comparable. Some reports use actuarial methods correcting the data for the age of the patients, thereby attempting to give a more accurate survival statistic. Some authors have used corrected 5-year survival rates by means of life tables to exclude deaths not due to carcinoma but caused by intercurrent disease. Presumably, this method is being used to increase the precision of reporting rather than making the survival statistics look better because, by definition, the corrected survival rate is always higher than the crude survival rate. Estimates of survival are commonly used in the literature to describe outcomes in patients treated for carcinoma. Terms such as carcinoma-specific and carcinoma-free survival are frequently quoted although often without clear definitions. Platell and Semmens (670) compared survival estimates on the same population of patients but using different definitions of what constitutes an event. This was to highlight some of the variation that can occur when different techniques are used to perform these calculations. The

patients (0.14%). Routine prophylactic anticoagulation did not decrease the incidence of pulmonary embolism. Significant thrombophlebitis at the intravenous cannula site occurred in 3.8%, wound infection in 2.1%, and postural peripheral nerve injury in the upper limbs occurred in 0.8%. Unscheduled operations were required in 2.7% of patients. A classification of anastomotic leak is suggested to assist in comparisons of this complication which remains a significant concern following extraperitoneal anastomoses.

& RESULTS In an effort to determine the survival rate following operations for colon carcinoma, review of the plethora of reports makes the reader quickly realize that the literature consists of a maze of information. In attempting to compare the results from various institutions, it rapidly becomes apparent that a host of reporting methods have been used. For example, some authors present overall survival rates of all patients who present with colon carcinoma. Others present data only for those who underwent an operation, while still others present data only for those who presum-

TABLE 11 & 5-Year Survival According to Dukes’ Staging Following Curative Resection Dukes’ A

Dukes’ B

Author(s) Corman, Veidenheimer, and Coller (7) (1979) Pihl et al. (671) (1980) Eisenberg et al. (686) (1982) Isbister and Fraser (675) (1985) Davis, Evans, and Cohen (677) (1987) Read et al. (684) (2002) Staib et al. (687) (2002) Morin et al. (685) (2006) a

Range depending on portion of colon affected.

225 109 101 172 24 73 184 80

Crude

Corrected

81

95 88 75–87a

64 71 99 82 84

96

Survival (%)

Survival (%)

Survival (%) No. of Patients

Dukes’ C

No. of Patients 332 208 274 427 125 151 388 147

Crude

Corrected

62

90 78 64–85a

58 65 87 74 76

87

No. of Patients 204 90 501 354 85 92 246 79

Crude

Corrected

35

55 60 39–43a

32 36 72 49 58

52

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

study included 497 patients with a mean age of 68 years, and a male to female ratio of 1.3 to 1. They were followed for a mean of 2.2 years. The various survivals at 5 years were: (1) overall survival, 55.6%; (2) carcinoma-specific survival, 67%; (3) carcinoma-free survival, 49.9%; (4) recurrence-free survival, 43.5%; and (5) relative survival, 73.4%. The 5-year survival calculations for this group of patients with colorectal carcinoma varied by as much as 30% depending on how the data were censored. This highlights that there needs to be a clear and accountable definition on how survival curves are calculated and presented in the literature to allow for meaningful interpretation and comparisons. It is not unreasonable to assume that figures quoted from major surgical centers would offer better survival rates because a higher standard of care is assumed. However, these figures may not be representative of the majority of regional hospitals. Statistics from tumor registries may be more representative. The crude 5-year survival rates reported from cancer registries have been less impressive, partly because a proportion of these patients almost certainly had only palliative excisions and partly because statistics for those patients who are alive 5 years after operation are expressed as a percentage of those submitted to surgical treatment and not of those surviving that treatment. Thus operative deaths would be included among the nonsurvivors, making the number of 5-year survivors correspondingly less. A more useful measure of surgical treatment is the overall or absolute survival rate, which expresses the number of patients alive and well after 5 years as a percentage of the total number of patients presenting to hospital with carcinoma of the colon in the first instance, not as a percentage of the immediate survivors of operation. The absolute survival rate automatically takes into account the resectability and operative mortality rates as well as the success of the operation in eradicating the carcinoma. Despite this confusing information and with full recognition of the limitations of the exercise, an effort has been made to extract a number of representative series of reasonable size, and these are presented in Tables 10 and 11. In a review of 22 series, Devesa, Morales, and Enriquez (688) determined that the corrected 5-year survival rates for patients with large bowel carcinoma operated on for cure varies from 44% to 68%. The Commission on Cancer Data from the National Cancer Data Base reports time trends in stage of disease, treatment patterns, and survival for patients with selected carcinomas. The l993 data for patients with colon carcinoma are described (689). Five calls for data yielded 3,700,000 cases of carcinoma for the years 1985 through 1993 from hospital cancer registries across the United States, including 36,937 cases of colon carcinoma from 1988 and 44,812 from 1993. Interesting trends are as follows: (1) the elderly ( > 80 years) present with earlier stage disease than do younger patients, (2) the National Cancer Institute recognized that cancer centers have more patients with advanced disease than do other types of hospitals, (3) all ethnic groups have generally similar stages of disease at presentation except for African-Americans, who have a slightly higher incidence of stage IV disease, (4) the proximal migration of the primary carcinoma continues; 54.7% of primary colon carcinoma arose in the right colon in 1993 compared with 50.9% in 1988, (5) an interaction

149

between grade and stage of carcinoma seems present, and (6) patients with stage III colon carcinoma who received adjuvant chemotherapy had a 5% improvement in 5-year relative survival (Fig. 61). The data suggest an important biologic role for grade of carcinoma. They also suggest that African-Americans and other ethnic groups have the same outcome as non-Hispanic whites but that access to medical care may still be less. Finally, the use of adjuvant therapy for stage III colon carcinoma may just beginning to be appreciated. The relative 5-year survival rates for patients with colon carcinoma are depicted in Fig. 62. Comparable data for those with rectal carcinoma are shown in Fig. 63. Survival rates for patients presenting with obstructing and perforating carcinomas are considerably more dismal. Representative reports are shown in Tables 12,13 and 14. The factors primarily contributing to this distressingly poor outlook are the low curability and survival rates for patients with large bowel obstruction secondary to colorectal carcinoma because of advanced disease at the time of diagnosis and treatment (321). Serpell, McDermott, and Katrivessis (695) reported a reduction in curative resection rate from 71% in patients without obstruction to 50% in those with obstruction. A review of the literature by Smithers et al. (486) revealed an operative mortality of 9% to 35% for emergency right hemicolectomy. Indeed, emergency operations of all kinds carry a higher operative mortality rate than do elective operations. Goodall and Park (500) reported on 40 patients with an obstructed left colon who underwent primary resection and anastomosis with a 5% mortality rate and a 40% complication rate. Mandava et al. (299) reported that in their series of 51 patients with perforated colorectal carcinoma, if the patients with metastatic disease and operative mortalities were excluded, there was a 58% 5-year survival rate in the remaining 32 patients. Scott, Jeacock, and Kingston (713) reported on risk factors in patients presenting in an emergency with colorectal carcinoma. Of 905 patients with colorectal carcinoma admitted to a single hospital, 272 (30%) were admitted as emergencies. Emergency patients had more advanced lesions (Dukes’ B and C, 96% vs. 88% of those admitted electively), a shorter history (median, 3 vs. 11 weeks), were less likely to be fully ambulatory (44% vs. 80%), and more likely to have abdominal pain (74% vs. 51%) and vomiting (40% vs. 10%). More emergency patients were given stomas (56% vs. 35%) and died in hospital (19% vs. 8%). Of those who survived to be discharged, patients admitted as an emergency spent longer in hospital (median stay, 16 vs. 13 days) and had a poor overall 5-year survival rate (29% vs. 39%). Emergency patients were significantly older (median, 74 vs. 72 years) and were much more likely to be widowed (41% vs. 27%) than those admitted for elective surgery. The authors concluded that if the personal and resource disaster of emergency colorectal carcinoma admission is to be reduced, screening strategies targeted by demographic characteristics require investigation. Anderson, Hole, and McArdle (713) conducted a prospective study of 570 patients presenting with colorectal carcinoma over a 6-year period. Of these, 363 were admitted electively and 207 presented as emergencies. In the elective group, the proportion of resected lesions was greater (77% vs. 64%), the operative mortality rate was

150

& PART I: COLORECTAL DISORDERS

FIGURE 61 & Relative survival for 1985 to 1988 colon carcinoma cases (combined AJCC stage group III) by treatment modality. Source: From Ref. 688.

lower (9% vs. 19%), and the 5-year disease-related survival rate was higher (37% vs. 19%). These differences may relate to the greater resection rates in the elective situation. Clemmensen and Sprechler (682) reported on the results of 803 patients with colorectal carcinoma, 273 of whom required emergency admission. Immediate operation was performed on 76 of these patients, 37 for obstruction, 15 for perforation, and 24 for other indications. The operative mortality rate for this group was 25%. Other factors to consider are advanced disease and high-risk elderly patients. Fitzgerald et al. (715) reviewed the perioperative mortality and long-term survival in elderly and high-risk patients with colorectal neoplasia. Elderly high-risk patients with localized disease were compared with those with advanced disease. Over a 5-year period, 82 high-risk (at least one major organ system disease) or elderly (age  70 years) patients underwent an operation for colorectal neoplasia. Overall 43 of 82 patients (52%) had advanced disease (obstruction, perforation, hemorrhage, or

FIGURE 62 & Relative survival for 1985–1988 colon carcinoma cases by combined AJCC stage group. Source: From Ref. 688.

FIGURE 63 & Relative five-year survival rate for carcinoma of rectum by combined pathologic and clinical AJCC staging. Percent survival by stage is shown in boxes below graph. Source: From Ref. 305.

metastatic disease), while 39 of 82 patients (48%) had localized disease. Pre-operative comorbid diseases included coronary atherosclerosis, 59 (72%); previous myocardial infarction, 17 (21%); previous arrhythmia, 10 (12%); emphysema, 32 (39%); renal failure, six (7%); and cirrhosis, three (4%). At the time of operation, 26 patients (32%) had metastatic disease. Six patients (7%) died in the perioperative period. There was no difference in major morbidity between patients operated on for localized and for advanced disease. The mean actuarial 18-month survival rate was less for patients with advanced disease. Sixty-eight patients (83%) were alive at a follow-up of 17.7  29 months postoperatively. The morbidity and mortality rates associated with resection of colorectal neoplasia in high-risk elderly patients are acceptable even in the presence of advanced disease. In select patients, resection offers the best palliation and may improve the quality of remaining life. Colorectal carcinoma in cirrhotic patients is different from that in patients without the liver disease. Gervaz et al. (716) retrospectively analyzed 72 patients operated on for colorectal adenocarcinoma with confirmed liver cirrhosis at the time of abdominal exploration. There were 43% Child A, 42% Child B, and 15% Child C. The median age was 70 years, and the mean duration of follow-up was 46 months. Postoperative death was 13%. The risk factors were an elevated bilirubin and prolonged prothrombin time. Liver metastases developed in 10%. For the whole group 1-, 3-, and 5-year survival rates were 69%, 49%, and 35%, respectively. Child A patients had a significantly better survival rate than the combined group of Child B and C patients. The risks for long-term survival were decreased albumin and prolonged prothrombin time. The Child’s classification, and not the TNM stage of the carcinoma, predicts the risk of postoperative death and long-term survival. Kotake et al. (717) examined trends of colorectal carcinoma in relation to age, gender, site, and survival during a 20-year period. The multi-institutional registry of the Japanese Society for Cancer of the Colon and Rectum offered 87,695 surgical cases with invasive adenocarcinoma. The number of cases showed a 2.5-fold increase with

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

151

TABLE 12 & Overall 5-Year Survival for Patients with Obstructing Carcinoma Survival (%) Author(s)

No. of Patients

Operative Mortality (%)

156 148 41 37 124 713 77 148 40 115 92 120 177

18 9 2 3 20 23

Kelley et al. (690) (1981) Ohman (321) (1982) Brief et al. (691) (1983) Crooms and Kovalcik (692) (1984) Umpleby and Williamson (327) (1984) Phillips et al. (693) (1985) Willett et al. (694) (1985) Serpell et al. (695) (1989) Ueyama et al. (324) (1991) Sjodahl, Franzen, and Nystrom (484) (1992) Mulcahy et al. (696) (1996) Chen et al. (697) (2000) Carraro et al. (683) (2001)

consistent male predominance confined to the distal colon and rectum. Colon carcinoma in the last 5-year period was more likely right-sided for females (odds ratio, 1.26) and males (odds ratio, 1.16) compared with the first period. Carcinoma in younger patients were more likely at stage III to IV in the late 1990s if the carcinoma were in the distal colon, the rectum (for both genders), or the proximal colon (for females). Survival was improved except for cases with proximal colon carcinoma of stage IV. In the multivariate analysis, hazard ratios for death in the postoperative five years were 0.77, 0.59, and 0.66 for proximal colon, distal colon, and rectal carcinomas respectively in the last period as compared with those in the first period. Reduced hazard ratio for females was the largest for proximal colon carcinoma with stage I and II. Although surgical outcome was largely improved, delayed presentation or diagnosis in younger patients remained a problem. Wang et al. (718) investigated the clinical features, diagnosis, treatment, and prognosis of 37 patients with multiple

Crude

Corrected

18 16 78 33 18 25 31

9 0 17 12 5 10

52 36 33 46

primary carcinomas. The incidence of multiple primary colorectal carcinomas was 2.7% in patients with primary colorectal carcinomas, 15 cases were patients with synchronous carcinomas and 22 cases were diagnosed as metachronous carcinomas. Most carcinomas were located in the right colon and rectum. Fifty-five percent of metachronous carcinomas were diagnosed within three years after resection of the initial lesion and 41% of metachronous carcinomas occurred after eight years. Radical resections were performed in all patients except for one case. The five-year survival rate of synchronous carcinomas was 72.7% and that of metachronous carcinomas after the first carcinoma and second carcinoma was 71.4% and 38.9%, respectively. Their results indicate the importance of complete preoperative examination and careful intraoperative exploration and periodic postoperative surveillance. Immunosuppression used in transplantation is associated with an increased incidence of various carcinomas. Papaconstantinou et al. (719) defined the characteristics and survival patterns of transplant patients developing

TABLE 13 & Results of Immediate Resection and Primary Anastomosis for Obstructing Carcinoma of the Left Colon

Author(s) Amsterdam and Krispin (482) (1985) Hughes et al. (698) (1985) Morgan et al. (699) (1985) White and Macfie (483) (1985) Feng, Hsu, and Chen (700) (1987) Halevy, Levi, and Orda (701) (1989) Slors et al. (702) (1989) Stephenson et al. (703) (1990) Runkel et al. (300) (1991) Antal et al. (704) (1991) Brief et al. (391) (1991) Tan et al. (705) (1991) Murray et al. (706) (1991) Sjodahl, Franzen, and Nystrom (484) (1992) Stewart, Diament, and Brennan (707) (1993) Arnaud and Bergamaschi (708) (1994) Lau, Lo, and Law (493) (1995) Carraro et al. (682) (2001)

Procedure

No. of Patients

Operative Mortality (%)

Complication Rate (%)

Segmental resection TAC or STC TAC or STC LHC or sigmoid resection STC or segmental resection STC TAC and IRA STC STC, LHC STC STC Segmental resection Segmental resection Not stated Segmental resection STC STC or segmental resection Segmental or STC

25 52 16 35 15 22 10 31 21 40 23 23 21 18 60 44 35 107

12 12 13 9 7 5 10 3 14 0 4 9 0 6 7 7 6 10

60

Abbreviations: STC, Subtotal colectomy; TAC, total abdominal colectomy; LHC, left hemicolectomy; IRA, ileorectal anastomosis.

29 27 36 30 6 46 25 29 43 60 17 7 31 –

152

& PART I: COLORECTAL DISORDERS

TABLE 14 & Overall 5-Year Survival for Patients with Perforated Carcinoma

Author(s)

No. of Patients

Operative Mortality (%)

5-Year Crude Survival (%)

27 42 34 36 20 51 13 54a 35a 13b 29b 48

33 38

7 23 44 40

Kelley et al. (690) (1981) Michowitz et al. (709) (1982) Willett et al. (694) (1985) Badia, Sitges-Serra, and Pia (710) (1987) Runkel et al. (299) (1991) Mandava et al. (298) (1996) Mulcahy et al. (696) (1996) Carraro et al. (711) (1998) Chen et al. (697) (2000)

Khan et al. (712) (2001)

14 30 12 12 17 9 31 48 14

32 12 – – 33 1 14

a

Perforation at site of carcinoma. Perforation proximal to carcinoma.

b

de novo colorectal carcinomas. A total of 150 transplant patients with de novo colorectal carcinoma were identified: 93 kidney, 29 heart, 27 liver, and 1 lung. Mean age of transplantation was 53 years. Age of transplantation of colorectal carcinoma diagnosis was not significant for gender, race, or stage of disease. Compared to National Cancer Institute Surveillance Epidemiology and End Results database, transplantation patients had a younger mean age of colorectal carcinoma diagnosis (58 years vs. 70 years) and a worse 5-year survival (overall 44% vs. 62%, Dukes’ A and B 74% vs. 90%, Dukes’ C 20% vs. 66%, and Dukes’ D 0% vs. 9%). Their data suggest that chronic immunosuppression results in a more aggressive biology. Frequent posttransplantation colorectal screening program may be warranted.

& PROGNOSTIC DISCRIMINANTS Over, the years, many factors that might influence survival rates have been studied, and different authors have placed varying degrees of confidence in certain ones. For example, in a large prospective analysis of 2524 patients who had undergone curable resection, the prognostic factors in order of importance were (1) lymph node status, (2) mobility of the carcinoma, (3) number of positive lymph nodes, (4) presence of bowel obstruction, and (5) depth of penetration of the primary lesions (720). In a review of the literature and combined with his experience, Jass (721) summarized the important variables as follows. The pathologist should provide the staging, completeness of excision (resection lines), extent of spread through the bowel wall, lymph node or satellite nodule involvement, especially involvement of nodes at the apex of the vascular pedicle, and the number of nodes. Other variables should be considered. For completeness of documentation, carcinoma of the large bowel should be typed as adenocarcinoma, mucinous adenocarcinoma, signet-ring cell carcinoma, undifferentiated carcinoma, or other. In addition, the grade of differentiation, well, moderate, or poor, should be stated. However, type and grade of carcinoma have little or no independent prognostic value. Classification of carcinoma as expanding or diffusely infiltrating has been shown to be of prognostic importance in a

number of multivariate studies. Most large bowel carcinomas are relatively well circumscribed. Approximately 25% show an irregular margin of growth with tongues of neoplastic cells dissecting between the normal structures of the bowel wall, making it difficult to define a clear border of the carcinoma. The term ‘‘diffuse infiltration’’ does not imply massive intramural spread, a type of spread very rarely seen in the large bowel. The presence of venous invasion by carcinoma should be recorded because this may have a bearing on the documentation of extent of spread. The presence of venous invasion is strongly correlated with distant spread. When cases with distant spread are excluded, the prognostic value of venous spread is greatly diminished and is absent in some multivariate studies. The presence of perineural and lymphatic spread influences prognosis adversely, but these factors were not found to be independent when Jass (721) examined 500 specimens from the prospectively collected Australian series. A conspicuous lymphocytic infiltrate, either within an inflammatory mantle at the growing edge of the carcinoma or arranged within nodular collections around small serosal vessels, confers a favorable prognosis that has been shown to be independent. The following discussion, although not totally comprehensive, elaborates on the factors that most often have been considered relevant. A survey of prognostic discriminants for colon carcinoma is presented in Table 15.

& CLINICAL FEATURES Age Survival rates of patients younger than 40 years of age are frequently believed to be lower than overall survival rates. The 5-year survival rates in this group of patients range from 16% to 43% (724–729,856). The poor prognosis in young patients has been attributed to the larger portion of poorly differentiated lesions, a larger number of mucinous lesions, and a lower potential for curative resection (724,725). Not all authors believe that young patients have a less favorable prognosis (856,857). Svendson et al. (919) reported that patients who are between 40 and 60 years of age at the time of diagnosis have a worse prognosis than both younger and older patients.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

153

TABLE 15 & Prognostic Discriminants Effect on Prognosisa Discriminants Clinical Features Age Young (< 40 years) Elderly (> 70 years) Sex Female Male History Asymptomatic Bleeding as presenting symptom > 2 presenting symptoms Delay in diagnosis (medical) Long duration of symptoms Short duration of symptoms Obstruction Perforation Adjacent organ involvement Presence of metastatic disease Systemic manifestations Technique of resection Left hemicolectomy vs. segmental No-touch technique Occlusion of bowel ends Intraoperative perforation of bowel High volume Colorectal trained surgeon Anastomotic leakage Perioperative blood transfusion Postoperative fever Previous appendectomy Obesity Pathologic Features Location Rectum Right colon Size Configuration Ulcerated Polypoid Annular Microscopy Well differentiated Moderately differentiated Poorly differentiated Mucin producing Inflammatory infiltrate Desmoplastic reaction Linitis plastica Depth of penetration Circumferential margin involvement Micro acinar ‘‘Budding’’ Signet ring cell Residual disease

Diminished

None

421, 722–729 674, 731, 732

7, 731, 856–863 696, 864–867

733 734

7, 696, 731, 738, 747, 775, 793 671, 696, 775

Improved

738, 894

737, 895, 896 731, 736, 739, 747 735 421 736 138 7, 327, 422, 583, 683, 697, 727, 731, 738–743, 922 298, 327, 421, 521, 696, 697, 711, 712, 727, 734, 738, 739, 745 422, 731, 747–750 All authors 422

774, 796, 868, 869

731

324

752, 870 397, 752 752 745, 747, 752–754 872

755–757 758–763

300, 753, 754, 897–903 754, 871, 904–909

872–874

764 765 766

875

910

732, 740, 764, 767 674, 731, 768 769–772

696, 720, 775 739, 775, 876 696, 731, 773, 793, 877

735 422, 740, 767

769, 773 769, 773, 911, 912 740, 774 421, 671, 696, 720, 731, 747, 769 422, 674, 714, 720, 731, 732, 747, 769, 771, 772, 775 237, 238, 696, 725, 776–781

742, 771, 773, 791, 878, 879 671, 769, 773, 913

782 783 720, 727, 732, 772, 782, 784 737 785 786–788 789–791 792

(Continued)

154

& PART I: COLORECTAL DISORDERS

TABLE 15 & Prognostic Discriminants (Continued ) Effect on Prognosisa Discriminants Staging Dukes A Dukes B Dukes C Lymph node status Positive 4 nodes Necrosis in metastatic lymph nodes Venous invasion Perineural invasion Perioperative disseminated malignant cells Biochemical and Special Investigations Diabetes mellitus High preoperative CEA levels Abnormal liver function tests DNA non-diploid Low serum protein levels Hemoglobin levels White blood cell count Erythrocyte sedimentation rate Genetic alteration 17p deletion DCC expression Loss of hMLHI expression Allelic loss chromosome 18q Allelic loss chromosome 5q p53 overexpression CD-44 v6 expression ras p21 overexpression high Ki67 expression Microsatellite instability DNA microarray–based gene expression Thymidylate synthetase overexpression Increased EGF expression Apoptotic index Sialyl Lewisx antigen PCNA expression Sialomucin staining Nuclear morphometry (>0.84)c High u-PA/t-PA (>0.22) Socioeconomic deprivation

Diminished

None

Improved All authorsb

All authorsb All authorsb 7, 230, 441, 720, 740, 741, 771, 773, 784, 793, 794, 795, 797

732 914

7, 230, 676, 693, 720, 731, 732, 740, 747, 752, 754, 772, 799–805 693, 754, 799, 800, 802, 804, 806, 807 808

809 747, 810–817 818 744, 819–824 747

771

880

687, 881 747 772, 775, 882–885 747, 775 747 747

825 886 915 825–830 826 831–838 839 840

886–889 772, 890–892 888

841 836, 842, 843

836 830, 916–918 918

893

844 847 845–847 848, 849 850–852 853 854 855

Abbreviations: CEA, carcinoembryonic antigen; DNA, deoxyribonucleic acid; PCNA, proliferating cell nuclear antigen. a All numbers in ‘‘Effect’’ columns correspond to reference numbers for this chapter. b ‘‘All authors’’ refers to unanimous agreement by all authors who address the subject. c See text for explanation.

Cusack et al. (729) conducted a retrospective review of 186 patients younger than 40 years of age. Regional lymph node metastases, distant metastases, or both were seen on first examination in 66% of young patients. The authors identified three biologic indicators of aggressive and potentially metastatic biology: signet-ring cell carcinoma (11.1%), infiltrating edge of the carcinoma (69%), and aggressive (poorly differentiated) grade (41%). Vascular invasion in stage II disease was also a significant negative

prognostic variable. These histologic measures of more aggressive disease in part account for the higher rate of advanced disease at presentation in patients younger than 40 years of age. Liang et al. (859) compared 138 consecutive patients with colorectal carcinoma aged less than 40 years with 339 patients aged 60 years or more. The younger patients with colorectal carcinoma had more mucin-producing (14.5% vs. 4.7%) and poorly differentiated (7.2% vs. 3.3%)

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

carcinomas, a higher incidence of synchronous (5.8% vs. 1.2%) and metachronous (4.0% vs. 0.6%) colorectal carcinomas, and more advanced stage than older patients. The operative mortality rate was lower (0.7% vs. 5.0%), and carcinoma-specific survival was similar (in stage I, II, and III disease) or better (in stage IV disease). There was a higher% of normal p53 expression (61.1% vs. 46.8%) and highfrequency of microsatellite instability (29.4% vs. 6.3%) and a similar family history of carcinoma (17.5% vs. 14.2%), compared with older patients. O’Connell et al. (730) performed the most comprehensive systematic review focusing on colorectal carcinoma in the young aiming to (1) characterize the disease in the young population and (2) determine how colorectal carcinoma in this population should be further addressed regarding detection and treatment. A Medline literature search chose 55 studies that examined 6425 patients less than 40 years old. Approximately 7% of all colorectal carcinomas consisted of patients less than 40 years old. They found that colorectal carcinoma in the young population appears to be more aggressive, to present with later stage (66% of patients < 40 years old presented with Dukes’ C and D lesions compared with 32% to 49% reported for patients > 40 years old), and to have poorer pathologic findings (higher prevalence of mucinous or poorly differentiated lesions including signet-ring carcinoma—one of the main distinctions between the disease in older vs. younger patients. Mucinous lesions constituted an average of 21% of carcinoma in younger patients compared with an average of 10% to 15% of patients of all ages with colorectal carcinoma. The average% of lesions found to be poorly differentiated was 27% for the young group compared with 2% to 29% of patients over the age of 40). The average overall 5year survival for young patients was 33% compared with 61% overall 5-year survival reflecting presentation with later stage disease thus appearing to have a poorer prognosis. However, if detected early, young patients with Dukes’ stage A or B lesions have better overall 5-year survival rates. Average-adjusted 5-year survival for Dukes’ A 94%; B 77%, and C 39%. A crucial issue is that close attention must be paid to young patients who present with common symptoms of colorectal carcinoma. O’Connell et al. (862) used a national-level, population-based cancer registry to compare rectal carcinoma outcomes between young versus older populations. All patients with rectal carcinoma in the Surveillance Epidemiology and End Results Cancer database from 1991 to 1999 were evaluated. Young (range 20– 40 years; n ¼ 466) and older groups (range 60–80 years; n ¼ 11,312) were compared for patient and carcinoma characteristics, treatment patterns, and 5-year overall and stagespecific survival. Mean ages for the groups were 34.1 and 70 years. The young group was comprised of more black and Hispanic patients compared with the older group. Young patients were more likely to present with late stage disease (young vs. older: stage III, 27% vs. 20% and stage IV, 17.4% vs. 13.6%, respectively). The younger group also had worse grade (poorly differentiated 24.3% vs. 14%, respectively). Although the majority of both groups received surgery (85% for each), significantly more young patients received radiation. Importantly, overall and stage-specific, 5-year survival rates were similar for both groups.

155

From a group of 2495 patients with malignancies of the colon and rectum, Turkiewicz et al. (860) identified 61 patients with colorectal carcinoma who were aged less than 40 years at presentation. Their clinical data were then compared with the larger group of older patients. A positive family history was the most consistent risk factor, present in 34% of patients. Despite this, only one patient out of 61 had been diagnosed as a result of a screening program. The overall 5-year survival among younger patients was 53%. The 5-year survival rates in younger patients were better than for older patients for the Australian Clinicopathological Staging A and B reaching statistical significance for both of these stages. Their results indicate that young patients with colorectal carcinoma have the potential to do just as well as older ones. For the prognosis of elderly patients, conflicting reports are found. A common notion is that aged patients have less biologically aggressive neoplasms. Newland et al. (732) found that in patients 75 years of age, the hazard ratio for survival was 1.98. Coburn, Pricolo, and Soderberg (920) compared 177 cases of colorectal carcinoma in patients > 80 years to 623 in patients < 80 years. There was no difference in operative mortality between the two groups. Octogenarians and nonagenarians more often displayed obstruction or perforation, elevated preoperative CEA levels, right-sided lesions, and solitary hepatic metastases. The actuarial 5-year survival rate was 32% in the older patients and 48% in the younger group. Others report that the prognosis for the elderly is not different from that of younger patients (864–866).

Sex Some authors have reported a slightly higher incidence of carcinoma of the colon in women than in men, but the 5-year survival rates are slightly higher in women (3). However, others have reported poorer survival rates in women (735). The hazard ratio for the male gender has been estimated at 1.27 (732). History As might be expected, patients who are asymptomatic have better survival rates than symptomatic patients do (736,895,896). Patients who have rectal bleeding as a presenting symptom have a better prognosis than do those presenting with other symptoms of colonic carcinoma (895). The number of symptoms is also a prognostic discriminant since patients with more than two symptoms have a poorer prognosis (735). There appears to be no correlation between the duration of symptoms and survival. In a study of 152 patients, Goodman and Irvin (869) found no difference in the survival rate for patients in whom the diagnosis was delayed >12 weeks from the onset of symptoms compared with those who presented early. They did find that patients with anemia and no abdominal symptoms had a significantly higher survival rate than those presenting with abdominal symptoms. Patients who ignore their symptoms tend to have biologically more favorable lesions. Ironically, in patients with a longer history of symptoms, there appears to be either no effect on survival or a better survival rate

156

& PART I: COLORECTAL DISORDERS

(774,796,921). Wiggers, Arends, and Volovics (747) found that patients who had symptoms for a very short time (<1 week) or a very long time ( > 6 months) had a poorer survival rate than those in the intermediate range, but this was not statistically significant. This apparent paradox may be explained by the fact that patients with aggressive lesions (i.e., annular, constricting, or poorly differentiated) are compelled to seek help early. Patients with less acute symptoms often exhibit lesions of average grade and may ultimately prove to have a better prognosis.

Obstruction The incidence of intestinal obstruction has been reported to be between 7% and 29% of all patients with colon and rectal cancer (922). The occurrence of acute obstruction diminishes the ultimate survival rate and increases the immediate hazard to the patient (739,740,742,895,922). In a review of 12 reports, Sugarbaker, Gunderson, and Wittes (422) found a median overall survival rate of 20% and a 5-year survival rate of 40% with curative surgery. The authors attributed this dismal prognosis to the fact that only about one half of the patients had potentially curative operations and that the operations were accompanied by high morbidity and mortality rates, with a median hospital mortality rate of 18% and complications occurring in one third to one half of the patients. Wang et al. (922) conducted a study to assess the longterm prognosis of patients with obstructing carcinoma of the right colon. The 256 patients who were status postcurative resection of right colon adenocarcinoma were classified as obstruction group (n ¼ 35) or nonobstruction group (n ¼ 221). The overall (49% vs. 22%), distant (40% vs. 18%), and local (14% vs. 5%) recurrence rates were significantly higher in obstructive patients than in nonobstructive patients. Long-term crude (36% vs. 77%) and carcinoma-specific survival rates (46% vs. 83%) were significantly lower in obstructed patients. Multivariate analysis demonstrated that obstruction and stage were both independent prognostic factors. Carraro et al. (683) reported on a series of 528 patients with colonic carcinoma, 34% of whom presented with obstruction. One-stage primary resection and anastomosis as curative treatment were performed in 107 obstructed and 256 nonobstructed patients. Three hundred thirty-six potentially cured survivors (94 in the former group and 242 in the latter) were followed for a median of 55 months. During follow-up, local recurrence occurred in 37 patients [12 obstructed (12.8%) and 25 nonobstructed (10.4%)], and metastatic disease in obstructed (27.6%) and nonobstructed (17.8%). Multivariate analysis of survival showed that age over 70 years, Dukes’ stage, histologic grade, and recurrence were the only prognostic factors. After one-stage emergency curative treatment, patients presenting with obstructing carcinomas of the colon have a smaller survival probability than that of patients with nonobstructing lesions. Chen et al. (697) reviewed the medical records of 1950 patients with colorectal carcinoma. Patients were grouped as follows: group 1, complete colonic obstruction without perforation (n ¼ 120); group 2, complete obstruction with perforation at the site of the carcinoma (n ¼ 35); group 3, complete obstruction with perforation proximal to the

carcinoma (n ¼ 13); and group 4, nonobstructing, nonperforated carcinomas (n ¼ 1682). When compared with group 4, group 1 had a more advanced Dukes’ stage, older age, greater incidence of colonic versus rectal carcinomas and a poorer carcinoma-free survival. Groups 2 and 3 had a greater incidence of colonic versus rectal carcinomas, and group 3 had a greater operative mortality. No significant differences were found between groups 1, 2, and 3. Independent factors favorable to carcinoma-free survival were female gender, well-differentiated pathology, uncomplicated cases, colon versus rectal location, and early stage. The perioperative mortality rate for perforated colorectal carcinoma at the site of the carcinoma was 9%; for obstructive colorectal carcinoma 5%. Perioperative mortality was much greater for perforations of the colon and rectum occurring proximal to the carcinoma (31%). Survival was worse for patients with obstruction (33%), or perforation proximal to the carcinoma (33%).

Perforation In comparison with obstruction, perforation has an even greater detrimental effect on the ultimate outcome for the patient. In a review of four reports, Sugarbaker, Gunderson, and Wittes (422) found a median overall 5-year survival rate of only 9% and a 5-year survival rate of 33% with curative surgery. The median curative operative rate was 55%, with a hospital mortality rate of 30%. For patients with a free perforation into the peritoneal cavity, the 5-year survival rate was a dismal 7.3%, whereas for a localized perforation, the 5-year survival rate was 41.4%. Another report shows an even higher postoperative mortality of 52% (327). Khan et al. (712) reviewed 48 patients presenting with acute colonic perforation associated with colorectal carcinoma. Thirty-six had perforation of the carcinoma, 11 proximal to the carcinoma, and one distal to the primary carcinoma. Patients who perforated proximal to the carcinoma were older (74.5 years vs. 64.7 years) and had a longer length of stay (46.8 days vs. 11.6 days). Fourteen patients had stage II disease, 19 stage III, and 15 stage IV. Thirty-day mortality was 14%. Of the 30-day survivors, 60% had curative resection (21 with local perforation and 9 with proximal perforation). Thirty-three percent had either unresectable or metastatic disease on exploration. One-year survival was 55%. Five-year disease-free survival was 14%. There were no long-term survivors after perforation proximal to the carcinoma although disease stage was comparable in both groups. Adjacent Organ Involvement Direct invasion of adjacent viscera does not preclude the possibility of a curative resection, and every effort must be made to perform a resection if technically possible. A review of 25 reports in the literature by Sugar baker, Gunderson, and Wittes (422) noted a median adjacent organ involvement in 9%, an operative mortality rate of 8%, and a salvage rate of 30% to 50%. This includes fistulization into an adjacent viscus such as the urinary bladder. Although adjacent organ involvement represents more advanced disease, curative resection may still be possible. However, extensive operation is associated with a higher morbidity and mortality.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

Presence of Metastatic Disease It is axiomatic that patients with metastatic disease bear a poorer prognosis than those who are free of disseminated disease. The rare exception is the situation in which the metastatic disease can be resected for cure. Systemic Manifestations Patients who develop symptoms of weight loss, anorexia, weakness, or anemia frequently do so in the presence of advanced disease, which does not bode well for the patient. Obesity To determine the relationship between body mass index and rates of sphincter preserving operations, overall survival, recurrence, and treatment-related toxicities, Meyerhardt et al. (766) evaluated a nested cohort of 1688 patients with stage II and III rectal carcinoma participating in a randomized trial of postoperative fluorouracil-based chemotherapy and radiation therapy. Obese patients were more likely to undergo an abdominoperineal resection than normal-weight patients (odds ratio, 1.77). Increasing adiposity in men was a strong predictor of having an abdominoperineal resection. Obese men with rectal carcinoma were also more likely than normal-weight men to have a local recurrence (hazard ratio 1.61). In contrast, obesity was not predictive of carcinoma recurrence in women nor was body mass index predictive of overall mortality in either men or women. Underweight patients had an increased risk of death (hazard ratio 1.43) compared with normal-weight patients but no increase in carcinoma recurrences. Among all study participants, obese patients had a significantly lower rate of grade 3 to 4 leukopenia, neutopenia, and stomatitis and a lower rate of any grade 3 or worse toxicity when compared with normal-weight individuals. Technique of Resection Various efforts have been made to diminish the risk of dissemination of malignant cells during the operation, but for the most part the results have not been conclusive. In their large multicenter trial, Phillips et al. (727) reported considerable surgeon-related variation with respect to local recurrence. The patients of surgeons with considerable experience had fewer recurrences of disease. Possible causes cited for local recurrence include inadequate resection, suture implantation of malignant cells, and development of a second primary carcinoma at the anastomosis site. The authors state that surgeons should be aware that a small group of our colleagues are obtaining results substantially better than those of the majority, and they conclude that these good results have been achieved by meticulous attention to detail. In a comprehensive review of the literature, Sugarbaker and Corlew (752) analyzed available data and concluded that there was no survival benefit with the use of radical left hemicolectomy instead of segmental resection, the use of the no-touch technique instead of conventional techniques, or the adoption of control of intraluminal spread of malignant cells. En bloc resection of an attached structure did seem important. The inadvertent intraoperative perforation of the bowel during a curative resection has a decidedly detrimental effect, both in

157

terms of survival and local recurrence (745). The 5-year survival rate with bowel disruption was 23%, and the rate fell as low as 14% when the carcinoma itself was disrupted. Local recurrence rose to 67% in Dukes’ B cases with spillage and as high as 87% when Dukes’ C carcinomas were perforated at the time of operation (745). Other authors have reported that spillage of malignant cells at the time of operation reduces 5-year survival after resection for cure from 70% to 44% (753). Akyol et al. (754) examined anastomotic leaks as a risk factor for recurrence. At a mean follow-up of 25 months. 46.9% of patients with leaks developed a recurrence compared with 18.5% for those with out a leak. Cancer-specific mortality rates at 24 months were higher for patients with leaks (36.9% vs. 12.6%). Fujita et al. (756) also found that for patients with anastomotic leakage, the incidence of local recurrence was higher (21.2% vs. 2.4%) and the disease-free survival rate was lower in Dukes’ A and B patients (55% vs. 80% at 5 years), but not in Dukes’ C patients than in patients with no leakage. In a study of 403 patients, Bell et al. (757) found after adjustment for lymph node metastases, the distal resection margin of resection, nontotal anatomical dissection of the rectum, and the level of the anastomosis, identified a significant association between anastomotic leakage and local recurrence (hazard ratio 3.8). They concluded that leakage following a colorectal anastomosis after potentially curative resection for carcinoma of the rectum is an independent predictor of local recurrence. Rouffet et al. (870) reviewed 270 consecutive patients randomly allotted to undergo either left hemicolectomy or left segmental colectomy. Left hemicolectomy removed the entire left colon along with the origin of the inferior mesenteric artery and the dependent lymphatic territory. Left segmental colectomy removed a more restrictive segment of the colon and left the origin of the inferior mesenteric artery unmolested. Both groups were similar with regard to preoperative risk factors. The number of early postoperative abdominal and extra-abdominal complications was similar in both groups. Overall, early postoperative mortality was 4% higher, in left hemicolectomy (6%) than in the left segmental colectomy (2%). Median survival was 10 years, and nearly equivalent in both groups. The two actuarial survival curves were similar. Bowel movement frequency was significantly increased after left hemicolectomy during the first postoperative year. Their results suggest that survival after left segmental colectomy is equivalent to that of left hemicolectomy. Inadvertent perforation of the bowel during curative resection for colon carcinoma has a definite adverse effect. Slanetz (745) reported a drop in 5-year survival rates from 29% to 14% when disruption occurred during dissection. Local recurrence developed in 75% of cases involving spillage of malignant cells.

Colorectal Specialization and Surgical Volume In an audit from a Scottish series of 646 anastomoses, the overall anastomotic leak rate was 4.8% (3.2% for colonic carcinoma and 8.9% after resection for rectal carcinoma) (300). Intersurgeon variation was scrutinized. When the anastomosis was performed by five of the 28 surgeons

158

& PART I: COLORECTAL DISORDERS

responsible for 50% of the patients, the leak rate was 4.2% vs. 14.3% for the others. The authors support the concept of specialized units. The impact of the variability in surgical skill among surgeons has also been reported in a multicenter study by Reinbach et al. (897) There was a significant difference between colorectal surgeons with regard to postoperative mortality rates, which varied from 8% to 30%, anastomotic leakage rates, which ranged from 0% to 25%, wound sepsis rates, which ranged from 6% to 35%, and local recurrence rates, which ranged from 0% to 29%. The 10-year survival rate varied from 0% to 63%. There are many studies that now stress the point that we cannot ignore the fact that there exists a variability of skill among surgeons. Volume Meyerhardt et al. (923) studied a nested cohort of 1330 patients with stage II and stage III rectal carcinoma participating in a multicenter adjuvant chemotherapy trial. They analyzed differences in rates of sphincter-preserving operations, overall survival, and carcinoma recurrence by hospital surgical volume. They observed a significant difference in the rates of abdominoperineal resections across tertiles of hospital procedure volume (46.3% for patients resected at low-volume, 41.3% at medium-volume, and 31.8% at high-volume hospitals). This higher rate of sphincter-sparing operation at high-volume centers was not accompanied by any increase in recurrence rates. Hospital surgical volume did not predict overall disease free, or local recurrence-free survival. Patients who did not complete the planned adjuvant chemotherapy, those who underwent operation at low-volume hospitals had a significant increase in carcinoma recurrence (hazard ratio 1.94) and a nonsignificant trend toward increased overall mortality and local recurrence. In contrast, no significant volume-outcome relation was noted among patients who did complete postoperative therapy. Shrag et al. (901) conducted a retrospective population-based cohort study utilizing the Surveillance, Epidemiology and End Results-Medicare linked database identified 2815 rectal carcinoma patients aged 65 and older. They found surgeon volume was better than hospital procedure volume as predicting long-term survival and can have a significant impact on survival for patients with rectal carcinoma. From the same database Shrag et al. (924) identified 24,166 colon carcinoma patients aged 65 and older. As opposed to their findings with rectal carcinoma, high-hospital procedure remained a strong predictor of low postoperative mortality rates for each outcome with and without adjustment for surgeon procedure volume. Surgeon-specific procedure volume was also an important predictor of surgical outcomes for 30-day mortality, and for 2-year mortality, although this effect was attenuated after adjusting for hospital volume. Hospital volume and surgeon volume were each an important predictor of the ostomy rate. Among high-volume institutions and surgeons, individual providers with unusually high ostomy rates could be identified. Both hospital and surgeonspecific procedure volumes predict outcomes following colon carcinoma resection; but hospital volume may exert a stronger effect. In an analysis of 600 patients undergoing

resections for rectal carcinoma, Hermanek and Hohenberger (903) reported that the patients of low-volume surgeons experienced an increased risk of local recurrence. Borowski et al. (925) examined surgeon-volume and specialization as defined as membership of the Association of Coloproctology of Great Britain and Ireland as independent prognostic factors for operative morbidity and mortality for patients undergoing operations for colorectal carcinoma. A total of 5948 patients in a regional center in the United Kingdom underwent operations with an operative mortality of 7.9%. Mortality risk was significantly reduced for surgeons who performed more than 20 operations per year while ACPGBI membership was not significant. Although membership demonstrated an interest, it did not necessarily represent specialty training. Surgeons with a high-case volume or specialized interest were more likely to achieve bowel continuity than low-volume surgeons and nonspecialists following rectal resection. There was no significant difference in anastomotic leak rate (5.1%). Martling et al. (902) compared outcomes in patients being operated upon by high-volume surgeons (more than 12 operations per year) with low-volume surgeons (12 operations or fewer per year). Forty-six surgeons operated on 652 patients. Five high-volume surgeons operated on 48% of the patients. In these, outcome was significantly better than in patients treated by low-volume surgeons (local recurrence rate 4% vs. 10%; rate of rectal carcinoma death 11% vs. 18%). Wibe (926) examined the influence of caseload on long-term outcome following standardization of rectal carcinoma surgery at a national level. Data relating to all 3388 Norwegian patients with rectal carcinoma treated for cure were recorded in a national database. Treating hospitals were divided into four groups according to their annual caseload: hospitals in group 1 carried out 30 or more procedures, those in group 2 performed 20 to 29 procedures, group 3 10 to 19 procedures, and group 4 fewer than 10 procedures. The 5-year local recurrence rates were 9.2%, 14.7%, 12.5%, and 17.5% and 5-year overall survival rates were 64.4%, 64.0%, 60.8%, and 57.8%, respectively, in the four hospital caseload groups. An annual hospital caseload of fewer than 10 procedures increased the risk of local recurrence compared with that in hospitals where 30 or more procedures were performed each year (hazard ratio 1.9). Overall survival was lower for patients treated at hospitals with an annual caseload of fewer than 10 versus hospital for 30 or more (hazard ratio 1.2). Colorectal Specialization Callahan et al. (904) examined the relationship of surgeon subspecialty training and interests to in-hospital mortality while controlling for both hospital and surgeon volume. A large Statewide Planning and Research Cooperative System was used to identify 48,582 in patients who underwent colectomy. Surgical subspecialty training and interests was defined as surgeons who were members of the Society of Surgical Oncology (training/interest n ¼ 68) or the American Society of Colon and Rectal Surgeons (training/ interest n ¼ 61). Overall mortality for colectomy patients was 4.6%; the adjusted mortality rate for subspecialty

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

versus nonspecialty-trained surgeons was 2.4% versus 4.8%, respectively. For colectomies, risk-adjusted mortality is substantially lower when performed by subspecialty interested and trained surgeons, even after accounting for hospital and surgeon volume and patient characteristics. These findings may have implications for surgical training programs and for regionalization of complex surgical procedures. Rosen et al. (905) examined variations in operative mortality among surgical specialists who perform colorectal surgery. Mortality rates were compared between six board-certified colorectal surgeons and 33 other institutional surgeons using comparable colorectal procedure codes and a validated database indicating patient severity of illness. Thirty-five ICD-9-CM procedure codes were used to identify 2805 patients who underwent colorectal surgery. Atlas, a state-legislated outcome database, was used by the hospital’s Quality Assurance Department to rank the Admission Severity Group of 1753 patients (higher ASG, 0–4, indicates increasing medical instability). Colorectal surgeons had an eight-year mean in-hospital mortality rate of 1.4% compared with 7.3% by other institutional surgeons. There was a significantly lower mortality rate for colorectal surgeons compared with other institutional surgeons in ASG2 (0.8% vs. 3.8%, respectively) and ASG3 (5.7% and 16.4%, respectively). Board-certified colorectal surgeons had a lower in-hospital mortality rate than other institutional surgeons as patients’ severity of illness increased. Platell et al. (906) reviewed patients with colorectal carcinoma managed in general surgery units versus a colorectal unit. These results were compared to a historical control group treated within general surgical units at the same hospital. There were 974 patients involved in the study with no significant differences in the demographic details for the three groups. Patients in the colorectal group were more likely to have rectal carcinoma and stage I carcinomas and less likely to have stage II carcinomas. Patients treated in the colorectal group had a significantly higher overall 5-year survival when compared with the general surgical group and the historical control group (56% vs. 45% vs. 40%, respectively). Survival regression analysis identified age, ASA scores, disease stage, adjuvant chemotherapy, and treatment in a colorectal unit (hazards ratio 0.67) as significant independent predictors of survival. The results suggest that there may be a survival advantage for patients with colon and rectal carcinoma being treated within a specialist colorectal unit. Read et al. (907) determined the effect of surgeon specialty on disease-free survival and local control in patients with carcinoma of the rectum. The records of 384 consecutive patients treated by colorectal surgeons (n ¼ 251) and noncolorectal surgeons (n ¼ 133) were reviewed independently by physicians in the Division of Radiation Oncology. Local recurrence was defined as pelvic recurrence occurring in the presence or absence of distant metastatic disease. Actuarial disease-free survival and local control rates at five years were 77% and 99% for colorectal surgeons versus 68% and 84% for noncolorectal surgeons. Multivariate analysis revealed that pathologic stage and background of the surgeon were the only independent predictors of disease-free survival and that pathologic stage, background of the surgeon, and proximal location of the carcinoma were independent

159

predictors of local control. Sphincter preservation was more common by colorectal surgeons (52%) than noncolorectal surgeons (30%). They concluded good outcome for patients with carcinoma of the rectum is associated with subspecialty training in colon and rectal surgery. Dorrance et al. (908) examined the effect of the surgeon’s specialty on patient outcome after potentially curative colorectal carcinoma surgery and to identify factors that may help explain differences in outcome among specialty groups. In a large teaching hospital, 378 patients underwent potentially curative operation for colorectal carcinoma by surgeons with specialty interests, vascular or transplant, general, and colorectal surgeons. At a median follow-up of 45 months the only factors associated with a significant reduced local recurrence rate were the length of the resection specimen (odds ratio, 0.56) and colorectal specialty. Patients operated on by a general surgeon were 3.42 times more likely to develop a local recurrence than those operated on by a colorectal surgeon. For overall recurrence, early stage disease, absence of vascular invasion, and colorectal specialty were the only factors associated with significantly improved outcome at multivariate analysis. These data show that surgeons with an interest in colorectal carcinoma achieve lower local and overall recurrence rates compared with vascular, transplant, or general surgeons. Martling et al. (927) evaluated the effects of an initiative to teach the TME technique on outcomes at five years after surgery. The study population comprised all 447 patients who underwent abdominal operations for rectal carcinoma in Stockholm County. Outcomes were compared with those in the Stockholm I (790 patients) and Stockholm II (542 patients) radiotherapy trials. The permanent stoma rate was reduced from 60.3% and 55.3% in the Stockholm I and II trials respectively to 26.5% in the TME project. Five-year local recurrence rates decreased from 21.9% and 19.1% to 8.2%, respectively. Five-year carcinoma-specific survival rates increased from 66.0% and 65.7% in the Stockholm trials to 77.3% in the TME project (hazard ratio 0.62). They concluded, a surgical teaching program had a major impact on rectal carcinoma outcome. McArdle and Hole (871) conducted a study to determine whether differences in survival following surgery for colorectal carcinoma were due to differences in caseload or degree of specialization. The outcome in 3200 patients who underwent resection for colorectal carcinoma was analyzed on the basis of caseload and degree of specialization of individual surgeons. Carcinoma-specific survival at five years following curative resection varied among surgeons from 53.4% to 84.6%; the adjusted hazard ratios varied from 0.48 to 1.55. Carcinoma-specific survival rate at five years following curative resection was 70.2%, 62.0%, and 65.9% for surgeons with a high, medium, and low case volume, respectively. There were no consistent differences in the adjusted hazard ratios by volume. Carcinoma-specific survival rate at five years following curative resection was 72.7% for those treated by specialists and 63.8% for those treated by nonspecialists; the adjusted hazard ratio for nonspecialists was 1.35. They concluded the differences in outcome following apparently curative resection for colorectal carcinoma among surgeons appear to reflect the degree of specialization rather than case volume. It is likely that increase specialization will lead to further improvements in survival.

160

& PART I: COLORECTAL DISORDERS

Perioperative Blood Transfusion Clinical and experimental studies indicate that transfusion of blood has immunomodulating properties and that the behavior of some neoplasms may be influenced by the immune system of the host. It has been suggested that blood transfusion in the perioperative period adversely affects the rate of carcinoma recurrence and is even associated with increased mortality (758,759). Leite et al. (760) noted a 5-year survival rate of 37% for patients who were transfused compared with 60% for nontransfused patients. Furthermore, it has been suggested that the number of units blood transfused perioperatively in patients operated on for colon carcinoma has a progressively strong negative influence on survival (761). It has been reported that the incidence of recurrence is higher in those patients who receive transfusion during the operation than in those who receive transfusion either before or after operation; however, the study indicated that factors influencing the need for blood transfusion during the operation had a greater bearing on prognosis than the receipt of the blood per se (762). As with other prognosis discriminants, various authors have presented conflicting views (872,873). To resolve some of the controversy of the degree of immunomodulation by perioperative blood transfusion and its effect on oncologic surgery, Chung, Steinmetz, and Gordon (762) reviewed all studies published between 1982 and 1990 using the statistical method of MantelHaentszel-Peto to determine a cumulative estimate of the direction and magnitude of this association. Some 20 papers were included in the analysis, representing 5236 patients. The cuumulative odds ratios (95% confidence interval) of disease recurrence, death from carcinoma, and death from any cause were 1.80 (1.30 to 2.51), 1.76 (1.15 to 2.66), and 1.63 (1.12 to 2.38), respectively. These results support the hypothesis that perioperative blood transfusion is associated with an increased risk of recurrence of colorectal carcinoma and death from this malignancy. Splenectomy has been considered a possible factor in the survival of patients operated on for colorectal carcinoma (928). The mechanism responseible for this adverse impact is undefined, but it may fall into the category of modulation of the immune response. Others disagree (929). Previous Appendectomy Armstrong et al. (765) studied a series of 519 patients presenting with carcinoma of the cecum in relation to a history with or without previous appendectomy. Previous appendectomy was associated with a higher incidence of local fixity, invasion of the abdominal wall, metastatic spread, and poor differentiation. These differences were reflected in a significantly lower resection rate for carcinomas in patients who had undergone appendectomy. The survival of patients who had previously had appendectomy was significantly reduced. Local recurrence was more common and often was noted to be in the old appendectomy wound itself. In this study appendectomy did not increase the risk of carcinogenesis in the cecum but worsened the prognosis for patients who subsequently developed carcinoma of the cecum.

& PATHOLOGIC FEATURES Numerous pathologic features have been studied in an effort to define and refine the prognosis of a given patient. In a very meticulous and thorough study Newland et al. (732) analyzed data from 579 patients collected prospectively during a follow-up ranging between 6 months and 21.5 years. Six variables showed significant independent effects on survival on multivariate analysis. In diminishing potency, these variables were apical lymph node involvement, spread involving a free serosal surface, invasion beyond the muscularis propria, location in the rectum, venous invasion, and high-grade malignancy. Significant independent effects also were shown for patient age and gender. The number of involved lymph nodes added no significant independent prognostic information. The authors recommend that all six independent variables be included in any future protocol for stratifying this prognostically diverse group of patients. Many of the individual discriminants are discussed below. Location Most reports suggest that rectal carcinoma has a poorer prognosis than colon carcinoma (732,740,742). In the report by Polissar, Sim, and Francis (735) colon carcinoma resulted in a significantly worse prognosis than rectal carcinoma. Newland et al. (732) reported a hazard ratio of 1.53. In contrast, the study by Martin et al. (876) found no difference in 5-year survival rates on the basis of site of the carcinoma. In patients with colon carcinoma, opinion has differed as to whether a right-sided lesion has a better prognosis (740), a left-sided lesion has a better prognosis (674), or the prognoses are equal. Several authors believe that right-sided lesions carry a poorer prognosis (768). Size In contrast to other malignancies, it has been reported that the size of a carcinoma of the colon bears little relationship to prognosis (732,877). Other authors have expressed the opposite opinion (769,770). Configuration The macroscopic features of a colon carcinoma appear to reflect its biologic activity. Polypoid lesions tend not to deeply invade the bowel wall, whereas ulcerating lesions more often penetrate the wall and are associated with a poorer prognosis. Lumen encirclement is a strong prognostic discriminant (742). Rate of survival with full circumferential involvement was found to be 29.7% at 5 years, whereas when less than half of the lumen was involved, the 5-year survival rate was 53.9% (740). Survival and local recurrence are significantly better for patients with exophytic (polypoid and sessile) carcinomas than for those with nonexophytic (ulcerated and flat raised lesions) (912). Exophytic lesions include significantly more stage T1 and fewer T2 and T3 carcinomas, and a significantly smaller proportion of carcinomas that show venous and lymphatic invasion than the nonexophytic lesions. Microscopy Although histologic grading is valuable in assessing prognosis, no uniform system of grading exists. Broders’ grade 1

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

carries a relatively good prognosis, while grades 3 and 4 have a poor prognosis. Approximately one half of patients fall into the grade 2 category, which makes this grading system of limited help. Furthermore, there is frequently a discrepancy between the preoperative and postoperative assessments. In their review, Sugarbaker, Gunderson, and Wittes (422) summarized the findings of many studies and concluded that malignancies of higher grades have less chance for cure than those of lower grades. Other microscopic features associated with a diminished chance for cure include more advanced primary lesions, cases involving increased frequency of venous invasion, increased frequency of distant metastases, increased frequency of perineural invasion, and increased frequency of metastases to lymph nodes. Preoperative biopsy is of limited prognostic value unless a poorly differentiated lesion is present, in which case the likelihood of lymphatic metastases is greater. Newland et al. (732) estimated that the hazard ratio for survival for patients with a high-grade carcinoma was 1.48. Carcinomas that secrete large amounts of mucus are associated with reduced survival. Yamamoto et al. (777) compared the clinicopathologic features of patients with a mucinous carcinoma (6.6% of their patients) to a nonmucinous carcinoma. They found that mucinous carcinomas were more likely to invade adjacent viscera (29% vs. 10%), show lymph node involvement beyond the pericolic region (50% vs. 26%), have a reduced rate of curative resection (34% vs. 69%), have a higher recurrence rate (27% vs. 19%), and result in a poorer 5-year survival rate (33% vs. 53%). Green et al. (878) found that stage for stage, the 5-year overall survival rate was the same for mucinous and nonmucinous carcinomas. However, the 5-year survival rate for mucinous carcinoma of the rectum was decidedly more poor than nonmucinous carcinoma (11% vs. 57%). In a review of 352 patients with colorectal carcinoma followed for a minimum of 5 years, Secco et al. (776) found that mucinous adenocarcinomas represented 11.1% and signet-ring cell carcinomas represented 1.1% of cases. Mucinous carcinomas were most frequently located in the rectum (61.5%) and in the sigmoid colon (15.3%). Patients presented with Dukes’ C and metastatic disease in 41% and 15% of cases, respectively. Disease recurrence was more frequently observed in patients with mucinous (51.7%) or signet-ring lesions (100%) compared with adenocarcinomas. Five-year survival rates were 45%, 28%, and 0% in patients with adenocarcinoma, mucinous adenocarcinoma, or signet-ring cell carcinomas, respectively. Signet-ring cell adenocardnomas (i.e., cells with abundant intracellular mucus) have a particularly poor prognosis. Chen et al. (789) identified 61 signet ring cell carcinoma patients and compared their clinical data and outcomes to those of 144 consecutive patients with nonsignet ring cell mucinous rectal carcinomas and 2414 consecutive patients with nonmucinous rectal carcinomas. The incidence of signet ring cell carcinomas was 1.39% of rectal carcinomas. Mean patient age at onset of signet-ring cell carcinomas (48.1 years) was significantly lower than that for nonsignet ring cell mucinous carcinomas (57.4 years) and nonmucinous carcinomas (62.6 years). The proportion of late stage (TNM III and IV) carcinomas was significantly higher in signet ring cell carcinomas (90%) than in nonsignet-ring mucinous carcinomas (69%) and nonmucinous carcinomas (48%). There were more carcinomas located in

161

the lower rectum in signet ring cell carcinomas (46%) than in nonsignet-ring mucinous carcinomas (34%) and nonmucinous carcinomas (29%). Signet ring cell carcinomas were significantly larger (5.7 cm) than nonsignet cell mucinous carcinoma (4.3 cm) and nonmucinous lesions (3.8 cm). A higher% of patients with signet ring cell carcinoma (42.6%) received abdominoperineal resection for treatment. In carcinomas with TNM stage IV the rate of spread via hematogenous route was significantly lower in signet ring cell carcinomas (18.5%) than in nonsignet cell mucin (43.5%) and in nonmucinous carcinomas (69%). The rate of spread via seeding to the peritoneum was lower in signet cell carcinomas (22.2%) than in nonsignet cell mucin carcinomas (43.5%) but higher than in nonmucinous carcinomas (2.7%). The rate of spread via the lymphatic route was higher in signet cell ring carcinomas (44.4%) than in nonsignet cell mucinous carcinomas (26.1%) and significantly higher than in nonmucinous carcinomas (12.3%). The 1-, 2-, and 5-year overall signet cell carcinoma survival rates were 73.9%, 36.3%, and 23.3%, respectively, which were significantly poorer than those of nonsignet cell mucin carcinomas and nonmucinous carcinomas. For the signet cell carcinomas, the 1-, 2-, and 5-year disease-free survival rates of signet cell carcinoma were 84%, 44.2%, and 30.3%, respectively which are comparable with general data of stage III rectal carcinoma in the world. They concluded diffuse infiltration of signet ring cells enhances the tendency of mucinous carcinomas of the rectum in more local extension and easier lymphatic spreading but not at peritoneal seeding. Nissan et al. (790) compared 46 patients with signet ring cell carcinomas with 3371 patients with primary nonsignet ring cell carcinomas. Lymphatic and peritoneal spread was more common among the signet cell ring carcinoma group. Approximately one-third of signet ring cell carcinoma patients presented with metastatic disease. Mean survival time of signet ring cell carcinoma group was 45.4 months compared with 78.5 months for the control patients group. The cumulative survival curve of patients with signet ring cell carcinoma resembles that of patients with poorly differentiated rectal carcinomas. The pathologic diagnosis of scirrhous carcinoma of the large bowel carries a very poor prognosis. Extent of penetration of the bowel wall is associated with a reduced prognosis. Greater extramural spread is associated with an increased incidence of nodal involvement (877). Hase et al. (786) examined 663 specimens from patients who underwent curative resection for colorectal carcinoma and identified small clusters of undifferentiated malignant cells ahead of the invasive front of the lesion, which they labeled ‘‘tumor budding.’’ The presence of this feature resulted in a diminished 5-year survival rate (22% vs. 71%). In a study of 138 patients, Tanaka et al. (787) reported recurrence in 48% of patients with ‘‘tumor budding’’ compared with 4.5% without this histologic feature. Cumulative disease-specific survivals at five years were 74% and 98%, respectively. In a review of 196 resected stage II and III colon carcinomas, Okuyama et al. (788) found budding detected significantly more frequently in lesions with lymph node metastases (stage III) than in lesions without it. Patients with budding-positive lesions had worse outcome than those with budding-negative lesions with 50.6% with

162

& PART I: COLORECTAL DISORDERS

budding-positive lesions and 8.1% with budding-negative lesions developing recurrence. Patients with budding-positive lesions had a worse prognosis than patients without it. Moreover, no significant difference in survival curves was observed between patients with budding-positive stage II lesions and those with stage III lesions. An additional histologic feature that may correlate with prognosis is the presence or absence of an inflammatory infiltrate. For patients whose resected specimens demonstrated infiltration of lymphocytes around blood vessels, together with hyperplasia of paracortical regions in lymph nodes, Pihl et al. (671) found a 5-year recurrence-free interval of 85%, while patients who lacked these characteristics exhibited a survival rate of 69%. Gagliardi et al. (785) studied the relationship between acinar growth patterns in 138 patients with rectal carcinoma and survival. Lesions were classified according to size, 28 microacinar and 110 macroacinar. Patients with microacinar (small regular tubules) had a significantly reduced 5-year survival rate compared with those with macroacinar (large irregular tubules) lesions (43% vs. 68%).

Residual Disease Local residual disease predicts poor patient survival after resection for colorectal carcinoma. Chan et al. (792) determined the prevalence of residual carcinoma in a line of resection in a large prospective series and identified other pathology variables that may influence survival in the absence of distant metastases. The overall prevalence of residual carcinoma in a line of resection was 5.9%. Of 12 pathology variables examined, only high grade and apical node metastases were independently associated with survival in the subset of 120 patients with residual disease in a line of resection but without distant metastases. The 2-year survival rate for patients with neither of these adverse features was 46.4% as compared with only 7.7% in those who had both. Dukes’ Staging The numerous eponymous modifications of the Dukes classification have not allowed any meaningful comparison to be made from one report to another. However, within each of these modifications a more advanced stage represents a poorer prognosis. If a staging system of local, regional, and distant categories were adopted, it might permit comparison and agreement on survival of a given category of patient. Histologic activity offers a means of estimating biologic behavior, and this association has been reflected in the Dukes classification. The effect of Dukes’ staging on survivorship was reported from the Lahey Clinic in Boston in a study of 344 patients treated for colorectal carcinoma. The uncorrected 5-year survivorship for Dukes’ A, B, and C patients was 85%, 65%, and 46%, respectively, while the corresponding corrected values were 100%, 78%, and 54% (740). Newland et al. (732) conducted a multivariate survival analysis on the depth of invasion. For spread beyond the muscularis propria, the authors found a hazard ratio of 1.68, and for spread involving the free serosal surface, the hazard ratio was 1.71. In another report, the same

authors found that the survival rates of patients with clinicopathologic stages A or B closely matched their expected survival as predicted from the general population (930). Males with stage B carcinomas were the only exception and their reduced survival rates were due to four clinical variables (cardiovascular complications, permanent stoma, urgent operation, or respiratory complications) and one pathologic variable (direct spread involving a free serosal surface). Greene et al. (751) proposed a new TNM staging strategy for node positive colon and rectal carcinoma because the current stage III designation of colon carcinoma excludes prognostic subgroups stratified for mural penetration (T1-4) or nodal involvement (N1 vs. N2). They analyzed 50,042 patients with stage III colon carcinoma reported to the National Cancer Data Base. Three distinct subcategories with a traditional stage III cohort of colon carcinoma were identified—IIIA: T1/2, N1; IIIB: T3/4, N1; and IIIC: any T, N2. Five-year observed survival rates for these three subcategories were 59.8%, IIIA; 42.0%, IIIB; and 27.3%, IIIC. Analysis of this large data set supports stratification into three subsets, confirming the benefit of adjuvant chemotherapy in each subgroup. They subsequently analyzed data entered in the National Cancer Data Base for 5987 stage III patients with rectal carcinoma (931). Five-year observed survival rates for stage III subcategories were 55.3% in IIIA; 35.3% in IIIB; and 24.5% in IIIC. Stratifying for treatment outcome, stage IIIA patients having operation alone had poorer observed 5-year survival (39%) than patients treated with operation and adjuvant chemotherapy or radiation therapy (60%). Similar outcomes occurred in IIIB (operation alone 21.7% and chemo/radiotherapy 40.9%) and in IIIC (operation alone 12.2% and chemo/radiotherapy 28.9%). The effect of postoperative adjuvant therapy was beneficial in all subsets.

Lymph Node Status The most important prognostic variable in colon carcinoma is the presence or absence of lymph node metastases. Patients with colorectal carcinoma found to have regional lymph node metastases after curative resection form a large and prognostically diverse group. In studies focusing on the level of lymph node involvement and the number of affected lymph nodes, the number of lymph nodes involved correlates with survival in some reports. In a report by Corman, Veidenheimer, and Coller (7), if more than three lymph nodes were positive, the overall survival rate was 18%, but with one to three nodes involved, the survival rate was 45% to 50%. At St. Mark’s Hospital in London, the 5-year survival rate is approximately 60% when one node is affected, 35% when two to five nodes are involved, and 20% when six or more nodes are affected (230). An analysis of the NSABP clinical trials by Wolmark, Fischer, and Wieand (784) revealed a relative risk of death of 1.9 and 3.4 for patients with one to four and five to nine positive lymph nodes, respectively. Gardner et al. (792) also found that prognosis worsened with an increased number of involved lymph nodes. Patients who had six or more involved nodes were 4.6 times as likely to die from the disease as patients with only one involved node. Cohen et al. (771) reported that when one to three nodes were involved, the 5-year

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

survival rate was 66%, but was reduced to 37% when four or more nodes were positive. Similarly, Tang et al. (795) found that the number of lymph nodes involved had an impact on survival. In a retrospective study of 538 patients, the 5-year survival rate for patients with one to three positive nodes was 69%, 44% for four to nine positive nodes, and 29% for 10 or more positive nodes. On the other hand, Newland et al. (733) reported that the number of involved lymph nodes added no significant independent prognostic information. Most potent in their analysis was apical node involvement, which they calculated to have a hazard ratio of 1.79 in their multivariate survival analysis. Such involvement was present in 9% of their node-positive patients. Malassagne et al. (794) found that both the number of lymph nodes involved as well as apical node involvement were prognosticators. The 5-year survival rates were 17% and 45% for patients with and without apical lymph node involvement, respectively, and 44% and 6% for those with four or fewer nodes involved compared with those with more than four positive nodes, respectively. Swanson et al. (932) examined data from the National Cancer Data Base to determine whether the number of examined lymph nodes is prognostic for T3N0 colon carcinoma. A total of 35,787 prospectively collected cases of T3N0 colon carcinomas that were surgically treated and pathologically reported as T3N0M0 were analyzed. The 5-year relative survival rate for T3N0M0 colon carcinoma varied from 64% if one or two lymph nodes were examined to 86% if > 25 lymph nodes were examined. Three strata of lymph nodes (1–7, 8–12, and 13) distinguished significantly different observed 5-year survival rates. These results demonstrate that the prognosis of T3N0 colon carcinoma is dependent on the number of lymph nodes examined. A minimum of 13 lymph nodes should be examined to label a T3 colon carcinoma as node negative. Fisher et al. (933) examined the presumably negative nodes of a larger cohort of patients for what they designated nodal mini micrometastases on parameters of survival. Mini micrometastases were detected by immunohistochemical staining of the original lymph node sections with anticytokeratin A1/A3 in a total of 241 Dukes’ A and B patients with rectal and 158 with colonic carcinoma. Nodal mini micrometastases were detected in 18% of patients in this cohort but this additional finding failed to exhibit any significant relationship to overall recurrence free survival. Other reports have recorded mini micrometastases in 19% to 39% of cases and this probably relates to the number of sections taken at the time of study. Sakuragi et al. (934) sought predictive markers of lymph node metastases to assist in management of 278 T1 colorectal carcinomas. Depth of submucosal invasion and lymphatic channel invasion were accurate predictive factors for lymph node metastases. The authors believe these two factors could be used in selecting appropriate cases for operation after endoscopic resection. Tepper et al. (935) analyzed data from 1664 patients with T3, T4, or node positive rectal carcinoma treated in a National intergroup trial of adjuvant therapy with chemotherapy and radiation therapy to assess the association between the number of lymph nodes found by the pathologist in the surgical specimen and the time to relapse and survival outcomes. No significant differences were found by quartiles among patients determined to be node

163

positive. Approximately 14 nodes need to be studied to define nodal status accurately. Examining greater number of nodes increases the likelihood of proper staging.

Venous Invasion In a personal study of more than 1000 operative specimens, Morson and Dawson (230) found regional venous involvement in 35% of cases. Submucosal venous spread occurred in 10%, and in 25% there was evidence of permeation of extramural vessels. In the former cases, there was little or no effect on prognosis, but extramural venous involvement reduced 5-year survival rates from 55% to approximately 30%. In the report by Corman, Veidenheimer, and Coller (7), patients with Dukes’ C lesions and blood vessel invasion had a 31% 5-year survival rate compared with 43% without blood vessel invasion. Comparable figures for Dukes’ B lesions were 55% and 70%, respectively. The former% is not significant, but the latter reached statistical significance. A subsequent report from the Lahey Clinic found a difference with and without blood vessel invasion (740). Minsky et al. (803) reviewed a series of 168 patients who underwent potentially curative resection. The authors found that patients who had extramural blood vessel invasion had a significantly decreased 5-year survival rate compared with patients who had intramural blood vessel invasion or no vascular invasion at all. When extramural and intramural invasion were combined, the difference disappeared. Krasna et al. (804) found that the 3-year survival rate decreased from 62.2% in patients without vascular invasion to 29.7% in patients demonstrating vascular invasion. In 128 operative specimens, Horn, Dahl, and Morild (800) identified venous invasion in 22%. The 5year survival rate in those with venous invasion was 32.9% vs. 84.3% for those without venous invasion. Newland et al. (732) identified venous invasion in 28.8% of their series, and in 81% of those patients extramural veins were involved. The authors calculated a hazard ratio of 1.49 in the multivariate survival analysis. In a very thorough study Sternberg et al. (801) investigated venous invasion as a predictor of prognosis in colorectal carcinoma. The reported incidence of venous invasion in colorectal carcinoma specimens varies between 10% and 89.9%, mainly as a result of the interobserver variability and differences in specimen processing. Their study goal was to assess and compare the incidence of venous invasion diagnosed on H&E-stained tissue versus tissue stained with both H&E and an elastic fiber stain. Venous invasion was assessed on sections from 81 colorectal carcinomas resected from patients with synchronous distant metastases. Only stage IV carcinomas were studied for the following reasons: (1) it can be assumed that in all patients with distant hematogenous metastases venous invasion had occurred, thus enabling the false negative rate to be calculated; (2) there can be no dispute about the clinical relevance of the various characteristics of venous invasion identified in the carcinomas of patients with synchronous distant hematogenous metastases; and (3) to eliminate the effect of variance in carcinoma staging on the incidence of venous invasion. Initially, H&E-stained sections were studied for venous invasion. Sections that were negative or questionable with regard to venous invasion were then stained with an elastic fiber stain and

164

& PART I: COLORECTAL DISORDERS

a second final search for venous invasion was carried out. Venous invasion was identified in 51.9% on H&E-stained sections. The addition of the elastic fiber stain enabled the diagnosis of venous invasion in 38.5% of the remaining specimens, increasing the overall incidence to 70.4%. Of the 57 positive specimens, venous invasion was minimal in 47.4%, intermediate in 8.8%, and massive in 43.9%. Only intramural veins were involved in 31.6%, only extramural veins in 45.6%, and both intramural and extramural veins in 22.8% of the positive specimens. The filling type of venous invasion was found in 71.9%, the floating type in 49.1%, and the infiltrating type in 10.5% of the positive specimens. There was no significant difference between the incidence of venous invasion in the colon (70%) versus rectal and rectosigmoid carcinomas (71.4%) nor in the incidence of venous invasion in patients with hepatic (70%) versus nonhepatic (72.7%) metastases. Only minimal venous invasion is required for the seeding of clinically relevant hematogenous metastases, and emphasizes the careful dedicated search for venous invasion that is required from the pathologist. Although extramural venous invasion was predominant in stage IV colorectal carcinomas, in a third of lesions only intramural venous invasion was found. This suggests that intramural venous invasion may also seed clinically relevant hematogenous metastases, and should therefore also be considered as an indicator of poor prognosis.

Perineural Invasion Perineural invasion has been shown to have a detrimental effect on prognosis. Its presence may be part of the overall penetration of the bowel wall. The association of disseminated disease has been reported (799). Krasna et al. (804) found that the 3-year survival rate decreased from 57.7% in patients without neural invasion to 29.6% in patients with neural invasion. Of 128 operative specimens examined by Horn, Dahl, and Morild (800), neural invasion was demonstrated in 32%. The 5-year survival rate in patients with neural invasion was 64.3% compared with 81.1% when neural invasion was not demonstrated. Ueno et al. (807) investigated perineural invasion in 364 patients who underwent curative resection for rectal carcinoma penetrating the muscular layer. A grading system was established based on the ‘‘intensity’’ (number of perineural invasion foci in a 20-power field) and ‘‘depth’’ (distance from the muscularis propria). PNI-0 was defined as without perineural invasion, PNI-1 as intensity of less than five foci and depth less than 10 mm, and PNI-2 as five or more foci or 10 mm or greater depth of invasion. Perineural invasion was observed in 14% and strongly correlated with pathological lymph node metastases. Five-year survival was related to the perineural invasion (74% in PNI-0, 50% in PNI-1, and 22% in PNI-2). The rate of local recurrence was also related to PNI stage: 43% in PNI-2 and 9% in PNI-0 and PNI-1. The PNI grading system may be useful in prognosis and may allow case selection for intensive postoperative adjuvant therapy. & BIOCHEMICAL AND SPECIAL INVESTIGATIONS Preoperative CEA Levels An analysis of data from 945 patients entered into the NSABP revealed a strong correlation between preoperative

CEA levels and the Dukes’ classes (813). The mean CEA level progressively increased with each Dukes’ category, and the mean value for each of the four classes was significantly different. Mean values ( SE) for Dukes’ A, B, C, and D (metastatic or contiguous disease) were 3.9  0.6, 9.3  1.4, 32.1  8.9, 251  84, respectively. The prognostic function was independent of the number of positive histologic lymph nodes and unrelated to the presence or absence of obstruction. Preoperative levels correlated with the degree of lumen encirclement by the carcinoma, with lesions involving more than one half the circumference being associated with significantly lower preoperative CEA levels. The relative risk of developing a treatment failure was associated with preoperative CEA in both Dukes’ B and C patients. For patients with Dukes’ B lesions, those with a CEA of 2.5 to 10 had 1.2 times the likelihood of developing a recurrence as those with less than 2.5, while those with a CEA level greater than 10 had 3.24 times the likelihood. For patients with Dukes’ C lesions the respective risks were 1.77 and 1.76. Although some authors have come to the same conclusion (813), others have found the correlation true only for patients with Dukes’ C lesions (812). An important caveat to note is that poorly differentiated carcinomas produce little CEA, and therefore a normal preoperative CEA in a patient with a poorly differentiated carcinoma does not suggest a favorable prognosis (817). Several authors have found CEA determinations to be of little prognostic significance (881). However, a clearer relationship has been demonstrated between persistently elevated levels of CEA in the postoperative period and early recurrence (936).

Liver Function Tests Abnormal results of liver function tests have been associated with a poor prognosis (818), but not uniformly so (747). Other Blood Tests Low serum protein levels have been reported to be associated with a poor prognosis. Factors found to have no effect on prognosis include hemoglobin level, white blood count, and erythrocyte sedimentation rate (747). DNA Distribution Several studies have reported that patients who exhibit an abnormal DNA pattern, that is, other than the normal diploid pattern, suffer a higher recurrence rate (819,822,937,744). Giaretti et al. (938) found DNA aneuploidy present in 31% of adenomas and 74% of adenocarcinomas. DNA ploidy correlated with the size and the degree of dysplasia but not with histologic type. From the same center, it was determined that fresh-frozen material gave a higher incidence of DNA aneuploidy than paraffin-embedded material (79% vs. 41%) (938) Armitage et al. (819) found that 55% of their patients had cells with abnormal DNA (aneuploid). Of this group, only 19% of patients survived 5 years, compared with 43% of patients with diploid neoplasms. In contrast, Jones, Moore, and Schofield (882) found that after the surgeon’s assessment of operability, the pathologic classification, and the patient’s age were considered, the DNA ploidy status conferred no

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

independent survival value. Halvorsen and Johannesen (939) reported a significant survival advantage in patients with diploid lesions compared to those with nondiploid lesions but no difference between carcinoma of the rectum and colon. The authors concluded that ploidy does not contribute to the explanation of why patients with rectal carcinoma had a poorer prognosis than those with colon carcinoma. Venkatesh, Weingart, and Ramanujam (823) compared two parameters in DNA analysis and found that the odds of survival were 3.7 times greater in patients with aneuploidy rather than aneuploidy plus an S-phase fraction > 20%.

Genetic Alteration Among the most recent factors considered in the galaxy of prognostic discriminants are molecular genetic alterations (826). Fractional allelic loss, a measure of allelic deletions, has provided independent prognostic information. Distant metastases were significantly associated with high fractional allelic loss and with deletions of 17p and 18q. Further associations were found between allelic losses and a family history of carcinoma, left-sided location of a carcinoma, and absence of extracellular mucin. Jen et al. (827) reported that patients with stage II disease have a 5-year survival rate of 93% when the carcinoma has no evidence of allelic loss of chromosome 18q, but only 54% when there was an allelic loss. In patients with stage III disease, 5-year survival is 52% without allelic loss and 38% with loss. An overall hazard ratio for death in patients with allelic loss of chromosome 18q is 2.83. It has also been reported that patients with CD44 v6-positive carcinomas have a poorer prognosis than those with negative lesions (839). Overexpression of ras p21 was reportedly associated with an increased incidence of lymphatic invasion, depth of invasion, incidence of liver metastases, and decreased operative curability and long-term survival (840). Overall expression of p53 has been found to be an independent predictor of recurrence in Dukes’ B and C carcinomas by some authors (832,833) and not a predictor by others (890). Auvinen et al. (829) reported patients with p53 overexpression had a corrected 5-year survival rate of 37% compared with that of 58% in patients with normal expression. Corresponding 10-year rates were 34% and 54%, respectively. TP53 gene mutations (topographic genotyping) have been associated with decreased survival (831). Shibata et al. (828) found that in patients with stage II disease whose carcinoma expressed DCC, the 5-year survival rate was 94.3%, whereas in patients with DCC-negative lesions, the survival rate was 61.6%. In patients with stage III disease, the survival rates were 59.3% and 33.2%, respectively. Wang et al. (841) used DNA chip technology to systematically identify new prognostic markers for relapse in Dukes’ B carcinoma patients. Gene expression profiling identified a 23-gene signature that predicts recurrence in Dukes’ B patients. The overall performance accuracy was 78%. Thirteen of 18 relapse patients and 15 of 18 diseasefree patients were predicted correctly, giving an odds ratio of 13. The clinical value of these markers is that the patients at a high predictive risk of relapse (13-fold risk)

165

could be upstaged to receive adjuvant therapy, similar to Dukes’ C patients. A series of other genetic alterations proposed as prognostic factors in colorectal carcinoma have been cited (230). Lim et al. (917) analyzed the association between MSI status and clinicopathological features and prognosis in 248 sporadic colorectal carcinoma patients of which 9.3% had MSIþ carcinomas. MSIþ sporadic colorectal carcinomas were found predominantly in the proximal colon and were associated with poor differentiation, a lower preoperative serum CEA level, and less frequent systemic metastases than MSI carcinomas. Low grade, low T-stage, no lymph node metastases, no systemic metastases, adjuvant chemotherapy, and MSIþ status were independent favorable prognostic factors for survival in sporadic colorectal carcinoma patients. Kohen-Corish et al. (940) undertook a detailed analysis of the prognostic significance of MSI-L and loss of MGMT protein expression in colon carcinoma in 183 patients with clinicopathologic stage C colon carcinoma who had not received adjuvant therapy. They showed that MSI-L defines a group of patients with poorer survival than MSS patients and that MSI-L was an independent prognostic indicator in stage III colon carcinoma. Loss of MGMT protein expression was associated with the MSI-L phenotype but was not a prognostic factor for overall survival in colon carcinoma. p-16 methylation was significantly less frequent in MSI-L than in MSI-H and MSS carcinomas and was not associated with survival. To derive a more precise estimate of the prognostic significance of MSI, Popat et al. (918) reviewed and pooled data from 32 eligible studies that reported survival in a total of 7642 cases, including 1277 with MSI. There was no evidence of publication bias. The combined hazard ratio estimate for overall survival associated with MSI was 0.65. This benefit was maintained restricting analyses to clinical trial patients (HR ¼ 0.69) and patients with locally advanced colorectal carcinoma (HR ¼ 0.67).

Sialomucin Staining Oncogenic transformation of colonic epithelium is accompanied by increased secretion of sialomucin at the expense of the normally predominant sulfomucins. For patients who exhibit a sialomucin-predominant pattern, there is an increased incidence for local recurrence and a predicted diminution of 5-year survival (850). Nuclear Morphometry In search of a reliable prognostic discriminant, Mitmaker, Be´gin, and Gordon (853) used nuclear morphometry to assess 100 cases of colorectal carcinoma in which patients who underwent curative resection were followed for at least 5 years. Each case was staged according to the Dukes’ classification and graded histologically. The nuclear shape factor was defined as the degree of circularity of the nucleus, with a perfect circle recorded as 1.0. A nuclear shape factor > 0.84 was associated with a poor outcome. This variable proved to be a highly significant predictor of survival and independent of the variables of sex, age, histologic grade, and Dukes’ classification.

166

& PART I: COLORECTAL DISORDERS

Plasminogen Activity Studies of tissue plasminogen activity have revealed that overall survival curves are related to the ratio of urokinasetype plasminogen activators to tissue-type plasminogen activators (854). A ratio > 0.22 in normal mucosa of patients with Dukes’ B and C carcinoma has a decreased probability of survival with a Cox’s hazard ratio of 2.8. Sialyl Lewisx Antigen Expression (Slex) Based on data of 114 patients who underwent curative resections, sialyl Lewisx antigen–positive patients had a higher incidence of recurrence in distant organs, especially in the liver, than that of sialyl Lewisx–negative patients (845). The 5-year disease-free survival rates of sialyl Lewisx–positive and –negative patients were 57.7% and 89.1%, respectively. PCNA Labeling Findings Proliferating cell nuclear antigen expressions of the invasive margin of a carcinoma was shown to be significantly higher in patients who were noted to have venous invasions, a higher potential for metastases to lymph nodes and liver, and less differentiated lesions (848). An effort has been made to tabulate the various prognostic discriminants that have been considered important (see Table 15). At a glance, the reader will recognize the controversy engendered in the efforts to sort out the important factors.

& RECURRENT DISEASE & FOLLOW-UP The most appropriate follow-up for patients who have been operated on for carcinoma of the colon has not been determined. Any follow-up program should focus on the detection of resectable anastomotic and locoregional failure, liver and lung metastases, and metachronous lesions. In a retrospective analysis of 5476 patients with colon or rectal carcinoma, Cali et al. (941) calculated the annual incidence for metachronous carcinomas to be 0.35%. Current recommendations for follow-up are described in Chapter 25. However, the wisdom of any type of follow-up has been questioned. The rationale for such a stance is the belief that little effective therapy is available when recurrences develop. Opponents argue that intensive follow-up is not worth the effort and expense, since as many as 62% of new lesions are detected when the patient presents with symptoms between scheduled follow-up sessions (936,942– 944). Proponents of intensive follow-up state that if a recurrent or metachronous lesion is detected when a patient is asymptomatic, the probability of cure by repeat resection will be increased, since the newly detected lesion will be at a more favorable stage than if the patient were symptomatic (945,946). In support of the latter argument, Buhler et al. (945) reported on a series in which patients with asymptomatic anastomotic recurrence had a re-resection rate of 66%, with a survival rate of 12 to 72 months, whereas none of the patients who were symptomatic had a resection for cure and the survival rate in this group was 1 to 24 months. In a report in which 1293 patients were

rigorously followed, 299 recurrences were detected in 168 patients (local recurrence, 40%; liver metastases, 29%; and other, 31%) (946). Of these patients, 51% with local recurrence and 47% with liver metastases were asymptomatic. Radical operation was performed in 50% of those with local recurrences and in 26% of those with liver metastases. The 3-year survival rate after reoperation was 35% in those with local recurrence and 33% in those with liver metastases. The 5-year survival rates were 23% and 15%, respectively. These results demonstrate the benefit of aggressive follow-up. Yamamoto et al. (947) reported the results of 974 patients who underwent a curative resection. Recurrence developed locally in 7.2%; liver, 4.8%; and lung, 3.6%. The%s of patients who underwent reoperation or curative resection were 77% and 24% of those with local recurrence, 34% and 38% of those with liver metastases, and 17% and 100% of those with pulmonary metastases, respectively. The 3- and 5-year survival rates were 13% and 9% after reoperation for local recurrence, 14% and 0% for liver metastases, and 53% and 53% after reoperation for pulmonary metastases, respectively. In an extensive review by Wade et al. (578) of 22,715 patients who underwent colectomy for carcinoma, 12,150 presented with metastatic disease. The estimated surveillance costs averaged $1.3 million per life saved by resection, or $203,000 per year of added life. Despite the apparent high price tag for postoperative studies, the authors believe that surveillance should continue. The costs of eliminating surveillance after curative colectomy would be paid every year by the patients who would die annually with recurrent carcinoma of the colon and rectum, by those who would lose 20 to 28 months of added life gained on average with resection of their isolated colorectal metastases, and by the patients whose cure would be sacrificed. There is a need to determine which tests and regimens can best identify metastatic disease at an early enough stage to allow curative treatment for those who will benefit from it.

& INCIDENCE In an excellent review, Devesa, Morales, and Enriquez (688) found that the incidence of recurrence varied because of multiple biases of classification and treatment, different methods of determining recurrence, and statistical manipulation. From their study it was noted that all series quote the incidence of recurrence by Dukes’ staging or by some modification of Dukes’ staging (e.g., Dukes’ A from 0% to 13%, Dukes’ B from 11% to 61%, and Dukes’ C from 32% to 88%). Although anastomotic recurrences are not uncommon with low anterior resection, such a development following a right hemicolectomy or intraperitoneal anastomosis is considered a rare entity. & CONTRIBUTING FACTORS Sugarbaker, Gunderson, and Wittes (422) have suggested several possible reasons that ‘‘curative’’ resections fail: (1) metastases in the lymphatic channels or nodes may result in unrecognized residual carcinoma, (2) malignant cells may be exfoliated from the primary lesion into veins prior to or during the operation, (3) malignant cells may persist at the circumferential margins of resection, and (4) malignant cells may be disseminated at the time of resection. Implantation is more likely to result in the development

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

of early recurrence at the suture line, whereas metachronous carcinogenesis provides a likely explanation for the development of late recurrences (948). Indeed, all four mechanisms may play a role to a lesser or greater extent.

& PATTERNS Depending on the location of the original resection, patterns of recurrence vary with respect to the development of local, anastomotic, regional, or distant failure as well as the time to recurrence (949,950). The definition of local recurrence generally includes recurrence in areas contiguous to the bed of the primary resection or recurrence at the site of anastomosis. Distant spread represents metastases to sites beyond the location of resection. Local pelvic failure is common in rectal carcinoma because of narrow radial margins defined by the anatomic limits of dissection. Colon carcinoma tends to fail in the peritoneal cavity, the liver, or distant sites, with a relatively small component of isolated local failure. This pattern explains the emphasis on radiotherapy as adjuvant therapy for rectal carcinoma and systemic chemotherapy for colonic carcinoma. In a review of several series, Devesa, Morales, and Enriquez (688) found that 30% to 50% of patients with recurrence of colon carcinoma manifested locoregional failure. Distant metastases are present in up to 80% of patients with recurrence. The liver is most often involved in 50% to 80% of autopsy studies, followed by lung, bone, and other sites. In a follow-up of 487 patients for a median of 48 months (range, 15 to 132 months) Bohm et al. (951) documented recurrence in 31%. Of those, distant metastases were found in 51%, only local recurrence in 31%, and both local and distant metastases in 18%. In the review by Obrand and Gordon (952) the reported recurrence rates after curative resection of large bowel adenocarcinoma varied widely from 3% to 50%. The patterns of recurrence from several selected series are presented in Table 16. It is often difficult to separate series of colon carcinoma from those of rectal carcinoma because the reports often are combined. Rodriguez-Bigas et al. (953) conducted a retrospective analysis of the prognostic significance of anastomotic recurrence in 50 patients with colorectal carcinoma. All

167

carcinomas were located above 10 cm from the anal verge. Forty anastomotic recurrences (80%) followed resection of sigmoid or proximal rectal lesions. The overall disease-free interval was 13 months, with 90% of recurrences diagnosed within 24 months of the primary resection. Forty-five recurrences (90%) were associated with synchronous or metachronous metastases. The overall median survival rate following the recurrence was 16 months (37 months if the anastomosis was the only recurrence site). Of five patients alive without evidence of disease, all were asymptomatic, and recurrence was confined to the anastomosis. The authors concluded that anastomotic recurrence following resection of colorectal carcinoma frequently heralds disseminated disease but can be potentially resected for cure if it is the only site in an otherwise asymptomatic patient. Willett et al. (694) reported that in patients with obstructing lesions, local failure developed in 42%, approximately, one third of whom had local failure only. In patients with perforated carcinoma, 44% developed local failure. The incidence of local failure and distant metastases in their control group was 14% and 21%, respectively. With respect to the time frame, Russell et al. (949) found that 70% of all recurrences of colon carcinoma were detected within 2 years of operation and 90% were detected within 4 years. A review of the literature by Devesa, Morales, and Enriquez (688) noted that 60% to 84% of recurrences became apparent within 2 years of the initial operation and 90% within 4 years (median, 22 months). With respect to location, Malcolm et al. (957) found overall recurrence rates of 24%, 10%, 11.5%, and 34%, respectively, for carcinoma of the right, transverse, left, and sigmoid colon. In a study to determine the incidence and patterns of recurrence after curative resection of colorectal carcinoma, Obrand and Gordon (952) conducted a retrospective review of 524 patients, 448 operated on with curative intent. The overall recurrence rate was 27.9%. The anastomotic recurrence rate was 11.7%. Locoregional recurrence rates, including anastomotic recurrences, were higher in patients with rectal lesions than colon lesions (20.3% vs. 6.2%). Distant metastases developed in 14.4% of patients (13.9% for colon carcinoma and 15.5% for rectal carcinoma). The average time to recurrence was 21.3 months (median, 17 months; range, 2 to 100 months). The average time for anastomotic

TABLE 16 & Patterns of Recurrence Following Curative Resection for Colon Carcinoma Recurrence (%) No. of Patients

Duration of Follow-Up (Yr)

Olson et al. (952) (1980) Malcolm et al. (893) (1981) Boey et al. (953) (1984) Russell et al. (948) (1984) Umpleby et al. (674) (1984) Willett et al. (954) (1984) Gunderson, Sosin, and Levitt (955) (1985) Galandiuk et al. (956) (1992)a

214 191 146 550 329 533 91 818

5.0 5.0

Obrand and Gordon (951) (1997)a

448

Author(s)

a

Includes colon and rectal carcinoma. 47% alive at average of 80 months.

b

4.0 5.0 5.0 2.0–11.5 1–15 (median, 70 mo)

Local

Distant

7 1 10 5 18 6 19 5

22 15 19 22 12 20 34

13

13

Local and Distant 16 5 10 10 7 13

Total

4

23 28 35 34 47 31 39 43

2

28

Time to Recurrence (Mo) 1–102 4–40 2–102

5-Year Survival (%) 49 21

27 63 0.5–98 (median, 17) 2–100 (median, 17)

47b

168

& PART I: COLORECTAL DISORDERS

recurrence was 16.2 months vs. 22.9 months for distant disease and 18.9 months for regional recurrence. Colon recurrences occurred at a median of 16 months vs. 17 months for rectal recurrences. Patients with Dukes’ A lesions had a 17.6% recurrence rate, those with Dukes’ B had a 23.4% rate, and those with Dukes’ C had 43.7%. Patients who did not undergo any intervention after diagnosis of recurrence survived an average of 28 months. Those who received palliative treatment survived an average of 39 months. Of those patients who underwent reoperation, 24% had re-resection for cure. Anastomotic reresections accounted for 20 of 30 resections. A majority of recurrences (69.4%) occurred within 24 months of the original operation and 95% recurred by 48 months. For those who received adjuvant therapy, the mean and median times to recurrence were 25.4 and 22.0 months, respectively. For those patients who did not receive any adjuvant treatment, the mean and median times to recurrence were 19.8 and 16.0 months, respectively. Neither of these reached statistical significance on multivariate analysis. Forty-seven percent of these patients were alive at a mean 8 months. Those who died of their disease did so at an average of 53 months. Positive predictive factors for recurrence included the site of the lesion (rectum vs. colon), stage, invasion of contiguous organs, and presence of perforation. Age, sex, degree of differentiation, mucin secretion, and gross morphology were not found to be predictive factors. Risk factors previously associated with increased recurrence rates include patient’s sex, age, Dukes’ stage, site of primary carcinoma (colon vs. rectum), infiltration of adjacent organs, perforation, and histology and size of the carcinoma, among others. Adverse prognostic factors reported by Galandiuk et al. (957) include males doing worse than females, rectum worse than colon, Dukes’ C worse than B, grades 3 and 4 worse than 1 and 2, adhesions and/or invasion worse than none, perforation worse than none, nondiploid worse than diploid. The clinical relevance of seeking factors, capable of predicting recurrence is to permit the physician to focus on subsets of patients who might most appropriately be targeted for aggressive adjuvant therapy and postoperative surveillance programs that will expedite the diagnosis of recurrent disease at a time when potentially curative therapy can be instituted. This is true even for patients who develop metastatic disease such as liver or lung metastases where curative resections are still favorable in selected circumstances. Disease recurrence in the abdominal wall from primary colorectal carcinoma has received renewed attention after the recognition of port site metastases in patients undergoing laparoscopic colorectal resections. Koea et al. (958) reviewed 31 patients presenting to Memorial SloanKettering Cancer Center with recurrent disease in the abdominal wall between 7 and 183 months after operation. Primary carcinomas were located in the right colon in 17 patients, left colon in 2 patients, sigmoid colon in 7 patients, and rectum in 3 patients. Nineteen percent of primary carcinomas were perforated, 45% were poorly differentiated, 92% were transmural (T3 or T4), and 51% had lymph node metastases at presentation. Twenty-two patients presented with a symptomatic abdominal wall mass, whereas recurrence in the abdominal wall was found incidentally in nine patients undergoing laparotomy. Four patients had isolated

abdominal wall disease whereas the remaining 27 were found to have associated intra-abdominal disease. Six patients who were left with residual intra-abdominal carcinoma after abdominal wall resection had a median survival of four months. Twenty-five patients underwent a histologically complete resection of recurrence restricted to the abdominal wall alone (n ¼ 4; median survival time, 18 months), abdominal wall and in continuity resection of adherent viscera (n ¼ 15; median survival time, 12.5 months), or resection of abdominal wall and intra-abdominal recurrence at a distant site (n ¼ 6, median survival time, 22 months, although only one patient remained alive with disease). The actual two-year and five-year survival rates were 16% and 3%, respectively. They concluded abdominal wall metastases are often indicators of recurrent intraabdominal disease; aggressive resection in patients with disease restricted to the abdominal wall and associated adherent viscera can result in local disease control.

& CLINICAL FEATURES The first suspicion of recurrence may be the insidious failure of general health signaled by malaise, weight loss, and anorexia. Vague discomfort as well as occasional bowel symptoms may be present. General physical examination is usually unrewarding, but as the disease progresses, a mass in the abdominal wall or within the abdominal cavity as well as ascites may be present. Anastomotic recurrences may be detected by endoscopy. & INVESTIGATIONS In the detection of local recurrence or peritoneal seeding, physical examination and radiologic tests are not sensitive. Barium enema examination may reveal the recurrence, but sometimes minor changes have been interpreted as a surgical tailoring defect and anastomotic recurrences are not likely to be detected early. Colonoscopy will more directly detect anastomotic recurrences, but mucosal disruption caused by locoregional recurrence is reported to occur in < 3% of patients (936). Barillari et al. (959) evaluated the effectiveness of routine colonoscopy in 481 patients who underwent curative resection. Approximately 10% of patients developed intraluminal recurrences, with more than half arising in the first 24 months. CT is not a reliable diagnostic test for low volume masses on peritoneal surfaces. Jacquet et al. (960) found an overall sensitivity of 79%. Sensitivity was 90% for nodules > 0.5 cm but only 28% for nodules < 0.5 cm. Sensitivity was lowest in the pelvis (60%). The newest technology available to detect recurrent disease is 18fluorodeoxyglucose (18FDG) PET. Delbeke et al. (961) assessed the accuracy of 18FDG-PET in patients with recurrent colorectal carcinoma in detecting liver metastases compared with CT and CT portography, detecting extrahepatic metastases compared with CT and evaluating the impact on patient management. Fifty-two patients previously treated for colorectal carcinoma presented on 61 occasions with suspected recurrence and underwent PET of the entire body. The final diagnosis was obtained by pathology (n ¼ 44) or clinical and radiological followup (n ¼ 17). A total of 166 suspicious lesions were identified. Of the 127 intrahepatic lesions, 104 were malignant

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

and of the 39 extrahepatic lesions, 34 were malignant. PET was more accurate (92%) than CT and CT portography (78% and 80%, respectively) in detecting liver metastases and more accurate than CT for extrahepatic metastases (92% and 71%, respectively). PET detected unsuspected metastases in 17 patients and altered surgical management in 28% of patients. They concluded PET is the most accurate noninvasive method for staging patients for recurrent metastatic colorectal carcinoma and plays an important role in management decisions in this setting. Libutti et al. (962) evaluated the PET scan and CEA scan as a means of localizing recurrent colorectal carcinoma. In 28 patients explored, disease was found at operation in 94%. Ten had unresectable disease. PET scan predicted unresectable disease in 90% of patients. CEA scans failed to predict unresectable disease in any patient. In 16 patients found to have resectable disease or disease that could be treated with regional therapy, PET scan predicted this in 81% and CEA scan in 13%. Desai et al. (963) determined the effect of PET on surgical decision-making in patients with metastatic or recurrent colorectal carcinoma. A total of 114 patients with advanced colorectal carcinoma were imaged with CT and PET scans. Forty-two of the 114 patients deemed to have resectable disease on the basis of CT, PET altered therapy in 40% on the basis of extrahepatic disease, bilobar involvement, thoracic involvement, retroperitoneal lymphadenopathy, bone involvement, and supraclavicular disease. In 25 patients with liver metastases, only PET found additional disease in 72%, extrahepatic disease, chest disease, retroperitoneal lymphadenopathy, and bone disease. Both scans underestimated small-volume peritoneal metastases discovered at laparotomy. Whiteford et al. (964) evaluated the records of 105 patients who underwent 101 CT and 109 PET scans for suspected metastatic colorectal carcinoma. Clinical correlation was confirmed at time of operation, histologically, or by clinical course. The overall sensitivity and specificity of PET scan in detection of clinically relevant carcinoma were higher (87% and 68%) than for CT plus other conventional diagnostic studies (66% and 59%). The sensitivity of PET scan in detecting mucous carcinoma was lower (58%) than for nonmucinous carcinoma (92%). The sensitivity of PET scanning in detecting locoregional recurrence was higher than for CT plus colonoscopy (90% vs. 71%, respectively). The sensitivity of PET in detecting hepatic metastases was higher than for CT (89% vs. 71%). The sensitivity of PET scanning in detecting extrahepatic metastases exclusive of locoregional recurrence was higher for CT plus other conventional diagnostic studies (94% vs. 67%). PET scanning altered clinical management in a beneficiary manner in 26% of cases when compared with evaluation of CT plus other conventional diagnostic studies. Johnson et al. (965) compared CT scan with PET scan in clinical decision making. A retrospective review of 41 patients with metastatic colorectal carcinoma in patients who had both CT and PET scans before operative exploration was performed. All patients underwent reexploration. Findings were divided into hepatic, extrahepatic, and pelvic regions of the abdomen. PET scan was found to be more sensitive than CT scan when compared with operative findings in the liver (100% vs. 69%), extrahepatic

169

region (90% vs. 52%), and abdomen as a whole (87% vs. 61%). Sensitivities of PET scan and CT scan were not significantly different in the pelvic region (87% vs. 61%). In each case specificity was not significantly different between the two examinations. However, PET scanning is more sensitive than CT scanning and more likely to give the correct result when actual metastatic disease is present.

& ROLE OF CARCINOEMBRYONIC ANTIGEN One area in which CEA may be of value is in the early detection of recurrence. However, this is not always the case, and when the CEA level is elevated, there is often other evidence of recurrence. In their review of the literature, Devesa, Morales, and Enriquez (688) found higher than normal blood CEA levels in < 50% of patients with early or localized failure and in approximately 75% of patients with widely disseminated disease nearly always involving the liver. The percentage of patients with lung metastases or peritoneal seeding who show elevated CEA levels is very low. False positive results are found in 6% to 25% of cases. Transient elevations of CEA have been reported to occur in 7% to 36% of patients without demonstrable recurrent carcinoma. In a remarkable series reported by Minton et al. (966) in which a CEA-directed second-look procedure was practiced on asymptomatic patients, approximately half of the patients were found to have a recurrence amenable to resection for cure. The 5-year survival rate for this group was 30%. In those patients with recurrences after a second-look procedure, a small select group was slated to undergo a thirdlook (and possibly a fourth-look) procedure in an attempt to make them disease-free. The authors justify this aggressive approach because of the unresponsiveness of colorectal carcinoma to other treatment modalities. Careful preoperative assessment must be performed to exdude those patients with unresectable metastatic disease. The surgeon must be aware that CEA levels may be elevated in nonmalignant conditions. Several authors have reported the benefit of a CEA-directed second-look operation (954,967,968). However, most surgeons have not encountered such uniformly encouraging results. In a selected review of four series comprising 203 patients, Wanebo and Stevens (809) found 80% with recurrent carcinoma. Disease was localized in 46%, and 54% had distant metastases. CEA levels at exploration ranged from 6.5 ng/mL (Ohio State University) to 25 ng/mL (Memorial Sloan-Kettering Cancer Center); 36% of the entire group underwent resection for cure, but the range was 7% (Roswell Park Memorial Institute) to 72% (Ohio State University). Hida et al. (968) reported on the usefulness of postoperative CEA monitoring for second-look operations. Seven hundred fifty-six patients with Dukes’ B and C, who had undergone curative resection, were monitored postoperatively using CEA and imaging techniques. A second-look operation was performed on any patient with a potentially resectable recurrence and, in addition, a second-look operation was performed when a persistently rising CEA value was detected. Recurrence developed in 18.8% of patients and 90.8% of the recurrences were detected within the first 3 years following curative resection. When comparing carcinomas of the colon with those

170

& PART I: COLORECTAL DISORDERS

of the rectum, the former were associated with significantly more hepatic and intra-abdominal recurrences, whereas the latter had significantly more locoregional and pulmonary recurrences. Seventy-two patients underwent a second-look operation. Of those patients, 54.2% had all of their disease resected and 1.4% had no detectable disease at the second look. Among the 142 patients with recurrence, 50% of patients underwent a second-look operation. The resectable group carried a significantly better survival than the unresectable recurrence group (41.3% vs. 5.2%). The authors concluded that complete removal of colorectal carcinoma recurrences by second-look operations, on the basis of postoperative, follow-up CEA, and imaging technique findings, results in improved survival.

& TREATMENT Operative Treatment Reluctance on the part of many surgeons to engage in the close follow-up of patients who previously have undergone resection for carcinoma of the colon and rectum lies in the pessimistic reports of the limited prospect for patients being amenable to reexcisional surgery. Bohm et al. (951) reported that 24% of patients could undergo further curative resection but only 25% of those (6% of all recurrences) were free of disease for more than 2 years. Nevertheless, some reports have been encouraging and it is well worthwhile to operate on some of these individuals (969). Follow-up is important for several reasons: (1) a second primary large bowel malignancy may be detected at an early stage of development, (2) patients who have had a colorectal carcinoma are at a higher risk of developing a primary malignancy in another organ (e.g., the breast or endometrium), and (3) recurrent disease may be diagnosed when it is localized and thus be amenable to curative therapy. Gwin, Hoffman, and Eisenberg (970) reviewed 28 patients with nonhepatic intra-abdominal recurrence of carcinoma of the colon and were able to report 15 patients who had a median actuarial survival of 25.5 months. Disease-free survival was prolonged for these patients when the time to recurrence was > 16 months. Patients who underwent palliative resection did better than those who had a bypass. Using CEA-directed second-look operations, Minton et al. (966) reported a re-resection for cure in 60% of patients with a 5-year survival of 30%. In a review of several series published in the 1980s, Herfarth, Schlag, and Hohenberger (971) found reoperation rates ranging from l8% to 60%, with an average of 31%. The value of operation for patients with incurable colorectal carcinoma is controversial. Law et al. (972) evaluated the outcomes of 180 patients undergoing operation for incurable colorectal carcinoma. Seventeen patients died in the postoperative period. Operative mortality was significantly higher in patients with nonresection procedures. Median survival of patients with resection was significantly longer than in those without resection (30 weeks vs. 17 weeks). Other independent factors that were significantly associated with poor survival were the presence of ascites, presence of bilobar liver metastases and absence of chemotherapy and/or radiation therapy. In the presence of these factors, the balance between the benefit and risk of operation should be carefully considered before decision for operative treatment.

Browne et al. (973) reported their experience with surgical resection for patients with locoregional recurrent colon carcinoma. A total of 744 patients with recurrent colon carcinoma were identified and 100 (13.4%) underwent exploration with curative intent for potentially resectable locoregional recurrence: 75 with isolated locoregional recurrence, and 25 with locoregional recurrence and resectable distant disease. The median follow-up for survivors was 27 months. Locoregional recurrence was classified into four categories: anastomotic; mesenteric/nodal; retroperitoneal; and peritoneal. Median survival for all patients was 30 months. Fifty-six patients had an R0 resection (including distant sites). Factors associated with prolonged disease-specific survival included R0 resection; age < 60 years; early stage of primary disease; and no associated distant disease. Poor prognostic factors included more than one site of recurrence and involvement of the mesenteric/ nodal basin. The ability to obtain an R0 resection was the strongest predictor of outcome, and these patients had a median survival of 66 months.

Intraoperative Radiotherapy Because of the inadequacy of operation or radiation therapy alone to treat recurrent locally advanced disease, a multimodality approach of intraoperative electron beam radiotherapy (IORT) combined with operation has been advocated (974). Willett et al. (974) reported that 5-year actuarial local control and disease-free survival rates for 30 patients undergoing this treatment program were 26% and 19%, respectively. Taylor et al. (975) reported on 100 colon carcinoma patients treated with combination therapy including surgical resection, chemotherapy, and external plus intraoperative radiotherapy. The 5-year survival was 24.7%. The 38 patients with recurrent disease whose disease was completely resected had a 37.4% 5-year survival. Endoscopic Laser Therapy For the recurrent lesion that is unresectable, relief of obstruction and control of bleeding or secretions may be obtained by the use of endoscopic laser therapy (976). This therapy offers no relief of pain nor would there be any expected improvement in survival. However, an important advantage of this mode of therapy is that it can be performed with little or no sedation, thus avoiding the risks of general anesthesia and major surgery. In addition, it can be performed on an outpatient basis in most cases and can be applied on a repetitive schedule because there is no limiting cumulative dose for laser energy. Finally, systemic side effects are few, and thus patient acceptance is good (977). The symptoms of obstruction or bleeding can be controlled in 80% to 90% of patients with a complication rate of < 10% and a mortality rate of approximately 1% (504,978–985). (Laser therapy is described in detail in Chapter 7.) Not all reports support the palliative virtues of lasers. In one report, two thirds of patients with large lesions showed little improvement and required alternative operative management (986). Courtney et al. (987) reported their experience with high powered diode laser to palliate 57 patients with inoperable colorectal carcinoma with neodymium:yttrium-aluminum-garnet (Nd:YAG). The median number of treatments received by

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

each patient was 3 (range, 1–16 treatments), with a median interval between treatments of 9.5 (range, 1–25) weeks. Lifelong palliation of symptoms occurred in 89% of patients. Major complications were two perforations and one hemorrhage, giving an overall complication rate of 5.3%. One of the patients who experienced perforation died, giving an overall mortality rate of 1.8% for the procedure. The median survival by laser therapy was 8.5 months with a probability of survival at 24 months of 15%. The role of metallic stenting has been described in detail on page 564. For the individual in whom laser therapy or stenting is unsuitable or unsuccessful and who ultimately presents with an obstruction secondary to recurrence, a stoma may be the last desperate effort at palliation.

Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy Hyperthermic intraoperative intraperitoneal chemotherapy has been recently proposed to treat peritoneal carcinomatosis arising from colon carcinoma, which is usually regarded as a lethal clinical entity. Pilati et al. (988) reviewed 46 patients treated for peritoneal carcinomatosis from colorectal carcinoma. Thirty-four patients were treated with complete cytoreductive surgery immediately followed by intraoperative hyperthermic intraperitoneal chemotherapy with mitomycin C and cisplatin. No operative deaths were reported. The postoperative morbidity rate was 35%. No severe locoregional or systemic toxicity was observed. The two-year overall survival was 31% and the median survival time and the median time to local disease progression were 18 and 13 months, respectively. Survival and local disease control in patients with well and moderately differentiated colon carcinoma were significantly better than in those with poorly differentiated lesions. Considering the dismal prognosis of this condition, hyperthermic intraoperative intraperitoneal chemotherapy seems to achieve encouraging results in a selected group of patients affected with resectable peritoneal carcinomatosis arising from colon carcinoma. Verwaal et al. (990) evaluated the outcome after the treatment of peritoneal carcinomatosis of colorectal carcinoma by cytoreduction and hyperthermic intraperitoneal chemotherapy. Recurrence within the study period of 7.5 years was 65%. For patients who had undergone a gross incomplete initial cytoreduction, the median duration of survival after recurrence was 3.7 months. If a complete cytoreduction had been accomplished initially, the median duration of survival after recurrence was 11.1 months. After effective initial treatment, a second surgical debulking for recurrence disease resulted in a median survival duration of 10.3 months and after treatment with chemotherapy it was 8.5 months. The survival was 11.2 months for patients who received radiotherapy for recurrent disease. Patients who did not receive further therapy survived 1.9 months. They concluded treatment of recurrence after cytoreduction and hyperthermic intraperitoneal chemotherapy is often feasible and seems worthwhile in selected patients. Management of Malignant Ureteral Obstruction The patient who develops a malignant ureteral obstruction poses a special and difficult problem in the

171

decision-making process. A thoughtful and reasoned approach to the subject was outlined by Smith and Bruera (991), from whom much of the following information has been obtained. The patient presents with an upper urinary tract obstruction with clinical manifestations including flank pain, hematuria, fever, sepsis, and pyuria. In cases of high-grade obstruction, oliguria, anuria, or uremia may be presenting features. The diagnosis may be suspected based on retroperitoneal lymphadenopathy seen on abdominal pelvic CT or hydronephrosis seen on abdominal ultrasonography. The imaging method of choice for proving ureteral obstruction is excretory urography. It is in the management of these patients that the decision-making process is difficult. An important consideration in the decision to treat these patients is their performance status. Some patients are already bedridden, severely symptomatic, and may not require treatment of their obstruction. Other patients may still be active with expressed, clearly defined personal goals that they can achieve with a few more months of life. For those patients deemed appropriate for aggressive therapy, both pharmacologic and urologic interventions may prove beneficial. It has clearly been shown that resection is contraindicated (568). Pharmacologic treatments include the use of agents with antineoplastic activity or agents capable of reducing edema. In the latter category, patients with acute or chronic renal failure are treated with intravenous rehydration and a trial of high-dose corticosteroids (intravenous dexamethasone, 10 mg every 6 hours for 48 hours). The role of urologic intervention in the management of malignant ureteral obstruction in advanced disease is not well defined. Some authors contend that active intervention is unwarranted with long-term progression of an underlying malignancy. Keidan et al. (991) reviewed 20 patients with advanced pelvic malignancy and concluded that before recommending percutaneous nephrostomy, the factors of in-hospital mortality (35% never left the hospital), limited survival (an additional 35% spent < 6 weeks at home before they died), significant morbidity (55% required multiple tube changes), and poor quality of life should be considered. Others argue that decision making should be guided by clinical indications and contraindications. Published indications for urinary diversion in malignant ureteral obstruction include bilateral hydronephrosis, unilateral ureteral obstruction with renal insufficiency, and unilateral pyelonephrosis. Contraindications include the evidence of rapid progression of underlying disease for which no further antineoplastic treatment is planned, other life-threatening medical problems for which no further treatment is planned, and asymptomatic unilateral malignant ureteral obstruction with normal stable renal function in patients whose previous ureteral stenting has failed. Endoscopic retrograde placement of double J (double pigtail)-type ureteral stents is generally considered the firstline urinary diversion procedure. Failing this, percutaneous nephrostomy tubes can be placed successfully in nearly all cases. Open nephrostomy, with its significant mortality and complication rates associated with prolonged hospitalization, makes this a less attractive option. Anterograde and bidirectional stenting, subcutaneous stenting, and cutaneous ureterostomy have a limited role. In the final analysis, the

172

& PART I: COLORECTAL DISORDERS

decision regarding the management of ureteral obstruction needs to be highly personalized and follow a careful discussion with the patient and/or his or her family.

Nonoperative Treatment For the most part, radiotherapy has not been used in the treatment of recurrent carcinoma of the colon because of the effects of radiation on the remaining abdominal viscera. However, specific localized areas of known recurrence can be treated by radiotherapy. Chemotherapy in the form of various drugs, dosage scheduling, and routes of administration has been used for the treatment of colon carcinoma. The most frequently used drug has been 5-FU, which offers expected response rates in the 15% range. In the past, drug combinations have not enjoyed major success when compared with 5-FU alone, but the combination of 5-FU and leucovorin has led several investigators to publish encouraging response rates in patients with advanced disease. Douillard et al. (992) investigated the efficacy of irinotecan in patients whose disease was refractory to fluorouracil. In a series of 387 patients that compared irinotecan plus fluorouracil and leucovorin to fluorouracil and leucovorin in a multicenter randomized trial, the response rate was higher in the irinotecan group (49% vs. 31%). Time to progression was longer (6.7 months vs. 4.4 months), the overall survival higher (median 17.4 months vs. 14.1 months) but grade 3 and 4 toxic effects were more frequent in the irinotecan group. In another multicenter trial, Saltz (993) compared the same groups plus a third group receiving irinotecan alone. The combination group with irinotecan resulted in longer progression free survival (median 7.0 months vs. 4.3 months), a higher response rate (39% vs. 21%), and longer overall survival (median 14.8 months vs. 12.6 months). Grade 3 diarrhea was more common with irinotecan. de Gramont et al. (994) studied fluorouracil and leucovorin with or without oxaliplatin as first-line treatment in advanced colorectal carcinoma. Patients allocated to the group with oxaliplatin had longer progression-free survival (median 9.0 months vs. 6.2 months), better response rate (50.7% vs. 22.3%). The overall survival did not reach significance (median 16.2 months vs. 14.7 months). Patients receiving oxaliplatin experienced higher frequencies of grade 3/4 neutropenia (41.7% vs. 5.3%), grade 3/4 diarrhea (11.9% vs. 5.3%), and grade 3 neurosensory toxicity (18.2% vs. 0%). In a multicenter phase II study, Sorbye et al. (995) evaluated the efficacy and safety of oxaliplatin combined with the Nordic bolus schedule of fluorouracil and folinic acid as firstline treatment in metastatic colorectal carcinoma. Eighty-five patients were treated with oxaliplatin (85 mg/m2) as a two-hour infusion on day 1 followed by a three-minute bolus injection with 5-FU (500 mg/m2) and 30 minutes later by a bolus injection of folinic acid (60 mg/m2) every second week. The same dose of 5-FU and folinic acid were also given on day 2. Of the 51 assessable patients the response rate was 62%. The estimated median time to progression was 7 months and the medial overall survival was 16.1 months in the intent-to-treat population. Neutropenia was the main adverse event with Grade 3 to 4 toxicity in 58% of patients. Febrile neutropenia developed in 8% and neuropathy in 54%. Hospitalization for treatment-related

toxicity was needed in 25% of patients. Nevertheless, they concluded oxaliplatin combined with bolus Nordic schedule of 5-FU/folic acid is a well tolerated, effective, and feasible bolus schedule as first-line treatment for metastatic colorectal carcinoma. Three agents with differing mechanisms of action are available for treatment of advanced colorectal carcinoma: fluorouracil, irinotecan, and oxaliplatin. Goldberg et al. (996) compared the activity and toxicity and three different two-drug combinations in patients with metastatic colorectal carcinoma who had not been treated previously for advanced disease. A total of 795 patients were concurrently randomly assigned to receive irinotecan and bolus fluorouracil plus leucovorin (IFL), oxaliplatin and infused fluorouracil plus leucovorin (FOLFOX) or irinotecan and oxaliplatin (IROX). A median time to progression of 8.7 months, response rate of 45% and median survival time of 19.5 months were observed for FOLFOX. These results were significantly superior to those observed for IFL for all end points (6.9 months, 31%, and 15.0 months, respectively) or for IROX (6.5 months, 35%, and 17.4 months, respectively) for time to progression and response. The FOLFOX regimen had significantly lower rates of severe nausea, vomiting, diarrhea, febrile neutropenia, and dehydration. Sensory neuropathy and neutropenia were more common with the regimens containing oxaliplatin. They concluded the FOLFOX regimen should be considered as standard therapy for patients with advanced colorectal carcinoma. Tournigard et al. (997) investigated folinic acid, 5-FU, and irinotecan (FOLFIRI) followed by folinic acid, 5-FU, and oxaliplatin (FOLFOX6), and FOLFOX6 followed by FOLFIRI in previously untreated patients. In first-line therapy, FOLFIRI achieved 56% response rate and 8.5 months median progression-free survival versus FOLFOX6 which achieved 54% response rate and 8 month median free survival. Second-line FOLFIRI achieved 4% response rate and 2.5-month median progression-free survival versus FOLFOX6 which achieved 15% response rate and a 4.2-month progression-free survival. In first-line therapy, National Cancer Institute Common Toxicity Criteria grade 3/4 mucositis, nausea/vomiting, and grade 2 alopecia were more frequent with FOLFIRI, and grade 3/4 neutropenia and neurosensory toxicity were more frequent with FOLFOX6. Both sequences achieved a prolonged survival and similar efficacy. Collucci et al. (998) in a multicenter Italian trial compared the irinotecan, leucovorin, and fluorouracil regimen (FOLFIRI) versus the oxaliplatin, leucovorin, and fluorouracil regimen (FOLFOX4) in 360 previously untreated patients with advanced colorectal carcinoma. They found no difference in overall response rates, median time to progression, and overall survival for patients treated with the FOLFIRI or FOLFOX4 regimen. Both therapies seem effective as first-line treatment in these patients. The difference between these two combination therapies is mainly in the toxicity profile. Ibrahim et al. (999) reported on a single multicenter randomized trial in 463 patients with metastatic colorectal carcinoma whose disease had recurred or progressed during or within six months of completion of therapy with a combination of bolus 5-FU, leucovorin, and irinotecan.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

Study arms included infusional 5-FU/leucovorin alone, oxaliplatin alone, and the combination of oxaliplatin infusional 5-FU/leucovorin. There were no complete responders. The partial response rates were 0%, 1%, and 9% for the 5-FU/leucovorin, oxaliplatin, oxaliplatin plus 5-FU/ leucovorin treatments, respectively. The median times to radiographic progression of disease were 2.7, 1.6, and 4.6 months, respectively. Common adverse events associated with combination treatment included peripheral neuropathy, fatigue, diarrhea, nausea, vomiting, stomatitis, and abdominal pain. Neutropenia was the major hematologic toxicity. A small but statistically significant improvement in response rate was observed in the combination arm of oxaliplatin and infusional 5-FU and leucovorin in a population that has no other treatment options. An interim analysis of radiographic time to progression, with approximately 50% of events, revealed that time to progression was longer in the combination arm. This improvement in response rate and the interim analysis showing a longer radiographic time to progression were the bases of approval for oxaliplatin. The trial results demonstrate that the efficacy of single agent oxaliplatin is similar to that of infusional 5-FU/leucovorin and that oxaliplatin should not be used alone in this patient population except in clinical trials. Caution should be exercised in choosing the regimen if bolus 5-FU is used together with oxaliplatin due to excessive toxicities. Most recently Kohne et al. (1000) reported that adding irinotecan to a standard weekly scheduled of high dose infusional fluorouracil and leucovorin can prolong progression-free survival. The median progression-free survival increased to 8.5 months from 6.4 months. The median overall survival time was increased from 16.9 to 20.1 months. The objective response rate was 62.2% in the irinotecan group and 34.4% in the other group. The limited response rates of current chemotherapeutic regimens underscore the need for newer modalities of therapy. Recognizing the role of the VEGF in promoting angiogenesis in neoplasms, Hurwitz et al. (1001) conducted a study using bevacizumab, a monoclonal antibody against vascular endothelial growth factor in combination with chemotherapy. Of 813 patients with previously untreated metastatic colorectal carcinoma, they randomly assigned 402 to receive irinotecan, bolus fluorouracil, and leucovorin (IFL) plus bevacizumab (5 mg/kg of body weight every two weeks) and 411 to receive IFL plus placebo. The median duration of survival was 20.3 months in the group given IFL plus bevacizumab as compared with 15.6 months in the group given IFL plus placebo, corresponding to a hazard ratio for death of 0.66. The median duration of progression-free survival was 10.6 months in the group with IFL plus bevacizumab as compared with 6.2 months in the group with IFL plus placebo (hazard ratio for progression 0.54); the corresponding rates of response were 44.6% and 34.8%. The median duration of the response was 10.4 months in the group given IFL plus bevacizumab as compared with 7.1 months in the group given IFL plus placebo (hazard ratio for progression 0.62). Grade 3 hypertension was more common during treatment with IFL plus bevacizumab than with IFL plus placebo (11.0% vs. 2.3%) but was easily managed. They concluded the addition of bevacizumab to fluorouracil-based combination chemotherapy

173

results in statistically significant and clinically meaningful improvement in survival among patients with metastatic colorectal carcinoma. Any recommendation for toxic regimens of chemotherapy should be made only after careful consideration of the high level of toxicity and the diminished quality of life for the patient. There has been renewed enthusiasm for the direct intra-arterial infusion of chemotherapeutic agents, specifically delivery into the liver. In their review of the literature, Blumgart and Fong (555) found six trials in which hepatic artery response rates were 48% to 62% compared with 0% to 21% for intravenous therapy. However, it must be pointed out that this increase in response rate translated to an improved survival in only one study and then only in subset analysis. Furthermore, potential complications associated with arterial infusional therapy included skin breakdown (1.0%); catheter break, misplacement, migration, leak, or kink (5.3%); pump pocket infection (1.2%); pump pocket hematoma or seroma (6.3%); and catheter-artery thrombosis or aneurysm (5.0%) (566). Added to this should be the chemotherapy complications, which included chemical hepatitis and sclerosing cholangitis (10.6%); gastritis, nausea, or vomiting (24.9%); ulcer (9.3%); and diarrhea (10.6%) (566). In my opinion, improved survival with attendant quality of life should be a benchmark to which to aspire. In their review of the literature on hepatic arterial therapy of hepatic metastases, Vauthey et al. (566) noted that although the progression of disease may best be controlled by hepatic artery chemotherapy, its clear-cut survival advantage has yet to be demonstrated. A metaanalysis of seven randomized trials evaluating the possible benefit of hepatic artery infusion using FUDR compared with intravenous chemotherapy showed a survival advantage only when untreated controls were included (1002). Since those meta-analyses, two other randomized trials of hepatic artery infusion have been published. The German Cooperative Group randomized 168 patients with unresectable liver metastases from colorectal carcinoma to hepatic artery infusion of 5-FUDR, hepatic artery infusion of 5-FU/leucovorin or intravenous 5-FU/leucovorin (1003). Response rates were higher in the two hepatic artery infusion arms but with no significant differences in time to progression. The Medical Research Council and the European Organization for the Research and Treatment of Cancer groups compared hepatic artery infusion (5-FU/ leucovorin) with intravenous 5-FU/leucovorin given as per the de Gramont regimen (1004). Response rates were assessed in 183 patients and were nearly identical (22% for hepatic artery infusion, 19% for intravenous 5-FU/leucovorin). No differences between the arms were noted for toxicity or progression-free or overall survival. Of note, both the above trials utilized subcutaneous ports rather than implantable pumps and had significant catheterrelated problems (36% of hepatic artery patients in the MRC/EORTC trial). The Cancer and Leukemia Group B (CALGB) recently completed trial 9481 which compared systemic 5-FU/leucovorin with hepatic artery infusion of FUDR, leucovorin, and dexamethasone (1005). Unfortunately, only 135 patients out of an accrual goal of 340 were randomized. The response rate (48% vs. 25%) was higher in

174

& PART I: COLORECTAL DISORDERS

the hepatic artery infusion group, though time to progression was not significantly different (5.3 months vs. 6.8 months), with time to hepatic progression better in the hepatic artery infusion arm (9.8 months vs. 7.3 months) and time to extrahepatic progression better in systemic arm (7.8 months vs. 23.3 months). The median overall survival time was statistically significantly better in the hepatic artery infusion arm (22.7 months vs. 19.9 months). In absolute terms, this three month advantage would appear marginal considering the hepatic artery infusion group required an operation with its attendant complications. Furthermore, even enthusiasts of hepatic artery infusion concede with the improved rates of response and survival reported with irinotecan and oxaliplatin-based regimens, a new standard of care for first-line treatment of metastatic colorectal carcinoma will have to be used as control arms in the evaluation of hepatic artery infusion trials (619). On the basis of trials performed, hepatic artery infusion does not represent the standard treatment option for patients with hepatic metastases from colorectal carcinoma. The most recent publication on the subject is by Keer et al. (1006) who conducted a randomized trial to compare an intrahepatic arterial fluorouracil and folinic acid regimen with the standard intravenous de Gramont fluorouracil and folinic acid regimen for patients with carcinoma of the colon or rectum, with metastases confined to the liver. They randomly allocated 290 patients from 16 centers to receive either intravenous chemotherapy (folinic acid 200 mg/m2, fluorouracil bolus 400 mg/m2 and 22-hour infusion (600 mg/m2, day 1 and 2, repeated every 14 days) or intra-arterial hepatic chemotherapy designed to be equitoxic (folinic acid 200 mg/m2, fluorouracil 400 mg/m2 over 15 minutes and 22-hour infusion 1600 mg/m2 day 1 and day 2, repeated every 14 days). For patients allocated to the intrahepatic arterial infusion, 37% did not start their treatment and another 29% had to stop before receiving six cycles of treatment because of catheter failure. The intrahepatic artery group received a median of two cycles (0–6), compared with 8.5 (6–12) for the intravenous group. Median overall survival was 14.7 months for the intrahepatic artery group and 14.8 months for the intravenous group. Their results showed no evidence of an advantage in progression-free survival or overall survival for the intrahepatic artery group. In a commentary on this article, Bonetti (1007) noted that since 1986, eight phase-3 trials have compared hepatic artery infusion with either intravenous chemotherapy or best supportive care in the treatment of unresectable hepatic metastases from colorectal carcinoma. The main conclusions of a meta-analysis of seven randomized trials were that hepatic artery infusion is associated with a higher response rate (41% vs. 14%) but similar median survival (16 months vs. 12.2 months). Martin et al. (1008) examined the survival and toxicity of hepatic arterial infusion pump following resection and/or radiofrequency ablation of all liver metastases. Patients received FUDR via the hepatic artery infusion pump at standard doses. Complications were graded according to a standard five-point grading scale. Thirty-four of 86 patients underwent placement of hepatic artery infusion pump at the time of hepatic resection or ablation. The hepatic artery group demonstrated a significantly greater number (median 5 vs. 2) and size (median 5 cm vs. 3 cm) of hepatic lesions

compared to the group without hepatic artery infusion pump. The hepatic artery infusion pump experienced a greater frequency of complications (53% vs. 33%), with 18% in the hepatic artery infusion pump group demonstrating biliary sclerosis. There were no deaths within 30 days of operation. Median survival was similar in both groups (hepatic artery infusion pump 20 months, no hepatic artery infusion pump 24 months). Adjuvant hepatic artery infusion chemotherapy was associated with significantly greater morbidity and given the availability of newer active systemic agents and regimens, the value of adjuvant hepatic artery infusion pump chemotherapy following hepatic resection or ablation remains controversial. Isolated reports have shown the efficacy of chemotherapy for recurrent malignancy in the pelvis by administration through a catheter introduced via the axillary artery (1009). Other forms of therapy tried have included external or internal liver irradiation, immunotherapy, hyperthermic therapy, hepatic artery ligation, and hepatic artery embolizaton—all without convincing evidence of sustained therapeutic advantage. The most recent effort has been intraperitoneal hyperthermic perfusion with mitomycin C for peritoneal metastases with a median response duration of 6 months (1010). For patients in whom no active treatment is available, it is imperative that the treating physician ensure that the patient is relieved of pain by the prescription of progressively strong analgesics as necessary. Chronic severe carcinoma pain often is not well controlled because of an inadequate understanding of the nature of pain. Common causes of chronic pain in patients with carcinoma include (1011): 1. Peripheral neuropathies due to radiation, chemotherapy (typically platinum, paclitaxel, and vincristine), erosion by the malignancy 2. Radiation fibrosis 3. Chronic postsurgical incisional pain 4. Phantom pain 5. Arthopathies and musculoskeletal pain due to changes in posture or mobility 6. Visceral pain due to damage to viscera or blockage due to the malignancy Most chronic pain in patients with carcinoma is neuropathic pain. Physicians often unnecessarily limit the dosage of analgesia because they have the ill-founded fear that the patient will become addicted to the drug. The basis of rational management is an appropriate analgesic given regularly in dosages adequate to suppress pain continuously. The stable of drugs might include traditional analgesics such as acetylsalicylic acid, acetaminophen, and pentazocine; anti-inflammatory agents such as acetylsalicylic acid, indomethacin, and phenylbutazone, valuable in the management of painful bony metastases; psychotropic analgesics such as tricyclic antidepressants (e.g., amitriptyline) and phenothiazine tranquilizers; and, ultimately, narcotics such as meperidine, methadone, codeine, and morphine. Drug selection climbs the ‘‘analgesic ladder’’ from nonopioid for mild pain to opioid with or without adjuvant medication for severe pain from carcinoma

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

FIGURE 64 &

(1012) (Fig. 64). A combination of agents is often useful. Since many of the narcotics result in constipation, patients should be prescribed laxatives at the same time (see Chapter 33). For patients in whom pain cannot be controlled by traditional oral and parenteral methods, Waterman, Hughes, and Foster (1013) have used epidural and intrathecal infusion with morphine. Excellent or good relief of pain was obtained in 70% of patients. It is a method of delivery that can be used in an outpatient setting. For patients in whom the pain is still not controlled, regional therapies should be considered, such as celiac plexus block or sympathetic blockade. Failing this, a neuroablative procedure should be offered, such as rhizotomy, neurolysis of primary afferent nerves or their ganglia, or cordotomy (1012). An often neglected aspect of the care of patients with advanced carcinoma is the anorexia and associated weight loss suffered by these terminally ill patients. Foltz et al. (1014) determined that nutritional counseling produced a significant increase in caloric intake. The augmented regimen included a target caloric intake with 25% of total calories derived from protein sources as well as zinc and magnesium supplementation. Increases in intake were not associated with significant weight gain or increased percent protein intake, but they had some effect on minimizing weight loss or stabilizing weight, even in patients with advanced disease who were undergoing systemic chemotherapy.

& RESULTS OF REOPERATION The virtue of re-resection of local recurrence for cure has been questioned, but it is undeniable that some patients with local recurrence are amenable to cure. Reports of success for the resection of local recurrence for cure have varied widely, but 5-year survivals in the 30% range have been recorded (966). Since many reports have documented that most recurrences occur within the first 2 years after resection of the

175

primary lesion (942), it seems reasonable to concentrate follow-up efforts during that period. Those opposed to intensive follow-up state that the poor reoperation rate for cure, ranging from 7% to 20%, negates the cost and effort applied (936,942–944). However, Buhler et al. (945) noted that repeat resections for cure were possible in 66% of patients who were asymptomatic at the time that the recurrent lesion was detected, whereas in symptomatic patients, the lesion was usually unresectable. In a series of 30 patients with anastomotic recurrence, Vassilopoulos et al. (1015) found that the majority of patients were diagnosed within 2 years of initial operation, but in this series the diagnosis was usually based on persistent signs and symptoms. Nevertheless, resection for cure was still feasible in 50% of the cases, with a 5-year survival rate of 49% and only a 3% operative mortality rate. In the report by Pihl et al. (1016) an anastomotic recurrence rate of 2.7% was detected with a re-resection rate of 40%. Other authors also believe that re-resection offers the best outlook with regard to survival (1017). Barillari et al. (959) detected a 10% intraluminal recurrence rate in a series of 481 patients. Of the 29 patients who underwent a second operation, 17 had a radical procedure and, in this group, the 5year survival rate was 70.6%. The authors concluded that asymptomatic patients more frequently underwent another operation for cure and thus had a better survival rate.

& INTESTINAL OBSTRUCTION DUE TO RECURRENT CARCINOMA A very difficult situation that challenges the clinical judgment of even the most experienced surgeon is the management of patients presenting with bowel obstruction after treatment of the primary malignancy when the feared cause is metastatic disease. Inappropriate operation does not improve outcome. Stellato and Shenk (1018) reviewed the literature on this subject and concluded that patients should be treated as any other patient manifesting intestinal obstruction, the rationale being that 26% to 38% of the bowel obstructions in patients with a history of malignancy are not secondary to recurrent or metastatic disease and that operative mortality rates for patients with carcinoma are comparable (9% to 15%) with those of patients presenting with obstruction without a history of malignancy. While 10% to 30% of patients obtain relief of obstruction by nasogastric decompression (two thirds within 3 days) 40% require operation. More than 35% of those with obstruction due to recurrent carcinoma obtain relief of their symptoms with operation. Predictors of poor outcome include the presence of shock, ascites, or an abdominal mass, with mortality rates of 100%, 70%, and 54%, respectively. Known carcinomatosis has a 40% 30-day mortality rate. Van Ooijen et al. (1019) reviewed the management of 59 patients with intestinal obstruction (38 patients with advanced carcinoma of the ovary and 21 patients with peritoneal carcinomatosis of other organs). The authors concluded that operative therapy for the relief of intestinal obstruction should only be considered in patients who do not present with manifest ascites or palpable masses and in patients with carcinoma of the ovary for whom effective chemotherapy is available. The rationale is simply that patients with masses or ascites had a median postoperative

176

& PART I: COLORECTAL DISORDERS

survival of only 36 days. Percutaneous gastrostomy should be the method of choice for other patients. Although the overall prognosis is poor in patients with malignant obstruction, the median survival being 6 months, operation still offers the best hope for palliation (1018). Lau and Lorentz (1020) believe that a more aggressive approach is appropriate. In a review of 30 patients with unresectable intra-abdominal disease, 63% had bowel function restored. Obstruction recurred after a mean symptom-free interval of 120 days in eight of 19 patients initially relieved. Another operation was performed in three of these patients. The authors believe that their results, with a median survival of 192 days for those who benefitted from operation, justify a more positive approach toward this problem and when conservatism fails, laparotomy should be undertaken in patients who are not terminally ill. Butler et al. (1021) also believe in an aggressive approach. The authors reviewed 54 patients with small bowel obstruction who had a previous diagnosis of carcinoma. Forty patients were initially treated nonoperatively, and 28% of those resolved after a mean 7 days of nasogastric suction. Five developed recurrent small bowel obstruction. Thirty-seven patients underwent laparotomy, at which time 68% were found to have obstruction due to recurrent carcinoma, with a mean survival of 5 months. Major postoperative complications occurred in 49% of patients, the most common being failure of resolution of the obstruction. The 30-day and in-hospital mortality rates of the 25 surgically treated patients were 24% and 28%, respectively. The authors concluded that (1) patients should be given an initial trial of nonoperative therapy, (2) patients with no known recurrence or a long interval to development of small bowel obstruction should be aggressively treated with early operation if nonoperative treatment fails, and (3) for patients with known abdominal recurrence in whom nonoperative therapy fails, the results of operative palliation are grim. Miller et al. (1022) conducted a study to determine the efficacy and long-term prognosis for operative versus nonoperative treatment of small bowel obstruction secondary to malignant disease. There were 32 patients accounting for 74 admissions. Colorectal neoplasm was the principal primary malignant disease that led to small bowel obstruction. The median time between diagnosis of the malignant disease and small bowel obstruction was 1.1 years. At their initial presentation, 80% of patients were treated by operation but 47% of these patients had an initial trial of nonoperative treatment. Re-obstruction occurred in 57% of patients who were operated on compared with 72% of patients who were not. The median time to re-obstruction was 17 months for patients who underwent operation compared with 2.5 months for patients who did not. Also, 71% of patients were alive and symptom free 30 days after discharge from the operative treatment compared with 52% after nonoperative treatment. Postoperative morbidity was 67%. Mortality was 13%, and 94% of patients eventually died from complications of their primary disease. They concluded that small bowel obstruction secondary to malignant disease usually indicates a grim prognosis. Operative treatment has better outcome than nonoperative management in terms of symptom free interval and re-obstruction rates. However, it is marked by high postoperative morbidity. They recommend that after a short trial of nasogastric decompression, patients with obstruc-

tion secondary to malignant disease be operated on if clinical factors indicate they will survive the operation. For patients with malignant large bowel obstruction, stent placement may successfully relieve obstruction and this subject has been described on page 562. Other methods to relieve malignant large bowel obstruction include NdYAG laser or balloon dilatation. Krouse et al. in their excellent review of palliative care of the patient with malignant obstruction detailed the wide ranged possible pharmacologic management (1023). It may include opioids for pain control, metoclopramide for nausea and vomiting; other antiemetics (prochloperazine, promethazine, and haloperidol to mention a few). Treatment of dehydration and nutritional depletion is controversial. Octreotide will be effective in the relief of symptoms of malignant bowel obstruction. Corticosteroids have been used in the hope of reducing edema around the malignancy. August et al. (1024) reported on the use of home parenteral nutrition in patients with inoperable malignant bowel obstruction. In a review of 17 patients so treated, most patients and families (82%) perceived therapy as beneficial, with a median survival of 90 days. In most circumstances, members of the nutritional support team agreed in that home parenteral nutrition facilitates compassionate home care for carefully selected patients. Despite their encouraging report, one should be circumspect about embarking on such a program. At times, because of diffuse involvement of the intestine by carcinomatosis, a percutaneous endoscopic gastroscopy may avoid the need for celiotomy. When used selectively, it can improve the quality of life of the carcinoma patient by relieving intractable vomiting and by providing an avenue for nutrition in a partially functioning gastrointestinal tract. Parker and Baines (1025) questioned the philosophy of the compulsion to operate on the assumption that obstruction always demands operative intervention. The authors point out a mortality rate of 13% with a median survival of 10 months, but in malnourished patients the mortality rate can climb to 72%. Emergency operation increases the mortality rate threefold and, although survival may be increased by a few months, the relief of symptoms lasts only 2 months overall. To counteract these depressing data, good prognostic factors include an early stage or low-grade initial lesion, a long interval from first operation, a well-nourished patient, and the fact that in one third of patients with previous carcinoma, the obstruction is caused by benign disease. The decision to operate must be made with as much information as possible. Barium studies, CT, MRI, ultrasonography, or endoscopic evaluation may be beneficial. When the decision is made not to operate, adequate analgesia, antiemetics, and antisecretogogues should be administered.

& COLORECTAL CARCINOMA COMPLICATING PREGNANCY Colorectal carcinoma in pregnancy is a rare condition and has been estimated to occur in 0.001% to 0.100% of pregnancies (1026). As of 1993, 205 cases were reported in the literature. Colorectal carcinoma that presents in pregnancy usually does so at an advanced stage because the individuals are young and the diagnosis is usually not entertained, while any symptoms that do arise are often attributed to the pregnancy

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

(1027). The most common presenting complaint is abdominal pain followed by nausea and vomiting, constipation, abdominal distention, rectal bleeding, fever, and backache. To better characterize the disease under this circumstance, Bernstein, Madoff, and Caushaj (1026) surveyed the membership of the American Society of Colon and Rectal Surgeons and identified 41 cases of large bowel carcinoma who presented during pregnancy or the immediate postpartum period. The mean age at presentation was 31 years (range, 16 to 41 years). Spatial distribution within the large bowel was as follows: right colon, 3; transverse colon, 2; left colon, 2; sigmoid colon, 8; and rectum, 26. Staging at presentation was Dukes’ A, 0; B, 16; C, 17; and metastatic, 6. Two patients were unstaged. The average follow-up was 41 months. The stagefor-stage survival was similar to patients with colorectal carcinoma in the general population. For those with rectal lesions, the 5-year survival rate for Dukes’ B was 83% and Dukes’ C, 27% and for the colon, 75% and 33%, respectively. Noteworthy was the distal distribution, with 64% in their series and 86% in the literature located in the rectum, in contradistinction to the changing distribution in the general population where a migration to the right has been documented. Also, patients presented at an advanced stage (60% had Dukes’ C or metastatic) and this probably accounts for the poor prognosis generally attributed to patients who develop colorectal carcinoma during pregnancy. A delay in diagnosis may in part account for the advanced stage of these lesions. It has been suggested that elevated levels of circulating estrogen and progesterone may stimulate growth of the carcinomas. The poor prognosis may in part be a reflection of the patient’s age, as many believe that patients younger than 40 years of age are destined to a poor outcome. Operative excision is the treatment of choice but, in pregnancy, operability, period of gestation, religious belief, and the patient’s desire for children might be considered. In the first two trimesters, the appropriate operation should be recommended, leaving the pregnancy intact (1027). Total abdominal hysterectomy has been recommended if the uterus is found to be involved (1028). If the lesion is unre-sectable or is obstructing, a colostomy or ‘‘hidden’’ colostomy should be performed to help provide time for the fetus to reach viability. In the third trimester, it has been suggested that treatment be delayed until fetal pulmonary maturity is demonstrated. At that time, labor may be induced or a cesarean section performed, often without the lesion being removed. Simultaneous removal is not usually recommended because of increased vascularity in the pelvis. The definitive operation can be delayed for a few weeks after vaginal delivery. In the event the lesion is inoperable, the pregnancy can be allowed to proceed until viability is ensured, following which palliative measures can be undertaken. Colorectal carcinoma coupled with pregnancy is a devastating combination yielding a poor prognosis. Whether this apparent poor prognosis reflects a delay in diagnosis, a biologically aggressive carcinoma in young women, or a hormonally driven carcinoma remains to be determined (1026).

& OVARIAN CARCINOMA INVOLVING THE COLON Involvement of the colon by advanced ovarian carcinoma is not uncommon. In these circumstances, the question arises as to the propriety of large bowel resection. Hertel et al.

177

(1029) reported on 100 patients with FIGO stage IIIc ovarian carcinoma who underwent pelvic en bloc resection with excision of the rectosigmoid colon as part of the primary or secondary cytoreductive operation. Malignant involvement of the rectum was confirmed histopathologically: infiltration of the serosa in 28% of patients, infiltration of the muscularis in 31% of patients and infiltration of the mucosa in 14% of patients. Histopathologically confirmed pelvic R0 resection was achieved in 85% of patients. Pelvic recurrence occurred in 4.7% of 85 optimally debulked patients compared to 60% of 15 patients with suboptimal pelvic resection status. End colostomy could be prevented in 94% of patients. They concluded pelvic en bloc surgery with rectosigmoid resection was justified by histopathological outcome since deperitonealization with preservation of the rectosigmoid would have left malignancy in-situ in 73% of patients with suspected cul-de-sac involvement. This recommendation of course can only be made if there is no other evidence of metastatic disease as is the usual situation with ovarian carcinoma.

& MALAKOPLAKIA AND COLORECTAL CARCINOMA Malakoplakia is a characteristic inflammatory condition which is usually seen in the urogenital tract. Gastrointestinal malakoplakia is seen in association with a variety of conditions such as ulcerative colitis, diverticular disease, adenomatous polyps, and carcinoma. Pillay and Chetty (1030) reported four cases of colorectal carcinoma associated with malakoplakia. Three of the cases were encountered in males and the patients ranged in age from 55 to 64 years. One case each occurred in the cecum/ascending colon and descending colon while the remaining two were located in the rectum. All four cases were Dukes’ stage B carcinoma. Furthermore, all four cases had spread to pericolonic fat and two had perforated. Microscopic examination showed the malakoplakia to be present at the infiltrating edge of the carcinoma. The draining lymph nodes were involved by malakoplakia to varying degrees in all cases. From their series and the literature review, malakoplakia associated with colorectal carcinoma tends to occur in elderly males in the rectum. The malakoplakia is found at the infiltrating front of the carcinoma and is not admixed with the neoplastic glands. Although lymph node involvement by malakoplakia has been reported only once previously, all four cases in this series showed evidence of involvement. The association does not appear to have any prognostic significance. OTHER MALIGNANT LESIONS

& CARCINOID The carcinoid neoplasm is one member of a collection of neoplasms grouped together because of a common biochemical function. These neoplasms all incorporate and store large amounts of amine precursor (5-hydroxytryptophan) and have the ability to decarboxylate this substrate, leading to the production of several biologically active amines; thus is derived the acronym APUD (amine precursor uptake and decarboxylation) (1031).

178

& PART I: COLORECTAL DISORDERS

& INCIDENCE Carcinoids arise from neuroectodermal derivatives. The gastrointestinal tract is the most common site and in decreasing order of frequency, the locations of carcinoids are the appendix, ileum, rectum, colon, and stomach (1032). Approximately 5% of all carcinoids are located in the colon (1032). The% of all gastrointestinal carcinoids has been reported to vary from 0.6% to 7.0% (1033). In the Connecticut Registry with 54 colonic carcinoids, 48% were in the cecum, 16% in the ascending colon, 6% in the transverse colon, 11% in the descending colon, 13% in the sigmoid colon, and 6% were not assigned (1034). The incidence of rectal carcinoids was 1.3% of noncarcinoid neoplasms of the rectum; for the colon, the incidence was 0.3% of noncarcinoid neoplasms of the colon (1035). In the Connecticut Registry, the age-adjusted incidence was 0.31 cases per 100,000 population/year (1034). Colonic involvement accounted for only 2.5% of all gastrointestinal carcinoids and 2.8% of all carcinoids (1036). Rectal carcinoids accounted for 12% to 15% of all carcinoids, and carcinoids of the remainder of the colon accounted for 7% of all carcinoids (1037). & CLINICAL FEATURES Carcinoids most commonly occur in the seventh and eighth decades of life, with a female preponderance of 2:1 (1031,1036). Colonic carcinoids may present as a simple polyp or as a gross malignancy that is indistinguishable from carcinoma radiologically and has an ‘‘apple core’’ appearance. These carcinoids may be entirely asymptomatic, found in 0.014% of rectal examinations, or they may produce symptoms indistinguishable from those of carcinoma. Colonic carcinoids are usually symptomatic (1036). Once they have been diagnosed, a search for other neoplasms should be made because the incidence of synchronous and metachronous neoplasms has been reported as high as 42% (1036). Gastrointestinal carcinoid is associated with a high incidence of second primary malignancy. Gerstle, Kaufman, and Koltun (1038) reviewed their experience with 69 patients with carcinoids of the gastrointestinal tract and found that 42% had second synchronous neoplasms and 4% had a metachronous neoplasm. The gastrointestinal tract was the site of 43% of these additional neoplasms with half of these being carcinomas of the colon and rectum. Tichansky et al. (1039) conducted a search of the National Cancer Institute Surveillance, Epidemiology, and End Result database from 1973 to 1996 and found 2086 patients with colorectal carcinoids. Patients with colorectal carcinoids had an increased rate of carcinoma in the colon and rectum, small bowel, esophagus/stomach, lung/bronchus, urinary tract, and prostate, when compared with the control population. Most of the gastrointestinal carcinomas were synchronous carcinomas whereas lesions outside the gastrointestinal tract were mostly metachronous neoplasms. After the diagnosis of colorectal carcinoid neoplasms, patients should undergo appropriate screening and surveillance for carcinoma at these other sites. Most gastrointestinal carcinoids are incidentally discovered at laparotomy or autopsy. The discovery of an asymptomatic gastrointestinal carcinoid during the operative treatment of another malignancy will usually only

require resection and has little effect on the prognosis of the individual. Carcinoids may be associated with multiple endocrine neoplasia, especially of the parathyroid, but most of these are associated with carcinoids of foregut origin (1040). Carcinoids of midgut and hindgut origin occur more frequently and produce significant endocrine relationships other than serotonin production. The diagnosis may be established by demonstrating elevated blood levels of serotonin or elevated urinary levels of 5-hydroxyindoleacetic acid. In the series reported by Rosenberg and Welch (1036), 44% of patients had signs of local spread, while 38% of patients had distant metastases. The liver was involved in 35.5% of the patients and the lung in 8%. In a review by Berardi (1041), 57% of patients with colonic carcinoids already had metastases, and of these, 42% had distant metastases. In the series reported by Gerstle, Kaufman, and Koltun (1038), the overall incidence of metastatic carcinoid at presentation was 32%. The most common sites of metastatic disease were lymph nodes in 82%, liver in 68%, satellite lesions to adjacent small bowel in 32%, peritoneum in 27%, and omentum in 18%. Only one of 18 appendiceal carcinoids metastasized and then only to the local lymph nodes. Four of nine colorectal carcinoids metastasized to the lymph nodes and liver. Other sites of metastatic disease included bone (1042).

& PATHOLOGY Macroscopically, carcinoids may vary in appearance from nodular thickening in the mucosa and submucosa to a sessile or pedunculated polypoid lesion, and they may have a yellowish tinge. Larger lesions may ulcerate and become annular, or they may obstruct and metastasize to regional lymph nodes or the liver (Fig. 65). The malignant character of these lesions correlates with size, location, and tissue invasion. Carcinoids of the appendix and rectum rarely metastasize (1037). Lesions < 2 cm rarely metastasize, whereas 80% of lesions > 2 cm in diameter do metastasize. Superficially invasive lesions have a better prognosis than do deeply penetrating ones. In his review Berardi (1041)

FIGURE 65 & Macroscopic features of a carcinoid of the large bowel. Lesion appears as a yellowish-tinged protruding mass.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

179

diagnosis and interventional management (1044). Doublecontrast studies still best define the primary neoplasms. The ‘‘spoke wheel’’ configuration of the desmoplastic mesenteric masses and lymph node metastases are best seen by CT, whereas hepatic metastases can be demonstrated by CT, CT angioportography, ultrasonography, MRI, and octreotide scintigraphy. Superior mesenteric angiography of the small bowel and cecum is useful when scanning procedures are not revealing. Percutaneous needle biopsy with radiologic guidance may confirm the diagnosis. Octreotide scintigraphy may have a fourfold impact on patient management (1045). It may detect resectable lesions that would be unrecognized with conventional imaging techniques, it may prevent operation in patients whose lesions have metastasized to a greater extent than can be detected with conventional imaging, it may direct the choice of therapy in patients with inoperable carcinoids, and in the future it may be used to select patients for radionuclide therapy. FIGURE 66 & Microscopic features of a carcinoid with uniform small cells in a trabecular anastomotic pattern. Although the lesion is not cytologically malignant, invasion into the muscle is evident. Source: Courtesy of H. Srolovitz, M.D., Montreal, Quebec.

found multiple carcinoids in 4.2% of patients with colonic carcinoids as compared with 30% of patients with ileal carcinoids. Associated malignancy of the colon was found in 2.5% of patients with colonic carcinoids as compared with 30% to 53% of patients with ileal carcinoids. Microscopically, carcinoids consist of uniform, small, round or polygonal cells with prominent round nuclei and eosinophilic cytoplasmic granules (Fig. 66). They usually demonstrate one of five histologic patterns: insular, trabecular, glandular, undifferentiated, or mixed. In decreasing order of median survival time in years, the growth patterns ranked as follows: mixed insular plus glandular, 4.4; insular, 2.9; trabecular, 2.5; mixed insular plus trabecular 2.3; mixed growth pattern, 1.4; glandular, 0.9; and undifferentiated, 0.5 (1043).

& IMAGING PROCEDURES The hypervascular nature of carcinoids and their metastases allows an aggressive role by the radiologist in

& CHEMICAL ACTIVITY Carcinoids secrete serotonin, a substance with pronounced pharmacologic effects, including flushing of the face, neck, anterior chest wall, and hands; increased peristalsis leading to diarrhea; constriction of bronchi presenting as wheezing; and cardiac valvular lesions with right-sided heart failure (pulmonary stenosis). Other components of the syndrome include a rise in pulmonary arterial pressure, hypotension, edema, pellagra-like skin lesions, peptic ulcers, arthralgia, and weight loss (1046). This constellation of symptoms, known as the carcinoid syndrome, usually occurs with metastases to the liver. Other products such as bradykinins, histamine, vasoactive intestinal, peptide, adrenocorticotropic hormone (ACTH), 5-hydroxytryptophan, and prostaglandins produce part of the syndrome complex. The syndrome occurs primarily with carcinoids of the small bowel but not with those of the colon or rectum. Foregut carcinoids, which are argentaffin negative and argyrophil positive, produce the serotonin precursor 5-hydroxytryptophan. Midgut carcinoids are usually both argyrophil positive and argentaffin positive, are frequently multicentric in origin, and may be associated with the carcinoid syndrome. Hindgut carcinoids are rarely argyrophil positive or argentaffin positive, are usually unicentric,

FIGURE 67 & (A) Microscopic appearance of a carcinoid with demonstration of argyrophilic granules by Fontana’s stain. (B) Microscopic appearance of a carcinoid demonstrating argentaffin tissue. Source: Courtesy of L.R. Be´gin, M.D., Montreal, Quebec.

180

& PART I: COLORECTAL DISORDERS

and are not usually associated with the carcinoid syndrome (Fig. 67). The carcinoid syndrome is a rare clinical entity that occurs with a prevalence of 1.6% in patients with carcinoids and almost only if liver metastases are present (1047). Berardi (1041) estimated that < 5% of colonic carcinoids cause the carcinoid syndrome. In a series by Rosenberg and Welch (1036), 4.2% of patients either had symptoms suggestive of the syndrome or elevated 5-hydroxyindoleacetic acid levels.

& TREATMENT Appendiceal carcinoids < 1 cm in diameter can be treated adequately by appendectomy, but if they are 2 cm in diameter, a right hemicolectomy should be performed (1032). The rationale for the latter recommendation is that the average rate of metastases is 30% (1032). For lesions between 1 and 2 cm, in which the risk of metastases is between 0% and 1%, the decision-making process is more difficult. Appendectomy is probably sufficient but additional criteria may be considered, such as extension to the mesoappendix or subserosal lymphatic invasion (1048). A more aggressive approach may be advised for younger patients (1032). Gouzi et al. (1049) recommended that other than having a size > 2 cm and base location, the presence of mucin production is a further indication for secondary right hemicolectomy. The recommended treatment for small bowel carcinoids is wide segmental resection. Because the average lymph node involvement is 44% (1032), relevant lymph node drainage should be included. Meticulous intraoperative examination is indicated because 20% to 40% of small bowel carcinoids are multicentric and simultaneous adenocarcinomas of other parts of the gastrointestinal tract occur at a rate of 8% to 29% (1032). For colonic carcinoids, the standard operation for adenocarcinoma should be performed. Metastatic disease occurs more frequently with carcinoids of the colon. If there is distant disease, resection of the primary lesion is still recommended to alleviate symptoms because a long survival period is possible (1037). Partial hepatectomy should be considered if technically feasible (1046,1050). Beaton, Homan, and Dineen (1051) have shown the value of aggressive operative debulking in reducing and sometimes obliterating the manifestations of the syndrome. For patients with unresectable metastatic carcinoids to the liver, combining either operative hepatic dearterialization or hepatic intra-arterial embolization with chemotherapy has reportedly been effective in inducing regression of the liver metastases (1044,1050,1052). Chemotherapeutic agents used have included 5-FU, streptozotocin delivered via hepatic artery or portal vein catheters, or FUDR and doxorubicin administered systematically (1052). A number of other pharmacologic and cytotoxic agents have been used to control the carcinoid syndrome (1046). Each is aimed at neutralizing one of the pharmacologically active products released by the carcinoid. Vinik and Moattari (1053) have reported the successful use of somatostatin analog in the management of the carcinoid syndrome. The symptoms of diarrhea, flushing, and wheezing can be dramatically reduced or even abolished. Ahlman et al. (1050) pursued an aggressive policy in the

management of patients with midgut carcinoid syndrome and bilobar disease. After primary operation to relieve symptoms of intestine obstruction and ischemia, the authors performed successful embolizations of hepatic arteries. Patients with remedial disease were treated by octreotide. In a series of 64 patients, the authors obtained a 70% 5-year survival rate.

& RESULTS Five-year survival rates for patients with colonic carcinoids are reported to be 52% (1035). Survival rates reported by Rosenberg and Welch (1036) were 51%, 25%, and 10% at 2, 5, and 10 years, respectively. In the Connecticut Registry, 2- and 5-year survival rates were 56% and 33%, respectively. The 5-year survival by Dukes’ staging was A, 83%; B, 43%; C, 35%; and metastatic, 21%. In a 25-year population-based study of 36 colonic carcinoids, Spread et al. (1033) found a perioperative mortality of 22%. Actuarial survival rates at 2 and 5 years were 34% and 26%, respectively. In the authors’ review, the size of the lesion and invasion into muscularis propria, the two major histopathologic prognostic factors for carcinoids, were not found to influence survival significantly. Stage, histologic pattern, differentiation, nuclear grade, and mitotic rate ( > 20 mitosis/10 hpf) proved to be prognostic factors. The 5-year survival rate in patients with appendiceal carcinoids was 90% to 100% (1032). The overall 5-year survival rate for small bowel carcinoids was 50% to 60% (1032). The 5-year survival rate decreases from 75% for local disease to 59% for patients with positive nodes to 20% to 35% if liver metastases are present.

& LYMPHOMA & INCIDENCE Lymphoma may occur as a primary lesion or as part of a generalized malignant process involving the gastrointestinal tract. As a primary lesion it constitutes only 0.5% of all cases of neoplastic disease of the colon, and yet it is the second most common malignant disease of the colon. Lymphoma comprises 6% to 20% of cases of primary gastrointestinal lymphoma (1037,1054–1057) and accounts for 5% to 10% of all non-Hodgkin’s lymphoma (1055). It most commonly involves the cecum (70%), with the rectum and ascending colon next in order of frequency (1055,1057). A more recent publication cited distribution sites of primary large bowel lymphoma as cecum 37.5%, descending colon 25%, ascending colon 25%, and rectum 12.5% (1056). It can occur at any age from 3 years to 81 years, but the average age is 50 years. Men are affected twice as often as women but a recent publication cited the reverse (1056). & PATHOLOGY Lymphomas represent a diverse group of neoplasms. At least six major classifications of non-Hodgkin’s lymphoma are in use, but there is no consensus among them. Three macroscopic types are seen (230). Annular or plaquelike thickenings are the most common type, followed by bulky protuberant growths, and, rarely, thickened and aneurysmal dilatations of the bowel wall. The cut surface has a

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

181

In one series of 15 patients, histologically 40% were classified as high grade and 60% as intermediate-grade non-Hodgkin’s lymphoma. The neoplasms usually presented at an advanced stage: in 87%, the lymphoma had spread to the adjacent mesentery, the regional lymph nodes, or both when first diagnosed (1057).

FIGURE 68 & Grayish fish-flesh appearance of a lymphoma of the large bowl. Note that the mucosa is intact.

uniform fleshy appearance (Fig. 68). Regional lymph nodes are involved in one half of all cases, but such involvement is not related to prognosis (230). Multiple primary foci are quite common. Malignant lymphomas may present as multiple polypoid protrusions of the entire colon and may mimic adenomatosis coli. Roentgenographically, 86% of colonic lymphomas are single lesions, 8% are multiple discrete lesions, and 6% show a diffuse colonic involvement (1037). Dawson, Cornes, and Morson (1059) presented these criteria for a primary lymphoma of the gastrointestinal tract: (1) no palpable peripheral lymphadenopathy, (2) normal roentgenographic findings except at the primary site, (3) normal white blood cell count and differential, (4) predominance of the alimentary tract lesion with only regional lymph node involvement, and (5) no involvement of the liver or spleen. The large series of large bowel lymphomas reported by Jinnai, Iwasa, and Watanuki (1055) were classified histologically in order of frequency as follows: histiocytic, lymphocytic, mixed, and Hodgkin’s disease. The incidence of each variety varies from series to series, but in the combined series it was as follows: histiocytic type, 43%; lymphocytic type, 29%; mixed type, 14%; Hodgkin’s disease, 3.5% (Fig. 69) (1054).

FIGURE 69 & Microscopic features of a poorly differentiated lymphoma of the colon demonstrating a somewhat nodular pattern, pleomorphism, and a degree of cellular necrosis. Source: Courtesy of L.R. Be´gin, M.D., Monteral, Quebec.

& CLINICAL FEATURES Lymphomas of the colon are characterized by abdominal pain in more than 90% of patients. Otherwise, the symptoms may be indistinguishable from those of carcinoma, with changes in bowel habits such as diarrhea or constipation, bleeding, weight loss, weakness, and possibly fever. Tender abdominal masses are present in 80% of patients on initial examination (1054). If ulceration supervenes, bleeding may be more prominent. Obstruction occurs in 20% to 25% of patients, but perforation is infrequent (1054). Multiple lesions constitute 8% of cases (1060). The radiologic signs observed during barium enema studies for non-Hodgkin’s lymphoma are as follows: a small nodular pattern frequently with multiple lesions (45.7%), a diffuse or infiltrating pattern (25.4%), a filling defect (22.9%), endoluminal and exoluminal images (17.8%), ulcerating patterns (3.4%), and a pure mesenteric form (0.8%) (1061). Lymphomas of the colon may produce the same radiologic appearance as carcinomas and similarly may be indistinguishable from carcinomas at laparotomy The colonoscopic appearance of a follicular lymphoma is seen in Fig. 70. Biopsy will clarify the diagnosis, but diagnosis still may be difficult because of the superficial nature of the biopsy. Once the diagnosis is made, staging should be performed through an adequate history, physical examination, barium enema, complete blood count, liver function tests, chest x-ray films, bone marrow assay, CT scan of the abdomen, and lymphangiography (1037).

FIGURE 70 & Colonospic appearance of a colonic lymphoma.

182

& PART I: COLORECTAL DISORDERS

& TREATMENT The treatment of primary lymphoma of the colon is resection. At laparotomy, appropriate staging with liver biopsy, lymph node biopsy, and splenectomy has been recommended (1037). Only one third of lymphomas are confined to the bowel wall at laparotomy (1060). For unresectable lesions, radiation is beneficial. Chemotherapy is recommended for systemic disease. & RESULTS The overall 5-year survival rate is approximately 40% (range, 20% to 55%) (1054,1058).When regional lymph nodes are involved, 5-year survivals fall to 12%. There is a marked difference in survival rate between patients who undergo resection only and those who receive supplementary radiotherapy, 83% vs. 16% (1060). In a series of 130 cases of primary lymphoma of the large intestine, Jinnai, Iwasa, and Watanuki (1055) reported a resection rate for cure of 55%, with most operations being abdominoperineal resections and others low anterior resections or hemicolectomies. Survival rates at 5 years and 10 years were 39.8% and 33.2%, respectively. Corresponding survival rates after curative resection were 44.2% and 40%. Prognosis was better when the lesion was 5 cm in diameter, intraluminal, and without lymph node metastases. When lymphomas were analyzed according to histologic type, 5-year and 10-year survival rates of curative resection were both 38.9% for histiocytic type, both 43% for lymphocytic type, 43.8% and 21.9% for the mixed type, and both 100% for Hodgkin’s disease. Overall 5-year survival rates were 25.4%, 33.2%, 35.4%, and 40%, respectively. The growth pattern for the lesions was intraluminal for about 50%, extramural in 15%, intramural in 25%, and unknown in the remainder. Corresponding 5-year survival rates were 47%, 20%, and 12%. The 10-year survival rates were the same for intraluminal and extramural lesions, but no patients in the intramural group survived > 7 years. The 5- and 10-year survival rates were 18.5% for patients with lymph node metastases, and 45.4% and 37.1%, respectively, for patients without metastases. Doolabh et al. (1062) reported their experience with seven cases of primary colonic lymphoma which represented 1.4% of all non-Hodgkin’s lymphoma, 14% of gastrointestinal non-Hodgkin’s lymphoma and 0.9% of all colonic malignancies diagnosed during the period of their study. The most common presentation was nonspecific abdominal pain. The lack of specific symptoms delayed diagnoses from 1 to 12 months. All patients underwent laparotomy with resection. The most common location of the lymphoma was the cecum (71%). Regional lymph nodes were affected in all but one patient. All lymphomas were B-cell lymphomas (five small noncleaved cell and two large cell). Six of seven patients received adjuvant chemotherapy. Of the six patients available for follow-up, four remained alive (12, 19, 23, and 25 months after diagnosis). In both patients who died, the disease recurred diffusely. Fan et al. (1063) identified 37 cases of primary colorectal lymphoma that comprised 0.48% of all cases of colon malignancies. The most common presenting sign and symptoms were abdominal pain (62%), abdominal mass (54%), and weight loss (43%). The most frequent site of involvement was the cecum (45%). Histologically, 78%

were classified as high-grade and 22% as intermediategrade to low-grade lymphoma. Fifty-seven percent of cases received adjuvant chemotherapy. The 5-year survival was 33% for all patients and 39% for patients treated with combination chemotherapy.

& SARCOMA There is a large variety of very uncommon neoplasms arising from mesenchymal tissues. The malignancies that fit into this category are classified as sarcomas. They include leiomyosarcoma, liposarcoma, hemangiosarcoma, fibrosarcoma, fibrous histiocytoma, neurofibrosarcoma, lymphangiosarcoma, and Kaposi’s sarcoma. Leiomyosarcoma of the colon is a rare pathologic entity, with only 58 recorded cases at the time of review of the literature by Suzuki et al. (1064) This type of sarcoma occurs two to six times more commonly in the rectum than in the colon in both sexes and most commonly appears in the fifth and sixth decades of life (1065). The lesion arises from the smooth muscle of the bowel. Macroscopically, it may range from a small nodule to a large mass, which is covered by mucosa in its early stages but eventually may become ulcerated (Fig. 71). The lesion may be intramural, endoenteric, exoenteric or dumbbell shaped (endoenteric and exoenteric) in position. It is usually a low-grade malignancy, and histologically it may be very difficult to differentiate from a benign leiomyoma (Fig. 72). Hematogenous spread results in metastases to the liver and the lung (1066). Regional lymph nodes are rarely involved. Early diagnosis is seldom accomplished before complications such as bleeding or obstruction occur (1066). Symptoms are similar to those of carcinoma: changes in bowel habit, rectal bleeding, passage of mucus, and, in the more advanced stage, weight loss. If the lesion is causing obstructive symptoms, abdominal pain may occur either as an ache or cramping. Rarely will an abdominal mass be palpable. Barium enema reveals a polypoid or constricting lesion that is usually indistinguishable from carcinoma. Colonoscopy and biopsy may be helpful if they are performed preoperatively. The treatment is resection as performed for carcinoma, and, in fact, the diagnosis will probably be made only after histiologic examination of the resected specimen. Curative resection has been reported in

FIGURE 71 & Macroscopic appearance of a transected leiomyosarcoma of the descending colon.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

FIGURE 72 & Microscopic features of the leiomyosarcoma shown in Figure 71. Relatively well-differentiated cells may be difficult to distinguish from those of a benign leiomyoma. Source: Courtesy of L.R. Be´gin, M.D., Monteral, Quebec.

45% of patients (1067). The 5-year survival rates are meaningless because of the paucity of cases. Pulmonary and hepatic metastases may occur many years later. The prognosis depends on the Broders grade, with grades 1 and 2 representing better prognoses than grades 3 and 4 (1067). Shiu et al. (1065) noted that endoenteric lesions carried a good prognosis and that exoenteric neoplasms invaded adjacent structures or perforated into the peritoneal cavity. Three clinicopathologic factors adversely affect prognosis: (1) lesions > 5 cm in diameter, (2) extraintestinal invasion or perforation, and (3) high histopathologic grade of malignancy. Patients rarely survive 5 years after operation and almost two thirds die within 1 year (1069). Radiation or chemotherapy, alone or in combination, have not been found to be effective (1069). For the most part, the other sarcomas in this group are pathologic curiosities, and their symptomatology and management are similar to those of leiomyosarcoma. For example a primary osteosarcoma arising in the colon has been reported (1070). The ultimate diagnosis of these conditions is probably also made from the resected specimen (1071).

& SQUAMOUS CELL CARCINOMA Primary squamous cell carcinoma of the large intestine is another exceedingly uncommon neoplasm. It is estimated that adenosquamous and squamous cell carcinoma of the large intestine comprise 0.05% to 0.10% of all large bowel malignancies (1072). Only 72 cases had been reported in the English-language literature until 1992 (1072). Subsequently, DiSario et al. (1) reported another 75 cases. In their review of the literature, Michelassi et al. (1073) found synchronous squamous cell carcinoma of the colon present in 3.2% of the collected cases; 10% had either antecedent, synchronous, or metachronous adenocarcinoma of the colon. The reported age range was 32 to 91 years. Mixed adenosquamous cell carcinoma occurs in men and women with equal frequency, but there are twice as many men in the squamous cell group. It has been suggested that a number of criteria must be satisfied before a diagnosis of primary squamous cell carcinoma of the large bowel is entertained (1074): (1) there must be no evidence in any

183

FIGURE 73 & Example of a squamous cell carcinoma of the large bowel, showing features similar to those of an adenocarcinoma.

other organ of a squamous cell carcinoma that might spread directly into the lower bowel or provide a source of intestinal metastases, (2) the affected bowel should not be involved in a fistulous tract lined with squamous cells (colocutaneous fistulas have been described in association with squamous cell carcinoma), and (3) when squamous cell carcinoma occurs in the rectum, care must be taken to exclude origin from the anal canal (i.e., there should be a lack of continuity between the lesion and the anal canal epithelium). Several mechanisms that have been proposed for the pathogenesis of this entity were reviewed by Vezeridis et al. (1075). They include (1) proliferation of uncommitted reserve or basal cells following mucosal injury, (2) squamous metaplasia of glandular epithelium, resulting from chronic irritation, (3) origin from embryonal nests of committed or uncommitted ectodermal cells remaining in an ectopic site after embryogenesis, (4) squamous metaplasia of an established colorectal adenocarcinoma, and (5) squamous differentiation arising in an adenoma. Associated conditions that have been described include ulcerative colitis, irradiated bowel, chronic colocutaneous fistulas, schistosomiasis, and colonic duplication (1073). Symptoms, investigations, and assessment are similar to those of colon carcinoma. Lesions are distributed throughout the large bowel with 25% to 50% of all reported cases located in the rectum (1072,1076). Coexistent disease has been reported—schistosomiasis, ulcerative colitis,

FIGURE 74 & Microscopic features of a squamous cell carcinoma with clusters of squamous cells in the stroma.

184

& PART I: COLORECTAL DISORDERS

colonic duplication, amebiasis, ovarian carcinoma previously treated with radiation, prostatic carcinoma, and ovarian teratoma. Macroscopic features may not be dissimilar to those of adenocarcinoma (Fig. 73). Microscopic features are demonstrated in Fig. 74. Of the cases in the literature, approximately half were pure squamous cell carcinoma, and the other half were mixed adenosquamous cell carcinoma. Treatment consists of resection of the affected segment. For rectal lesions, it has been suggested that multimodality therapy, as described by Nigro, be the first line of treatment and that only when this therapy fails should extensive operative procedures be used (1074). On the other hand, the Nigro protocol may be used as adjuvant therapy. The 5-year survival rate was 50% for patients with Dukes’ B lesions and 33% for those with Dukes’ C lesions; no patients with metastatic disease survived (1073).

& ADENOSQUAMOUS CARCINOMA This malignancy contains both adenocarcinomatous and squamous carcinomatous elements that are separate although contiguous thus making a distinction from the common squamous metaplasia in an adenocarcinoma (230). It may be found in all parts of the large bowel but mostly in the right colon and rectum. They are very aggressive neoplasms that may have a worse prognosis than the more common form of colonic adenocarcinoma. Furthermore, the squamous component, in particular, may have a greater potential for metastasizing and can do so as an undifferentiated-appearing carcinoma (1077). Frizelle et al. (1078) searched the Mayo Clinic Tissue Registry for all primary cases of squamous and adenosquamous carcinoma of the colon and rectum. Cases were divided into pure squamous cell carcinoma (n ¼ 11), mixed adenosquamous carcinoma (n ¼ 31), and adenocarcinoma with benign-appearing squamous metaplasia (adenoacanthoma; n ¼ 2). Right-sided lesions were the most common (43%). Metastatic disease was evident at presentation in 49% of patients, the most common sites in order being liver, peritoneal, and lung. The 5-year overall survival was 34%, stage I to III disease had a 65% 5-year survival rate, and stage IV mean survival time was 8.5 months. For nodepositive and node-negative disease, 23% and 85%, respectively, survived five years.

& PLASMACYTOMA Primary plasmacytoma involving the colon is an exceedingly rare lesion (1079,1080). Presenting symptoms are nonspecific for gastrointestinal disease and may include abdominal pain, rectal bleeding, weight loss, nausea, vomiting, and anorexia. The lesion may be single or multiple and consists of polypoid or nodular protrusions. In the presence of intestinal involvement, appropriate scans and bone marrow biopsy should be obtained to rule out bone and marrow involvement. Microscopically, the plasmacytoma lesion is composed of many plasma cells. In most cases treatment has consisted of resection of the involved colon. An 80% 10-year survival can be expected (1037). However, if the diagnosis can be made by colonoscopic biopsy, treatment options include chemotherapy and radiotherapy (1081).

& MELANOMA Primary melanoma of the colon is a distinctly uncommon entity. Indeed, since melanoblasts are necessary for the development of melanoma and since they are found in tissues of ectodermal-origin (not in the large intestine above the mucocutaneous junction), it is questionable whether primary intestinal melanoma occurs at all. When the colon is involved, the disease is usually metastatic in origin. Tessier et al. (1082) conducted a review of the literature in which they identified 88 cases of metastatic melanoma to the colon to which they added 24 patients. The mean age of their patients was 63.9 years. The interval time between diagnosis of the primary and metastatic disease to the colon was 7.5 years. Presenting symptoms included bleeding (51%), obstruction (29%), pain (20%), weight loss (11%), and perforation (7%). The frequency of these symptoms is comparable to those in the literature. Colonoscopy was the most commonly used diagnostic test (58%), followed by exploratory laparotomy (25%), autopsy (8%), and barium enema (8%). Resection was performed in 61% of patients with 39% having positive lymph nodes. The 1-year survival rate was 60% and the 5-year survival rate was 33%. Much of the following information has been obtained from the review by Tessier et al. Radiologic studies of the gastrointestinal tract have shown a wide variation of abnormalities with the small intestine having the most diverse findings. A ‘‘bull’s eye’’ or ‘‘target’’ sign on barium studies is a welldescribed finding in the small intestine, stomach, and duodenum. Findings of colonic involvement on barium studies may include multiple submucosal nodules, intussusception, large ulcerative lesions, and extrinsic masses compressing the colon. Macroscopically, the lesions are characteristically mucosal or submucosal, may be polypoid or infiltrative, single or multiple, melanotic or amelanotic. On endoscopic examination these lesions may appear amelanotic but may have enough pigmentation to be easily recognizable on gross or microscopic examination. Intussusception of the colon, multiple colonic polyps, and fungating masses resembling colon carcinoma have been described on endoscopy. Colonoscopy is not only the most reliable study but also offers the benefit of obtaining tissue for diagnosis. Special stains such as nuclear S-100 and cytoplasmic HMB-45 may be necessary to secure the diagnosis. There has been much debate as to the benefit of operation in patients with metastatic disease to the gastrointestinal tract from malignancy. Resection was performed in 75% of patients in the series of Tessier et al. and in 61% of patients from historic data. The average time until death after operation was 27.5 months. Nonoperative candidates died within 7.8 months after diagnosis. Patients with negative nodes had an average survival of 34.7 months whereas those with positive nodes lived an average of 20.4 months. In their group, 87.5% of patients had another organ involved at the time of presentation highlighting the rarity of isolated colonic metastases. From the literature review the long-term survival for isolated colonic metastases was 58.7 months. It would appear, therefore, that resection of isolated metastases to the colon is beneficial and negative nodal status is indicative of a favorable prognosis. Over 90% of patients who underwent operative resection of gastrointestinal metastases have also reported improvement in

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

their symptoms further supporting a role for operative intervention in symptomatic patients. In the series of Tessier et al., patients presenting with obstruction or perforation had a dismal prognosis with no patient surviving longer than 10 months.

& LEUKEMIC INFILTRATION Although a solid neoplasm is not involved in leukemia of the colon, it might be appropriate to include this disease in this section. The following account draws heavily from the work of Moir, Scudamore, and Benny (1083). The underlying pathology of leukemia of the colon is a neutropenic enterocolitis with primarily cecal involvement (typhlitis). The edema and hemorrhagic infarction may be seen elsewhere in the gastrointestinal tract. Neutropenic enterocolitis occurs particularly in patients with acute myelogenous leukemia undergoing high-dose cytosine arabinoside chemotherapy but also is seen in patients with other hematologic disorders (1084). Four pathophysiologic mechanisms have been proposed: (1) enteric vascular ischemia caused by stasis or shock contributes to mucosal ulceration and subsequent invasion by pathogens, (2) mucosal necrosis caused by intramural neoplastic infiltrate alone or in combination with necrosis induced by chemotherapy allows entry of organisms into the bowel wall, (3) bleeding into the mucosa or submucosa secondary to thrombocytopenia causes mucosal disruption and subsequent invasion by colonic flora, and (4) focal fecal ulceration provides entry for pathogens (1085). The clinical triad is comprised of neutropenia, sepsis, and abdominal pain (1086). The onset, which is heralded by prodromal fever, watery or bloody diarrhea, and abdominal distention, occurs during the phase of severe neutropenia. Symptoms may localize in the right lower quadrant with an associated increase in systemic toxicity. The diagnosis can be confirmed by several reexaminations, abdominal radiographs (showing partial small bowel obstruction, thickened and irregular mucosal folds, and air within the bowel wall), ultrasonography, or radionucleotide scans (gallium-labeled or indium-labeled white blood cells). CT findings include an edematous cecum and/or right colon, spiculation, and inflammation of the pericolic fat, and pneumatosis—all thought to be pathognomonic for neutropenic typhlitis (1086). The mainstay of management is complete bowel rest with total parenteral nutrition, nasogastric suction, broad-spectrum antibiotics, and avoidance of laxatives or antidiarrheal agents. Granulocyte support may be helpful. Patients with a history of typhlitis should have prophylactic bowel rest and total parenteral nutrition instituted at the beginning of further chemotherapy. Patients with ongoing severe systemic sepsis who do not respond to chemotherapy and those with overt perforation, obstruction, massive hemorrhage, or abscess formation will require operative intervention. All necrotic material must be resected, usually by right hemicolectomy, ileostomy, and mucous fistula. Depending on the extent of bowel involvement, a more extensive resection may be required. Anastomosis is not advised. To prevent recurrence, elective right hemicolectomy has been suggested if additional courses of chemotherapy are required (1086).

185

& NEUROENDOCRINE LESIONS OF THE COLORECTUM In their review of the subject, Vilor et al. (1087) noted that neuroendocrine proliferations have been classified into three types: benign (glicentin, PP- or peptide YY-producing carcinoid), low-grade malignant (serotonin-producing carcinoid), and high-grade malignant (neuroendocrine carcinoma). Neuroendocrine carcinomas have further been subdivided into the oat cell type and intermediate cell type. The incidence in neuroendocrine carcinomas is < 0.1% of all malignancies at this site. Thomas and Sobin (1088) noted that only 38 of 108,303 colonic and 18 of 46,618 rectal malignancies were small cell carcinomas. The preferred sites of occurrence are the cecum and rectum (1089). Microscopically, neuroendocrine neoplasms are recognized by their characteristically cytologic appearance with lack of tubule formation and submucosal growth pattern. They are argyrophilic and diffusely immunoreactive for neuron-specific enolase and synaptophysin. Most reported cases show metastases to the lymph nodes or distant organs at the time of diagnosis. Only approximately 10% of patients survive 1 year. The most recent report on neuroendocrine carcinomas of the colon and rectum was by Bernick et al. (1090). They identified 38 patients with neuroendocrine carcinomas from a database comprising 6495 patients (0.6%). These endocrine carcinomas did not include carcinoids. Average patient age was 57 years—44.7% males and 55.3% females. Location of the carcinomas was as follows: 17 colon, 14 rectum, 6 anal canal, and 1 appendix. Pathology was reviewed and carcinomas were categorized as small cell carcinomas (n ¼ 22) or large cell neuroendocrine carcinoma (n ¼ 16). Eighty percent stained positive by means of immunohistochemistry for neuroendocrine markers, including chromogranin, synaptophysin, and/or neuron-specific enolase. Metastatic disease was detected at the time of diagnosis in 69.4% of patients. As a group, these carcinomas had a poor prognosis with a mean survival of 10.4 months. One, two, and three year survival was 46%, 26%, and 13%, respectively. There was no significant difference in survival based on pathological subtypes. Median follow-up time was 9.4 months.

& MEDULLARY CARCINOMA OF THE COLON Wick et al. (1091) studied 68 sporadic colorectal carcinomas with medullary features and compared them with 35 poorly differentiated purely ‘‘enteric’’ colorectal carcinomas and 15 purely neuroendocrine carcinomas of grades II and III, all in patients lacking a family history of colorectal carcinoma. Medullary carcinomas were significantly more common in the ascending colon than were enteric carcinomas, but there was no significant dissimilarity to neuroendocrine carcinomas. Purely enteric carcinomas occurred more often in the rectosigmoid than medullary carcinomas or neuroendocrine carcinomas. Medullary carcinomas arose in older patients and a marked sex difference also was noted. Despite an infiltrated growth pattern, medullary carcinoma was less likely than enteric carcinomas to manifest with stage III or IV disease, but there was no stage-related difference from neuroendocrine carcinomas. Although the histologic images of medullary carcinomas were evocative of neuroendocrine differentiation,

186

& PART I: COLORECTAL DISORDERS

chromogranin positivity and synaptophysin reactivity in that group did not differ meaningfully from enteric colorectal carcinomas but was dissimilar to the 100% labeling of neuroendocrine carcinomas. p53 immunolabelling was similar in the three groups of carcinomas. Follow-up data in the study cases showed that 5-year mortality was 40% for medullary carcinomas, 59% for enteric carcinomas, and 93% for neuroendocrine carcinomas. Medullary colorectal carcinoma seems to be a distinct clinicopathologic variant of colorectal carcinoma, which does not have a neuroendocrine lineage. The biologic behavior of medullary carcinoma was better than that for enteric carcinomas or neuroendocrine carcinomas.

& CARCINOSARCOMA A unique case of carcinosarcoma of the colon has been reported (1092). It invaded the bowel wall deeply, metastasized widely, resisted multiagent chemotherapy, and caused the patient’s death four years later. It was composed of adenosquamous carcinoma admixed with sarcoma showing osseous, cartilaginous, and nonspecific spindlecell differentiation.

& SCHWANNOMA Schwannomas of the colon and rectum are uncommon. Miettinen et al. (1093) identified 20 colorectal schwannomas from the files of the Armed Forces Institute of Pathology. The schwannomas occurred equally in men and women in a wide age range (18–87 years). The most common location was cecum, followed by sigmoid and rectosigmoid, transverse colon, descending colon and rectum. The lesions commonly presented as polypoid intraluminal protrusions often with mucosal ulceration. Rectal bleeding, colonic obstruction, and abdominal pain were the most common presenting symptoms. The most common histologic variant (n ¼ 15) was a spindle-cell schwannoma with a trabecular pattern and vague or no Verocay bodies. These neoplasms ranged from 0.5 to 5.5 cm in diameter. A lymphoid cuff with germinal centers typically surrounded these lesions and focal nuclear atypia was often present, but mitotic activity never exceeded 5 per 50 HPF. All lesions were strongly positive for S-100 protein and negative for CD117 (KIT), neurofilament proteins, smooth muscle actin, and desmin. Colorectal schwannomas behaved in a benign fashion with no evidence of aggressive behavior or connection with neurofibromatosis 1 or 2, based on follow-up information on 18 patients.

& ANGIOSARCOMA Colorectal sarcomas are rare, accounting for less than 0.001% of all colorectal malignancies. Brown et al. (1094) recently reviewed the literature on the subject and the following information has been extracted from that review. Thirteen cases of colonic angiosarcoma have been reported in the literature. The majority (61%) of the patients were female. The locations of the sarcoma were: sigmoid (n ¼ 5),

cecum (n ¼ 4), rectum (n ¼ 2), descending colon (n ¼ 1), and multiple colonic sites (n ¼ 1). Most patients presented with rectal bleeding (n ¼ 7), abdominal pain (n ¼ 6), abdominal mass (n ¼ 5), and/or weight loss (n ¼ 3). Although chronic lymphedema, radiation, thorium dioxide (Thorotrast) exposure, and a number of syndromes have been cited as risk factors for angiosarcoma, none of these were noted in the cases of colonic angiosarcoma reported in the literature. The presence of a foreign body, a predisposing factor noted in other cases of angiosarcoma, were seen in only one patient with colonic angiosarcoma who had a surgical sponge left in the abdomen after previous operation. The histomorphology revealed dissecting, atypical vascular channels with plump and layered endothelial cells and areas of solid and spindled cells with an infiltrative and destructive growth pattern typical of angiosarcoma. The differential diagnosis would include sarcomatoid carcinoma, metastatic melanoma, and other sarcomas. Size has been shown to be an independent prognostic factor in angiosarcoma. In this series, five of six patients with a colonic angiosarcoma < 5 cm in largest diameter were alive at last followup (13–24 months postoperative). Conversely, only 1 of 6 patients with size 5 cm was alive at last follow-up. Patient age ranged from 16 to 77 years. Six of seven patients older than aged 60 years had rapid progression of their disease leading to death, whereas four of six patients younger than aged 60 years were still alive at the last reported follow-up (13–36 months postoperatively). A review of the current literature on colorectal angiosarcoma revealed that surgical excision is the only management shown to result in long-term survival. All survivors had surgical resection of the lesion and none received adjuvant radiotherapy or chemotherapy for the primary lesion. The patient with the longest recorded survival is a 16-yearold female who had multiple peritoneal metastases at the time of her original operation. Despite incomplete resection of the lesion and no adjuvant therapy, the patient was alive and well at 36 months follow-up. These early findings suggest that the patient’s age at the time of diagnosis may influence the prognosis of the disease. With so few reported cases of angiosarcoma of the colon and rectum, the role of adjuvant therapy is unclear. Generally, there seems to be little or no survival benefit with adjuvant chemotherapy in the treatment of sarcoma and limited experience in angiosarcoma has shown similarly disappointing results. However, doxorubicin-based regimens have shown response rates of 25% in subset analysis of a randomized controlled trial of chemotherapy in sarcoma. The role of adjuvant radiotherapy is unclear.

& CHORIOCARCINOMA Primary choriocarcinoma of the colon is a very rare neoplasm with only seven reported cases in the world literature, all but two of which was associated with an adjacent adenocarcinoma (1095). This has led to the suggestion that colonic choriocarcinomas may arise from the more typical adenocarcinoma a process of further de-differentiation. The overall poor prognosis may reflect the late diagnosis and the high volume of metastatic disease.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

& METASTASES FROM OTHER SOURCES Patients who present with a suspected colonic neoplasm and a past history of another malignancy should be considered to have possible metastatic disease, especially if the constriction appears extramucosal and the lesion is at the splenic flexure. If the colonic lesion is a squamous carcinoma, an extracolonic source should be sought because primary squamous cell carcinoma of the colon is extremely rare. Metastases from carcinoma of the lung (1096) or breast (1097,1098) presenting as primary colonic neoplasms have been reported. In a review of the literature, Washington and McDonagh (1099) found the most common sources of metastases to the colon and rectum to be melanoma, lung, and breast. From their own series of surgical autopsy specimens, the authors added gynecologic, bladder, prostate, and pancreas. Survival after the development of gastrointestinal involvement is generally poor, with most patients surviving < 1 year. However, long-term palliation may be achieved in a small subset of patients, chiefly with single small bowel deposits of melanoma or those with breast carcinoma responsive to tamoxifen. For patients with locally advanced ovarian carcinoma with contiguous extension to or encasement of the reproductive organs, pelvic peritoneum, cul-de-sac, and sigmoid colon, Bristow et al. (1100) reported on 31 consecutive patients undergoing radical oophorectomy and en bloc rectosigmoid colectomy with primary stapled anastomosis. All patients had advanced stage epithelial ovarian carcinoma; International Federation of obstetrics and Gynecology (FIGO) stage III B (6.5%), stage III C (64.5%), and stage IV (29%). There was one anastomotic breakdown requiring reoperation and colostomy. Complete clearance of macroscopic pelvic disease was achieved in all cases. Overall, 87.1% of patients were left with optimal (1 cm) residual disease and 61.3% were visibly disease free. There were no postoperative deaths but major and minor postoperative morbidity occurred in 12.9% and 35.5% of patients, respectively. They concluded resection of locally advanced ovarian carcinoma contributes significantly to a maximal cytoreductive effort.

& REFERENCES 1. DiSairo JA, Burt RW, Kendrick ML et al. Colorectal cancers of rare histologic types compared with adenocarcinomas. Dis Colon Rectum 1994; 37:1277–1280. 2. Jemal A, Murray T, Ward E, et al. MS Cancer Statistics 2005, CA 55:10–30. 3. McLaughlin JR, Dryer D, Mao Y, et al. Canadian Cancer Statistics Toronoto, Canada: National Cancer Institute of Canada, 2005. 4. Surveillance, Epidemiology, and End Results (SEER). SEER Cancer Statistics Review 1975–2001. Seer.cancer.gov. Accessed July 4, 2004. 5. Hayne D, Brown RS, McCormack M, Quinn MJ, Payne HA, Babb P. Current trends in colorectal cancer: site, incidence, mortality and survival in England and Wales. Clin Oncol (R Coll Radiol) 2001; 13(6):448–52. 6. Correa P, Haenszel W. The epidemiology of large bowel cancer. Adv Cancer Res 1978; 26:1–141. 7. Corman ML, Veidenheimer MC, Coller JA. Colorectal carcinoma: a decade of experience at the Lahey Clinic. Dis Colon Rectum 1979; 22:477–479. 8. Axtell LM, Cutler SJ, Myers MH, eds. End results in cancer, report no. 4. National Institutes of Health. Pub. no. 73–272. Bethesda, Md.: US Department of Health, Education and Welfare, 1972:217. 9. Fuchs CS, Giovannucci EL, Colditz GA, et al. A prospective study of family history and the risk of colorectal cancer. N Engl J Med 1994; 331: 1669–1674. 10. Slattery ML, Mori M, Gao R, et al. Impact of family history of colon cancer on development of multiple primaries after diagnosis of colon cancer. Dis Colon Rectum 1995; 38:1053–1058.

187

11. Hall NR, Bishop DT, Stephenson BM, et al. Hereditary susceptibility to colorectal cancer. Relatives of early onset cases are particularly at risk. Dis Colon Rectum 1996; 39:739–743. 12. St. John DJB, McDermott FT, Hoppes JL, et al. Cancer risk in relatives of patients with common colorectal cancer. Ann Intern Med 1993; 118:785–790. 13. Genetics of colorectal cancer (PDQR), Nov 19, 2003. http://www.cancer.gov/ cancerinfo/pdq/genetics/colorectal. 14. Mamazza J, Gordon PH. The changing distribution of large intestinal cancer. Dis Colon Rectum 1982; 25:558–562. 15. Obrand DI, Gordon PH. The continued change in the distribution of colorectal carcinoma. Br J Surg 1998; 85:246–248. 16. Qing SH, Rao KY, Jiang HY, Wexner SD. Racial differences in the anatomical distribution of colorectal cancer: a study of differences between American and Chinese patients. World J Gastroenterol 2003; 9(4):721–725. 17. Cucino C, Buchner AM, Sonnenberg A. Continued rightward shift of colorectal cancer. Dis Colon Rectum 2002; 45:1035–1040. 18. Takada H, Ohsawa T, Iwamoto S, et al. Changing site distribution of colorectal cancer in Japan. Dis Colon Rectum 2002; 45:1249–1254. 19. Bang KM, White JE, Gause BL, et al. Evaluation of recent trends in cancer mortality and incidence among Blacks. Cancer 1988; 61:1255–1261. 20. Lynch HT, Rubinstein WS, Locker GY. Cancer in Jews: introduction and overview. Fam Cancer 2004; 3(3–4):177–192. 21. Kune GA, Kune S, Watson LF. Perceived religiousness is protective for colorectal cancer: data from the Melbourne Colorectal Cancer Study. J R Soc Med 1993; 86:645–647. 22. Vobecky J, Devroede G, Caro J. Risk of large bowel cancer in synthetic fiber manufacture. Cancer 1984; 54:2537–2542. 23. de Verdier MG, Plato N, Steinbeck G, et al. Occupational exposures and cancer of the colon and rectum. Am J Ind Med 1992; 22:291–303. 24. Homa DM, Garabrand DH, Gillespie BW. A meta-analysis of colorectal cancer and asbestos exposure. Am J Epidemiol 1994; 139:1210–1212. 25. Demers RY, Burns PB, Swanson GM. Construction occupations, asbestos exposure, and cancer of the colon and rectum. J Occup Med 1994; 36:1027–1031. 26. Spiegelman D, Wegman DH. Occupational-related risks for colorectal cancer. J Natl Cancer Inst 1985; 75:813–821. 27. Avni A, Feuchtwanger MM. Juvenile versus adult colonic cancer: distinct different etiologic factors. Dis Colon Rectum 1984; 27:842. 28. Acquavella JF, Douglass TS, Philips SC. Evaluation of excess colorectal cancer incidence among workers involved in the manufacture of polypropylene. J Occup Med 1988; 30:438–442. 29. Lewis RJ, Schnatter AR, Lerman SE. Colorectal cancer incidence among polypropylene manufacturing workers. An update. J Occup Med 1994; 36:652–659. 30. Walker Am, Cohe AJ, Loughlin JE, et al. Mortality from cancer of the colon or rectum among workers exposed to ethyl acrylate and methyl methacrylate. Scand J Work Environ Health 1991; 17:7–19. 31. Lennard-Jones JE, Melville DM, Morson BC, et al. Precancer and cancer in extensive ulcerative colitis: findings among 401 patients over 22 years. Gut 1990; 31:800–806. 32. Greenstein AJ, Sachar DB, Smith H, et al. A comparison of cancer risk in Crohn’s disease and ulcerative colitis. Cancer 1981; 48:2742–2745. 33. Ekbom A, Helmick C, Zack M, et al. Ulcerative colitis in colorectal cancer: a population based study. N Engl J Med 1990; 323:1228–1233. 34. Munkholm P. Review article: the incidence and prevalance of colorectal cancer in inflammatory bowel disease. Aliment Pharmacol Ther 2003; 18(Suppl 2):1–5. 35. Ekbom A, Helmick C, Zack M, et al. Increased risk of large bowel cancer in Crohn’s disease with colonic involvement. Lancet 1990; 336:357–359. 36. Allen JI. Molecular biology of colorectal cancer: a clinician’s view. Perspect Colon Rectal Surg 8:181–202, 195. 37. Knudson A. Hereditary cancer, oncogenes, and antioncogenes. Cancer Res 1985; 45:1437. 38. Lynch HT, Smyrk T. Hereditary non-polyposis colorectal cancer (Lynch syndrome): an updated review. Cancer 1996; 78:1149–1167. 39. Foulkes WD, Thiffault I, Gruber SB, et al. The founder mutation MSH2
1906G– > C is an important cause of hereditary nonpolyposis colorectal cancer in the Ashkenazi Jewish population. Am J Hum Genet 2002; 71:1395–412. 40. Guillem JG, Rapaport BS, Kirchhoff T, et al. A636P is associated with early-onset colon cancer in Ashkenazi Jews. J Am Coll Surg 2003; 196: 222–225. 41. Nystrom-Lahti M, Kristo P, Nicolaides NC, et al. Founding mutations and alu-mediated recombination in hereditary colon cancer. Nature Med 1995; 1: 1203–1206. 42. Peltomaki P. Role of DNA mismatch repair defects in the pathogenesis of human cancer. J Clin Oncol 2003; 21:1174–1179. 43. Burt RW, Bishop T, Cannon LA, et al. Dominant inheritance of adenomatous polyps and colorectal cancer. N Engl J Med 1985; 312:1540–1544. 44. Solomon E, Voss R, Hall V. Chromosome 5 allele loss in human colorectal carcinomas. Nature 1987; 328:616–619. 45. Law DS, Olschwang S, Monpezat JP, et al. Concerted non-syntenic allelic loss in human colorectal carcinoma. Science 1988; 241:961–965. 46. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990; 61:759–767.

188

& PART I: COLORECTAL DISORDERS

47. Wang L, Patel U, Gosh L, et al. Mutation in the NM23 gene is associated with metastasis in colorectal cancer. Cancer Res 1993; 53:717–720. 48. Aaltonen LA, Peltomaki P, Leach FS, et al. Clues to the pathogenesis of familial colorectal cancer. Science 1993; 260:812–816. 49. Lu SL, Kawabata M, Imamura T, et al. HNPCC associated with germline mutation in the TGF-beta type II receptor gene. Nat Genet 1998; 19:17–18. 50. Liu B, Nicolaides NC, Markowitz S, et al. Mismatch repair gene defects in sporadic colorectal camcers with microsatellite instability. Nat Genet. 1995; 7:48–55.. 51. DeFrancisco J, Grady WM. Diagnosis and management of hereditary non-polyposis colon cancer. Gastrointest Endosc 2003; 58:390–408. 52. Lawes DA, SenGupta S, Boulos PB. The clinical importance and prognostic implications of microsatellite instability in sporadic cancer. Eur J Surg Oncol 2003; 29:201–212. 53. Piepoli A, Santoro R, Cristofaro G, et al. Linkage analysis identifies gene carriers among members of families with hereditary non-polyposis colorectal cancer. Gastroenterology 1996; 110:1404–1409. 54. Vasen HFA, Mecklin JP, Khan PM, et al. The International Colaborative Group on Hereditary Non-Polyposis Colorectal Cancer (ICG-HNPCC). Dis Colon Rectum 1991; 34:424–425. 55. Aaltonen LA, Sankila R, Mecklin JP, et al. A novel approach to estimate the proportion of hereditary non-polyposis colorectal cancer of total colorectal cancer burden. Cancer Detect Prev 1994; 18:57–63. 56. Boland CR, Brown EF, Evans RM, Goldberg A, Short MP. Identifying and Managing Risk for Hereditary Nonpolyposis Colorectal Cancer and Endometrial Cancer (HNPCC) American Medical Association and America Gastroenterological Association Continuing Medical Edueda 2001; 1–21. 57. Umar A, Boland CR, Terdiman JP, et al. Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst 2004; 96(4):261–268. 58. Pinol V, Castells A, Andreu M, et al. Gastrointestinal oncology group of the Spanish gastroenterological association. Accuracy of revised Bethesda guidelines, microsatellite instability, and Immunohistochemistry for identification of patients with hereditary non polyposis colorectal cancer. JAMA 2005; 293(16):1986–1994. 59. De Jong AE, Morreau H, Van Pujjenbroek M, et al. The role of mismatch repair gene defects in the development of adenomas in patients with HNPCC. Gastroenterology 2004; 126(1):42–48. 60. Cunningham C, Dunlop MG. Molecular genetic basis of colorectal cancer susceptibility. Br J Surg 1996; 83:321–329. 61. Mecklin JP, Jarvinen HJ. Tumor spectrum in Cancer Family Syndrome (hereditary non-polyposis colorectal cancer). Cancer 1991; 68:1109–1112. 62. Aarnio M, Mecklin JP, Aaltonen LA, Nystrom-Lahti M, Jarvinen HJ. Life-time risk of different cancers in hereditary non-polyposis colorectal cancer (HNPCC) syndrome. Int J Cancer 1995; 64:430–433. 63. Brezden-Masley C, Aronson MD, Bapat B, et al. Hereditary nonpolyposis colorectal cancer-molecular basis. Surgery 2003; 134:29–33. 64. Plaschke J, Engel C, Kruger S, et al. Lower incidence of colorectal and later age of disease onset in 27 families with pathogenic MSH6 germline mutations compared with familes with MLH1 or MSH2 mutations: the German Hereditary Nonpolyposis Colorectal Cancer Consortium. J Clin Oncol 2004; 22(22):4486–4494. 65. Lindor NM, Rabe K, Petersen GM, et al. Lower cancer incidence in Amsterdam-I criteria families without mismatch repair deficiency: familial colorectal cancer type X. JAMA 2005; 293(16):1979–1985. 66. Jass JR. Colorectal adenomas in surgical specimens from subjects with hereditary non-polyposis colorectal cancer. Histopathology 1995; 27:263–267. 67. Mecklin JP, Sipponen P, Jarvinen H. Histopathology of colorectal carcinomas and adenomas in cancer family syndrome. Dis Colon Rectum 1986; 29:849–853. 68. Sankila R, Aaltonen LA, Jarvinen HJ, et al. Better survival rates in patients with MLH1-associated hereditary colorectal cancer. Gastroenterology 1996; 110:682–687. 69. Tomoda H, Baba H, Oshiro T. Clinical manifestations in patients with hereditary non-polyposis colorectal cancer. J Surg Oncol 1996; 61:262–266. 70. Rodriguez-Bigas MA, Lee PHV, O’Malley L, et al. Establishment of a hereditary non-polyposis colorectal cancer registry. Dis Colon Rectum 1996; 39:649–653. 71. Hampel H, Frankel WL, Martin E, et al. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med 2005; 352(18): 1851–1860. 72. Lynch HT, Smyrk T, Lynch JF. Overview of natural history, pathology, molecular genetics, and management of HNPCC (Lynch syndrome). Int J Cancer (Pred Oncol) 1996; 36:38–43. 73. Jarvinen HJ, Aarnio M, Mustonen H, et al. Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer. Gastroenterology 2000; 118:829–834. 74. de Vos tot Nederveen Cappel WH, Nagengast FM, Griffioen G, Menko FH, Taal BG, Kleibeuker JH, Vasen HF. Surveillance for hereditary nonpolyposis colorectal cancer: a long-term study on 114 families. Dis Colon Rectum 2002; 45(12):1588–1594. 75. Mecklin JP, Jarvinen H. Treatment and follow-up strategies in hereditary nonpolyposis colorectal carcinoma. Dis Colon Rectum 1993; 36:927–929.

76. Itoh H, Houlsten RS, Harocopos C, et al. Risk of cancer death in first-degree relatives of patients with hereditary non-polyposis cancer syndrome. (Lynch type II): a study of 130 kindreds in the United Kingdom. Br J Surg 1990; 77:1367–1370. 77. Brown GJ, St John DJ, Macrae FA, Aittomaki K. Cancer risk in young women at risk of hereditary nonpolyposis colorectal cancer: implications for gynecologic surveillance. Gynecol Oncol 2001; 80:346–349. 78. Watson P, Butzow R, Lynch HT, et al. International Collaborative Group on HNPCC. The clinical features of ovarian cancer in hereditary nonpolyposis colorectal cancer. Gynecol Oncol 2001; 82:223–228. 79. Meijers-Heijboer H, Wijnen J, Vasen H, ET AL. The CHEK2 1100delC mutation identifies families with a hereditary breast and colorectal cancer phenotype. Am J Hum Genet 2003; 72:1304–1314. 80. Wong N, Lasko D, Rabelo R, Pinsky L, Gordon PH, Foulkes W. Genetic counselling and interpretation of gentic tests in familial adenomatous polyposis and hereditary nonpolyposis colorectal cancer. Dis Colon Rectum 2001; 44:271–279. 81. Rabelo R, Foulkes W, Gordon PH, et al. Role of molecular diagnostic testing in familial adenomatous polyposis and hereditary nonpolyposis colorectal cancer families. Dis Colon Rectum 2001; 44:437–446. 82. Dunlop MG, Farrington SM, Carothers AD, et al. Cancer risk associated with germline DNA mismatch repair gene mutations. Hum Mol Genet 1997; 6:105–110. 83. Xia L, Shen W, Ritacca F, et al. A truncated hM SH2 transcript occurs as a common variant in the population: implications for genetic testing. Cancer Res 1996; 56:2289–2292. 84. Keller M, Jost R, Kadmon M, et al. Acceptance of and attitude toward genetic testing for hereditary nonpolyposia colorectal cancer: a comparison of participants and nonparticipants in genetic counseling. Dis Colon Rectum 2004; 47:153–162. 85. Gritz ER, Peterson SK, Vernon SW, et al. Psychological impact of genetic testing for hereditary nonpolyposis colorectal cancer. J Clin Oncol 2005; 23(9): 1902–1910. 86. ASHG Statement. Professional disclosure of familial genetic information. The American Society of Human Genetics Social Issues Subcommittee on Familial Disclosure. Am J Hum Genet 1998; 62:474–483. 87. Lerman C, Marshall J, Audrian J, et al. Genetic testing for colon cancer susceptibility: anticipated reactions of patients and challenges to providers. Int J Cancer (Pred Oncol) 1996; 69:58–61. 88. Hadley DW, Jenkins JF, Dimond E, de Carvalho M, Kirsch I, Plamer CGS. Colon cancer screening practices after genetic counseling and testing for hereditary nonpolyposis colorectal cancer. J Clini Oncol 2004; 22:39–44. 89. Offit K. Clinical Cancer Genetics. Risk counseling and management. 1st ed. New York: Wiley-Liss, 1998:301. 90. Jass JR, Whitehall VL, Young J, Leggett BA. Emerging concepts in colorectal neoplasia. Gastroenterology 2002; 123:862–876. 91. Lamprecht SA, Lipkin M. Chemoprevention of colon cancer by calcium, vitamin D and folate: molecular mechanisms. Nat Rev Cancer 2003; 3:601–614. 92. Nagao M, Sugimura T. Carcinogenic factors in food with relevance to colon cancer development. Mutat Res 1993; 290:43–51. 93. Miller AB, Howe GR, Jain M, et al. Food items and food groups as risk factors in a case-control study of diet and colo-rectal cancer. Int J Cancer 1983; 32: 155–161. 94. Willett WC, Stampfer MJ, Colditz GA, et al. Relation of meat, food, and fiber intake to the risk of colon cancer in a prospective study amongst women. N Engl J Med 1990; 323:164–172. 95. Nigro ND, Singh DV, et al. Effect of dietary beef fat on intestinal cancer formation in rats. J Natl Cancer Inst 1975; 54:439–442. 96. Reddy BS, Maruyana H. Effect of dietary fish oil on colon carcinogenesis in rats. Cancer Res 1986; 46:3367–3370. 97. Chao A, Thun MJ, Connell CJ. Meat consumption and risk of colorectal cancer. JAMA 2005; 293(2):233–234. 98. Larsson SC, Rafter J, Holmberg L, Bergvist L, Wolk A. Red meat consumption and risk of cancers of the proximal colon, distal colon and rectum: the Swedish Mammography Cohort. Int J Cancer 2005; 113(5):829–834. 99. Norat T, Bingham S, Ferrari P, et al. Meat, fish, and colorectal cancer risk: the European Prospective Investigation into cancer and nutrition. J Natl Cancer Inst 2005; 97:906–916. 100. Burkitt DP. Epidemiology of cancer of the colon and rectum. Cancer 1971; 28:3–13. 101. Fleiszer D, MacFarlane J, Murray D, et al. Protective effect of dietary fiber against chemically induced bowel tumours in rats. Lancet 1978; 2:552–553. 102. Nigro ND, Bull AW, Klopfer BA, et al. Effect of dietary fiber on intestinal carcinogenesis in rats. J Natl Cancer Inst 1979; 62:1097–1102. 103. Greenwald P, Lanza E, Eddy GA. Dietary fiber in the reduction of colon cancer. J AM Diet Assoc 1987; 87:1178–1188. 104. Trock B, Lanza E, Greenwald P. Dietary fiber, vegetables and colon cancer: critical review and meta-analysis of the epidemiological evidence. J Natl Cancer Inst 1990; 82:650–661. 105. Freudenheim JL, Graham S, Horvath PJ, et al. Risks associated with source of fibre and fibre components in cancer of the colon and rectum. Cancer Res 1990; 52:3295–3300.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

106. Frentzel-Beyme R, Cheng-Claude J. Vegetarian diets and colon cancer: the German experience. Am J Clin Nutr 1994; 59(suppl):1143S–1152S. 107. Asano T, McLeod RS. Dietary fibre for the prevention of colorectal adenomas and carcinomas. Cochrane Database Syst Rev 2002(2):CD003430. 108. Fuchs CS, Giovannucci EL, Colditz GA, et al. Dietary fiber and the risk of colorectal cancer and adenoma in women. N Engl J Med 1999; 340:169–176. 109. Michels KB, Giovannucci, Joshipura KJ, et al. Prospective study of fruit and vegetable consumption and incidence of colon and rectal cancers. J Natl Cancer Inst 2000; 92:1740–1752. 110. Terry P, Giovannucci E, Michels KB, et al. Fruit, vegetables, dietary fiber, and risk of colorectal cancer. J Natl Cancer Inst 2001; 93:525–533. 111. Peters U, Sinha R, Chatterjee N, et al. Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Project Team. Dietary fibre and colorectal adenoma in a colorectal cancer early detection programme. Lancet 2003, 361: 1491–1495. 112. Meyer F, White E. Alcohol and nutrients in relation to colon cancer in middleaged adults. Am J Epidemiol 1993; 138(4):225–236. 113. Howe GR, Benito E, Castelleto R, et al. Dietary intake of fiber and decreased risk of cancers of the colon and rectum: evidence from the combined analysis of 13 case-control studies. J Natl Cancer Inst 1992; 84:1887–1896. 114. Bingham SA, Day NE, Luben R, et al. European Prospective Investigation into Cancer and Nutrition. Dietary fibre in food and protection against colorectal cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC): an observational study. Lancet 2003 361:1496–1501. 115. Slattery ML, Sorenson AW, Ford MH. Dietary calcium intake as a mitigating factor in colon cancer. Am J Epidemiol 1988; 128:504–514. 116. Garland C, Shekelle RD, Barrett-Connor E, et al. Dietary vitamin D and calcium and risk of colorectal cancer: a 19 year prospective study in men. Lancet 1985; 1:307–309. 117. Rozen P, Fireman Z, Wax Y, et al. Oral calcium suppresses colonic mucosal proliferation of persons at risk for colorectal neoplasia. Gastroenterology 1987; 92:1603. 118. Wargovich MJ, Lynch PM, Levin B. Modulating effects of calcium in animal models of colon carcinogenesses and short-term studies in subjects at increased risk for colon cancer. Am J Clin Nutr 1991; 54:2025–2055. 119. Baron JA, Beach M, Mandel JS, et al. Calcium supplements for the prevention of colorectal adenomas, Calcium Polyp Prevention Study Group. N Engl J Med 1999; 340:101–117. 120. Wu K, Willett WC, Fuchs CS, Colditz GA, Giovannucci EL. Calcium intake and risk of colon cancer in women and men. J Natl Cancer Inst 2002; 94:437–446. 121. Wallace K, Baron JA, Cole BF, et al. Effect of calcium supplementation on the risk of large bowel polyps. J Natl Cancer Inst 2004; 96(12):921–925. 122. Larsson SC, Bergkvist L, Wolk A. Magnesium intake in relation to risk of colorectal cancer in women. JAMA 2005; 293(1):86–89. 123. Vernie LN. Selenium in carcinogenesis. Biochim Biophys Acta 1984; 738: 203–217. 124. Rumi G, Imre I, Sulle G, et al. Selenium in the blood of patients with colorectal cancer and neoplastic polyp. Acta Physiol Hung 1992; 80:275–279. 125. Finley JW. Reduction of cancer risk by consumption of selenium–enriched plants: enrichment of broccoli with selenium increases the anticarcinogenic properties of broccoli. J Med Food 2003; 6:19–26. 126. Nelson RL, Davis FG, Sutter E, et al. Serum selenium and colonic neoplastic risk. Dis Colon Rectum 1995; 38:1306–1310. 127. Jacobs ET, Jiang R, Alberts DS, et al. Selenium and colorectal adenoma: results of a pooled analysis. J Natl Cancer Inst 2004; 96(22):1669–1675. 128. Nigro ND, Bull AW. Prospects for the prevention of colorectal cancer. Dis Colon Rectum 1987; 30:751–754. 129. Wargovich MJ. New dietary anticarcinogens and prevention of gastrointestinal cancer. Dis Colon Rectum 1988; 31:72–75. 130. Wattenberg LW. Chemoprevention of cancer. Cancer Res 1985; 45:1–8. 131. Nigro ND, Bull AW, Wilson PS, et al. Combined inhibitors of carcinogenesis or intestinal cancer in rats. J Natl Cancer Inst 1982; 69:103–107. 132. Colacchio TA, Memoli VA. Chemoprevention of colorectal neoplasms. Arch Surg 1986; 121:1421–1424. 133. Lognecker MP. A case-control study of alcoholic beverage consumption in relation to risk of cancer of the right colon and rectum in men. Cancer Causes Control 1990; 1:5–14. 134. Pollack ES, Nomura AMY, Heilbrun LK, et al. Prospective study of alcohol consumption and cancer. N Engl J Med 1984; 310:617–621. 135. Newcomb PA, Storer BE, Marcus PM. Cancer of the large bowel in women in relation to alcohol consumption: a case-control study in Wisconsin (United States). Cancer Causes Control 1993; 4:405–411. 136. Kato I, Tominaga S, Ikari A. A case-control study of male colorectal cancer in Aichi Prefecture, Japan: with special reference to occupational activity level, drinking habits and family history. Jpn J Cancer Res 1990; 81:115–121. 137. Riboli E, Corne´e J, Macquart-Moulin G, et al. Cancer and polyps of the colorectum and lifetime consumption of beer and other alcoholic beverages. Am J Epidemiol 1991; 133:157–166. 138. Serralva MS, Anjos J, Vilaca F. Colorectal carcinoma in patients older than 65 years: prognostic factors. Br J Surg 1995; 82(suppl 1):35–36. 139. Carstensen JM, Bygren LO, Hatscheck T. Cancer incidence among Swedish brewery workers. Int J Cancer 1990; 45:393–396.

189

140. Maekawa SJ, Aoyama N, Shirasaka D, et al. Excessive alcohol intake enhances the development of synchronous cancerous lesion in colorectal cancer patients. Int J Colorectal Dis 2004; 18:171–175. 141. Sharpe CR, Siemiatycki J, Rachet B. Effects of alcohol consumption on the risk of colorectal cancer among men by anatomical subsite (Canada). Cancer Causes Control 2002; 13:483–491. 142. Kikendall JW, Bowen PE, Burgess MB, et al. Cigarettes and alcohol as independent risk factors for colonic adenomas. Gastroenterology 1989; 97: 660–664. 143. Giovannucci E, Rimm EB, Stampfer MJ, et al. A prospective study of cigarette smoking and risk of colorectal adenoma and colorectal cancer in U.S. men. J Natl Cancer Inst 1994; 86:183–191. 144. Giovannucci E, Colditz GA, Stampfer MJ, et al. A prospective study of cigarette smoking and risk of colorectal adenoma and colorectal cancer in U.S. women. J Natl Cancer Inst 1994; 86:192–199. 145. Chao A, Thun MJ, Jacobs EJ, Henley SJ, Rodriguez C, Calle EE. Cigarette smoking and colorectal cancer mortality in the cancer prevention study II. J Natl Cancer Inst 2000; 92:1888–1896. 146. Jain M, Cook GM, Davis FG, et al. A case-control study of diet and colorectal cancer. Int J Cancer 1980; 26:757–768. 147. Potter JD, McMichael AJ. Diet and cancer of the colon and rectum: a casecontrol study. J Natl Cancer Inst 1986; 76:557–569. 148. Kritchevsky D. Epidemiology of fibre, resistant starch and colorectal cancer. Eur J Cancer Prev 1995; 4:345–352. 149. Phillips RL, Snowdon DA. Dietary relationships with fatal colorectal cancer among Seventh-Day Adventists. J Natl Cancer Inst 1985; 74:307–317. 150. Stemmerman GN, Nomura MY, Heilbrun LK. Dietary fat and the risk of colorectal cancer. Cancer Res 1984; 44:4633–4637. 151. Berry EM, Zimmerman J, Ligumsky M. The nature of dietary fat and plasma lipids in relation to the development of polyps and carcinoma of the colon. Gastroenterology 1985; 88:1323. 152. Freudenheim JL, Graham S, Marshall JR, et al. A case-control study of diet and rectal cancer in western New York. Am J Epidemiol 1990; 131: 612–624. 153. Giovannucci E, Stampfer MJ, Colditz GA, et al. Multivitamin use, folate, and colon cancer in women in the Nurses’ Health Study. Ann Intern Med 1998; 129:517–524. 154. Fernandez E, Chatenoud L, La Vecchia C, Negri E, Franceschi S. Fish consumption and cancer risk. Am J Clin Nutr 1999; 70:85–90. 155. Jao SW, Beart RW Jr., Reiman HM, et al. Colon and anorectal cancer after pelvic irradiation. Dis Colon Rectum 1987; 30:953–958. 156. Matsuo T, Ito M, Sekine I, et al. Mucosal de novo cancer of the rectum following radiation therapy for uterine cancer. Intern Med 1993; 32:427–429. 157. Tamai O, Nozato E, Miyazato H, et al. Radiation-associated rectal cancer: report of four cases. Dig Surg 1999; 16(3):238–243. 158. Martins A, Sternberg SS, Attiyeh FF. Radiation-induced carcinoma of the rectum. Dis Colon Rectum 1980; 23:572–575. 159. Shirouzu K, Isomoto H, Morodomi T, et al. Clinicopathologic characterisitics of large bowel cancer developing after radiotherapy for uterine cervical cancer. Dis Colon Rectum 1994; 37:1245–1249. 160. Hareyama M, Okubo O, Oouchi A, et al. A case of carcinoma of the rectum after radiotherapy for carcinoma of the cervix. Radiat Med 1989; 7:197–200. 161. Baxter NN, Tepper JE, Durham SB, Rothenberger DA, Virnig BA. Increased risk of rectal cancer after prostate radiation: a population-based study. Gastroenterology 2005; 128(4):819–824. 162. Labow SB, Hoexter B, Walrath DC. Colonic adenocarcinomas in patients with ureterosigmoidostomies. Dis Colon Rectum 1979; 22:157–158. 163. Sheldon CA, McKinley CR, Hartig PR, et al. Carcinoma at the site of ureterosigmoidostomy. Dis Colon Rectum 1983; 26:55–58. 164. Van Driel MF, Zwiers W, Grand S, et al. Juvenile polyps at the site of a ureterosigmoidostomy. Report of 5 cases. Dis Colon Rectum 1988; 31: 553–557. 165. Guy RJ, Handa A, Traill Z, Mortensen NJ, et al. Rectosigmoid carcinoma at previous ureterosigmoidostomy site in a renal transplant recipient: report of a case. Dis Colon Rectum 2001; 44(10):1534–1539. 166. Kalble T, Tricker AR, Friedl P, et al. Ureterosigmoidostomy: long-term results, risk of carcinoma, and etiological factors for carcinogenesis. J Urol 1990; 144:1110–1114. 167. Kliment J, Luptak J, Lofaj M, et al. Carcinoma of the colon after ureterosigmoidostomy and trigonosigmoidostomy for extrophy of the bladder. Int Urol Nephrol 1993; 25:339–343. 168. Husmann DA, Spence HM. Current status of tumor of the bowel following ureterosigmoidostomy. A review. J Urol 1990; 144:607–610. 169. Zaridze DG. Environmental etiology of large-bowel cancer. J Natl Cancer Inst 1983; 70:389–400. 170. Hill MJ. Mechanism of colorectal carcinogenesis. Joosens JV, Hill MJ, Geboers J, eds. Diet and Human Carcinogenesis. Amsterdam: Elsevier Science, 1986:149–164. 171. Jorgensen T, Rafaelsen S. Gallstones and colorectal cancer––There is a relationship, but it is hardly due to cholecystectomy. Dis Colon Rectum 1992; 35:24–28.

190

& PART I: COLORECTAL DISORDERS

172. Schernhammer ES, Leitzmann MF, Michaud DS et al. Cholecystectomy and the risk for developing colorectal cancer and distal colorectal adenomas. Br J Cancer 2003; 88:78–83. 173. Legergren J, Ye W, Ekbom A. Intestinal cancer after cholecystectomy: is bile involved in carcinogenesis? Gastroentelogy 2001; 121(3):542–547. 174. Wynder El, Reddy BS. Metabolic epidemiology of colorectal cancer. Cancer 1974; 34:801–806. 175. Moorehead RJ, McKelvey STD. Cholecystectomy and colorectal cancer. Br J Surg 1989; 76:250–253. 176. Gafa M, Sarli L, Sansebastiano G, et al. Prevention of colorectal cancer. Role of association between gallstones and colorectal cancer. Dis Colon Rectum 1987; 30:692–696. 177. Hickman MS, Salinas HC, Schwesinger WH. Does cholecystectomy affect colonic tumorigenesis? Arch Surg 1987; 12:334–336. 178. McMichael A, Potter JD. Host factors in carcinogenesis: certain bile acid metabolic profiles that selectively increase the risk of colon cancer. J Natl Cancer Inst 1985; 75:185–191. 179. Nielsen GB, Theodoro A, Tulinius A, et al. Cholecystectomy and colorectal carcinoma: a total-population historical prospective study. Am J Gastroenterol 1991; 88:1486–1490. 180. McFarlane MJ, Welch KE. Gallstones, cholecystectomy, and colorectal cancer. Am J Gastroenterol 1993; 88:1994–1999. 181. Abrams JS, Anton JR, Dreyfuss DC. The absence of a relationship between cholecystectomy and the subsequent occurrence of cancer of the proximal colon. Dis Colon Rectum 1983; 26:141–144. 182. Kune GA, Kune S, Watson LF. Large bowel cancer after cholecystectomy. Am J Surg 1988; 156:359–362. 183. Kaibara N, Wakatsuki T, Mizusawa K, et al. Negative correlation between cholecystectomy and the subsequent development of large bowel carcinoma in a low risk Japanese population. Dis Colon Rectum 1986; 29:644–646. 184. Ekbom A, Yuen J, Adami HO, et al. Cholecystectomy and colorectal cancer. Gastroenterology 1993; 105:142–147. 185. Neugent AI, Murray TI, Garbowski GC, et al. Cholecystectomy as a risk factor for colorectal adenomatous polypoid carcinoma. Cancer 1991; 68: 1644–1647. 186. Boulos PB, Cowin AP, Karamanolis DG, et al. Diverticula, neoplasia, or both? Early detection of carcinoma in sigmoid diverticular disease. Ann Surg 1985; 202:607–609. 187. Morini S, Angelis P, Manurita L, et al. Association of colonic diverticula with adenomas and carcinomas. A colonoscopic experience. Dis Colon Rectum 1988; 31:793–796. 188. Persky V, Andrianopoulos G. The etiology of cancer of the colon. It it all in the diet? Nelson RL, ed. Problems in Current Surgery. Controversies in Colon Cancer. Philadelphia: JB Lippincott, 1987:11–23. 189. Thune I, Lund E. Physical activity and risk of colorectal cancer in men and women. Br J Cancer 1996; 73:1134–1140. 190. Slattery ML, Edwards S, Curtin K, et al. Physical activity and colorectal cancer. Am J Epidemiol 2003; 158:214–224. 191. Terry MB, Neugut AI, Bostick RM, et al. Risk factors for advanced colorectal adenomas: a pooled analysis. Cancer Epidemiol Biomarkers Prev 2002; 11:622–629. 192. Colbert LH, Hartman TJ, Malila N, et al. Physical activity in relation to cancer of the colon and rectum in a cohort of male smokers. Cancer Epidemiol Biomakers Prev 2001; 10:265–268. 193. Longnecker MP, de Verdier MG, Frumkin H, et al. A case-control study of physical activity in relation to risk of cancer of the right colon and rectum in men. Int J Epidemiol 1995; 24:42–50. 194. White E, Jacobs EJ, Daling JR. Physical activity in relation to colon cancer in middle-aged men and women. Am J Epidemiol 1996; 144:42–50. 195. Quadrilatero J, Hoffman-Geotz L. Physical activity and colon cancer. A systematic review of potential mechanisms. J Sports Med Phys Fitness 2003; 43(2):121–138. 196. Neugent AI, Santos J. The association between cancers of the small bowel and large bowel cancers. Epidemiol Biomarkers Prev 1993; 2:551–553. 197. Sonnenberg A, Mu¨ller AD. Constipation and cathartics as risk factors of colorectal cancer: a meta-analysis. Pharmacology 1993; 47(suppl 1):224–233. 198. Singh S, Morgan BF, Broughton M, et al. A 10 year prospective audit of outcome of surgical treatment for colorectal carcinoma. Br J Surg 1995; 82: 1486–1490. 199. Brodeurs MJM, Lambe M, Bacon JA, et al. History of child-bearing and colorectal cancer risk in women aged less than 60: an analysis of Swedish routine registry data 1960–1984. Int J Cancer 1996; 66:170–175. 200. Newcomb PA, Storer BE. Postmenopausal hormone use and risk of large bowel cancer. J Natl Cancer Inst 1995; 87:1067–1071. 201. Calle EE, Miracle-McMahill HL, Thun MJ, et al. Estrogen replacement therapy and risk of fatal colon cancer in a prospective cohort of postmenopausal women. J Natl Cancer Inst 1995; 87:517–523. 202. Jacobs EJ, White E, Weiss NS. Exogeneous hormone reproductive history, and colon cancer (Seattle, Washington, U.S.). Cancer Causes Control 1994; 5: 359–366.

203. Marcus PM, Newcomb PA, Young T, et al. The association of reproductive and menstrual characteristics, and colon and rectal cancer risk in Wisconsin women. Ann Epidemiol 1995; 5:303–309. 204. Grodstein F, Newcomb PA, Stampfer MJ. Postmenopausal hormone therapy and the risk of colorectal cancer: a review and meta-analysis. Am J Med 1999; 106:574–582. 205. Baris D, Gridley G, Ron E, et al. Acromegaly and cancer risk: a cohort study in Sweden and Denmark. Cancer Causes Control 2002; 13:395–400. 206. Mullen PJ, Wilson HK, Majury CW, et al. Bile acids and the increased risk of colorectal tumors after truncal vagotomy. Br J Surg 1990; 77:1085–1090. 207. Fisher SG, Davis F, Nelson R, et al. Large bowel cancer following gastric surgery for benign disease: a cohort study. Am. J. Epidemiol. 1994; 139:684–92. 208. Little J, Owen RW, Fernandez F, et al. Asymptomatic colorectal neoplasia and fecal characteristics: a case-control study of subjects participating in the Nottingham fecal occult blood screening trial. Dis Colon Rectum 2002; 45: 1233–1241. 209. Babbs CF. Free radicals and the etiology of colon cancer. Free Radic Biol Med 1990; 8:191–200. 210. Gorbach SL, Goldin BL. The intestinal microflora and the colon cancer connection. Rev Infect Dis 1990; 12:S252. 211. Knekt P, Reunanen A, Takkunen H, et al. Body iron stores and risk of cancer. Int J Cancer 1994; 56:379–382. 212. Vesterinen E, Pukkala E, Timonen T, et al. Cancer incidence among 78,000 asthmatic patients. Int J Epidemiol 1993; 22:976–982. 213. Howden CW, Hornung CA. A systematic review of the association between Barrett’s esophagus and colon neoplasm. Am J Gastoenterol 1995; 90: 1814–1819. 214. Vaughan TL, Kiemeney LALM, McKnight B. Colorectal cancer in patients with esophageal adenocarcinoma cancer. Epidemiol Biomar Prev 1995; 4:93–97. 215. Seigers CP, von Hertzberg-Lottin E, Otte M, et al. Anthanoid laxative abuse— A risk for colorectal cancer? Gut 1993; 34:1099–1101. 216. Younes M, Katikaneni RR, Lechago J. Association between mucosal hyperplasia of the appendix and adenocarcinoma of the colon. Histopathology 1995; 26:33–37. 217. Erlinger TP, Platz EA, Rifai N, Helzlsouer KJ. C-reactive protein and the risk of incident colorectal cancer. JAMA 2004; 291:585–590. 218. Woolcott CG, King WD, Marrett LD. Coffee and tea consumption and cancers of the bladder, colon and rectum. Eur J Cancer Prev 2002; 11:137–145. 219. Kavan MG, Engdahl BE, Kay S. Colon cancer: personality factors predictive of onset and stage of presentation. J Psychosom Res 1995; 39:1031–1039. 220. Galloway DJ, Indran M, Carr K, et al. Dietary manipulation during experimental carcinogenesis: a morphological study in the rat. Int J Colorectal Dis 1987; 2:193–200. 221. Shatzkin A, Kelloff G. Chemo-and dietary prevention of colorectal cancer. Eur J Cancer 1995; 31A:1198–1204. 222. Cassidy A, Bingham SA, Cummings JH. Starch intake and colorectal cancer risk; an international comparison. Br J Cancer 1994; 69:937–942. 223. Thun MJ, Namboodiri MM, Heath CW. Aspirin use and reduced risk of fatal colon cancer. N Engl J Med 1991; 325:1593–1596. 224. Rosenberg L, Palmer JR, Zauber AG, et al. A hypothesis: non-steroidal antiinflammatory drugs reduce the incidence of large bowel cancer. J Natl Cancer Inst 1991; 83:355–358. 225. Smalley W, Ray WA, Daugherty J, Griffin MR. Use of nonsteroidal antiinflammatory drugs and incidence of colorectal cancer: a population-based study. Arch Intern Med 1999; 159:161–166. 226. Baron JA, Cole BF, Sandler RS, et al. A randomized trial of aspirin to prevent colrectal adenomas. N Engl J Med 2003; 348:891–899. 227. Sturmer T, Glynn RJ, Lee IM, Manson JE, Buring JE, Hennekens CH. Aspirin use and colorectal cancer: post-trial follow-up data from the Physicians’ Health Study. Ann Intern Med 1998; 128:713–720. 228. Burke CA, Bauer WM, Lashner B. Chemoprevention of colorectal cancer: slow, steady progress. Cleve Clin J Med 2003; 70:346–350. 229. Greenwald P, Kellofff G, Burch-WhitmanC, et al. Chemoprevention. CA Cancer J Clin 1995; 45:31–49. 230. Morson BC, Dawson IMP. Gastrointestinal Pathology. Day DW, Jass JR, Price AB, Shepherd NA, Sloan JM, Talbot JC, Warren BF, Williams GT. 4th Ed. Malden Oxford: Blackwell Science 2003. 231. Papp JP, Levine EJ, Thomas FB. Primary linitis plastica carcinoma of the colon and rectum. Am J Gastroenterol 1995; 90:141–145. 232. Broders AC. Grading of carcinoma. Minn Med 1925; 8:726–730. 233. Dukes CE. The classification of cancer of the rectum. J Pathol Bacteriol 1932; 35:323–332. 234. Jass JR, Atkin WS, Cuzick J, et al. The grading of rectal cancer: historical perspectives and multivariate analysis of 447 cases. Histopathology 1986; 10: 437–459. 235. Gibbs NM. Undifferentiated carcinoma of the large intestine. Histopathology 1977; 1:77–84. 236. Sundblad AS, Paz RA. Mucinous carcinomas of the colon and rectum and their relation to polyps. Cancer 1982; 50:2504–2509.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

237. Bonello JC, Sternberg SS, Quan SHQ. The significance of signet cell variety of adenocarcinoma of the rectum. Dis Colon Rectum 1980; 23:180–183. 238. Umpleby HC, Ranson DL, Williamson RCN. Peculiarities of mucinous colorectal carcinoma. Br J Surg 1985; 72:715–718. 239. Okuno M, Ikehara T, Nagayama M, et al. Mucinous colorectal carcinoma: clinical pathology and prognosis. Ann Sure 1988; 54:681–685. 240. Anthony T, George R, Rodriguez-Bigas M, et al. Primary signet-ring cell carcinoma of the colon and rectum. Ann Surg Oncol 1996; 3:344–348. 241. Shirouza K, Isomoto H, Morodomi T, et al. Primary linitis plastica carcinoma of the colon and rectum. Cancer 1994; 74:1863–1868. 242. Hase K, Shatney CH, Mochizuki H, et al. Long-term results of curative resection of ‘‘minimally invasive’’ colorectal cancer. Dis Colon Rectum 1995; 38:19–26. 243. Novello P, Duvillard P, Goandjouan S, et al. Carcinomas of the colon with multidirectional differentiation. Report of two cases and review of fhe literature. Dig Dis Sci 1995; 40:100–106. 244. Begin LR, Gordon PH, Alpert LC. Endophytic malignant transformation within flat adenoma of the colon: a potential diagnostic pitfall. Virchows Arch A Pathol Anat 1993; 442:415–418. 245. Kudo S, Tamura S, Nakajima T, et al. Depressed type of colorectal cancer. Endoscopy 1995; 27:54–57. 246. Kubota O, Kino I, Kimura T, et al. Nonpolypoid adenomas and adenocarcinomas found in background mucosa of surgically resected colons. Cancer 1996; 77:621–626. 247. Minamoto T, Sawaguchi K, Ohta T, et al. Superficial-type adenomas and adenocarcinomas of the colon and rectum: a comparative morphological study. Gastroenterology 1994; 106:1436–1443. 248. Tada S, Yao T, Iida M, et al. A clinicopathologic study of flat colorectal carcinoma. Cancer 1994; 74:2430–2435. 249. Iishi H, Kitamura S, Nakaizumi A, et al. Clinicopathologic features and endoscopic diagnosis of superficial early adenocarcinomas of the large intestine. Dig Dis Sci 1993; 38:1333–1337. 250. Togashi K, Konishi F, Koinuma K, et al. Flat and depressed lesions of the colon and rectum: pathogenesis and clinical management. Ann Acad Med Singapore 2003; 32:152–158. 251. Nivatvongs S. Surgical management of early colorectal cancer. World J Surg 2000; 24:1052–1055. 252. Mulsow J, Winter DC, O’Keane JC, O’Connell PR. Sentinel lymph node mapping in colorectal cancer. Br J Surg 2003; 90(6):659–667. 253. Bilchik AJ, Nora DT, Sobin LH, et al. Effect of lymphatic mapping on the new tumor-node-metastasis classification for colorectal cancer. J Clin Oncol 2003; 21:668–672. 254. Saha S, Dan AG, Berman B, et al. Lymphazurin 1% versus 99 mTc sulfur colloid for lymphatic mapping in colorectal tumors: a comparative analysis. Ann Surg Oncol 2004; 11:21–26. 255. Bembenek A, Rau B, Moesta T, et al. Sentinel lymph node biopsy in rectal cancer—not yet ready for routine clinical use. Surgery 2004; 135: 498–505. 256. Bertagnoli M, Redston M, Miedma B, et al. Sentinel node staging of resectable colon cancer: results of CALG B 80001. Proc Am Soc Clin Oncol 2004; 22:2465 (abstr. 3506; suppl). 257. Read TE, Fleshman JW, Caushaj PF. Sentinel lymph node mapping for adenocarcinoma of the colon does not improve staging accuracy. Dis Colon Rectum 2005; 48:80–85. 258. Stojadinovic A, Allen PJ, Protic M, et al. Colon sentinel lymph node mapping: practical surgical applications. J Am Coll Surg 2005; 201(2):297–313. 259. Cimmino VM, Brown AC, Szocik JF, et al. Allergic reactions to isosulfan blue during sentinel node biopsy-a common event. Surgery 2001; 130(5): 439–442. 260. Gutman M, Fidler IJ. Biology of human colon cancer metastases. World J Surg 1995; 19:226–234. 261. Griffiths JD, McKinna JA, Rowbotham HD, et al. Carcinoma of the colon and rectum: circulating malignant cells and five-year survival. Cancer 1973; 31:226–236. 262. Weiss L, Grundmann E,.Torhorst J, et al. Hematogenous metastatic patterns in colonic carcinoma: an analysis of 1541 necropsies. J Pathol 1986; 150:195–203. 263. Taylor I. Liver metastases from colorectal cancer: lessons from the past and present clinical studies. Br J Surg 1996; 83:456–460. 264. Killingback M, Wilson E, Hughes ESR. Anal metastases from carcinoma of the rectum and colon. Aust N Z J Surg 1965; 34:178–187. 265. Norgren J, Svensson JD. Anal implantation metastases from carcinoma of the sigmoid colon and rectum. A risk when performing anterior resection with the EEA stapler. Br J Surg 1985; 72:602. 266. Rollinson PD, Dundas SAC. Adenocarcinoma of the sigmoid colon seeding into preexisting fistula-in-ano. Br J Surg 1984; 71:664–665. 267. Rosenberg IL. The etiology of colonic suture-line recurrence. Ann R Coll Surg Engl 1979; 61:251–257. 268. Scott NA, Taylor BA, Wolff BG, et al. Perianal metastases from a sigmoid carcinoma—objective evidence of a clonal origin. Report of a case. Dis Colon Rectum 1988; 31:68–70.

191

269. Thomas DJ, Thompson MR. Implantation metastasis from adenocarcinoma of sigmoid colon into fistula-in-ano. J R Soc Med 1992; 85:361. 270. Murakami S, Terakado M, Hashimoto T, Tsuji Y, Okubo K, Hirayama R. Adrenal metastasis from rectal cancer: report of a case. Surg Today 2003:126–130. 271. Conti JA, Kemeny N, Klimstra D, et al. Colon carcinoma metastatic to the trachea. Report of a case and a review of the literature. Am J Clin Oncol 1994; 17:227–229. 272. Araki K, Kobayashi M, Ogata T, et al. Colorectal; carcinoma metastatic to skeletal muscle. Hepatogastroenterology 1994; 41:405–408. 273. Kupfer HWEM, Theunissen P, Delaere KPJ. Urethral metastasis from a rectal carcinoma. Acta Urol Belg 1995; 63:31–32. 274. Bhutani MS, Pacheco J. Metastatic colon carcinoma to oral soft tissues. Spec Care Dentist 1992; 12:172–173. 275. Vasilevsky CA, Alou-Khalel A, Rochon L, et al. Carcinoma of the colon presenting as tonsillar metastases. J Otolaryngol 1997; 26:325–326. 276. Gabriel WB, Dukes CE, Bussey HJR. Lymphatic spread in cancer of the rectum. Br J Surg 1935; 23:395–413. 277. Turnbull RB, Kyle K, Watson FR, et al. Cancer of the colon: the influence of the no-touch isolation technique on survival rates. Ann Surg 1967; 166:420– 427. 278. Wong JH, Severino R, Honnebier MB, Tom P, Namiki TS. Number of nodes examined and staging accuracy in colorectal carcinoma. J Clin Oncol 1999; 17(9):2896–2900. 279. Jass JR, Chapuis PH, Dixon MF, et al. Symposium on staging of colorectal cancer. Int J Colorectal Dis 1987; 2:123. 280. Greene FL, Page DL, Floming ID, et al. American Joint Committee for Cancer. Cancer Staging Handbook. 6th ed. New York: Springer-Verlag, 2002:127–138. 281. Astler VB, Coller FA. The prognostic significance of direct extension of carcinoma of the colon and rectum. Ann Surg 1954; 139:846–852. 282. Chapuis PH, Dent MF, Newland RC, et al. An evaluation of the American Joint Committee pTNM staging method for cancer of the colon and rectum. Dis Colon Rectum 1986; 29:6–10. 283. Chapuis PH, Dixon MF, Fielding LP, et al. Staging of colorectal cancer (Symposium). Int J Colorectal Dis 1987; 2:123–138. 284. Davis NC, Newland RC. Terminology and classification of colorectal adenpearcinoma. The Australian clinicopathological staging system. Aust N Z J Surg 1983; 53:211–221. 285. Gastrointestinal Tumor Study Group. Adjuvant therapy of colon cancer. Results of a prospectively randomized trial. N Engl J Med 1984; 310:737–743. 286. Terrazas JM, Val-Bernal JF, Buelta L. Staging of carcinoma of the colon and rectum. Surg Gynecol Obstet 1987; 165:255–259. 287. Goligher JC. The Dukes’ A, B, and C categorization of the extent of spread of carcinomas of the rectum. Surg Gynecol Obstet 1976; 146:793–794. 288. Kirklin JW, Dockerty MB, Waugh JM. The role of the peritoneal reflection in the prognosis of carcinoma of the rectum and sigmoid colon. Surg Gynecol Obstet 1949; 88:326–331. 289. Zinkin LD. A critical review of the classifications and staging of colorectal cancer. Dis Colon Rectum 1983; 26:37–43. 290. Newland RC, Chapuis PH, Smyth EJ. The prognostic value of substaging colorectal carcinoma. A prospective study of 1117 cases with standardized pathology. Cancer 1987; 60:852–857. 291. Spratt JS Jr. Gross rates of growth of colonic neoplasms and other variables affecting medical decisions and prognosis. In: Burdette WJ, ed. Carcinoma of the Colon and Antecedent Epithelium. Springfield, III.: Charles C Thomas, 1970:66–77. 292. Finlay IG, Brunton GF, Meek D, et al. Rate of growth of hepatic metastases in colorectal carcinoma. Br J Surg 1982; 69:689. 293. Havelaar I, Sugarbaker PH. Rate of growth of intraabdominal metastases from colon and rectal cancer followed by serial EOE CT. Cancer 1984; 54:163–171. 294. Bolin S, Nilsson E, Sjodahl R. Carcinoma of the colon and rectum—growth rate. Ann Surg 1983; 198:151–158. 295. Burnett KR, Greenbaum El. Rapidly growing carcinoma of the colon. Dis Colon Rectum 1981; 24:282–286. 296. Matsui T, Tsuda S, Yao K, Iwashita A, Sakurai T, Yao T. Natural history of early colorectal cancer: evolution of a growth curve. Dis Colon Rectum 2000; 43(suppl 10):S18–S22. 297. Aldridge MC, Phillips RKS, Hittinger R, et al. Influence of tumour site on presentation, management and subsequent outcome in large bowel cancer. Br J Surg 1986; 73:663–670. 298. Mandava N, Kumar S, Pizzi WF, et al. Perforated colorectal carcinomas. Am J Surg 1996; 172:236–238. 299. Runkel NS, Schlag P, Schwarz V, et al. Outcome after emergency surgery for cancer of the large intestine. Br J Surg 1991; 78:183–188. 300. The consultant surgeons and pathologists of the Lothian and Borders health boards. Lothian and Borders large bowel cancer project: immediate outcome after surgery. Br J Surg 1995; 82:888–890.

192

& PART I: COLORECTAL DISORDERS

301. Beart RW Jr., Melton LJ, Maruta M, et al. Trends in right and left sided colon cancer. Dis Colon Rectum 1983; 26:393–398. 302. Farrands PA, Hardcastle JD. Colorectal screening by a self-completion questionnaire. Gut 1984; 25:445–447. 303. Ramsay JA, Rose TH, Ross T. Colonic carcinoma presenting as an appendiceal abscess in a young woman. Can J Surg 1996; 39:53–56. 304. Beart RW, Steele GD Jr., Menck HR, et al. Management and survival of patients with adenocarcinoma of the colon and rectum: a national survey of the commission on cancer. J Am Coll Surg 1995; 181:225–236. 305. Acher PL, Al-Mishlab T, Rahman M, Bates T. Iron-deficiency anaemia and delay in the diagnosis of colorectal cancer. Colorectal Dis 2003; 5:145–148. 306. Church J, McGannon E. Family history of colorectal cancer: how often and how accurately is it recorded? Dis Colon Rectum 2000; 43:1540–1544. 307. Rosato FE, Shelly WB, Fitts WT Jr., et al. Non-metastatic cutaneous manifestations of cancer of the colon. Am J Surg 1969; 117:277–281. 308. Halak M, Hazzan D, Kovacs Z, Shiloni E. Synchronous colorectal and renal carcinomas: a noteworthy clinical entity. Report of five cases. Dis Colon Rectum 2000; 43:1314–1315. 309. Vasilevsky CA, Gordon PH. Colonoscopy in the follow-up of patients with colorectal carcinoma. Can J Surg 1988; 31:188–190. 310. Evers BM, Mullins RJ, Mathews TH. Multiple adenocarrinomas of the colon and rectum: an analysis of incidences and current trends. Dis Colon Rectum 1988; 31:518–522. 311. Pinol V, Andreu M, Castells A, et al. Synchronous colorectal neoplasms in patients with colorectal cancer: predisposing individual and familial factors. Dis Colon Rectum 2004; 47:1192–1200. 312. Adloff M, Arnaud JP, Bergamaschi R, et al. Synchronous carcinoma of the colon and rectum: prognostic and therapeutic implications. Am J Surg 1989; 157:299–302. 313. Langevin JM, Nivatvongs S. The true incidence of synchronous cancer of the large bowel: a prospective study. Am J Surg 1984; 147:330–333. 314. Rider JA, Kirsner JB, Moeller HC, et al. Polyps of the colon and rectum. JAMA 1959; 170:633. 315. Slater G, Fleshner P, Aufses AH. Colorectal cancer location and synchronous adenomas. Am J Gastroenterol 1988; 83:832–836. 316. Chu DZ, Giacoo G, Martin RG, Guinee VF. The significance of synchronous carcinoma and polyps in the colon and rectum. Cancer 1986; 57:445–450. 317. Lee TK, Barringer M, Myers RT, et al. Multiple primary carcinoma of the colon and associated extracolonic primary malignant tumors. Ann Surg 1982; 195:501–507. 318. Tanaka H, Hiyama T, Hanai A, et al. Second primary cancers following colon and rectal cancer in Osaka Japan. Jpn J Cancer Res 1991; 82:1356–1365. 319. Schoen RE, Weissfeld JL, Kuller LH. Are women with breast, endometrial, or ovarian cancer at increased risk for colorectal cancer? Am J Gastroenterol 1994; 89:835–842. 320. De Dombal FT, Matharu SS, Staniland JR, et al. Presentation of cancer to hospital as ‘‘acute abdominal pain.’’ Br J Surg 1980; 67:413–416. 321. Ohman U. Prognosis in patients with obstructing colorectal carcinoma. Am J Surg 1982; 143:742–747. 322. Wolloch Y, Zer M, Lurie M, et al. Ischemic colitis proximal to obstructing carcinoma of the colon. Am J Proctol 1979; 30:17–22. 323. Saegesser F, Sandblom P. Ischemic lesions of the distended colon. A complication of obstructive colorectal cancer. Am J Surg 1975; 129:309–315. 324. Ueyama T, Yao T, Nakamura K, et al. Obstructing carcinoma of the colon and rectum: clinicopathologic analysis of 40 cases. Jpn J Clin Oncol 1991; 21: 100–109. 325. Seow-Choen F, Chua TL, Goh HS. Ischemic colitis and colorectal cancer: some problems and pitfalls. Int J Colorectal Dis 1993; 8:210–212. 326. Wolmark N, NSABP Investigators. The prognostic significance of tumour location and bowel obstruction in Dukes’ B and C colorectal cancer. Ann Surg 1983; 198:743–750. 327. Umpleby HC, Williamson RCN. Survival in acute obstructing colorectal carcinoma. Dis Colon Rectum 1984; 27:299–304. 328. Panwalker AP. Unusual infections associated with colorectal carcinoma. Rev Infect Dis 1988; 10:347–364. 329. Lam S, Greenberg R, Bank S. An unusual presentation of colon cancer: purulent pericarditis and cardiac tamponade due to Bacteroides fragilis. Am J Gastroenterol 1995; 90:1518–1520. 330. Belinkie SA, Narayanan NC, Russell JC, et al. Splenic abscess associated with streptococcus bovis septicemia and neoplastic lesions of the colon. Dis Colon Rectum 1983; 26:823–824. 331. Silver SC. Streptococcus bovis endocarditis and its association with colonic carcinoma. Dis Colon Rectum 1984; 27:613–614. 332. Legier J. Streptococcus salivarius meningitis and colonic carcinoma. South Med J 1991; 84:1058–1059. 333. Lonardo A, Grisendi A, Pulvirenti M, et al. Right colon adenocarcinoma presenting as bacteroides liver abscesses. J Clin Gastroenterol 1992; 14:335–338. 334. Teitz S, Guidetti-Sharon A, Monro H, et al. Pyeogenic liver abscess: warning indicator of silent colonic cancer. Dis Colon Rectum 1995; 58:1220–1223.

335. Kudsk KA. Occult gastrointestinal malignancies producing metastatic clostridium septicum infections in diabetic patients. Surgery 1992; 112:765–772. 336. Lorimer JW, Eldus LB. Invasive clostridium septicum infection in association with colorectal carcinoma. Can J Surg 1994; 37:245–249. 337. Poleski MH, Gordon PH. Screening for carcinoma of the colon: pitfalls of the hemoccult test. In: Nelson RL, ed. Problems in Current Surgery. Controversies in Colon Cancer. Philadelphia: JB Lippincott, 1987:1–10. 338. Nivatvongs S, Fryd DS. How far does the proctosigmoidoscope reach? N Engl J Med 1980; 303:380–382. 339. Marks G, Gathright JB, Boggs HW, et al. Guidelines for use of the flexible fiberoptic sigmoidoscope in the management of the surgical patient. Dis Colon Rectum 1982; 23:187–190. 340. Traul DG, Davis CB, Pollock JC, et al. Flexible fiberoptic sigmoidoseopy—the Monroe Clinic experience. A prospective study of 5,000 examinations. Dis Colon Rectum 1983; 26:161–166. 341. Bernard D, Tasse D, Morgan S, et al. Is preoperative colonoscopy in carcinoma a realistic and valuable proposition? Can J Surg 1987; 30:87–89. 342. Isler JT, Brown PC, Lewis FG, et al. The role of preoperative colonoscopy in colorectal cancer. Dis Colon Rectum 1987; 30:435–439. 343. Finan PJ, Ritchie JK, Hawley PR. Synchronous and early metachronous carcinomas of the colon and rectum. Br J Surg 1987; 74:945–947. 344. Reilly JC, Rusin LC, Theuerkauf FJ Jr. Colonoscopy: its role in cancer of the colon and rectum. Dis Colon Rectum 1982; 25:532–538. 345. Sollenberger LL, Eisenstat TE, Rubin RJ, et al. Is preoperative colonoscopy necessary in carcinoma of the colon and rectum. Am Surg 1988; 54: 113–115. 346. Thomas RD, Fairhurst JJ, Frost RA. Wessex regional radiology audit: barium enema in colorectal carcinoma. Clin Radiol 1995; 50:647–650. 347. Peel AL, Benyon L, Grace RH. The value of routine preoperative urological assessment in patients undergoing elective surgery for diverticular disease or carcinoma of the large bowel. Br J Surg 1980; 67:42–47. 348. Vezeridis MP, Petrelli NJ, Mittelman A. The value of routine preoperative urologic evaluation in patients with colorectal carcinoma. Dis Colon Rectum 1987; 30:758–760. 349. Phillips R, Hittinger R, Saunders V, et al. Pre-operative urography in large bowel cancer: a useless investigation? Br J Surg 1983; 70:425–427. 350. Tartter PI, Steinberg BM. The role of preoperative intravenous pyelogram in operations performed for carcinoma of the colon and rectum. Surg Gynecol Obstet 1986; 163:65–69. 351. Grace RH, Hale M, Mackie G, et al. Role of ultrasound in the diagnosis of liver metastases before surgery for large bowel cancer. Br J Surg 1987; 74:480–481. 352. Rafaelsen SR, Kronborg O, Larsen C, et al. Intraoperative ultrasonography in detection of hepatic metastases from colorectal cancer. Dis Colon Rectum 1995; 38:355–360. 353. Meijer S, Pavel MA, Cuesta MA, et al. Intra-operativc ultrasound in detection of liver metastases. Eur J Cancer 1995; 31A:1210–1211. 354. Machi J, Isomoto H, Kurohiji T, et al. Accuracy of intraoperative ultrasonography in diagnosing liver metastases from colorectal cancer: evaluation with postoperative follow up results. World J Surg 1991; 15: 551–557. 355. Takeuchi N, Ramirez JM, Mortensen NJM, et al. Intraoperative ultrasonography in the diagnosis of hepatic metastases during surgery for colorectal cancer. Int J Colorectal Dis 1996; 11:92–95. 356. Kerner BA, Oliver GC, Eisenstat TE, et al. Is preoperative computerized tomography useful in assessing patients with colorectal carcinoma? Dis Colon Rectum 1993; 36:1050–1053. 357. Theoni RF, Rogalla P. CT for the evaluation of carcinomas in the colon and rectum. Semin Ultrasound CT MR 1995; 16:112–126. 358. Mauchley DC, Lynge DC, Langdale LA, Stelzner MG, Mock CN, Billingsley KG. Clinical utility and cost-effectiveness of routine preoperative computed tomography scanning in patients with colon cancer. Am J Surg 2005; 189(5):512–517. 359. Tempero M, Brand R, Holderman K, et al. New imaging techniques in colorectal cancer. Semin Oncol 1995; 22:448–471. 360. Zerhouni EA, Rntter C, Hamilton SR, et al. CT and MR imaging in the staging of colorectal carcinoma: report of the Radiology Diagnostic Oncology Group II. Radiology 1996; 200:443–451. 361. Johnson K, Bakhsh A, Young D, Martin TE Jr., Arnold M. Correlating computed tomography and positron emission tomography scan with operative findings in metastatic colorectal cancer. Dis Colon Rectum 2001; 44: 354–357. 362. Bertsch DJ, Burak WE, Young DC, et al. Radioimmunoguided surgery for colorectal cancer. Ann Surg Oncol 1996; 3:310–316. 363. Arnold MW, Hitchcock CL, Young DC, et al. Intra-abdominal patterns of disease dissemination in colorectal cancer identified using radioimmunoguided surgery. Dis Colon Rectum 1996; 39:509–513. 364. Dominguez JM, Wolff BE, Nelson H, et al. 111In-CYT-103 scanning in recurrent colorectal cancer—does it affect standard management? Dis Colon Rectum 1996; 39:514–519.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

365. Moffat FL, Pinsky CM, Hammershaimb L, et al. Clinical utility of external immunoscintigraphy with the IMMU-4 technetium-99m Fab’ antibody fragment in patients undergoing surgery for carcinoma of the colon and rectum: results of a pivotal, phase III trial. J Clin Oncol 1996; 14: 2295–2305. 366. Corman ML, Galandiuk S, Block GE, et al. Immunoscintigraphy with 111Insatumomab pendetide in patients with colorectal adenocarcinoma: performance and impact on clinical management. Dis Colon Rectum 1994; 37: 129–137. 367. Galandiuk S. Immunoscintigraphy in the surgical management of colorectal cancer. J Nucl Med 1993; 34:541–544. 368. Limberg B. Diagnosis and staging of colonic tumors by conventional abdominal sonography as compared with hydrocolonic sonography. N Engl J Med 1992; 327:65–69. 369. Leveson SH, Wiggins PA, Giles GR, et al. Deranged liver blood flow patterns in the detection of liver metastases. Br J Surg 1985; 72:128–130. 370. Dian-Yuan Z, Fu-cai F, Ya-li Z, et al. Comparison of Sham’s test for rectal mucus to an immunological test for fecal occult blood in large intestinal carcinoma screening. Analysis of a checkup of 6480 asymptomatic patients. Chin Med J 1993; 106:739–742. 371. Sakamoto K, Muranti M, Ogawa T, et al. Evaluation of a new test for colorectal neoplasms: a prospective study of a symptomatic population. Cancer Biother 1993; 81:49–54. 372. Chen YL. The diagnosis of colorectal cancer with cytologic brushings under direct vision at fiberoptic colonoscopy. A report of 59 cases. Dis Colon Rectum 1987; 30:342–344. 373. Rosman AS, Federman O, Feinman L. Diagnosis of colonic cancer by lavage cytology with an orally administered balanced electrolyte solution. Am J Gastroenterol 1994; 89:51–56. 374. Gold P, Freedman SO. Demonstration of tumor-specific antigens in human colonic cardnomata by immunological tolerance and absorption techniques. J Exp Med 1965; 121:439–462. 375. Gold P. The carcinoembryonic antigen (CEA): discovery and three decades of study. Perspect Colon Rectal Surg 1996; 9(2):1–47. 376. Kyzer S, Mitmaker B, Gordon PH, et al. Proliferative activity of colonic mucosa at different distances from primary adenocarcinoma as determined by the presence of statin: a non proliferation-specific nuclear protein. Dis Colon Rectum 1992; 35:879–883. 377. Stein R, Juweid M, Mattes MJ, Goldbenberg DM. Carcinoembryonic antigen as a target for radioimmunotherapy of human medullary thyroid carcinoma: antibody processing, targeting, and experimental therapy with 131I and 90Y labeled MAbs. Cancer Biother Radiopharm 1999; 14(1):37–47. 378. Bockhorn M, Frilling A, Rewerk S, et al. Lack of elevated serum carcinoembryonic antigen and calcitonin in medullary thyroid carcinoma. Thyroid 2004; 14: 468–470. 379. Moorehead RJ, Hoper M, McKelvey STD. Assessment of ornithine decarboxylase activity in rectal mucosa as a marker for colorectal adenomas and carcinoma. Br J Surg 1987; 74:364–365. 380. Narisawa T, Takahashi M, Niwa M, et al. Increased mucosal ornithine decarboxylase activity in large bowel with multiple tumors, adenocarcinoma and adenomas. Cancer 1989; 63:1572–1576. 381. Gelister JSK, Jass RJ, Mahmoud M, et al. Role of urokinase in colorectal neoplasia. Br J Surg 1987; 74:460–463. 382. Fazio VW, Tekkis PP, Remzi F, Lavery IC. Assessment of operative risk in colorectal cancer surgery: the Cleveland Clinic Foundation colorectal cancer model. Dis Colon Rectum 2004; 47(12):2015–2024. 383. Portnoy J, Kagan E, Gordon PH, et al. Prophylactic antibiotics in elective colorectal surgery. Dis Colon Rectum 1983; 26:310–313. 384. Rubin RJ, White RA, Eisenstat TE, et al. Left subcostal transverse incision combined with the right lateral position for excising the splenic flexure: a reintroduction. Perspect Colon Rectal Surg 1988; l(2):41–47. 385. Thomas WM, Morris DL, Hardcastle JD. Contact ultrasonography in the detection of liver metastases from colorectal cancer: an in vitro study. Br J Surg 1987; 74:955–956. 386. Busuttil RW, Foglia RP, Longmire WP. Treatment of carcinoma of the sigmoid colon and upper rectum. A comparison of local segmental resection and left hemicolectomy. Arch Surg 1977; 112:920–923. 387. Dwight RW, Higgins GA, Keehn RJ. Factors influencing survival after resection in cancer of the colon and rectum. Am J Surg 1969; 117:512–522. 388. Grinnell RS. Results of ligation of inferior mesenteric artery at the aorta in resections of carcinoma of the descending and sigmoid colon and rectum. Surg Gynecol Obstet 1965; 170:1031–1046. 389. Pezim ME, Nicholls RJ. Survival after high or low ligation of the inferior mesenteric artery during curative surgery for rectal cancer. Ann Surg 1984; 200:729–733. 390. Enker WE, Laffer VT, Black GE. Enhanced survival of patients with colon and rectal cancer is based upon wide anatomic resection. Ann Surg 1979; 190:350–360. 391. Brief DK, Brener BJ, Goldenkrantz R, et al. Defining the role of subtotal colectomy in the treatment of carcinoma of the colon. Ann Surg 1991; 213:248–252.

193

392. Fisher ER, Turnbull RB Jr. The cytologic demonstration and significance of tumor cells in the mesenteric venous blood in patients with colorectal carcinoma. Surg Gynecol Obstet 1955; 100:102–108. 393. Cole WH, Packard D, Southwick HW. Carcinoma of the colon with special reference to the prevention of recurrence. JAMA 1954; 155:1549–1553. 394. Turnbull RB Jr., Kyle K, Watson FR, et al. Cancer of the colon: the influence of the no-touch technique on survival rates. Ann Surg 1967; 166:420–427. 395. Cohn I Jr., Floyd CE, Atik M. Control of tumor implantation during operations of the colon. Ann Surg 1963; 157:825–838. 396. Long RTL, Edwards RH. Implantation metastases as a cause of local recurrence of colorectal carcinoma. Am J Surg 1989; 157:194–201. 397. Wiggers T, Jeekel J, Arends JW. No-touch isolation technique in colon cancer: a controlled prospective trial. Br J Surg 1988; 75:409–415. 398. Gordon PH, Dalrymple S. The use of staples for reconstruction after colonic and rectal surgery. In: Ravitch MM, Steichen FM, eds. Principles and Practice of Surgical Stapling. Chicago: Year Book Medical Publishers, 1987:402–431. 399. Welter R, Charlier A, Psalmon F. Personal communication. In: Steichen FM, Ravitch MM, eds. Stapling in Surgery. Chicago: Year Book Medical Publishers, 1988:271. 400. Chassin JL, Rifkind KM, Turner JW. Errors and pitfalls in stapling gastrointestinal anastomoses. Surg Clin North Am 1984; 64:441–459. 401. Kyzer S, Gordon PH. The stapled functional end-to-end anastomoses following colonic resection. Int J Colorectal Dis 1992; 7:125–131. 402. Chassin JL, Rifkind KM, Sussman B, et al. The stapled gastrointestinal anastomosis: incidence of postoperative complications compared with the sutured anastomosis. Ann Surg 1978; 188:689–696. 403. Fortin CL, Poulin EC, Leclerc V. Evaluation de utilisation des appareils d’autosuture en chirurgie digestive. Can J Surg 1979; 22:580–582. 404. Brodman RF, Brodman HR. Staple suturing of the colon above the peritoneal reflection. Arch Surg 1981; 116:191–192. 405. Reuter MJP. Les sutures me´caniques en chirurgie digestive et pulmonaire. Thesis. Unive´rsity Louis Pasteur, Faculte´ de Me´decine de Strasbourg, France, 1982. 406. Scher KS, Scott-Conner C, Jones CW, et al. A comparison of stapled and sutured anastomoses in colonic operations. Surg Gynecol Obstet 1982; 155: 489–493. 407. Steichen FM, Ravitch MM. Stapling in Surgery. Chicago: Year Book Medical Publishers, 1984:271. 408. Tuchmann A, Dinstl K, Strasser K, et al. Stapling devices in gastrointestinal surgery. Int Surg 1985; 70:23–27. 409. Kracht M, Hay JM, Fagniez PL, et al. Ileocolonic anastomosis after right hemicolectomy for carcinoma: stapled or hand-sewn? A prospective, multicenter, randomized trial. Int J Colorectal Dis 1993; 8:29–33. 410. Corman ML. Colon and Rectal Surgery. 2nd ed Philadelphia: JB Lippincott, 1989:417. 411. Abcarian H, Pearl RK. Simple technique for high ligation of the inferior mesenteric artery and vein. Dis Colon Rectum 1991; 34:1138. 412. Heald RJ. Anterior resection of the rectum. In: Fielding LP, Goldberg SM, eds. Rob and Smith’s Operative Surgery. Surgery of the Colon, Rectum, and Anus. 5th ed. London: Butterworth-Heinemann Ltd, 1993:456–471. 413. Birnkrant A, Sampson J, Sugarbaker PH. Ovarian metastases from colorectal cancer. Dis Colon Rectum 1986; 29:767–771. 414. Graffner HOL, Aim POA, Oscarson JEA. Prophylactic oophorectomy in colorectal carcinoma. Am J Surg 1983; 146:233–235. 415. Sielezneff I, Salle E, Antoine K, Thirion X, Brunet C, Sastre B. Simultaneous bilateral oophorectomy does not improve prognosis of postmenopausal women undergoing colorectal resection for cancer. Dis Colon Rectum 1997; 40(11):1299–1302. 416. Young-Fadok TM, Wolff B, Nivatvongs S, et al. Prophylactic oophorectomy in colorectal carcinoma: preliminary results. Dis Colon Rectum 1998; 41: 277–285. 417. Cutait R, Lesser ML, Enker WE. Prophylactic oophorectomy in surgery for large bowel cancer. Dis Colon Rectum 1983; 26:6–11. 418. O’Brien PH, Newton PB, Metcalf JS, et al. Oophorectomy in women with carcinoma of the colon and rectum. Surg Gynecol Obstet 1981; 153:827–830. 419. MacKeigan JM, Ferguson JA. Prophylactic oophorectomy and colorectal cancer in premenopausal patients. Dis Colon Rectum 1979; 22:401–405. 420. Blamey S, McDermott F, Pihl E, et al. Ovarian involvement in adenocarcinoma of the colon and rectum. Surg Gynecol Obstet 1981; 153:42–44. 421. Steele G, Augenlicht L, Begg C, et al. National Institutes of Helth consensus development conference statement—adjuvant therapy for patients with colon and rectal cancer. JAMA 1990; 264:1444. 422. Sugarbaker PH, Gunderson LL, Wittes RE. Colorectal cancer. In: DeVita VT Jr., Hellman S, Rosenberg SA, eds. Cancer Principles and Practices of Oncology. 2nd ed. Philadelphia: JB Lippincott, 1985:795–884. 423. Duttenhaver JR, Hoskins RB, Gunderson LL, et al. Adjuvant postoperative radiation therapy in the management of adenocarcinoma of the colon. Cancer 1986; 57:955–963. 424. Ghossein NA, Samala EC, Alpert S, et al. Elective postoperative radiotherapy after incomplete resection of a colorectal cancer. Dis Colon Rectum 1981; 24:252–256.

194

& PART I: COLORECTAL DISORDERS

425. Wong CS, Harwood AR, Cummings BJ, et al. Postoperative local abdominal irradiation for cancer of the colon above the peritoneal reflection. Int J Radiat Oncol Biol Phys 1985; 11:2067–2071. 426. Kopelson G. Adjuvant postoperative radiation therapy for colorectal carcinoma above the peritoneal reflection. I. Sigmotd colon. Cancer 1983; 51:1593–1598. 427. Kopelson G. Adjuvant postoperative radiation therapy for colorectal carcinoma above the peritoneal reflection. II. Antimesenteric wall, ascending and descending colon and cecum. Cancer 1983; 52:633–636. 428. Minsky BD. Adjuvant radiation therapy for colon cancer. Cancer Treat Rev 1995; 2:407–414. 429. Schabel FM Jr. Rationale for perioperative anticanccr treatment. Recent Results Cancer Res 1985; 98:1–10. 430. Gastrointestinal Tumor Study Group. Adjuvant therapy of colon cancer: results of a prospectively randomized trial. N EngI J Med 1984; 310:737–743. 431. Gilbert JM, Hellman K, Evans M, et al. Randomized trial of oral adjuvant raz˙oxane (ICRF 159) in resectable colorectal cancer: five year follow-up. Br J Surg 1986; 73:446–450. 432. Grage TB, Moss SE. Adjuvant chemotherapy in cancer of the colon and rectum: demonstration of effectiveness of prolonged 5-FU chemotherapy in a prospectively controlled randomized trial. Surg Clin North Am 1981; 61:1321–1329. 433. Higgins GA Jr., Amadeo JH, McElhinney J, et al. Efficacy of prolonged intermittent therapy with combined 5-fluorouracil and methyl CCNU following resection for carcinoma of the large bowel. A Veterans Administration Surgical Oncology Group report. Cancer 1984; 53:1–8. 434. Laurie J, Moertel C, Fleming T, et al. Surgical adjuvant therapy of poor prognostic colorectal cance˙r with levamisole alone or combined levamisole and 5-fluorouracil. A North Central Cancer Treatment Group and Mayo Clinic study. Proc Am Soc Clin Oncol 1986; 5:316. 435. Panettiere FJ, Rogers AR. SWOG large bowel post-operative program. Proceedings of the Fourth International Conference on the Adjuvant Therapy of Cancer [abstr T7]. 1984:94. 436. Buyse M, Zelenuick-Jacquolte A, Chalmers TC. Adjuvant therapy for colorectal cancer, why we still don’t know. JAMA 1988; 259:3571–3578. 437. Windle A, Bell PRF, Shaw D. Five year results of a randomized trial of adjuvant 5-fluorouracil and levamisole in colorectal cancer. Br J Surg 1987; 74:569–572. 438. Laurie J, Moertel C, Fleming T, et al. Surgical adjuvant therapy of large bowel carcinoma: an evaluation of levamisole and the combination of levamisole and 5-fluorouracil. A study of the North Central Cancer Treatment Group and Mayo Clinic. J Clin Oncol 1989; 7:1447–1456. 439. Moertel C, Fleming TR, Macdonald JS, et al. Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 1990; 322:352–358. 440. Wolmark N, Rocketter H, Mamounas E, et al., Clinical trial to assess the relative efficacy of fluorouracil and leucovorin, fluorouracil and levamisole, and fluorouracil, leucovorin and levamisole in patients with Dukes B and C carcinoma of the colon: Results from National Surgical Adjuvant Breast and Bowel Project C-04. J. Clin. Oncol. 1999; 17: 3553–9. 441. Le Voyer TE, Sigurdson ER, Hanlon AL, et al. Colon cancer survival is associated with increasing number of lymph nodes analyzed: a secondary survey of intergroup trial INT-0089. J Clin Oncol 2003; 21:2912–2119. 442. Poplin EA, Benedetti JK, Estes NC, et al. Phase III Southwest oncology group 6415/Intergroup 0153 randomized trial of fluorouracil, leucovorin, and levamisole versus fluorouracil continuous infusion and lovamisole for adjuvant treatment of stage III and high-risk stage II colon cancer. J Clin Oncol 2005; 23(9):1819–1825. 443. Thirion P, Michiels S, Pignon JP,et al. Modulation of fluorouracil by leucovorin in patients with advance colorectal cancer: an updated meta-analysis. J Clin Oncol 2004; 22(18):3766–3775. 444. Moertel CG, Flaming TR, Macdonald JS, et al. Fluorouracil plus levamisole as effective adjuvant therapy after resection of stage III colon carcinoma: a final report. Ann Intern Med 1995; 122:321–326. 445. Wolmark N, Fisher B, Rockette H, et al. Postoperative adjuvant chemotherapy or BCG for colon cancer. Results from NSABP protocol C-01. J Natl Cancer Inst 1988; 80:30–36. 446. Wolmark N, Rockette H, Fisher B, et al. The benefit of leucovorin-modulated fluorouracil as postoperative adjuvant therapy for primary colon cancer: results from National Surgical Adjuvant Breast and Bowel Project Protocol C-03. J Clin Oncol 1993; 11:1879–1887. 447. Piedbois P, Buyse M, Rustum Y, et al. Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: evidence in terms of response rate. J Clin Oncol 1992; 10:896–903. 448. Wolmark N, Bryant J, Smith R, et al. Adjuvant 5-fluorouracil and leucovorin with or without interferon alfa-2a in colon carcinoma: National Surgical Adjuvant Breast and Bowel Project protocol C-05. J Natl Cancer Inst 1998; 90: 1810–1816. 449. Mamounas EP, Wieand S, Wolmark N, et al. Comparative efficacy of adjuvant chemotherapy in patients with Dukes’ B versus Dukes’ C colon cancer: results for four National Surgical Adjuvant Breast and Bowel Project adjuvant studies (C-01, C-02, C-03, and C-04). J Clin Oncol 1999; 17: 1349–1355.

450. International Multicentre Pooled Analysis of B2 Colon Cancer Trials (IMPACT B2) investigators. Efficacy of adjuvant fluorouracil and folinic acid in B2 colon cancer. J Clin Oncol 1999; 17:1356. 451. International Multicentre Pooled Analysis of Colon Cancer Trials (IMPACT} investigators. Efficacy of adjuvant fluorouracil and folinic acid in colon cancer. Lancet 1995; 345:939–944. 452. Dube´ S, Heyen F, Jenicek M. Adjuvant chemotherapy in colorectal carcinoma. Results of a meta-analysis. Dis Colon Rectum 1997; 40:35–41. 453. Sakamoto J, Ohashi Y, Hamada C, Buyse M, Burzykowski T, Piedbois P, Meta-Analysis Group of the Japanese Society for Cancer of the colon and Rectum; Meta-Analysis Group in Cancer. Efficacy of oral adjuvant therapy after resection of colorectal cancer: 5-year results from three randomized trials. J Clin Oncol 2004; 22:484–492. 454. Andre T, Boni C, Mounedji-Boudiaf L, et al. Multicenter International Study of Oxaliplatin/5-Fluorouracil/Leucovorin in the Adjuvant Treatment of Colon Cancer (MOSAIC) Investigators. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med 2004; 350:2343–2351. 455. Martenson JA Jr., Willett CG, Sargent DJ, et al. Phase III study of adjuvant chemotherapy and radiation therapy compared with chemotherapy alone in the surgical adjuvant treatment of colon cancer: results of intergroup protocol 0130. J Clin Oncol 2004; 22:3277–3283. 456. Weiss L, Grundmann E, Torhost J, et al. Hematogenous metastatic patterns in colonic carcinoma: an analysis of 1,541 necropsies. J Pathol 1986; 150:195–203. 457. Taylor L, Machin D, Mullee M, et al. A randomized control trial of adjuvant portal vein cytotoxic perfusion in colorectal cancer. Br J Surg 1985; 72:359–363. 458. Swiss Group for Clinical Cancer Research (SAKK). Long-term results of a single course of adjuvant intraportal chemotherapy for colorectal cancer. Lancet 1995; 345:349–353. 459. Wolmark N, Bockette H, Wickerham DL, et al. Adjuvant therapy for Dukes’ A, B, and C adenocarcinoma of the colon with portal vein. Fluorouracil hepatic infusion: preliminary results of National Surgical Adjuvant Breast and Bowel Project Protocol C-02. J Clin Oncol 1990; 8:1466–1475. 460. Wereldsma J, Bruggink E, Meijer W, et al. Adjuvant portal liver infusion in colorectal cancer with 5-fluorouracil/heparin vs. urokinase vs. control. Results of a prospective randomized clinical trial (colorectal adenocarcinoma trial). Cancer 1990; 65:425. 461. Piedbois P, Buyse M, Gray R, et al. Portal vein infusion is an effective adjuvant treatment for patients with colorectal cancer. Proc Am Soc Clin Oncol 1995; 14:192. 462. Chau I, Chan S, Cunninghan D. Overview of preoperative and postoperative therapy for colorectal cancer: the European and United States perspectives. Clin Colorectal Cancer 2003; 3(1):19–33. 463. Haydon A. Adjuvant chemotherapy in colon cancer: what is the evidence? Intern Med J 2003; 33:119–124. 464. Chapuis PH, Dent OF, Bokey EL, Newland RC, Sinclair G. Adverse histopathological findings as a guide to patient management after curative resection of node-positive colonic cancer. Br J Surg 2004; 91:349–354. 465. Figueredo A, Charette ML, Maroun J, Brouwers MC, Zuraw L. Adjuvant therapy for stage II colon cancer: a systematic review from the Cancer Care Ontario Program in evidence-based care’s gastrointestinal cancer disease site group. J Clin Oncol 2004; 22:3395–3407. 466. Benson AB 3rd, Schrag D, Somerfield MR, et al. American Society of Clinical Oncology recommendations on adjuvant chemotherapy for stage II colon cancer. J Clin Oncol 2004; 22:3408–3419. 467. Ribic CM, Sargent DJ, Moore MJ, et al. Tumor microsatellite-instability status as a predictor of benefit from fluorouracil-based adjuvant chemotherapy for colon cancer. N Engl J Med 2003; 349:247–257. 468. Culakova E, Khorama A, Kuderer M, Crawford J, Dale DC, Lyman GH. For the ANC study group. Hospitalization with febrile neutropenia in colorectal cancer patients. J Clin Oncol 2004; 22:145; 295. 469. Arora A, Potter J. Older patients with colon cancer: is adjuvant chemotherapy safe and effective? J Am Geriatr Soc 2003; 51:567–569. 470. Rothenberg ML, Meropol NJ, Poplin EA, Van Custem E, Wadler S. Morality associated with irinotecan plus bolus fluorouracil/leucovorin: summary findings of an independent panel. J Clin Oncol 2001; 19:3801–3807. 471. Colorectal Cancer Chemotherapy Study Group of Japan–The 2nd Trial. Results of a randomized trial with or without 5-FU-based preoperative chemotherapy followed by postoperative chemotherapy in resected colon and rectal carcinoma. Jpn J Clin Oncol 2003; 33:288–296. 472. Douillard JY, Hoff PM, Skillings JR, et al. Multicenter phase III study of uracil/tegafur and oral leucovorin versus fluorouracil and leucovorin in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2002; 20:3605–3616. 473. Kabbinavar FF, Hambleton J, Mass RD, Hurwitz HI, Bergsland E, Sarkar S. Combined analysis of efficacy: the addition of bevacizumab to fluorouracil/ leucovorin improves survival for patients with metastatic colorectal cancer. J Clin Oncol 2005; 23(16):3706–3712. 474. Hurwitz HI, Fehrenbacher L, Hainsworth JD, et al. Bevacisumab in combination with fluorouracil and leucovorin: an active regimen for first-line metastatic colorectal cancer. J Clin Oncol 2005; 23(15):3502–3508.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

475. Lise M, Gerard A, Nitti D, et al. Adjuvant therapy for colorectal cancer. The EORTC experience and a review of the literature. Dis Colon Rectum 1987; 30:847–854. 476. Gray BN, Walker C, Andrewartha L, et al. Melbourne trial of adjuvant immunotherapy in operable large bowel cancer. Aust N Z J Surg 1988; 58:43–46. 477. Wiesenfeld M, O’Connell MJ, Wieand HS, et al. Controlled clinical trial of interferon-a vs. postoperative surgical adjuvant therapy for colon cancer. J Clin Oncol 1995; 13:2324–2329. 478. Reithmulller G, Schnieder-Godicke E, Schlimok G, et al. Randomized trial of monoclonal antibody for adjuvant therapy of resected Dukes’ C colorectal carcinoma. Lancet 1994; 343:1177–1183. 479. Guillou PJ. Potential impact of immunobiotechnology on cancer therapy. Br J Surg 1987; 74:705–710. 480. McArdle CS, Hole DJ. Emergency presentation of colorectal cancer is associated with poor 5-year survival. Br J Surg 2004; 91:605–609. 481. Jestin P, Nilsson J, Heurgren M, Pahlman L, Glimelius B, Gunnarsson U. Emergency surgery for colonic cancer in a defined population. Br J Surg 2005; 92:94–100. 482. Amsterdam E, Krispin M. Primary resection with colostomy for obstructive carcinoma of the left side of the colon. Am J Surg 1985; 150:558–560. 483. White CM, Macfie J. Immediate colectomy and primary anastomoses for acute obstruction due to carcinoma of the left colon and rectum. Dis Colon Rectum 1985; 28:155–157. 484. Sjodahl R, Franzen T, Nystrom PO. Primary versus staged resection for acute obstructing colorectal carcinoma. Br J Surg 1992; 79:685–688. 485. Nozoe T, Yasuda M, Honda M, Inutsuka S, Korenaga D. Obstructing carcinomas of the colon and rectum have a smaller size compared with those of nonobstructing carcinomas. Oncol Rep 2001; 8:1313–1315. 486. Smithers BM, Theile DE, Cohen JR, et al. Emergency right hemicolectomy in colon carcinoma: a prospective study. Aust N Z J Surg 1986; 56:749–752. 487. Lee YM, Law WL, Chu KW, Poon RT. Emergency surgery for obstructing colorectal cancers: a comparison between right-sided and left-sided lesions. J Am Coll Surg 2001; 192(6):719–725. 488. Deans GT, Krukowski ZH, Irwin ST. Malignant obstruction of the left colon. Br J Surg 1994; 81:1270–1276. 489. Porter JA, Salvati EP, Rubin RJ, et al. Complications of coiostomies. Dis Colon Rectum 1989; 32:299–303. 490. Wong SK, Eu KW, Lim SL, et al. Total colectomy removes undetected proximal synchronous lesions in acute left-sided colonic obstruction. Tech Coloproctol 1996; 4:87–88. 491. Stephenson BM, Shandall AA, Farouk R, et al. Malignant left-sided bowel obstruction managed by subtotal/total colectomy. Br J Surg 1990; 77:1098–1102. 492. Perez MD, de Fuenmayor ML, Calvo N, et al. Morbidity of emergency subtotal colectomy in obstructing carcinoma of the left colon. Br J Surg 1995; 82(Suppl):33. 493. Lau PW, Lo CY, Law WL. The role of one stage surgery in acute left-sided colonic obstruction. Am J Surg 1995; 169:406–409. 494. Chrysos E, Athanasakis E, Vassilakis JS, Zoras O, Xynos E. Total colectomy and J-pouch ileorectal anastomosis for obstructed tumors of the rectosigmoid junction. ANZ J Surg 2002; 72:92–94. 495. Tan SG, Nambiar R. Resection and anastomosis of obstructed left colonic cancer: primary or staged? Aust N Z J Surg 1995; 65:728–731. 496. Stewart J, Diament RH, Brennan TG. Management of obstructing lesions of the left colon by resection on table lavage and primary anastomosis. Surgery 1993; 114:502–505. 497. Kressner U, Antonsson J, Ejerblad S, et al. Intraoperative colonic lavage and primary anastomosis—an alternative to the Hartmann procedure in emergency surgery of the left colon. Eur J Surg 1994; 160:287–292. 498. Scotia Study Group. Single-stage treatment for malignant left-sided colonic obstruction: a prospective randomized clinical trial comparing subtotal colectomy with segmental resection following intraoperative irrigation. Br J Surg 1995; 82:1622–1627. 499. Chiappa A, Zbar A, Biella F, Staudacher C. One-stage resection and primary anastomosis following acute obstruction of the left colon for cancer. Am Surg 2000; 66:619–622. 500. Goodall RG, Park M. Primary resection and anastomosis of lesions obstructing the left colon. Can J Surg 1988; 31:167–168. 501. Ravo B. Colorectal anastomotic healing and intracolonic bypass procedure. Surg Clin North Am 1988; 68:1267–1294. 502. Kronborg O. The missing randomized trial of two surgical treatments for acute obstruction due to carcinoma of the left colon and rectum. Int J Colorectal Dis 1986; 1:162–166. 503. Fielding LP, Stewart-Brown S, Blesovsky L. Large bowel obstruction caused by cancer. A prospective study. Br Med J 1979; 2:515–517. 504. Eckhauser ML, Mansour EG. Endoscopic laser therapy for obstructing and/ or bleeding colorectal carcinoma. Am Surg 1992; 58:358–362. 505. Dohmoto M. New method-endosopic implantation of rectal stent in palliative treatment of malignant stenosis. Digestiva 1991; 3:1507–1512. 506. Tejero E, Mainar A, Fernandez L, Tobio R, De Gregorio MA. New procedure for the treatment of colorectal neoplastic obstructions. Dis Colon Rectum 1994; 37:1158–1159.

195

507. Dauphine CE, Tan P, Beart W Jr., Vukasin P, Cohen H, Corman ML. Placement of self-expanding metal stents for acute malignant large-bowel obstruction: a collective review. Ann Surg Oncol 2002; 9:574–579. 508. Khot UP, Lang AW, Murali K, Parker MC. Systematic review of the efficacy and safety of colorectal stents. Br J Surg 2002; 89:1096–1102. 509. Law WL, Choi HK, Lee YM, Chu KW. Palliation for advanced malignant colorectal obstruction by self-expanding metallic stents: prospective evaluation of outcomes. Dis Colon Rectum 2004; 47:39–43. 510. Saida Y, Sumiyama Y, Nagao J, Uramatsu M. Long-term prognosis of preoperative ‘‘bridge to surgery’’ expandable metallic stent insertion for obstructive colorectal cancer: comparison with emergency operation. Dis Colon Rectum 2003; 46(suppl 10):S44–S49. 511. Martinez-Santos C, Lobato RF, Fradejas JM, Pinto I, Ortega-Deballon P, MorenoAzcoita M. Self-expandable stent before elective surgery versus. emergency surgery for the treatment of malignant colorectal obstructions: comparison of primary anastomosis and morbidity rates. Dis Colon Rectum 2002; 45: 401–406. 512. Johnson R, Marsh R, Corson J, Seymour K. A comprison of two methods of palliation of large bowel obstruction due to irremovable colon cancer. Ann R Coll Surg Engl 2004; 86:99–103. 513. Meisner S, Hensler M, Knop FK, West F, Wille-Jorgensen P. Self-expanding metal stents for colonic obstruction: experiences from 104 procedures in a single center. Dis Colon Rectum 2004; 47:444–450. 514. Suzuki N, Saunders BP, Thomas-Gibson S, Akle C, Marshall M, Halligan S. Colorectal stenting for malignant and benign disease: outcomes in colorectal stenting. Dis Colon Rectum 2004; 47:1207–1207. 515. Tomiki Y, Watanabe T, Ishibiki Y, et al. Comparison of stent placement and colostomy as palliative treatment for inoperatble malignant colorectal obstruction. Surg Endosc 2004; 18:1572–1977. 516. Sebastian S, Johnston S, Geoghegan T, Torreggiani W, Buckley M. Pooled Analysis of the efficacy and safety of self-expanding metal stenting in malignant colorectal obstruction. Am J Gastroenterol 2004; 99(10):2051–2057. 517. Carne PW, Frye JN, Robertson GM, Frizelle FA. Stents or open operation for palliation of colorectal cancer: a retrospective, cohort study of perioperative outcome and long-term survival. Dis Colon Rectum 2004; 47(9): 1455–1461. 518. Tsai MA, Yang YC, Leu FJ. Obstructive colitis proximal to partially obstructive colonic carcinoma: a case report and review of the literature. Int J Colorectal Dis 2004; 19:268–272. 519. Rowe VL, Frost DB, Huang S. Extended resection for locally advanced colorectal carcinoma. Ann Surg Oncol 1997; 4:131–136. 520. Glass RE, Fazio VW, Jagelman DG, et al. The results of surgical treatment of cancer of the colon at the Cleveland Clinic from 1965–1975. A classification of the spread of colon cancer and long term survival. Int J Colorectal Dis 1986; 1:33–39. 521. Gall FP, Tonak J, Altendorf A. Multivisceral resections in colorectal cancer. Dis Colon Rectum 1987; 30:337–341. 522. Hunter JA, Ryan JA, Schultz P. En bloc resection of coln cancer adherent to other organs. Am J Surg 1987; 154:67–71. 523. Orkin BA, Dozois RR, Beart RW Jr., Patterson DE, Gunderson LL, Ilstrup DM. Extended resection for locally advanced primary adenocarcinoma of the rectum. Dis Colon Rectum 1989; 32:286–292. 524. Eisenberg SB, Kraybill WG, Lopez MJ. Long-term results of surgical resection of locally advanced coiorectal carcinoma. Surgery 1990; 108:779–786. 525. Montesani C, Ribotta G, DeMilito R, et al. Extended resection in the treatment of colorectal cancer. Int J Colorectal Dis 1991; 6:161–164. 526. Curley SA, Carlson GW, Shumato CR, et al. Extended resection for locally advanced colorectal carcinoma. Am J Surg 1992; 163:553–559. 527. Izbicki JR, Hosch SB, Knoefel WT, et al. Extended resections are beneficial for patients with locally advanced colorectal cancer. Dis Colon Rectum 1995; 38:1251–1256. 528. Carne PW, Frye JN, Kennedy-Smith A, et al. Local invasion of the bladder with colorectal cancers: surgical management and patterns of local recurrence. Dis Colon Rectum 2004; 47:44–47. 529. Nakafusa Y, Tanaka T, Tanaka M, Kitajima Y, Sato S, Miyazaki K. Comparison of multivisceral resection and standard operation for locally advanced colorectal cancer: analysis of prognostic factors for short-term and long-term outcome. Dis Colon Rectum 2004; 47(12):2055–2063. 530. Curley SA, Evans DB, Ames FC. Resection for cure of carcinoma of the colon directly involving the duodenum or pancreatic head. J Am Coll Surg 1994; 179:587–592. 531. Koea JB, Conlon K, Paty PB, Guillem JG, Cohen AM. Pancreatic or duodenal resection or both for advanced carcinoma of the right colon: is it justified? Dis Colon Rectum 2000; 43:460–465. 532. Talamonti MS, Shumate CR, Carlson GW, et al. Locally advanced carcinoma of the colon and rectum involving the urinary bladder. Surg Gynecol Obstet 1993; 177:481–487. 533. Kroneman H, Castelein A, Jeekel J. En bloc resection of colon carcinoma adherent to other organs: an efficacious treatment? Dis Colon Rectum 1991; 34:780–783. 534. Poeze M, Houbiers JGA, van de Velde CJH, et al. Radical resection of locally advanced colorectal cancer. Br J Surg 1995; 82:1386–1390.

196

& PART I: COLORECTAL DISORDERS

535. Landercasper J, Stolee RT, Steenlage E, et al. Treatment and outcome of right colon cancers adherent to adjacent organs or the abdominal wall. Arch Surg 1992; 127:841–846. 536. Yamada K, Ishizawa T, Niwa K, Chuman Y, Aikou T. Pelvic exenteration and sacral resection for locally advanced primary and recurrent rectal cancer. Dis Colon Rectum 2002; 45:1078–1084. 537. Lopez MJ, Monafo WW. Role of extended resection in the initial treatment of locally advanced colorectal carcinoma. Surgery 1993; 113:365–372. 538. Campbell SC, Church JM, Fazio VW, et al. Combined radical retropubic prostatectomy and proctosigmoidectomy for en bloc removal of locally invasive carcinoma of the rectum. Surg Gynecol Obstet 1993; 176:605–608. 539. McNamara DA, Fitzpatrick JM, O’Comnnell PR. Urinary tract involvement by colorectal cancer. Dis cancer Rectum 2003; 46:1266–1276. 540. Cady B, Monson DO, Swinton NW Sr. Survival of patients after colonic resection for carcinoma with simultaneous liver mects-tases. Surg Gynecol Obstet 1970; 131:697–700. 541. Joffe J, Gordon PH. Palliative resection for colorectal carcinoma. Dis Colon Rectum 1981; 24:355–360. 542. Liu SEM, Church JM, Lavery IC, et al. Operation in patients with incurable colon cancer—is it worthwhile? Dis Colon Rectum 1997; 40:11–14. 543. Takaki HS, Ujiki GT, Shields TS. Palliative resection in the treatment of primary colorectal cancer. Am J Surg 1977; 133:548–550. 544. Goslin R, Steele G, Zamcheck N, et al. Factors influencing survival in patients with hepatic metastases from adenocardnoma of the colon or rectum. Dis Colon Rectum 1982; 25:749–754. 545. Makela J, Haukipuro K, Laitinen S, et al. Palliative operations for colorectal cancer. Dis Colon Rectum 1990; 33:846–850. 546. Feqiz G, Ramacciato G, Indinnimeo M, et al. Synchronous large bowel cancer: a series of 47 cases. Ital J Surg Sci 1989; 19:23–28. 547. Bussey HJR, Wallace MH, Morson BC. Metachronous carcinomas of the large intestine and intestinal polyps. Proc R Soc Med 1967; 60:208–210. 548. Chu DZJ, Giacco G, Martin RG, et al. The significance of synchronous carcinoma and polyps in the colon and rectum. Cancer 1986; 57:445–450. 549. Passman MA, Pommier RF, Vetto JT. Synchronous colon primaries have the same prognosis as solitary colon cancers. Dis Colon Cancer 1996; 39:329–334. 550. Fogler R, Weiner E. Multiple foci of colorectal carcinoma. Argument for subtotal colectomy. N Y State J Med 1980; 80:47–51. 551. Gervaz P, Bucher P, Neyroud-Caspar I, Soravia C, Morel P. Proximal location of colon cancer is a risk factor for development of metachronous colorectal cancer: a population-based study. Dis Colon Rectum 2005; 48:227–232. 552. Shitoh K, Konishi F, Miyakura Y, Togashi K, Okamoto T, Nagai H. Microsatellite instability as a marker in predicting metachronous multiple colorectal carcinomas after surgery: a cohort-like study. Dis Colon Rectum 2002; 45:329–333. 553. Finlay IG, McArdle CS. Occult hepatic metastases in colorectal carcinoma. Br J Surg 1986; 73:732–735. 554. Machi J, Isomoto H, Kurohiji T, et al. Detection of unrecognized liver metastases from colorectal cancer by routine use of operative ultrasonography. Dis Colon Rectum 1986; 29:405–409. 555. Blumgart LH, Fong Y. Surgical options in the treatment of hepatic metastases from colorectal cancer. Curr Probl Surg 1995; 32:335–421. 556. Finlay IG, Meek D, Branton F, et al. Growth rate of hepatic metastases in colorectal carcinoma. Br J Surg 1998; 75:641–644. 557. Wigmore SJ, Madhavan K, Redhead DN, Currie EJ, Garden OJ. Distribution of colorectal liver metastases in patients referred for hepatic resection. Cancer 2000; 89:285–287. 558. Van Ooijen B, Oudkerk M, Schmitz PIM, et al. Detection of liver metastases from colorectal carcinoma: is there a place for routine computed tomography arteriography. Surgery 1996; 119:511–516. 559. Strasberg SM, Dehdashti F, Siegel BA, Drebin JA, Linehan D. Survival of patients evaluated by FDG-PET before hepatic resection for metastatic colorectal carcinoma: a prospective database study. Ann Surg 2001; 233:293–299. 560. Rodgers MS, Collinson R, Desai S, Stubbs RS, McCall JL. Risk of dissemination with biopsy of colorectal liver metastases. Dis Colon Rectum 2003; 46:454–458. 561. Fuhrman GM, Curley SA, Hohn DC, et al. Improved survival after resection of colorectal liver metastases. Ann Surg Oncol 1995; 2:537–541. 562. Bengtsson G, Carlsson G, Hafstrom L, et al. Natural history of patients with treated liver metastases from colorectal cancer. Am J Surg 1981; 141:586–589. 563. Nagorney DM. Hepatic resection for metastases from colorectal cancer. Nelson RL, ed. Problems in Current Surgery. Controversies in Colon Cancer. Philadelphia: JB Lippincott, 1987:83–92. 564. Palmer M, Petrelli NJ, Herrera L. No treatment option for liver metastases for colorectal adenocarcinoma. Dis Colon Rectum 1989; 32:698–701. 565. Stangl R, Altendorf-Hofmann A, Charnley RM, et al. Factors influencing the natural history of colorectal liver metastases. Lancet 1994; 343:1405–1410. 566. Vauthey JN, Marsch R, de W, et al. Arterial therapy of hepatic colorectal metastases. Br J Surg 1996; 83:447–445. 567. Hughes KS, Scheele J, Sugarbaker PH. Surgery for colorectal cancer metastatic to the liver. Optimizing the results of treatment. Surg Clin North Am 1989; 69:339–359.

568. Schlag P, Hohenberger P, Herforth C. Resection of liver metastases in colorectal cancer—competitive analysis of treatment results in synchronous versus metachrounous metastases. Eur J Surg Oncol 1990; 16:360–365. 569. Petrelli N, Gupta B, Piedmonte M, et al. Morbidity and survival of liver resection for colorectal adenocarcinoma. Dis Colon Rectum 1991; 34: 889–904. 570. Doci R, Gennari L, Bignami P, et al. One hundred patients with hepatic metastases from colorectal cancer treated by resection: analysis of prognostic determinants. Br J Surg 1991; 78:797–801. 571. Rosen CB, Nagorney DM, Taswell HF, et al. Perioperative blood transfusion and determinants of survival after liver resection for metastatic colorectal carcinoma. Ann Surg 1992; 216:492–505. 572. Nakamura S, Yokoi Y, Suzuki S, et al. Results of extensive surgery for liver metastasis in colorectal carcinoma. Br J Surg 1992; 79:35–38. 573. Van Ooijen B, Wiggers T, Meijer S, et al. Hepatic resections for colorectal cancer metastases in the Netherlands. A multi-institutional 10-years study. Cancer 1992; 70:28–34. 574. Grayowski TJ, Iwatsuki S, Madariaga JR, et al. Experience in resection for metastatic colorectal cancer: analysis of clinical and pathologic risk factors. Surgery 1994; 116:703–711. 575. Scheele J, Stang R, Altendorf-Hofmann A, et al. Resection of colorectal liver metastases. World J Surg 1995; 19:59–71. 576. Hananel N, Garzon J, Gordon PH. Hepatic resection for colorectal livery metastases. Am Surg 1995; 61:444–447. 577. Rougier PH, Milan C, Lazorthes F, et al. Prospective study of prognostic factors in patients with unresected hepativ metastases from colorectal cancer. Br J Surg 1995; 82:1397–1400. 578. Wade TP, Virgo KS, Li MJ, et al. Outcomes after detection of metastatic carcinoma of the colon and rectum in a national hospital system. J Am Coll Surg 1996; 182:353–361. 579. Wanebo HJ, Chu QD, Vezeridis MP, et al. Patient selection for hepatic resection of colorectal metastases. Arch Surg 1996; 131:322–329. 580. Ohlsson B, Stenram U, Tranberg KG. Resection of colorectal liver metastases: 25-year experience. World J Surg 1998; 22:268–276. 581. Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999; 230:309–318. 582. Buell JF, Rosen S, Yoshida A, et al. Hepatic resection: effective treatment for primary and secondary tumors. Surgery 2000; 128:686–693. 583. Elias D, Ouellet JF, Bellon N, Pignon JP, Pocard M, Lasser P. Extrahepatic disease does not contraindicate hepatectomy for colorectal liver metastases. Br J Surg 2003; 90:567–574. 584. Kato T, Yasui K, Hirai T, et al. Therapeutic results for hepatic metastasis of colorectal cancer with special reference to effectiveness of hepatectomy: analysis of prognostic factors for 763 cases recorded at 18 institutions. Dis Colon Rectum 2003; 46(suppl):S22–S31. 585. Teh CS, Ooi LL. Hepatic resection for colorectal metastases to the liver: The National Cancer Centre/Singapore General Hospital experience. Ann Acad Med Singapore 2003; 32:196–204. 586. Weber JC, Bachellier P, Oussoultzoglou E, Jaeck D. Simultaneous resection of colorectal primary tumour and synchronous liver metastases. Br J Surg 2003; 90:956–962. 587. Lefor AT, Hughes KS, Shiloni E, et al. Intra-abdominal extrahepatic disease in patients with colorectal hepatic metastases. Dis Colon Rectum 1988; 31: 100–103. 588. Wagner JS, Adson MA, VanHeerden JA, et al. The natural history of hepatic metastases from colorectal cancer. Ann Surg 1984; 199:502–508. 589. Luna-Perez P, Rodriguez-Coria DF, Arroyo B, Gonzalez-Macouzet J. The natural history of liver metastases from colorectal cancer. Arch Med Res 1998; 29(1):319–324. 590. Ruo L, Gougoutas C, Paty PB, Guillem JG,Cohen AM, Wong WD. Elective bowel resection for incurable stage IV colorectal cancer: prognostic variables for asymptomatic patients. J Am Coll Surg 2003; 196:722–728. 591. Kuo LJ, Leu SY, Liu MC, Jian JJM, Cheng SH, Chen CM. How aggressive should we be in patients with stage IV colorectal cancer? Dis Colon Rectum 2003; 46:1646–1652. 592. de Santibanes E, Lassalle FB, McCormack L, et al. Simultaneous colorectal and hepatic resections for colorectal cancer: postoperative and long term outcomes. J Am Coll Surg 2002; 195:196–202. 593. Chua HK, Sondenna K, Tsiotos GG, Larson DR, Wolff BG, Nagorney DM. Concurrent vs. staged colectomy and hepatectomy for primary colorectal cancer with synchronous hepatic metastases. Dis Colon Rectum 2004; 47:1310–1316. 594. Tocchi A, Mazzoni G, Brozzetti S, Miccini M, Cassini D, Bettelli E. Hepatic resection in stage IV colorectal cancer: prognostic predictors of outcome. Int J Colorectal Dis 2004 19: 580–5. 595. Tanaka K, Shimada H, Matsuo K, et al. Outcome after simultaneous colorectal and hepatic resection for colorectal cancer with synchronous metastases. Surgery 2004; 136:650–659. 596. Attiyeh FF, Wichern WA. Hepatic resection for primary metastatic tumours. Am J Surg 1988; 156:368–373.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

597. Adson MA. Resection of liver metastases – when is it worthwhile? World J Surg 1987; 11:511–520. 598. Bozzetti F, Doci R, Bignami P, et al. Patterns of failure following surgical resection of colorectal cancer, liver metastases. Rationale for a multimodal approach. Ann Surg 1987; 205:264–270. 599. Tanaka K, Adam R, Shimada H, Azoulay D, Levi F, Bismuth H. Role of neoadjuvant chemotherapy in the treatment of multiple colorectal metastases to the liver. Br J Surg 2003; 90:963–969. 600. Allen PJ, Kemeny N, Jarnagin W, Dematteo R, Blumgart L, Fong Y. Importance of response to neoadjuvant chemotherapy in patients undergoing resection of synchronous colorectal liver metastases. J Gastrointest Surg 2003; 7:109–115. 601. Iwatsuki S, Dvorchik I, Madariaga JR, et al. Hepatic resection for metastatic colorectal adenocarcinoma: a proposal of a prognostic scoring system. J Am Coll Surg 1999; 189:291–299. 602. Smith DL, Soria JC, Morat L, et al. Human telomerase reverse transcriptase (hTERT) and Ki-67 are better predictors of survival than established clinical indicators in patients undergoing curative hepatic resection for colorectal metastases. Ann Surg Oncol 2004; 11:45–51. 603. Imamura H, Seyama Y, Kokudo N, et al. Single and multiple resections of multiple hepatic metastases of colorectal origin. Surgery 2004; 135:508–517. 604. Jaeck D. The significance of hepatic pedicle lymph nodes metastases in surgical management of colorectal liver metastases and of other liver malignancies. Ann Surg Oncol 2003; 10:1007–1011. 605. Elias D, Sideris L, Pocard M, et al. Results of R0 resection for colorectal liver metastases associated with extrahepatic disease. Ann Surg Oncol 2004; 11:274–280. 606. Martin R, Paty P, Fong Y, et al. Simultaneous liver and colorectal resections are safe for synchronous colorectal liver metastasis. J Am Coll Surg 2003; 197:223–241. 607. Nelson RL, Freels S. A systematic review of hepatic artery chemotherapy after hepatic resection of colorectal cancer metastatic to the liver. Dis Colon Rectum 2004; 47:739–745. 608. Clancy TE, Dixon E, Perlis R, Sutherland FR, Zinner MJ. Hepatic arterial infusion after curative resection of colorectal cancer metastases: a metaanalysis of prospective clinical trials. J Gastrointest Surg 2005; 9:198–206. 609. Bines SD, Doolas A, Jenkins L, et al. Survival after repeat hepatic resection for recurrent colorectal hepatic metastases. Surgery 1996; 120:591–596. 610. Wanebo HJ, Chu QD, Avradopoulos KA, et al. Current perspectives on repeat hepatic resection for colorectal carcinoma: a review. Surgery 1996; 119: 361–371. 611. Pinson CW, Wright JK, Chapman WC, et al. Repeat hepatic surgery for colorectal cancer metastases to the liver. Am Surg 1996; 223:765–776. 612. Fernandez-Trigo V, Shamsa F, Sugarbaker PH. Repeat liver resections from colorectal metastasis. Surgery 1995; 117:296–304. 613. Oue FG, Nagorney DM. Resection of ‘‘recurrent’’ colorectal metastases to the liver. Br J Surg 1994; 81:255–258. 614. Takahashi S, Inoue K, Konishi M, Nakagouri T, Kinoshita T. Prognostic factors for poor survival after repeat hepatectomy in patients with colorectal liver metastases. Surgery 2003; 133:627–634. 615. Suzuki S, Sakaguchi T, Yokoi Y, et al. Impact of repeat hepatectomy on recurrent colorectal liver metastases. Surgery 2001; 129:421–428. 616. Oshowo A, Gillams A, Harrison E, Lees WR, Taylor I. Comparison of resection and radiofrequency ablation for treatment of solitary colorectal liver metastases. Br J Surg 2003; 90:1240–1243. 617. Berber E, Pelley R, Siperstein AE. Predictors of survival after radiofrequency thermal ablation of colorectal cancer metastases to the liver: a prospective study. J Clin Oncol 2005; 23(7):1358–1364. 618. Ueno H, Mochizuki H, Hashiguchi Y, Hastsuse K, Fujimoto H, Hase K. Predictors of extrahepatic recurrence after resection of colorectal liver metastases. Br J Surg 2004; 91:327–333. 619. Cohen DA, Kemeny EN. An update on hepatic arterial infusion chemotheraphy for colorectal cancer. The Oncologist 2003; 8:553–556. 620. Gray BN, Anderson JE, Burton MA, et al. Regression of liver metastases following treatment with yttrium-90 microspheres. Aust N Z J Surg 1992; 62:105–110. 621. Stubbs RS, Cannan RJ, Mitchell AW. Selective internal radiation therapy with 90yttrium microspheres for extensive colorectal liver metastases. J Gastrointest Surg 2001; 5(3):294–302. 622. Lang EK, Brown CL. Colorectal metastases to the liver: selective chemoembolization. Radiology 1993; 189:417–422. 623. Stuart K. Chemoembolization in the manangement of liver tumors. The Oncologist 2003; 8:425–437. 624. Weaver ML, Atkinson D, Zemel R. Hepatic cryosurgery in the treatment of unresectable metastases. Surg Oncol 1995; 4:231–236. 625. Ruers TJ, Joosten J, Jager GJ, Wobbes T. Long-term results of treating hepatic colorectal metastases with cryosurgery. Br J Surg 2001; 88:844–849. 626. Scheele J, Altendorf-Hofmann A, Stangl R, et al. Pulmonary resection for metastatic colon and upper rectum cancer: is it useful? Dis Colon Rectum 1990; 33:745–752. 627. Saclarides TJ, Krueger BL, Szeluga DJ, et al. Thoracotomy for colon and rectal cancer metastases. Dis Colon Rectum 1993; 36:425–429.

197

628. Brister SJ, de Varennes B, Gordon PH, et al. Contemporary operative management of pulmonary metastases of colorectal origin. Dis Colon Rectum 1988; 31:786–792. 629. Kanemitsu Y, Kato T, Hirai T, Yasui K. Preoperative probability model for predicting overall survival after resection of pulmonary metastases from colorectal cancer. Br J Surg 2004; 91:112–120. 630. Watanabe I, Arai T, Ono M, et al. Prognostic factors in resection of pulmonary metastasis from colorectal cancer. Br J Surg 2003; 90(11):1436–1440. 631. Negri F, Musolino A, Normon AR, Landos G, Pastorino U, Chong E, Cunningham D. The development for preoperative chemotherapy strategy for patients selected to undergo pulmonary metastectomy from colorectal cancer. J Clin Oncol 2004; 145:202S. 632. King J, Glenn D, Clark W, et al. Percutaneous radiofrequency ablation of pulmonary metastases in patients with colorectal cancer. Br J Surg 2004; 91: 217–223. 633. Ike H, Shimada H, Ohki S, Togo S, Yamaguchi S, Ichikawa Y. Results of aggressive resection of lung metastases from colorectal carcinoma detected by intensive follow-up. Dis Colon Rectum 2002; 45:468–473. 634. Ishikawa K, Hashiguchi Y, Mochizuki H, Ozeki Y, Ueno H. Extranodal cancer deposit at the primary tumor site and the number of pulmonary lesions are useful prognostic factors after surgery for colorectal lung metastases. Dis Colon Rectum 2003; 46:629–636. 635. Zink S, Kayser G, Gabius HJ, Kayser K. Survival, disease-free interval, and associated tumor features in patients with colon/rectal carcinomas and their resected intra-pulmonary metastases. Eur J Cardiothorac Surg 2001; 19:908–913. 636. Irshad K, Ahmad F, Morin JE, Mulder DS. Pulmonary metastases from colorectal cancer: 25 years of experience. Can J Surg 2001; 44:217–221. 637. Vogelsang H, Haas S, Hierhoizer C, Berger U, Siewert JR, Prauer H. Factors influencing survival after resection of pulmonary metastases from colorectal cancer. Br J Surg 2004; 91:1066–1071. 638. Mineo TC, Ambrogi V, Tonini G, et al. Longterm results after resection of simultaneous and sequential lung and liver metastases from colorectal carcinoma. J Am Coll Surg 2003; 197:386–391. 639. Ike H, Shimada H, Togo S, Yamaguchi S, Ichikawa Y, Tanaka K. Sequential resection of lung metastasis following partial hepatectomy for colorectal cancer. Br J Surg 2002; 89:1164–1168. 640. Nagakura S, Shirai Y, Yamato Y, Yokoyama N, Suda T, Hatakeyama K. Simultaneous detection of colorectal carcinoma liver and lung metastases does not warrant resection. J Am Coll Surg 2001; 193:153–160. 641. Traina TA, Loonard GD, Tang L, Paty PB, Maki RG. Problems in colon cancer and a child with Renal Lymphoma: CASE 1. Metastatic colon cancer to the ovaries in a Krukenberg-like pattern. J Clin Oncol 2005; 23(22): 5255–5256. 642. Herrera-Ornelas L, Natarajan N, Tsukada Y, et al. Adenocarcinoma of the colon masquerading as primary ovarian neoplasia. An analysis of ten cases. Dis Colon Rectum 1983; 26:377–380. 643. Morrow M, Enker WE. Late ovarian metastases in carcinoma of the colon and rectum. Arch Surg 1984; 119:1385–1388. 644. Huang PP, Weber TK, Mendoza C, Rodriguez-Bigas MA, Petrelli NJ. Longterm survival in patients with ovarian metastases from colorectal carcinoma. Ann Surg Oncol 1998; 5(8):695–698. 645. Besbeas S, Stearns MW Jr. Osseous metastases from carcinomas of the colon and rectum. Dis Colon Rectum 1978; 21:266–268. 646. Bonnheim DC, Petrelli NJ, Herrera L, et al. Osseous metastases from colorectal carcinoma. Am J Surg 1986; 151:457–459. 647. Scuderi G, Macri A, Sfuncia G, et al. Sternal metastasis as initial presentation of a unknown rectal cancer. Int J Colorectal Dis 2004; 19:292–293. 648. Wong ET, Berkenblit A. The role of topotecan in the treatment of brain metastases. The Oncologist 2004; 9:68–79. 649. Alden TD, Gianino JW, Saclarides TJ. Brain metastases from colorectal cancer. Dis Colon Rectum 1996; 39:541–545. 650. Hammond MA, McCutcheon IE, Elsouki R, et al. Colorectal carcinoma and brain metastases: distribution, treatment and survival. Ann Surg Oncol 1996; 3:453–463. 651. Farnell GF, Buckner JC, Cascino TL, et al. Brain metastases from colorectal carcinoma. The long-term survivors. Cancer 1996; 78:74–76. 652. Knorr C, Reingruber B, Meyer T, Hohenberger W, Stremmel C. Peritoneal carcinomatosis of colorectal cancer: incidence, prognosis, and treatment modalities. Int J Colorectal Dis 2004; 19:181–187. 653. Glehen O, Kwiatkowski F, Sugarbaker PH, et al. Cytoreductive surgery combined with perioperative intraperitoneal chemotherapy for the management of peritoneal carcinomatosis from colorectal cancer: a multi-institutional study. J Clin Oncol 2004; 22:3284–3292. 654. Culliford AT 4th, Brooks AD, Sharma S, et al. Surgical debulking and intraperitoneal chemotherapy for established peritoneal metastases from colon and appendix cancer. Ann Surg Oncol 2001; 8(10):787–795. 655. Verwaal VJ, van Ruth S, de Bree E, et al. Randomized trial of cytoreduction and hyperthermic intraperitoneal chemotherapy versus systemic chemotherapy and palliative surgery in patients with peritoneal carcinomatosis of colorectal cancer. J Clin Oncol 2003; 21:3737–3743.

198

& PART I: COLORECTAL DISORDERS

656. Verwaal VJ, van Ruth S, Witkamp A, Boot H, van Slooten G, Zoetmulder FA. Long-term survival of peritoneal carcinomatosis of colorectal origin. Ann Surg Oncol 2005; 12:65–71. 657. Shen P, Hawksworth J, Lovato J, et al. Cytoreductive surgery and intraperitoneal hyperthermic chemotherapy with mitomycin C for peritoneal carcinomatosis from nonappendiceal colorectal carcinoma. Ann Surg Oncol 2004; 11:178–186. 658. Elias D, Delperro JR, Sideris L, et al. Treatment of peritoneal carcinomatosis from colorectal cancer: impact of complete cytoreductive surgery and difficulties in conducting randomized trials. Ann Surg Oncol 2004; 11(5):518–521. 659. Mainprize KS, Berry AR. Solitary splenic metastasis from colorectal carcinoma. Br J Surg 1997; 84:70. 660. Rendi MH, Dhar AD. Cutaneous metastasis of rectal adenocarcinoma. Dermatol Nurs 2003; 15:131–132. 661. Tsai HL, Huang YS, Hsieh JS, Huang TJ, TSai KB. Signet-ring cell carcinoma of the rectum with diffuse and multiple skin metastases–a case report. Kaohsiung J Med Sci 2002; 18:359–362. 662. Gabriele R, Borghese M, Conte M, Bosso L. Sister Mary Joseph’s nodule as a first sign of cancer of the cecum: report of a case. Dis Colon Rectum 2004; 47:115–117. 663. Yan BK, Nyam DC, Ho YH. Carcinoma of the rectum with a single penile metastasis. Singapore Med J 2002; 43:39–40. 664. Tutton MG, George M, Hill ME, Abulafi AM. Solitary pancreatic metastasis from a primary colonic tumor detected by PET scan: report of a case. Dis Colon Rectum 2001; 44:288–290. 665. Chagpar A, Kanthan SC. Vaginal metastasis of colon cancer. Am Surg 2001; 67:171–172. 666. Charles W, Joseph G, Hunis B, Rankin L. Problems in colon cancer and a child with renal lymphoma: CASE 2. metastatic colon cancer to the testicle presenting as testicular hydrocele. J Clin Oncol 2005; 23(22):5256–5257. 667. Tiong HY, Kew CY, Tan KB, Salto-Tellez M, Leong AF. Metastatic testicular carcinoma from the colon with clinical, immunophenotypical, and molecular characterization: report of a case. Dis Colon Rectum 2005; 48:582–585. 668. Radhakrishnan CN, Bruce J. Colorectal cancers in children without any predisposing factors. A report of eight cases and review of the literature. Eur J Pediatr Surg 2003; 13:66–68. 669. Killingback M, Barron P, Dent O. Elective resection and anastomosis for colorectal cancer: a prospective audit of mortality and morbidity 1976–1998. ANZ J Surg 2002; 72:689–698. 670. Platell CF, Semmens JB. Review of survival curves for colorectal cancer. Dis Colon Rectum 2004; 47(12):2070–2075. 671. Pihl E, Hughes ESR, McDermott FT, et al. Carcinomas of the colon—cancer specific long-term survival. A series of 615 patients treated by one surgeon. Ann Surg 1980; 192:114–117. 672. Stefanini P, Castrini G, Pappalardo G. Surgical treatment of cancer of the colon. Int Surg 1981; 66:125–131. 673. Zhou YG, Yu BM, Shen YX. Surgical treatment and late results in 1226 cases of colorectal cancer. Dis Colon Rectum 1983; 26:250–256. 674. Umpleby HC, Bristol JB, Rainey JB, et al. Survival of 727 patients with single carcinoma of the large bowel. Dis Colon Rectum 1984; 27:803–810. 675. Isbister WH, Fraser J. Survival following resection for colorectal cancer. A New Zealand national study. Dis Colon Rectum 1985; 28:725–727. 676. Wied U, Nilsson T, Knudson JB, et al. Postoperative survival of patients with potentially curable cancer of the colon. Dis Colon Rectum 1985; 28:233–235. 677. Davis NC, Evans EB, Cohen JR. Colorectal cancer: a large unselected Australian series. Aust N Z J Surg 1987; 57:153–159. 678. Moreaux J, Catala M. Carcinoma of the colon: long term survival and prognosis after surgical treatment in a series of 798 patients. World J Surg 1987; 11:804–809. 679. Brown SGW, Walsh S, Sykes PA. Operative mortality rate and surgery for colorectal cancer. Br J Surg 1988; 75:645–647. 680. Enblad P, Adami HO, Bergstrom R, et al. Improved survival of patients with cancers of the colon and rectum. J Natl Cancer Inst 1988; 80:586–591. 681. Jatzko G, Lisborg P, Wette V. Improving survival rates for patients with colorectal cancer. Br J Surg 1992; 79:588–591. 682. Clemmensen T, Sprechler M. Recording of patients with colorectal cancer on a database: results and advantages. Eur J Surg 1994; 160:175–178. 683. Carraro PG, Segala M, Cesana BM, Tiberio G. Obstructing colonic cancer: failure and survival patterns over a ten-year follow-up after one-stage curative surgery. Dis Colon Rectum 2001; 44:243–250. 684. Read TE, Mutch MG, Chang BW, et al. Locoregional recurrence and survival after curative resection of adenocarcinoma of the colon. J Am Coll Surg 2002; 195:33–40. 685. Morin NA, Obrand DF, Shrier I, Gordon PH (unpublished data). 2006. 686. Eisenberg B, Decosse JJ, Harford F, et al. Carcinoma of the colon and rectum: the natural history reviewed in 1704 patients. Cancer 1982; 49:1131–1134. 687. Staib L, Link KH, Blatz A, Beger HG. Surgery of colorectal cancer: surgical morbidity and five- and ten-year results in 2400 patients–monoinstitutional experience. World J Surg 2002; 26:59–66.

688. Devesa JM, Morales V, Enriquez JM, et al. Colorectal cancer. The bases for a comprehensive follow-up. Dis Colon Rectum 1988,31: 636–52. 689. The National Cancer Data Base Report on colon cancer, Nov 23, 2004. www.facs.org. 690. Kelley WE, Brown PW, Lawrence W Jr., et al. Penetrating, obstructing, and perforating carcinomas of the colon and rectum. Arch Surg 1981; 116: 381–384. 691. Brief DK, Brener BJ, Goldenkranz R, et al. An argument for increased use of subtotal colectomy in the management of carcinoma of the colon. Am Surg 1983; 49:66–72. 692. Crooms JW, Kovalcik PJ. Obstructing left-sided colon carcinoma. Appraisal of surgical options. Ann Surg 1984; 50:15–19. 693. Phillips RKS, Hittinger R, Fry JS, et al. Malignant large bowel obstruction. Br J Surg 1985; 72:296–302. 694. Willett C, Tepper JE, Cohen A, et al. Obstructive and perforative colonic carcinoma: patterns of failure. J Clin Oncol 1985; 3:379–384. 695. Serpell JW, McDermott FT, Katrivessis H, et al. Obstructing carcinomas of the colon. Br J Surg 1989; 76:965–969. 696. Mulcahy HE, Skelly MM, Husain A, et al. Long-term outcome following curative surgery for malignant large bowel obstruction. Br J Surg 1996; 83:46–50. 697. Chen HS, Sheen-Chen SM. Obstruction and perforation in colorectal adenocarcinoma: an analysis of prognosis and current trends. Surgery 2000; 127:370–376. 698. Hughes ESR, McDermott FT, Polglasse A, et al. Total and subtotal colectomy for colonic obstruction. Dis Colon Rectum 1985; 28:162–163. 699. Morgan WP, Jenkins N, Lewis P, et al. Management of obstructing carcinoma of the left colon by extended right hemicolectomy. Am J Surg 1985; 149:327–329. 700. Feng YS, Hsu H, Chen SS. One-stage operation for obstructing carcinomas of the left colon and rectum. Dis Colon Rectum 1987; 30:29–32. 701. Halevy A, Levi J, Orda R. Emergency subtotal colectomy. A new trend for treatment of obstructing carcinoma of the left colon. Ann Surg 1989; 210:220–223. 702. Slors JFM, Taat CW, Mallonga ET, et al. One-stage colectomy and ileorectal anastomosis for complete left-sided obstruction of the colon. Neth J Surg 1989; 41:1–4. 703. Stephenson BM, Shandall AA, Farouk R, et al. Malignant left-sided large bowel obstruction managed by subtotal/total colectomy. Br J Surg 1990; 77:1098–1102. 704. Antal SC, Kovacs ZG, Feigenbaum V, et al. Obstructing carcinoma of the left colon: treatment by extended right hemicolectomy. Int Surg 1991; 76: 161–163. 705. Tan SG, Nambiar B, Rauff A, et al. Primary resection and anastomosis in obstructed descending colon due to cancer. Arch Surg 1991; 126:748–751. 706. Murray J, Schoetz DJ Jr., Coller JA, et al. Intraoperative colonic lavage and primary anastomosis in non-elective colon resection. Dis Colon Rectum 1991; 34:527–531. 707. Stewart J, Diament RH, Brennan TG. Management of obstructing lesions of the left colon by resection, on table lavage, and primary anastomosis. Surgery 1993; 114:502–505. 708. Arnaud JP, Bergamaschi R. Emergency subtotal/total colectomy with anastomosis for acutely obstructed carcinoma of the left colon. Dis Colon Rectum 1994; 37:685–688. 709. Michowitz M, Avnieli D, Lazarovici J, et al. Perforation complicating carcinoma of the colon. J Surg Oncol 1982; 19:18–21. 710. Badia JM, Sitges-Serra A, Pia J. Perforation of colonic neoplasms. Review of 36 cases. Int J Colorectal Dis 1987; 2:187–189. 711. Carraro PG, Segala M, Orlotti C, Tiberio G. Outcome of large-bowel perforation in patients with colorectal cancer. Dis Colon Rectum 1998; 41(11): 1421–1426. 712. Khan S, Pawlak SE, Eggenberger JC, Lee CS, Szilagy EJ, Margolin DA. Acute colonic perforation associated with colorectal cancer. Am Surg 2001; 67: 261–264. 713. Scott NA, Jeacock J, Kingston RD. Risk factors in patients presenting as an emergency with colorectal cancer. Br J Surg 1995; 82:321–323. 714. Anderson JH, Hale D, McArdle CS. Elective versus emergency surgery for patients with colorectal cancer. Br J Surg 1992; 79:706–709. 715. Fitzgerald SD, Longo WE, Daniel GL, et al. Advanced colorectal neoplasia in the high-risk elderly patient: is surgical resection justified? Dis Colon Rectum 1993; 36:161–166. 716. Gervaz P, Pakart R, Nivatvongs S, Wolff BG, Larson D, Ringel S. Colorectal adenocarcinoma in cirrhotic patients. J Am Coll Surg 2003; 196:874–879. 717. Kotake K, Honjo S, Sugihara K. Changes in colorectal cancer during a 20-year period: an extended report from the multi-institutional registry of large bowel cancer, Japan. Dis Colon Rectum 2003; 46(suppl 10):S32–S43. 718. Wang HZ, Huang XF, Wang Y, Ji JF, Gu J. Clinical features, diagnosis, treatment and prognosis of multiple primary colorectal carcinoma. World J Gastroenterol 2004; 10(14):2136–2139. 719. Papaconstantinou HT, Sklow B, Hanaway MJ, et al. Characteristics and survival patterns of solid organ transplant patients developing de novo colon and rectal cancer. Dis Colon Rectum 2004; 47(11):1898–1903.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

720. Phillips RKS, Hittinger R, Blesovsky L, et al. Large bowel cancer and surgical pathology and its relationship to survival. Br J Surg 1984; 71:604–610. 721. Jass JR. Pathologists’ perspective on colorectal cancer. Perspect Colon Rectal Surg 1991; 4:327–332. 722. Adkins RB, Delozier JB, McKnight WG, et al. Carcinoma of the colon in patients 35 years of age and younger. Ann Surg 1987; 53:141–145. 723. Behbehani A, Sakwa M, Erlichman R, et al. Colorectal carcinoma in patients under age 40. Ann Surg 1985; 202:610–614. 724. Koh SJ, Johnson WW. Cancer of the large bowel in children. South Med J 1986; 79:931–935. 725. Rao BN, Pratt CB, Fleming ID, et al. Colon carcinoma in children and adolescents. Cancer 1985; 55:1322–1326. 726. Okuno M, Ikehara T, Nagayama M, et al. Colorectal carcinoma in young adults. Am J Surg 1987; 154:265–268. 727. Phillips RK, Hittinger R, Blesovsky L, et al. Local recurrence following curative resection for large bowel cancer. I. The overall picture. Br J Surg 1984; 71:12–16. 728. Cusack JC, Giacco GG, Cleary K, et al. Survival factors in 186 patients younger than 40-years old with colorectal adenocarcinoma. J Am Coll Surg 1996; 183:105–112. 729. Palmer ML, Herrera L, Petrelli NJ. Colorectal adenocarcinoma in patients less than 40 years of age. Dis Colon Rectum 1991; 34:343–345. 730. O’Connell JB, Maggard MA, Livingston EH, Yo CK. Colorectal cancer in the young. Am J Surg 2004; 187:343–348. 731. Chapuis PH, Dent OF, Fisher R, et al. A multivariate analysis of clinical and pathological variables in prognosis of resection of large bowel cancer. Br J Surg 1985; 72:698–702. 732. Newland RC, Dent OF, Lyttle MNB, et al. Pathologic determinants of survival associated with colorectal cancer with lymph node metastases. Cancer 1994; 73:2076–2082. 733. de Mello J, Struthers L, Turner R, et al. Multivariate analysis as aid to diagnosis and assessment of prognosis in gastrointestinal cancer. Br J Surg 1983; 48:341–348. 734. McArdle CS, McMillan DC, Hole DJ. Male gender adversely affects survival following surgery for colorectal cancer. Br J Surg 2003; 90:711–715. 735. Polissar L, Sim D, Francis A. Survival of colorectal cancer patients in relation to duration of symptoms and other prognostic factors. Dis Colon Rectum 1981; 24:364–369. 736. Mzabi R, Himal HS, Demers R, et al. A multiparametric computer analysis of carcinoma of the colon. Surg Gynecol Obstet 1976; 143:959–964. 737. Adam IJ, Mohamdee MO, Martin IG, et al. Role of circumferential margin involvement in the local recurrence of rectal cancer. Lancet 1994; 344: 707–711. 738. Griffin MR, Bergstralh EJ, Coffey RJ, et al. Predictors of survival after curative resection of carcinoma of the colon. Cancer 1987; 60:2318–2324. 739. Steinberg SM, Barkin JS, Kaplan RS, et al. Prognostic indicators of colon tumors: the Gastrointestinal Tumour Study Group experience. Cancer 1986; 57:1866–1870. 740. de Leon ML. Schoetz DJ, Coller JA, et al. Colorectal cancer. Lahey Clinic experience, 1972–1976. An analysis of prognostic indicators. Dis Colon Rectum 1987; 30:237–242. 741. Fielding LP, Phillips RKS, Frye JS, et al. The prediction of outcome after curative for large bowel cancer. Lancet 1986; 2:904–907. 742. Wolmark N, Wieand HS, Rockette HE, et al. The prognostic significance of tumour location and bowel obstruction in Dukes’ B and C colorectal cancer: findings from the NSABP clinical trials. Ann Surg 1983; 198:743–752. 743. McArdle CS, Hole DJ. Emergency presentation of colorectal cancer is associated with poor 5-year survival. Br J Surg 2004; 91:605–609. 744. Venkatesh KS, Weingart DJ, Ramanujam PJ. Comparison of double and single parameters in DNA analysis for staging and as a prognostic indicator in patients with colon and rectal carcinoma. Dis Colon Rectum 1994; 37:1142–1147. 745. Slanetz CA Jr. The effect of inadvertent intraoperative perforation on survival and recurrence in colorectal cancer. Dis Colon Rectum 1984; 27:792–797. 746. James RD, Donaldson D, Gray R, Northover JM, Stenning SP, Taylor I. AXIS collaborators. Randomized clinical trial of adjuvant radiotherapy and 5-fluorouracil infusion in colorectal cancer (AXIS). Br J Surg 2003; 90: 1200–1212. 747. Wiggers T, Arends JW, Volovics A. Regression analysis of prognostic factors in colorectal cancer after curative resection. Dis Colon Rectum 1988; 31:33–41. 748. Bonfanti G, Bonzetti F, Doci E, et al. Results of extended surgery for cancer of the rectum and sigmoid. Br J Surg 1982; 69:305–307. 749. Habib NA, Peck MA, Sawyer CN, et al. An analysis of the outcome of 301 malignant colorectal tumors. Dis Colon Rectum 1983; 26:601–605. 750. Wood CB, Gills CR, Hole D, et al. Local tumour invasion as a prognostic factor in colorectal cancer. Br J Surg 68:326–328. 751. Greene FL, Stewart AK, Norton HJ. A new TNM staging strategy for nodepositive (stage III) colon cancer: an analysis of 50,042 patients. Ann Surg 2002; 236:416–421. 752. Sugarbaker PH, Corlew S. Influence of surgical techniques in patients with colorectal cancer: a review. Dis Colon Rectum 1982; 25:545–557. 753. Zirngibl H, Husemann B, Hemanek P. Intraoperative spillage ot tumor cells in surgery for rectal cancer. Dis Colon Rectum 1990; 33:610–614.

199

754. Porter GA, Soskolne CL, Yakimets WW, Newman SC. Surgeon related factors and outcome in rectal cancer. Ann Surg 1998; 227:157–167. 755. Akyol AM, Mcgregor JR, Galloway DJ, et al. Anastomotic leaks in colorectal cancer surgery: a risk factor for recurrence? Int J Colorectal Dis 1991; 6: 179–183. 756. Fujita S, Teramoto T, Watanabe M, et al. Anastomotic leakage after colorectal cancer surgery: a risk factor for recurrence and poor prognosis. Jpn J Clin Oncol 1993; 23:299–302. 757. Bell SW, Walker KG, Rickard MJ, et al. Anastomotic leakage after curative anterior resection results in a higher prevalence of local recurrence. Br J Surg 2003; 90:1261–1266. 758. Parrott NR, Lennard TWJ, Taylor RMR. Effect of perioperative blood transfusion on recurrence of colorectal cancer. Br J Surg 1986; 73:970–973. 759. Wobbes T, Joosen KHG, Kuypers HHC, et al. The effect of packed cells and whole blood transfusion on survival after curative resection for colorectal carcinoma. Dis Colon Rectum 1989; 32:743–748. 760. Leite JPM, Granjo MEM, Martins MI, et al. Effect of perioperative blood transfusions on survival of patients after radical surgery for colorectal cancer. Int J Colorectal Dis 1993; 8:1129–1133. 761. Arnoux R, Corman J, Peloquin A, et al. Adverse effect of blood transfusions on patient survival after resection of rectal cancer. Can J Surg 1998; 31:121–126. 762. Francis DMA, Judson RT. Blood transfusion and recurrence of cancer of the colon and rectum. Br J Surg 1987; 74:26–30. 763. Chung M, Steinmetz OK, Gordon PH. Perioperative blood transfusion and outcome after resection for colorectal carcinoma. Br J Surg 1993; 80:927–932. 764. Nowacki MP, Szymandera JJ. The strongest prognostic factors in colorectal carcinoma. Surgicopathologic stage of disease and postoperative fever. Dis Colon Rectum 1985; 26:263–268. 765. Armstrong CP, Ahsan Z, Hinchley G, et al. Appendicectomy and carcinoma of the caecum. Br J Surg 1989; 76:1049–1053. 766. Meyerhardt JA, Tepper JE, Niedzwieki D, et al. Impact of body mass index on outcomes and treatment-related toxicity in a patients with stage II and III rectal cancer: findings from Intergroup Trial 0114. J Clin Oncol 2004; 22:648–657. 767. Halvorsen TB, Seim E. Tumor site: a prognostic factor in colorectal cancer? a multivariate analysis. Scand J Gastoenterol 1987; 22:124–128. 768. Alley PG, McNee RK. Age and sex differences in right colon cancer. Dis Colon Rectum 1986; 29:227–229. 769. Bjerkeset T, Morild J, Mark S, et al. Tumor characteristics of colorectal cancer and their relationship to treatment and prognosis. Dis Colon Rectum 1987; 30:934–938. 770. Steinberg SM, Barwick KW, Stablein DM. Importance of tumour pathology and morphology in patients with surgically resected colon cancer. Findings from the Gastrointestinal Tumor Study Group. Cancer 1986; 58:1340–1345. 771. Cohen AM, Tremiterra S, Candeh F, et al. Prognosis of node-positive colon cancer. Cancer 1991; 67:1859–1861. 772. Ogiwara H, Nakamura T, Baba S. Variables related to risk of recurrence in rectal cancer without lymph node metastases. Ann Surg Oncol 1994; 1:199–104. 773. Schmitz-Moormann P, Himmelmann GW, Baum U, et al. Morphological predictors of survival in colorectal carcinoma: Univariate and multivariate analysis. J Cancer Res Clin Oncol 1987; 113:586–592. 774. McDermott FT, Hughes ESR, Pihl E, et al. Prognosis in relation to symptom duration in colon cancer. Br J Surg 1981; 68:846–849. 775. Lindmark G, Gerdin B, Pahlman L, et al. Prognostic predictors in colorectal cancer. Dis Colon Rectum 1994; 37:1219–1227. 776. Secco GB, Pardelli B, Campora E, et al. Primary mucinous adenocarcinomas and signet-ring cell carcinomas of colon and rectum. Oncology 1994; 51: 30–34. 777. Yamamoto S, Mochizuki H, Hase K, et al. Assessment of clinicopathologic features of colorectal mucinous adenocarcinoma. Am J Surg 1993; 166: 257–261. 778. Sadahiro S, Ohumura T, Saito T, et al. An assessment of the mucous component in carcinoma of the colon and rectum. Cancer 1989; 64:1113–1116. 779. Nozoe T, Anai H, Nasu S, Sugimachi K. Clinicopathological characteristics of mucinous carcinoma of the colon and rectum. J Surg Oncol 2000; 75: 103–107. 780. Consorti F, Lorenzotti A, Midiri G, Di Paola M. Prognostic significance of mucinous carcinoma of colon and rectum: a prospective case-control study. J Surg Oncol 2000; 73:70–74. 781. Kanemitsu Y, Kato T, Hirai T, et al. Survival after curative resection for mucinous adenocarcinoma of the colorectum. Dis Colon Rectum 2003; 46: 160–167. 782. Halvorsen TB, Seim E. Association between invasiveness, inflammatory reaction, desmoplasia and survival in colorectal cancer. J Clin Pathol 1989; 42:162–166. 783. Nadel L, Mori K, Shinya H. Primary linitis plastica of the colon and rectum. Report of two cases. Dis Colon Rectum 1983; 26:738–742. 784. Wolmark N, Fischer B, Wieand HS. The prognostic value of the modifications of the Dukes’ C classification of colorectal cancer. An analysis of the NSABP clinical trials. Ann Surg 1986; 203:115–122. 785. Gagliardi G, Stepniewska KA, Hershman MJ, et al. New grade related prognostic variable in rectal cancer. Br J Surg 1995; 82:599–602.

200

& PART I: COLORECTAL DISORDERS

786. Hase K, Shatney C, Johnson D, et al. Prognostic value of tumor ‘‘budding’’ in patients with colorectal cancer. Dis Colon Rectum 1993; 36:627–635. 787. Tanaka M, Hashiguchi Y, Ueno H, Hase K, Mochizuki H. Tumor budding at the invasive margin can predict patients at high risk of recurrence after curative surgery for stage II, T3 colon cancer. Dis Colon Rectum 2003; 46:1054–1059. 788. Okuyama T, Nakamura T, Yamaguchi M. Budding is useful to select high-risk patients in stage II well-differentiated or moderately colon adenocarcinoma. Dis Colon Rectum 2003; 46:1400–1406. 789. Chen JS, Hsieh PS, Hung SY, et al. Clinical significance of signet ring cell rectal carcinoma. Int J Colorectal Dis 2004; 19:102–107. 790. Nissan A, Guillem JG, Paty PB, Wong WD, Cohen AM. Signet-ring cell carcinoma of the colon and rectum: a matched control study. Dis Colon Rectum 1999; 42:1176–1180. 791. Kang H, O’Connell JB, Maggard MA, Sack J, Ko CY. A 10-year outcomes evaluation of mucinous and signet-ring cell carcinoma of the colon and rectum. Dis Colon Rectum 2005; 46:1161–1168. 792. Chan CL, Chafai N, Rickard MJ, Dent OF, Chapuis PH, Bokey EL. What pathologic features influence survival in patients with local residual tumor after resection of colorectal cancer? J Am Coll Surg 2004; 199:680–686. 793. Gardner B, Feldman J, Spivak Y, et al. Investigations and factors influencing the prognosis of colon cancer. Am J Surg 1987; 153:541–544. 794. Malassagne B, Valleur P, Serra J, et al. Relationship of apical lymph node involvement to survival in resected colon carcinoma. Dis Colon Rectum 1993; 36:645–653. 795. Tang R, Wang JY, Chen JS, et al. Survival impact of lymph node metastases in TNM stage III carcinoma of the colon and rectum. J Am Coll Surg 1995; 180:705–712. 796. Khubchandani M. Relationship of symptom duration and survival in patients with carcinoma of the colon and rectum. Dis Colon Rectum 1985; 28:585–587. 797. Wong JH, Steinemann S, Tom P, Morita S, Tauchi-Nishi P. Volume of lymphatic metastases does not independently influence prognosis in colorectal cancer. J Clin Oncol 2002; 20:1506–1511. 798. Rothenberg KH. Genetic information and health insurance: state legislative approaches. J Law Med Ethics 1995; 23:312–319. 799. Martin EW, Joyce S, Lucas J, et al. Colorectal carcinoma in patients less than 40 years of age. Pathology and prognosis. Dis Colon Rectum 1981; 24:25–28. 800. Horn A, Dahl O, Morild I. Venous and neural invasion as predictors of recurrence in rectal adenocarcinoma. Dis Colon Rectum 1991; 34:798–804. 801. Sternberg A, Amar M, Alfici R, Groisman G. Conclusions from a study of venous invasion in stage IV colorectal adenocarcinoma. J Clin Pathol 2002; 55:17–21. 802. Horn A, Dahl O, Morild I. The role of venous and neural invasion on survival in rectal adenocarcinoma. Dis Colon Rectum 1990; 33:598–601. 803. Minsky BD, Mies C, Recht A, et al. Resectable adenocarcinoma of the rectosigmoid and rectum. II. The influence of blood vessel invasion. Cancer 1988; 61:1417–1424. 804. Krasna MJ, Flanebaum L, Cody RP, et al. Vascular and neural invasion in colorectal carcinoma. Incidence and prognostic significance. Cancer 1988; 61:1018–1023. 805. Tissot E, Naouri A, Naouri C, et al. Potentially curative surgery of colon carcinoma: influence of blood vessel invasion. Br J Surg 1995; 82(suppl 1):36. 806. Shirouzu K, Isomoto H, Kakegana T. Prognostic evaluation of perinueral invasion in rectal cancer. Am J Surg 1993; 165:233–237. 807. Ueno H, Hase K, Mochizuki H. Criteria for extramural perineural invasion as a prognostic factor in rectal cancer. Br J Surg 2001; 88:994–1000. 808. Bosch B, Guller U, Schnider A, et al. Perioperative detection of disseminated tumour cells is an independent prognostic factor in patients with colorectal cancer. Br J Surg 2003; 90(7):882–888. 809. Meyerhard JA, Catalano PJ, Haller DG, et al. Impact of diabetes mellitus on outcomes in patients with colon cancer. J Clin Oncol 2003; 21:433–440. 810. Hojo K, Koyama Y. Postoperative follow-up studies on cancer of the colon and rectum. Am J Surg 1982; 143:293–295. 811. Lewi H, Blumgart LH, Carter DC, et al. Preoperative carcinoembryonic antigen and survival in patients with colorectal cancer. Br J Surg 1984; 71: 206–208. 812. Wanebo HJ, Stevens W. Surgical treatment of locally recurrent colorectal cancer. In: Nelson RL, ed. Problems in Current Surgery. Controversies in Colon Cancer. Philadelphia: JB Lippincott, 1987:115–129. 813. Wolmark N, Fisher B, Wieand HS, et al. The prognostic significance of preoperative carcinoembryonic antigen levels in colorectal cancer. Results from NASBP clinical trials. Ann Surg 1984; 199:375–382. 814. Wang JY, Tang R, Chiag JM. Value of carcinoembryonic antigen in the management of colorectal cancer. Dis Colon Rectum 1994; 37:272–277. 815. Wiratkapun S, Kraemer M, Seow-Choen F, Ho YH, Eu KW. High preoperative serum carcinoembryonic antigen predicts metastatic recurrence in potentially curative colonic cancer: results of a five-year study. Dis Colon Rectum 2001; 44:231–235. 816. Wang WS, Lin JK, Chiou TJ, et al. Preoperative carcinoembryonic antigen level as an independent prognostic factor in colorectal cancer: Taiwan experience. Jpn J Clin Oncol 2000; 30:12–16.

817. Steele G, Zamcheck N. The use of carcinoembryonic antigen in the clinical management of patients with colorectal cancer. Cancer Detect Prev 1985; 8:421–427. 818. Lahr CJ, Soong SJ, Cloud G, et al. A multifactoria analysis of prognostic factors in patients with liver metastases from colorectal carcinoma. J Clin Concol 1983; 1:720–726. 819. Armitage NC, Robins RA, Evans DF, et al. The influence of tumour DNA abnormalities on survival in colorectal cancer. Br J Surg 1985; 72: 828–830. 820. Banner BF, Tomas-de la Vega JE, Roseman DL, et al. Should flow cytometric DNA analysis precede definitive surgery for colon carcinoma. Ann Surg 1986; 202:740–744. 821. Emdin SO, Stenling R, Roos G. Prognostic value of DNA content in colorectal carcinoma. A flow cytometric study with some methodologic aspects. Cancer 1987; 60:1282–1287. 822. Kokal W, Sheibani, K, Terz J, et al. Tumor DNA content in the prognosis of colorectal carcinoma. JAMA 1986; 255:3123–3127. 823. Giaretti W, Danova M, Geido G, et al. Flow cytometric DNA index in the progress of colorectal cancer. Cancer 1991; 67:1921–1927. 824. Garrity MM, Burgart LJ, Mahoney MR, et al. North Central Cancer Treatment Group Study. Prognostic value of proliferation apoptosis, defective DNA mismatch repair, and p53 overexpression in patients with resected Dukes’ B2 or C colon cancer: a North Central Cancer Treatment Group Study. J Clin Oncol 2004; 22:1572–1582. 825. Diep CB, Thorstensen L, Meling GI, Skovlund E, Rognum TO, Lothe RA. Genetic tumor markers with prognostic impact in Dukes’ stages B and C colorectal cancer patients. J Clin Oncol 2003; 21:820–829. 826. Kern SE, Fearon ER, Tersmette KWF, et al. Allelic loss in colorectal carcinoma. JAMA 1989; 261:3099–3103. 827. Jen J, Kim H, Piantadosi S, et al. Allelic loss of chromosome 18q and prognosis in colorectal cancer. N Engl J Med 1994; 331:213–221. 828. Shibata D, Reale MA, Lavin P, et al. The DCC protein and prognosis in colorectal cancer. N Engl J Med 1996; 335:1727–1732. 829. Jernvall P, Makinen MJ, Karttunen TJ, Makela J, Vihko P. Morphological and genetic abnormalities in prediction of recurrence in radically operated colorectal cancer. Anticancer Res 1999; 19:1357–1362. 830. Sarli L, Bottarelli L, Bader G, et al. Associated between recurrence of sporadic colorectal cancer, high level of microsatellite instability, and loss heterozygosity at chromosome 18q. Dis Colon Rectum 2004; 47(9): 1467–1482. 831. Pricolo VE, Finkelstein SD, Wu TT, et al. Prognostic value of TP53 and K-ras-2 mutational analysis in stage III carcinoma of the colon. Am J Surg 1996; 171:41–46. 832. Diez M, Gonzalez A, Enriquez JM, et al. Prediction of recurrence in B-C stages of colorectal cancer by p-53 expression. Br J Surg 1995; 82(Suppl 1):26. 833. Auvinen A, Isola J, Visakorpi T, et al. Overexpression of p53 and long-term survival in colon carcinoma. Br J Cancer 1994; 70:293–296. 834. Sun XF, Carstensen JM, Stall O, et al. Prognostic significance of p53 expression in relation to DNA ploidy in colorectal adenocarcinoma. Virchows Arch A Pathol Anat 1993; 423:443–448. 835. Yamaguchi A, Nakagawara G, Kurosake Y, et al. p53 immunoreaction in endoscopic biopsy specimens of colorectal cancer, and its prognostic significance. Br J Cancer 1993; 68:399–402. 836. Allegra CJ, Paik S, Colangelo LH, et al. Prognostic value of thymidylate synthase, Ki-67, and p53 in patients with Dukes’ B and C colon cancer: a National Cancer Intitute-National Surgical Adjuvant Breast and Bowel Project collaborative study. J Clin Oncol 2003; 21:241–250. 837. Petersen S, Thames HD, Nieder C, Petersen C, Baumann M. The results of colorectal cancer treatment by p53 status: treatment-specific overview. Dis Colon Rectum 2001; 44:322–333. 838. Gervaz P, Bouzourene H, Cerottini JP, et al. Colorectal cancer: distinct genetic categories and clinical outcome based on proximal or distal tumor location. Dis Colon Rectum 2001; 44:364–372. 839. Herrlich P, Pals S, Ponta H. CD 44 in colon cancer. Eur J Cancer 1995; 31A:1116–1122. 840. Miyahara M, Saito T, Kaketani K, et al. Clinical significance of ras p21 over expression for patients with an advanced colorectal cancer. Dis Colon Rectum 1991; 34:1097–1102. 841. Wang Y, Jatkoe T, Zhang Y, et al. Gene expression profiles and molecular markers to predict recurrence of Dukes’ B colon cancer. J Clin Oncol 2004; 22: 1564–1571. 842. Gonen M, Hunmer A, Zervoudakis A, et al. Thymidylate synthase expression in hepatic tumors is a predictor of survival and progression in patients with resectable metastatic colorectal cancer. J Clin Oncol 2003; 21:406–412. 843. Broll R, Busch P, Duchrow M, et al. Influence of thymidylate synthase and p53 protein expression on clinical outcome in patients with colorectal cancer. Int J Colorectal Dis 2005; 20(2):94–102. 844. Kopp R, Rothbauer E, Mueller E, Schildberg FW, Jauch KW, Pfeiffer A. Reduced survival of rectal cancer patients with increased tumor epidermal growth factor receptor levels. Dis Colon Rectum 2003; 46:1391–1399.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

845. Nakmori S, Kameyama M, Imaoka S, et al. Increased expression of sialyl Lewisx antigen correlates with poor survival in patients with colorectal carcinoma: clinicopathological and immunohistochemical study. Cancer Res 1993; 53:3632–3637. 846. Nakayama T, Watanabe M, Katsumata J, et al. Expression of Sialyl Lewisx as a new prognostic factor for patients with advanced colorectal carcinoma. Cancer 1995; 75:2051–2052. 847. Nakagoe T, Fukushima K, Hirota M, et al. Immunohistochemical expression of sialyl Lex antigen in relation to survival of patients with colorectal carcinoma. Cancer 1993; 72:2323–2330. 848. Teixeira CR, Tanaka S, Haruma K, et al. Proliferating cell nuclear antigen expression at the invasive tumor margin predicts malignant potential of colorectal carcinomas. Cancer 1994; 73:575–579. 849. Al-Sheneber IF, Shibata HR, Sampalis J, et al. Prognostic significance of proliferating cell nuclear antigen expression in colorectal cancer. Cancer 1993; 71:1954–1959. 850. Habib NA, Dawson PM, Bradfield JWB, et al. Sialomucins at resection margin and likelihood of recurrence in colorectal carcinoma. Br Med J 1986; 293:521–523. 851. Villias K, Markidis P, Simatus G, et al. Mucin production and survival in colorectal carcinoma. Br J Surg 1995; 82 (Suppl 1):37. 852. Bonatsos G, Velmahos GC, Davaris P, et al. The role of sialomucin of resection margins on local recurrence and survival of patients radically operated on for colorectal carcinoma. Res Surg 1992; 4:68–70. 853. Mitmaker B, Begin LR, Gordon PH. Nuclear shape as a prognostic discriminant in colorectal carcinoma. Dis Colon Rectum 1991; 34:249–259. 854. Verspaget HW, Sier CPM, Ganesh S, et al. Prognostic value of plasminogen activities and their inhibitors in colorectal cancer. Eur J Cancer 1995; 31A: 1105–1109. 855. Hole DJ, McArdle CS. Impact of socioeconomic deprivation on outcome after surgery for colorectal cancer. Br J Surg 2002; 89(5):586–590. 856. Adloff M, Arnaud JP, Schloegel M, et al. Colorectal cancers in patients under 40 years of age. Dis Colon Rectum 1986; 29:322–325. 857. Enblad G, Enblad P, Adami HO, et al. Relationship between age and survival in cancer of the colon and rectum with special reference to patients less than 40 years of age. Br J Surg 1990; 77:611–616. 858. Lee PY, Fletcher WS, Sullivan ES, et al. Colorectal cancer in young patients: characteristics and outcome. Am Surg 1994; 60:607–612. 859. Liang JT, Huang KC, Cheng AL, Jeng YM, Wu MS, Wang SM. Clinicopathological and molecular biological features of colorectal cancer in patients less than 40 years of age. Br J Surg 2003; 90:205–214. 860. Turkiewicz D, Miller B, Schache D, Cohen J, Theile D. Young patients with colorectal cancer: how do they fare? ANZ J Surg 2001; 71:707–710. 861. Chung YF, Eu KW, Machin D, et al. Young age is not a poor prognostic marker in colorectal cancer. Br J Surg 1998; 85(2):1255–1259. 862. O’Connell JB, Maggard MA, Liu JH, Etzioni DA, Ko CY. Are survival rates different for young and older patients with rectal cancer? Dis Colon Rectum 2004; 47(12):2064–2069. 863. Paraf F, Jothy S. Colorectal cancer before the age of 40: a case-control study. Dis Colon Rectum 2000; 43:1222–1226. 864. Irvin TT. Prognosis of colorectal carcinoma in the elderly. Br J Surg 1988; 75:419–421. 865. Arnaud JP, Schloegel M, Ollier JC, et al. Colorectal cancer in patients over 80 years of age. Dis Colon Rectum 1991; 34:896–898. 866. Mulcahy HE, Patchett SE, Daly L, et al. Prognosis of elderly patients with large bowel cancer. Br J Surg 1994; 81:736–738. 867. Barrier A, Ferro L, Houry S, Lacaine F, Huguier M. Rectal cancer surgery in patients more than 80 years of age. Am J Surg 2003; 185:54–57. 868. Schillaci A, Cavallora A, Nicolanti V. The importance of symptom duration in relation of prognosis of carinoma of the large intestine. Surg Obstet Gynecol 1984; 158:423–426. 869. Goodman D, Irvin TT. Delay in diagnosis and prognosis of carcinoma of the right colon. Br J Surg 1993; 80:1327–1329. 870. Rouffet F, Hay JM, Vacher B, et al. Curative resection for left colonic carcinoma: hemicolectomy vs. segmental colectomy. A prospective, controlled, multicenter trial. French Association for Surgical Research. Dis Colon Rectum 1994; 37:651–659. 871. McArdle CS, Hole DJ. Influence of volume and specialization on survival following surgery for colorectal cancer. Br J Surg 2004; 91:610–617. 872. Crowson MC, Hallissey MT, Kiff RS. Blood transfusion in colorectal cancer. Br J Surg 1989; 76:522–523. 873. Weiden PL, Bean MA, Schultz P. Perioperative blood transfusion does not increase the risk of colorectal cancer recurrence. Cancer 1987; 60:870–874. 874. Sibbering DM, Locker AP, Hardcastle JD, et al. Blood transfusion and survival in colorectal cancer. Dis Colon Rectum 1994; 37:358–363. 875. Fucini C, Bandettini L, D’Elia M, et al. Are postoperative fever and/or septic complications prognostic factors in colorectal cancer resected for cure. Dis Colon Rectum 1985; 28:94–95. 876. Martin MB, Frontier T, Jarman W, et al. Colon and rectal carcinoma. Forty years and 1400 cases. Am Surg 1987; 53:146–148. 877. Wolmark N, Fisher B, Wieand HS, et al. The relationship of depth of penetration and tumour size to number of positive nodes in Dukes’ C colorectal carcinoma. Cancer 1984; 53:2707–2712.

201

878. Green JB, Timmcke AE, Mitchell WT, et al. Mucinous carcinoma—just another colon cancer? Dis Colon Rectum 1993; 36:47–54. 879. Minsky B. Clinicopathologic impact in colorectal carcinoma. Dis Colon Rectum 1990; 33:714–719. 880. Bessa X, Pinol V, Castellvi-Bel S, et al. Prognostic value of postoperative detection of blood circulating tumor cells in patients with colorectal cancer operated on for cure. Ann Surg 2003; 237:368–375. 881. Moertel CG, O’Fallon JR, Go VL, et al. The preoperative carcinoembryonic antigen test in the diagnosis, staging, and prognosis of colorectal cancer. Cancer 1986; 58:603–611. 882. Jones DJ, Moore M, Schofield PF. Prognostic significance of DNA ploidy in colorectal cancer. A prospective flow cytometric study. Br J Surg 1988; 75:28–33. 883. Enker WE. Flow cytometric determination of tumor cell DNA content and proliferative index as prognostic variables in colorectal cancer. Perspect Colon Rectal Surg 1990; 3:1–28. 884. Michelassi F, Ewing CH, Montag A, et al. Prognostic significance of polidy determination in rectal cancer. Hepatogastroenterology 1992; 39:222–225. 885. Deans GT, Williamson K, Heatley M, et al. The role of flow cytometry in carcinoma of the colon and rectum. Surg Gynecol Obstet 1993; 177: 377–382. 886. Morgan M, Koorey D, Painter D, et al. p53 and DCC immunohistochemistry in curative rectal cancer surgery. Int J Colorectal Dis 2003; 18:188–195. 887. Iacopetta B. TP53 mutation in colorectal cancer. Hum Mutat 2003; 21:271–276. 888. Allega CJ, Parr Al, Wold LE, et al. Investigation of the prognostic and predictive value of thymidylate synthase, p53, and Ki-67 in patients with locally advanced colon cancer. J Clin Oncol 2002; 20:1735–1743. 889. Gallego MG, Acenero MJ, Ortega S, Delgado AA, Cantero JL. Prognostic influence of p53 nuclear overexpression in colorectal carcinoma. Dis Colon Rectum 2000; 43:971–975. 890. Grewal H, Guillem JG, Klimstra DS, et al. p53 Nuclear over-expression may not be an independent prognostic market in early colorectal cancer. Dis Colon Rectum 1995; 38:1176–1181. 891. Mulder JWR, Baas IO, Polak MM, et al. Evaluation of p53 protein expression as a marker for long-term prognosis in colorectal carcinoma. Br J Cancer 1995; 71:1257–1262. 892. Suzuki H, Matsumoto K, Koide A, et al. Correlation of p53 with the clinicopathologic features and prognosis of colorectal carcinoma. Jpn J Surg 1994; 24:85–87. 893. Johnston PG, Benson AB 3rd, Catalano P, Rao MS, O’Dwyer PJ, Allegra CJ. Thymidylate synthase protein expression in primary colorectal cancer: lack of correlation with outcome and response to fluorouracil in metastatic disease sites. J Clin Oncol 2003; 21:815–819. 894. Malcolm AW, Perencevich NP, Olson RM, et al. Analysis of recurrence patterns following curative resection for carcinoma of the colon and rectum. Surg Gynecol Obstet 1981; 152:131–136. 895. Pescatori M, Maria G, Betrani B, et al. Site, emergency, and duration of symptoms in prognosis of colorectal cancer. Dis Colon Rectum 1982; 25:33–40. 896. Sanfelippo PM, Beahrs OH. Factors in the prognosis of adenocarcinoma of the colon and rectum. Arch Surg 1972; 104:401–406. 897. Reinbach DH, McGregor JR. Murray GD, et al. Effect of surgeon’s speciality interest on the type of resection performance for colorectal cancer. Dis Colon Rectum 1994; 37:1020–1023. 898. Myerson RJ, Michalski JM, King ML, et al. Adjuvant radiation therapy for rectal carcinoma: predictors of outcome. Int J Radiat Oncol Biol Phys 1995; 32:41–50. 899. Harmon JW, Tang DG, Gordon TA, et al. Hospital volume can serve as a surrogate for surgeon volume for achieving excellent outcomes in colorectal resection. Ann Surg 1999; 230(3):404. 900. Ko Cy, Chang JT, Chaudhry S, Kominski G. Are high-volume surgeons and hospitals the most important predictors of in-hospital outcome for colon cancer resection? Surgery 2002; 132:268–273. 901. Schrag D, Panageas KS, Riedel E, et al. Hospital and surgeon procedure volume as predictors of outcome following rectal cancer resection. Ann Surg 2002; 236:583–592. 902. Martling A, Cedermark B, Johansson H, Rutqvist LE, Holm T. The surgeon as a prognostic factor after the introduction of total mesorectal excision in the treament of rectal cancer. Br J Surg 2002; 89:1008–1013. 903. Hermanek P, Hohenberger W. The importance of volume in colorectal cancer surgery. Eur J Surg Oncol 1996; 22:213–215. 904. Callahan MA, Christos PJ, Gold HT, Mushlin AI, Dally JM. Influence of surgical subspecialty training on in-hospital mortality for gastrectomy and colectomy patients. Ann Surg 2003; 238:629–636. 905. Rosen L, Stasik JJ Jr., Reed JF 3rd, Olenwine JA, Aronoff JS, Sherman D. Variations in colon and rectal surgical mortality. Comparison of specialties with a state-legislated database. Dis Colon Rectum 1996; 39:129–135. 906. Platell C, Lim D, Tajudeen N, Tan JL, Wong K. Does surgical sub-specialization influence survival in patients with colorectal cancer? World J Gastroenterol 2003; 9:961–964. 907. Read TE, Myerson RJ, Fleshman JW, Fry RD, Birnbaum EH, Walz BJ, Kodner IJ. Surgeon speciality is associated with outcome in rectal cancer treatment. Dis Colon Rectum 2002; 45:904–914.

202

& PART I: COLORECTAL DISORDERS

908. Dorrance HR, Docherty GM, O’Dwyer PJ. Effect of surgeon speciality interest on patient outcome after potentially curative colorectal cancer surgery. Dis Colon Rectum 2000; 43:492–498. 909. Smith JAE, King PM, Lane RHS, Thompson MR. Evidence of the effect of ‘specialization’ on the management, surgical outcome and survival from colorectal cancer in Wessex. Br J Surg 2003; 90:583–592. 910. Hafstrom L, Holmin T. Relationship between postoperative temperature and survival in patients resected for colorectal cancer. Am J Surg 1978; 135: 312–314. 911. Michelassi F, Vannucci L, Montag A, et al. Importance of tumor morphology for the long term prognosis of rectal adenocarcinoma. Am Surg 1988; 54: 376–379. 912. Chambers WM, Khan U, Gagliano A, Smith RD, Sheffield J, Nicholls RJ. Tumor morphology as a predictor of outcome after local excision of rectal cancer. Br J Surg 2004; 91:457–459. 913. Jass JR, Love SB, Northover JMA. A new prognostic classification of rectal cancer. Lancet 1987; 1:1303–1306. 914. Fulmes M, Setrakian S, Raj PK, Bogard BM. Cancer biology and necrotic changes in metastatic lymph nodes and survival of colon cancer patients. Am J Surg 2005; 189:364–368. 915. Smyth EF, Sharma A, Sivarajasingham N, Hartley J, Manson JR, Cawkwell L. Prognostic implications of hMLH1 and p53 immunohistochemical status in right-sided colon cancer. Dis Colon Rectum 2004; 47(12):2086–2091. 916. Hemminki A, Mecklin JP, Jarvinen H, Aaltonen LA, Joensuu H. Microsatellite instability is a favorable prognostic indicator in patients with colorectal cancer receiving chemotherapy. Gastroenterology 2000; 119:921–928. 917. Lim SB, Jeong SY, Lee MR, et al. Prognostic significance of microsatellite instability in sporadic colorectal cancer. Int J Colorectal Dis 2004; 19(6): 533–537. 918. Popat S, Hubner R, Houlston RS. Systematic review of microsatellite instability and colorectal cancer prognosis. J Clin Oncol 2005; 23(3):609–618. 919. Svendson LB, Sorensen C, Kjersgaard P, et al. The influence of age upon the survival of curative operation for colorectal cancer. Int J Colorectal Dis 1989; 4:123–127. 920. Coburn MC, Pricolo VE, Soderberg CH. Factors affecting the prognosis and management of carcinoma of the colon and rectum on patients more than eighty years of age. J Am Coll Surg 1994; 179:65–69. 921. Nilsson E, Bolin S, Sjodeh R. Carcinoma of the colon and rectum. Delay in diagnosis. Acta Chir Scand 1982; 148:617–622. 922. Wang SH, Lin JK, Moul YC, et al. Long-term prognosis of patients with obstructing carcinoma of the right colon. Am J Surg 2004; 187:497–500. 923. Meyerhardt JA, Tepper JE, Niedzwiecki D, et al. Impact of hospital procedure volume on surgical operation and long-term outcomes in high-risk curatively resected rectal cancer: findings from the Intergroup 0114 study. J Clin Oncol 2004; 22:166–174. 924. Schrag D, Panageas KS, Riedel E, et al. Surgeon volume compared to hospital volume as a predictor of outcome following primary colon cancer resection. J Surg Oncol 2003; 83:68–78. 925. Borowski DW, Kelly SB, Ratcliffe AA, et al. Small caseload in colorectal cancer surgery is associated with adverse outcome. Br J Surg 2004; 91(suppl 1):73. 926. Wibe A, Eriksen MT, Syse A, Tretli S, Myrvold HE, Soreide O. Norwegian rectal cancer group. Effect of hospital caseload on long-term outcome after standardization of rectal cancer surgery at a national level. Br J Surg 2005; 92:217–224. 927. Martling A, Holm T, Rutqvist LE, et al. Impact of a surgical training programme on rectal cancer outcomes in Stockholm. Br J Surg 2005; 92(2):225–229. 928. Davis CJ, Ilstrup DM, Pemberton JH. Influence of splenectomy on survival rate of patients with colorectal cacner. Am J Surg 1988; 155:173–179. 929. Varty PP, Linehan IP, Boulos PB. Does concurrent splenectomy at colorectal cancer resection influence survival? Dis Colon Rectum 1993; 36:602–606. 930. Newland RC, Dent OF, Chapuis PH, et al. Survival after curative resection of lymph node negative colorectal carcinoma. Cancer 1995; 76:564–571. 931. Greene FL, Stewart AK, Norton HJ. New tumor-node-metastasis staging strategy for node-positive (stage III) rectal cancer: an analysis. J Clin Oncol 2004; 22:1778–1784. 932. Swanson RS, Compton CC, Stewart AK, Bland KI. The prognosis of T3N0 colon cancer is dependent on the number of lymph nodes examined. Ann Surg Oncol 2003; 10:65–71. 933. Fisher ER, Colangeio L, Wieand S, Fisher B, Wolmark N. Lack of influence of cytokeratin-positive mini micrometastases in ‘‘Negative Node’’ patients with colorectal cancer: findings from the national surgical adjuvant breast and bowel projects protocols R-01 and C-01. Dis Colon Rectum 2003; 46: 1021–1025. 934. Sakuragi M, Togashi K, Konishi F, et al. Predictive factor for lymph node metastasis in T1 stage colorectal carcinomas. Dis Colon Rectum 2003; 46: 1626–1632. 935. Tepper JE, O’Connell MJ, Niedzwiecki D, et al. Impact of number of nodes retrieved on outcome in patients with rectal cancer. J Clin Oncol 2001; 19:157–163. 936. Beart RW Jr., O’Connell MJ. Postoperative follow-up of patients with carcinoma of the colon. Mayo Clin Proc 1983; 58:361–363.

937. Araki Y, Isomoto H, Morodomi T, et al. Survival of rectal carcinoma patients studied with flow cytometric DNA analysis. Kurume Med J 1990; 37:277–283. 938. Giaretti W, Sciallero S, Bruno S, et al. DNA flow cytometry of endoscopically examined colorectal adenomas and adenocarcinomas. Cytometry 1988; 9:238–244. 939. Halvorsen TB, Johannesen E. DNA ploidy, tumour site, and prognosis in colorectal cancer. A flow cytometric study of paraffin-embedded tissue. Scand J Gastroenterol 1990; 25:141–148. 940. Kohen-Corish MR, Daniel JJ, Chan C, et al. Low microsatellite instability is associated with poor prognosis in stage C colon cancer. J Clin Oncol 2005; 23(11):2318–2324. 941. Cali RL, Pitsch RM, Thorson AG, et al. Cumulative incidence of metachronous colorectal cancer. Dis Colon Rectum 1993; 36:388–393. 942. Tornqvist A, Ekelund G, Leandoer L. The value of intensive follow-up after curative resection for colorectal carcinoma. Br J Surg 1982; 69:725–728. 943. Cochrane JP, Williams JT, Faber RG, et al. Value of outpatient follow-up after curative surgery for carcinoma of the large bowl. Br Med J 1980; 280: 593–595. 944. Ekman CA, Gustavson J, Henning A. Value of a follow-up study of recurrent carcinoma of the colon and rectum. Surg Gynecol Obstet 1977; 145:895–897. 945. Buhler H, Seefeld U, Deyhle P, et al. Endoscopic follow-up after colorectal cancer surgery. Early detection of local recurrence? Cancer 1984; 54:791–793. 946. Wenzl E, Wunderlich M, Herbst F. Results of a rigorous follow-up system in colorectal cancer. Int J Colorectal Dis 1988; 3:176–180. 947. Yamamoto Y, Imai H, Iwamoto S, et al. Surgical treatment of the recurrence of colorectal cancer. Jpn J Surg 1996; 26:164–168. 948. Umpleby HC, Williamson RCN. Anastomotic recurrence in large bowel cancer. Br J Surg 1987; 74:873–878. 949. Russell A, Tong D, Dawson LE, et al. Adenocarcinoma of the proximal colon: Sites of initial dissemination and patterns of recurrence following surgery alone. Cancer 1984; 53:360–367. 950. Gunderson LL, Tepper JE, Dosoretz DE, et al. Patterns of failure after treatment of gastrointestinal cancer. In: Cox JD, ed. Proceedings of CROS-NCI Conference on Patterns of Failure after Treatment of Cancer. Vol. 2. Cancer Treatment Symposium, 1983. 951. Bohm B, Schwenk W, Hucke HP, et al. Does methodic long-term follow-up affect survival after curative resection of colorectal carcinoma? Dis Colon Rectum 1993; 36:280–286. 952. Obrand DI, Gordon PH. Incidence and patterns of recurrence following curative resection for colorectal carcinoma. Dis Colon Rectum 1997; 40:15–24. 953. Rodriguez-Bigas MA, Stulc JP, Davidson B, et al. Prognostic significance of anastomotic recurrence from colorectal adenocarcinoma. Dis Colon Rectum 1992; 35:838–842. 954. Boey J, Cheung HC, Lai CK, et al. A prospective evaluation of serum carcinoembryonic antigen (CEA) levels in the management of colorectal carcinoma. World J Surg 1984; 8:279–286. 955. Willett CG, Tepper JE, Cohen AM, et al. Failure patterns following curative resection of colonic carcinoma. Ann Surg 1984; 200:685–690. 956. Gunderson LL, Sosin H, Levitt S. Extrapelvic colon–Areas of failure in reoperation series: implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 1985; 11:731–741. 957. Galandiuk S, Wieand HS, Moertel CG, et al. Patterns of recurrence after curative resection of carcinoma of the colon and rectum. Surg Gynecol Obstet 1992; 174:27–32. 958. Koea JB, Lanouette N, Paty PB, Guilem JG, Cohen AM. Abdominal wall recurrence after colorectal resection for cancer. Dis Colon Rectum 2000; 43:628–632. 959. Barillari P, Ramacciato G, Manetti G, et al. Surveillance of colorectal cancer. Effectiveness of early detection of intraluminal recurrences on prognosis and survival of patients treated for cure. Dis Colon Rectum 1996; 38:388–393. 960. Jacquet P, Jelinek JS, Stenes MA, et al. Evaluation of computed tomography in patients with peritoneal carcinomatosis. Cancer 1993; 72: 1631–1636. 961. Delbeke D, Vitola JV, Sandler MP, et al. Staging recurrent metastatic colorectal carcinoma with PET. J Nucl Med 1997; 38:1196–1201. 962. Libutti SK, Alexander HR Jr., Choyke P, et al. A prospective study of 2-[18F] fluoro-2-deoxy-D-glucose/positron emission tomography scan 99mTclabeled arcitumomab (CEA-scan), and blind second-look laparotomy for detecting colon cancer recurrence in patients with increasing carcinoembryonic antigen levels. Ann Surg Oncol 2001; 8(10):779–786. 963. Desai DC, Zervos EE, Arnold MW, Burak WE Jr., Mantil J, Martin EW Jr. Positron emission tomography affects surgical management in recurrent colorectal cancer patients. Ann Surg Oncol 2003; 10:59–64. 964. Whiteford MH, Whiteford HM, Yee LF, et al. Usefulness of FDG-PET scan in the assessment of suspected inetastatic or recurrent adenocarcinoma of the colon and rectum. Dis Colon Rectum 2000; 43:759–767. 965. Johnson K, Bakhsh A, Young D, Martin TE Jr, Arnold M. Correlating computed tomography and positron emission tomography scan with operative findings in metastatic colorectal cancer. Dis Colon Rectum 2001; 44: 354–357.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

966. Minton JP, Hoehn JL, Gerber DM, et al. Results of a 400 patient carcinoembryonic antigen second-look colorectal cancer study. Cancer 1985; 55: 1284–1290. 967. Staab HJ, Anderer FA, Stumpf E. Eighty-four potential second-look operations based on sequential carcino-embryonic antigen determinations and clinical investigations in patients with recurrent gastrointestinal cancer. Am J Surg 1985; 149:198–204. 968. Hida J, Yasutomi M, Shindoh K, et al. Second-look operation for recurrent colorectal cancer based on carcino-embryonic antigen and imaging techniques. Dis Colon Rectum 1996; 39:74–79. 969. Waldron RP, Donovam JA. Clinical follow-up and treatment of locally recurrent colorectal cancer. Dis Colon rectum 1987; 30:428–430. 970. Gwin JL, Hoffman JP, Eisenberg BL. Surgical management of nonhepatic intra-abdominal recurrence of carcinoma of the colon. Dis Colon Rectum 1993; 36:540–544. 971. Herfarth C, Schlag P, Hohenberger P. Surgical strategies in locoregional recurrences of gastrointestinal carcinoma. World J Surg 1987; 11:504–510. 972. Law NL, Chan WF, Lee YM, Chu KW. Non-curative surgery for colorectal cancer: critical appraisal of outcomes. Colorectal Dis 2004; 19:197–202. 973. Bowne WB, Lee B, Wong WD, et al. Operative salvage for locoregional recurrent colon cancer after curative resection: an analysis of 100 cases. Dis Colon Rectum 2005; 48:897–909. 974. Willett CG, Shelleto PC, Teppa JE, et al. Intraoperative electron beam radiation therapy for recurrent locally advanced rectal or rectosigmoid carcinoma. Cancer 1991; 67:1504–1508. 975. Taylor WE, Donohue JH, Gunderson LL, et al. The Mayo clinic experience with multimodality treatment of locally advanced or recurrent colon cancer. Ann Surg Oncol 2002; 9:177–185. 976. Wodnicki H, Goldeberg RI, Kaplan SR, et al. The laser. An alternative for palliative treatment of obstructing intraluminal lesions. Am Surg 1988; 54: 227–230. 977. Buchi KN. Endoscopic laser surgery in the colon and rectum. Dis Colon Rectum 1958; 31:739–748. 978. Bown SG, Barr H, Mathewson K, et al. Endoscopic treatment of inoperable colorectal cancers with the Nd:YAG laser. Br J Surg 1986; 73:949–952. 979. Faintuch JS. Endoscopic laser therapy in colorectal carcinoma. Hematol Oncol Clin North Am 1989; 3:155–170. 980. Krasner N. Lasers in the treatment of colorectal disease. Symposium. Int J Colorectal Dis 1989; 4:1–29. 981. Tan CC, Ifeikhar SY, Allan A, et al. Local effects of colorectal cancer are well palliated by endoscopic laser therapy. Eur J Surg Oncol 1995; 21:648–652. 982. Mandava N, Petrelli N, Herrera L, et al. Laser palliation for colorectal carcinoma. Am J Surg 1991; 162:212–215. 983. Chia YC, Ngoi SS, Goh PMY. Endoscopic Nd:YAG laser in the palliative treatment of advanced low rectal carcinoma in Singapore. Dis Colon Rectum 1991; 34:1093–1096. 984. Arrigoni A, Pernazio M, Spandre M, et al. Emergency endoscopy: recanalization of intestinal obstruction caused by colorectal cancer. Gastrointest Endosc 1994; 40:576–580. 985. Dittrich K, Armbruster C, Hoffer F, et al. Nd:YAG laser treatment of colorectal maliganancies: an experience of 412 years. Lasers Surg Med 1992; 12:199–203. 986. Bright N, Hale P, Mason R. Poor palliation of colorectal malignancy with the neodymium yttrium-aluminum-garnet laser. Br J Surg 1992; 79:308–309. 987. Courtney ED, Raja A, Leicester RJ. Eight-years experience of high-powered endoscopic diode laser therapy for palliation of colorectal carcinoma. Dis Colon Rectum 2005; 48:845–850. 988. Pilati P, Mocellin S, Rossi CR, et al. Cytoreductive surgery combined with hyperthermic intraperitoneal intraoperative chemotherapy for peritoneal carcinomatosis arising from colon adenocarcinoma. Ann Surg Oncol 2003; 10:508–513. 989. Verwaal VJ, Boot H, Aleman BM, van Tinteren H, Zoetmulder FA. Recurrences after peritoneal carcinomatosis of colorectal origin treated by cytoreduction and hyperthermic intraperitoneal chemotherapy: location, treatment, and outcome. Ann Surg Oncol 2004; 11:375–379. 990. Smith P, Bruera E. Management of malignant ureteral obstruction in the palliative care setting. J Pain Symptom Manage 1995; 10:481–486. 991. Keidan RD, Greenberg RE, Hoffman JP, et al. Is precutaneous nephrostomy for hydronephrosis appropriate in patients with advanced cancer? Am J Surg 1988; 156:206–208. 992. Douillard JY, Cunningham D, Roth AD, et al. Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: a multicentre randomised trial. Lancet 2000; 355: 1041–1047. 993. Saltz LB,.Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000; 343: 905–914. 994. de Gramont A, Figer A, Seymour M, et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18:2938–2947. 995. Sorbye H, Glimelius B, Berglund A, et al. Multicenter phase II study of Nordic fluorouracil and folinic acid bolus schedule combined with oxalipla-

996.

997.

998.

999. 1000.

1001.

1002.

1003.

1004.

1005.

1006.

1007. 1008.

1009. 1010.

1011. 1012. 1013. 1014.

1015.

1016.

1017. 1018. 1019.

1020.

1021.

203

tin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2004; 22:31–38. Goldberg RM, Sargent DJ, Morton RF, et al. A randomized controlled trial of fluorouracil plus leucovorin, irinotecan, and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004; 22:22–30. Tournigand C, Andre T, Achille E, et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol 2004; 22:229–237. Colucci G, Gebbia V, Paoletti G, et al. Phase III randomized trial of FOLFIRI vs FOLFOX4 in the treatment of advanced colorectal cancer: A multicenter study of the gruppo Oncologico Dell’italia Meridionale. J Clin Oncol 2005; 23:4866–4875. Ibrahim A, Hirschfeld S, Cohen MH, Griebel DJ, Williams GA, Pazdur R. FDA drug approval summaries: Oxaliplatin. The Oncologist 2004; 9:8–12. Kohne CH, van Custem E, Wils J, et al. Phase III study of weekly high-dose infusional fluorouracil plus folinic acid with or without irinotecan in patients with metastatic colorectal cancer: European Organization for Research and Treatment of Cancer Gastrointestinal Group (GI) study 40986. J Clin Oncol 2005; 23(22):4856–4865. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350:2335–2342. Meta-analysis Group in Cancer. Reappraisal of hepatic arterial infusion in the treatment of nonresectable liver metastases from colorectal cancer. J Natl Cancer Inst 1996; 88:252–258. Lorenz M, Muller HH. Randomized, multicenter trial of fluorouracil plus leucovorin administered either via hepatic arterial or intravenous infusion versus fluorodeoxyuridine administered via hepatic arterial arterial infusion in patients with nonresectable liver metastases from colorectal carcinoma. J Clin Oncol 2000; 18:243–254. Kerr DJ, McArdle CS, Ledermann J, et al. Intrahepatic arterial versus intravenous fluorouracil and folinic acid for coloreactal cancer liver metastases: a multicentre randomised trial. Lancet 2003; 361:368–373. Kemeny N, Neidwiecki D, Hollis D, et al. Hepatic arterial infusion (HAI) versus systemic therapy for hepatic metastases from colorectal cancer: a CALGB randomized trial of efficacy, quality of life (QOL), cost effectiveness, and molecular markers. Proc Am Soc Clin Oncol 2003; 22:252a. Keer DJ, McArdle CS, Ledermann J, et al. Medical Research Council’s colorectal cancer study group; European Organisation for Research and Treatment of Cancer colorectal cancer study group. Intrahepatic arterial versus intravenous fluorouracil and folinic acid for colorectal cancer liver metastases: a multicentre randomized trial. Lancet 2003; 361:368–73. Bonetti A. Hepatic artery infusion for liver metastases from colorectal cancer. Lancet 2003; 361:358–359. Martin RC, Edwards MJ, McMasters KM. Morbidity of adjuvant hepatic arterial infusion pump chemotherapy in the management of colorectal cancer metastatic to the liver. Am J Surg 2004; 188(6):714–721. Duprat G, Chalaoui J, Sylvestre J, et al. Intra-arterial infusion chemotherapy in rectal cancer. Can J Surg 1984; 27:57–59. Schneebaum S, Arnold MW, Staubus A, et al. Intraperitoneal hyperthermic perfusion with mitomycin C for colorectal cancer with peritoneal metastases. Ann Surg Oncol 1996; 3:44–50. Ballantyne JC. Chronic pain following treatment for cancer: the role of opioids. The Oncologist 2003; 8:567–575. Cherny NI, Portenoy RK. The management of cancer pain. CA Cancer J Clin 1994; 44:262–303. Waterman NG, Hughes S, Foster WS. Control of cancer pain by epidural infusion of morphine. Surgery 1991; 110:612–616. Foltz A, Besser P, Ellenberg S, et al. Effectiveness of nutritional counseling on caloric intake, weight change and percent protein intake in patients with advanced colorectal and lung cancer. Nutrition 1987; 3:263–271. Vassilopoulos PP, Yoon JM, Ledesma EJ, et al. Treatment of recurrence of adenocarcinoma of the colon and rectum at the anastomotic site. Surg Gynecol Obstet 1981; 152:777–780. Pihl E, Hughes ESR, McDermott FT, et al. Recurrence of carcinoma of the colon and rectum at the anastomotic suture line. Surg Gynecol Obstet 1981; 53:495–496. Makela J, Kairaluoma MI. Reoperation for colorectal cancer. Acta Chir Scand 1986; 152:151–155. Stellato TA, Shenk RR. Gastrointestinal emergencies in the oncologic patient. Semin Oncol 1989; 16:521–531. Van Ooijen B, van der Burg MEL, Planting ASTH, et al. Surgical treatment of gastric drainage only for intestinal obstruction in patients with carcinoma of the ovary or peritoneal carcinomatosis of other origin. Surg Gynecol Obstet 1993; 176:469–474. Lau PWK, Lorentz TG. Results of surgery for malignant bowel obstruction in advanced unresectable, recurrent colorectal cancer. Dis Colon Rectum 1993; 36:61–64. Butler JA, Cameron BL, Morrow M, et al. Small bowel obstruction in patients with a prior history of cancer. Am J Surg 1991; 162:624–628.

204

& PART I: COLORECTAL DISORDERS

1022. Miller G, Boman J, Shrier I, Gordon PH. Small-bowel obstruction secondary to malignant disease: an 11-year audit. Can J Surg 2000; 43:353–358. 1023. Krouse RS, McCahill LE, Easson AM, Dunn GP. When the sun can set on an unoperated bowel obstruction: management of malignant bowel obstruction. J Am Surg 2002; 195:117–128. 1024. August DA, Thorn D, Fisher RL, et al. Home parenteral nutrition for patients with inoperable malignant bowel obstruction. J Parenter Enteral Nutr 1991; 15:323–327. 1025. Parker MC, Baines MJ. Intestinal obstruction in patients with advanced malignant disease. Br J Surg 1996; 83:1–2. 1026. Bernstein MA, Madoff RD, Caushaj PF. Colon and rectal cancer in pregnancy. Dis Colon Rectum 1993; 36:172–178. 1027. Van Voorhis B, Cruikshank DP. Colon carcinoma complicating pregnancy. A report of two cases. J Reprod Med 1989; 34:923–927. 1028. Nesbitt JC, Moise KJ, Sawyers JL. Colorectal carcinoma in pregnancy. Arch Surg 1985; 120:636–639. 1029. Hertel H, Diebolder H, Herrmann J, Kohler C, Kuhne-Heid R, Possover M, Schneider A. Is the decision for colorectal resection justified by histopathologic finding: a prospective study of 100 patients with advanced ovarian cancer. Gynecol Oncol 2001; 83:481–484. 1030. Pillay K, Chetty R. Malakoplakia in association with colorectal carcinoma: a series of four cases. Pathology 2002; 34(4):332–335. 1031. Nakano PH, Bloom RR, Brown BC, et al. Apudomas. Am Surg 1987; 53: 505–509. 1032. Stinner B, Kisker O, Zielke A, et al. Surgical management for carcinoid tumors of small bowel, appendix, colon and rectum. World J Surg 1996; 20:183–188. 1033. Spread C, Berkel H, Jewell L, et al. Colonic cacinoid tumors. A populationbased study. Dis Colon Rectum 1994; 37:482–491. 1034. Ballantyne GH, Savoca PE, Flannery JT, et al. Incidence and mortality of carcinoids of the colon. Cancer 1992; 69:2400–2405. 1035. Godwin JD II. Carcinoid tumors. An analysis of 2837 cases. Cancer 1975; 36:560–569. 1036. Rosenberg JM, Welch JP. Carcinoids of the colon: a study of 72 patients. Am J Surg 1985; 149:775–779. 1037. Walker MJ. Rare tumours of the colon and rectum. In: Nelson RL, ed. Problems in Current Surgery. Controversies in Colon Cancer. Philadelphia: JB Lippincott1987:99:141–153. 1038. Gerstle JT, Kaufman GL, Koltun WA. The incidence, management, and outcome of patients with gastrointestinal carcinoids and secondary primary malignancies. J Am Coll Surg 1995; 180:427–432. 1039. Tichansky DS, Cagir B, Borrazzo E, et al. Risk of second cancers in patients with colorectal carcinoids. Dis Colon Rectum 2002; 45:91–97. 1040. Duh QY, Hybarger CP, Geist R. Carcinoids associated with multiple endocrine neoplasia syndromes. Am J Surg 1987; 154:142–148. 1041. Berardi RS. Carcinoid tumors of the colon (exclusive of the rectum): review of the literature. Dis Colon Rectum 1972; 15:383–391. 1042. Jolles PR. Rectal carcinoid metastatic to the skeleton. Scintigraphic and radiographic correlation. Clin Nucl Med 1994; 19:108–111. 1043. Johnson LA, Lavin P, Moertel CG, et al. Carcinoids: the association of histologic growth pattern and survival. Cancer 1983; 51:882–889. 1044. Wallace S, Ajani JA, Charn Sangavej C, et al. Carcinoid tumors: imaging procedures and interventional radiology. World J Surg 1996; 20:147–156. 1045. Kwekkeboom DJ, Krenning EP. Somatostatin receptor scintigraphy in patients with carcinoid tumors. World J Surg 1996; 20:157–161. 1046. Wood HF, Bax NDS, Smith JAR. Small bowel carcinoid tumors. World J Surg 1985; 9:921–929. 1047. Creutzfeldt N. Carcinoid tumors: development of our knowledge. World J Surg 1996; 20:126–131. 1048. Bowman GA, Rosenthal D. Carcinoid tumors of the appendix. Am J Surg 1983; 146:700–703. 1049. Gouzi JL, Laigneau P, Delalonde JP, et al. Indications for right hemicolectomy in carcinoid tumors of the appendix. Surg Gynecol Obstet 1993; 176: 543–547. 1050. Ahlman H, Westberg G, Wangberg B, et al. Treatment of liver metastases of carcinoid tumors. World J Surg 1996; 20:196–202. 1051. Beaton H, Homan W, Dineen P. Gastrointestinal carcinoids and the malignant carcinoid syndrome. Surg Genecol Obstet 1981; 152:268–272. 1052. Azizkhan RG, Tegtmeyer CJ, Wanebo HJ. Malignant rectal carcinoid: a sequential multidisciplinary approach for successful treatment of hepatic metastases. Am J Surg 1985; 149:210–214. 1053. Vinik A, Moattari AR. Use of somatostatin analog in the management of carcinoid syndrome. Dig Dis Sci 1989; 34(Suppl):14S–27S. 1054. Henry CA, Berry RE. Primary lymphoma of the large intestine. Am Surg 1988; 54:262–266. 1055. Jinnai D, Isawa Z, Watanuki T. Malignant lymphoma of the large intestine— operative results in Japan. Jpn J Surg 1983; 13:331–336. 1056. Pandey M, Kothari KC, Wadhwa MK, Patel HP, Patel SM, Patel DD. Primary malignant large bowel lymphoma. Am Surg 2002; 68:121–126. 1057. Zighelboim J, Larson MV. Primary colonic lymphoma. Clinic presentation, histopathologic features, and outcome with combination chemotherapy. J Clin Gastroenterol 1994; 18:291–297.

1058. Zinzani PL, Magagnoli M, Pagliani G, et al. Primary intestinal lymphoma: clinical and therapeutic features of 32 patients. Haematologica 1997; 82: 305–308. 1059. Dawson IMP, Cornes JS, Morson BC. Primary malignant lymphoid tumours of the intestinal tract. Report of 37 cases with study of factors influencing prognosis. Br J Surg 1961; 49:80–89. 1060. Contreary K, Nance FC, Becker WF. Primary lymphoma of the gastrointestinal tract. Ann Surg 1990; 191:593–598. 1061. Bruneton JN, Thyss A, Bourry J, et al. Colonic and rectal lymphoma. A report of 6 cases and review of the literature. Fortschr Rontgenstr 1983; 138:283–287. 1062. Doolabh N, Anthony T, Simmang C, et al. Primary colonic lymphoma. J Surg Oncol 2000; 74:257–262. 1063. Fan CW, Changchien CR, Wang JY, Chen JS, Hsu KC, Tang R, Chiang JM. Primary colorectal lymphoma. Dis Colon Rectum 2000; 43:1277–1282. 1064. Suzuki A, Fukuda S, Tomita S, et al. An unusual case of colonic leiomyosarcoma presenting with fever. Significant uptake of radioactivity of gallium-67 in the tumor. Gastroenterol Jpn 1984; 19:486–492. 1065. Shiu MH, Farr GH, Egeli RA, et al. Myosarcomas of the small and large intestine: a clinicopathologic study. J Surg Oncol 1983; 24:67–72. 1066. Stavorovsky M, Jaffa AJ, Papo J, et al. Leiomyosarcoma of the colon and rectum. Dis Colon Rectum 1980; 23:249–254. 1067. Akwari OE, Dozois RR, Weiland LN, et al. Leimyosarcoma of the small and large intestine. Cancer 1978; 42:1375–1384. 1068. Berkley KM. Leiomyosarcoma of the large intestine, excluding the rectum. Int Surg 1981; 66:177–179. 1069. Nuessle WR, Magill TR III. Leiomyosarcoma of the transverse colon. Report of a case with discussion. Dis Colon Rectum 1990; 33:323–326. 1070. Shimazu K, Funata N, Yamamoto Y, Mori T. Primary osteosarcoma arising in the colon: report of a case. Dis Colon Rectum 2001; 44:1367–1370. 1071. Smith JA, Bhathal PS, Cuthbertson AM. Angiosarcoma of the colon. A report of a case with long-term survival. Dis Colon Rectum 1990; 33: 330–333. 1072. Schneider TA, Birkett DH, Vernava AM. Primary adenosquamous and squamous cell carcinoma of the colon and rectum. Int J Colorectal Dis 1992; 7: 144–147. 1073. Michelassi F, Mishlove LA, Stipa F, et al. Squamous cell carcinoma of the colon. Experience of the University of Chicago. Review of the literature, report of two cases. Dis Colon Rectum 1988; 31:228–235. 1074. Lafreniere R, Ketcham AS. Primary squamous carcinoma of the rectum. Report of a case and review of the literature. Dis Colon Rectum 1985; 25:967–972. 1075. Vezeridis M, Herrera LO, Lopez GE, et al. Squamous cell carcinoma of the colon and rectum. Dis Colon Rectum 1983; 26:188–191. 1076. Juturi JV, Francis B, Koontz PW, Wilkes JD. Squamous-cell carcinoma of the colon responsive to chemotherapy. Dis Colon Rectum 1999; 42:102–109. 1077. Cerezo L, Alvarez M, Edwards O, Price G. Adenosquamous carcinoma of the colon. Dis Colon Rectum 1985; 28:597–603. 1078. Frizelle FA, Hobday KS, Batts KP, Nelson H. Adenosquamous and squamous carcinoma of the colon and upper rectum: a clinical and histopathologic study. Dis Colon Rectum 2001; 44:341–346. 1079. Sidani MS, Campos MM, Joseph JI. Primary plasmacytomas of the colon. Dis Colon Rectum 1983; 26:182–187. 1080. Lattuneddu A, Farneti F, Lucci E, Garcea D, Ronconi S, Saragoni L. A case of primary extramedullary plasmacytoma of the colon. Int J Colorectal Dis 2004; 19:289–291. 1081. Sperling RI, Fromowitz FB, Castellano TJ. Anaplastic solitary extramedullary plasmacytoma of the cecum. Report of a case confirmed by immunoperoxidase staining. Dis Colon Rectum 1987; 30:894–898. 1082. Tessier DJ, McConnell EJ, Young-Fadok T, Wolff BG. Melanoma metastatic to the colon: case series and review of the literature with outcome analysis. Dis Colon Rectum 2003; 46:441–447. 1083. Moir CR, Scudamore CH, Benny WB. Typhlitis: selective surgical management. Am J Surg 1986; 151:563–566. 1084. Taylor AJ, Dodds WJ, Gonyu JE, et al. Typhlitis in adults. Gastrointest Radiol 1985; 10:363–369. 1085. McClenathan JA. Metastatic melanoma involving the colon. Report of a care. Dis Colon Rectum 1989; 32:70–72. 1086. Keidan RD, Fanning J, Gatenby RA, et al. Recurrent typhlitis. A disease resulting from aggressive chemotherapy. Dis Colon Rectum 1989; 32:206–209. 1087. Vilor M, Tsutsumi Y, Osamura RY, et al. Small cell neuroendocrine carcinoma of the rectum. Case report. Pathol Int 1995; 45:605–608. 1088. Thomas R, Sobin L. Gastrointestinal cancer. Cancer 1995; 75(Suppl): 154–169. 1089. Saclarides TJ, Szeluga D, Staren ED. Neuroendocrine cancers of the colon and rectum: results of 10 year experience. Dis Colon Rectum 1994; 37:635–642. 1090. Bernick PE, Klimstra DS, Shia J, et al. Neuroendocrine carcinomas of the colon and rectum. Dis Colon Rectum 2004; 47:163–169. 1091. Wick MR, Vitsky JL, Ritter JH, et al. Sporadic medullary carcinoma of the colon: A clinicopathologic comparison with nonhereditary poorly differentiated enteric-type adenocarcinoma and neuroendocrine colorectal carcinoma. Am J Clin Pathol 2005; 123(11):56–65.

CHAPTER 3: MALIGNANT NEOPLASMS OF THE COLON &

1092. Weidner N, Zekan P, Carcinosarcoma of the colon. Report of a unique case with light and immunohistochemical studies. Cancer 1986; 58: 1126–1130. 1093. Miettinen M, Shekitka KM, Sobin LH. Schwannomas in the colon and rectum: a clinicopathologic and immunohistochemical study of 20 cases. Am J Surg Pathol 2001; 25:846–855. 1094. Brown CJ, Falck VG, MacLean A. Angiosarcoma of the colon and rectum: report of a case and review of the literature. Dis Colon Rectum 2004; 47(12): 2202–2207. 1095. Le DT, Austin RC, Payne SN, Dworkin MJ, Chappell ME. Choriocarcinoma of the colon: report of a case and review of the literature. Dis Colon Rectum 2003; 46:264–266.

205

1096. Carr CS, Boulos DB. Two cases of solitary metastases from carcinoma of the lung presenting as primary colonic tumours. Br J Surg 1996; 83:647. 1097. Voravud N, El-Naggai AK, Balch CM, et al. Metastatic lobular breast carcinoma simulating primary colon cancer. Am J Clin Oncol 1992; 15:365–369. 1098. Law WL, Chu KW. Scirrhous colonic metastasis from ductal carcinoma of the breast: report of a case. Dis Colon Rectum 2003; 46:1424–1427. 1099. Washingtom K, McDonagh D. Secondary tumors of the gastrointestinal tract: surgical pathologic findings and comparison with autopsy survey. Mod Pathol 1995; 8:427–433. 1100. Bristow RE, del Carmen MG, Kaufman HS, Montz FJ. Radical oophorectomy with primary stapled colorectal anastomosis for resection of locally advanced epithelial ovarian cancer. J Am Coll Surg 2003; 197:565–574.

4

Malignant Neoplasms of the Rectum Philip H. Gordon

& & & &

Concomitant Pelvic Organ Excision, 260 Palliative Therapy Advanced Rectal Carcinoma, 261 Concomitant Pelvicfor Organ Excision, 260 Hartmann’s Procedure, 264 Palliative Therapy for Advanced Rectal Carcinoma, 261 Unresectable Procedure, Carcinoma 264 of the Rectum, 264 Hartmann’s High LigationCarcinoma of InferiorofMesenteric Artery, Unresectable the Rectum, 264 265 Marking the Rectum, 265Mesenteric Artery, 265 High Ligation of Inferior Marking the Rectum, 265 & Adjuvant Therapy for Carcinoma of the Rectum, 266 Radiotherapy, 266 for Carcinoma of the Rectum, 266 & Adjuvant Therapy Preoperative 266 Irradiation, 266 Radiotherapy,

ADENOCARCINOMA, 208 ADENOCARCINOMA, 208 Mechanisms of Spread of Rectal Carcinoma, 208 Mechanisms of Spread of Rectal Carcinoma, 208 Direct Extension, 208 Direct Extension, 208 Transperitoneal Spread, 208 Transperitoneal Spread, 208 Implantation, 208 Implantation, 208 Lymphatic Spread, 208 Lymphatic Spread, 208 Retrograde Retrograde Retrograde Retrograde

Intramural Metastases, 208 Intramural Metastases, 208 Extramural Metastases, 208 Extramural Metastases, 208

Venous Spread, 208 Venous Spread, 208 & Clinical Features, 209 & Clinical Features, 209 Symptoms, 209 Symptoms, 209 General and Abdominal Examination, 209 General and Abdominal Examination, 209

PostoperativeIrradiation, Radiotherapy, Preoperative 266 271

Digital Rectal Examination, 209 Digital Rectal Examination, 209

&

& Investigations, 209 & Investigations, 209

& &

Endoscopy, 209 Endoscopy, 209 Routine Laboratory Blood Work, 209 Routine Laboratory Blood Work, 209 Radiology, 209 Radiology, 209

&

Chest X-Ray Examination, 209 Chest X-Ray Examination, 209 Barium Enema, 209 Barium Enema, 209 Intravenous Pyelography and Cystoscopy, 210 Intravenous Pyelography and Cystoscopy, 210 Imaging Techniques, 210 Imaging Techniques, 210

& & & &

Preoperative Preparation, 213 Preoperative Preparation, 213 Radical Extirpative Operations, 214 Radical Extirpative Operations, 214 Assessment of Resectability, 214 Assessment of Resectability, 214 Selection of Appropriate Operation, 214 Selection of Appropriate Operation, 214 Operative Procedures, 216 Operative Procedures, 216

Chemotherapy, Postoperative 274 Radiotherapy, 271 Combination Chemoradiotherapy, 274 Chemotherapy, 274 Immunotherapy, 279 Combination Chemoradiotherapy, 274 Summary, 280 279 Immunotherapy, Summary, 280 Postoperative Complications, 280 Postoperative Complications, 280 Recurrent Disease, 280 Follow-up, Disease, 280 Recurrent 280 Incidence, 280 Follow-up, 280 Factors Contributing to Recurrence, 281 Incidence, 280 Patterns of Recurrence, 281 Factors Contributing to Recurrence, 281 Clinical Features, 282 281 Patterns of Recurrence, Investigations, 282282 Clinical Features, Treatment of Recurrent Disease, 283 Investigations, 282 OperativeofTreatment, Treatment Recurrent 283 Disease, 283 Management of Sacral283 and Perineal Defects, 288 Operative Treatment, Local Ablation, Management of 289 Sacral and Perineal Defects, 288 Nonoperative 290 Local Ablation,Treatment, 289

Low Anterior Resection, 216 Low Anterior Resection, 216 Abdominoperineal Resection, 226 Abdominoperineal Resection, 226 Restoration of Continuity after Sphincter Preserving Restoration of Continuity after Sphincter Preserving Operations, 230 Operations, 230 Other Procedures, 236 Other Procedures, 236 Double Dynamic Graciloplasty, 236 Double Dynamic Graciloplasty, 236

& & & &

Postoperative Care, 236 Postoperative Care, 236 Results, 236 Results, 236 & Local Forms of Therapy, 245 & Local Forms of Therapy, 245 Rationale, 245 Rationale, 245 Procedures, 245 Procedures, 245

& &

Local Excision, 245 Local Excision, 245 Electrocoagulation, 252 Electrocoagulation, 252 Intracavitary Radiotherapy, 252 Intracavitary Radiotherapy, 252 Cryotherapy, 253 Cryotherapy, 253 Laser Therapy, 253 Laser Therapy, 253 External Beam Radiotherapy, 253 External Beam Radiotherapy, 253

& & & & & & &

& Special Considerations, 253 & Special Considerations, 253

Distal Margins, 253 Distal Margins, 253 Circumferential Margins, 254 Circumferential Margins, 254 Total Mesorectal Excision, 254 Total Mesorectal Excision, 254 Radical Lymphadenectomy, 258 Radical Lymphadenectomy, 258

& & &

207

Results of Reoperation, 290290 Nonoperative Treatment, Results of Reoperation, 290 OTHER MALIGNANT LESIONS OF THE RECTUM, 290 OTHER MALIGNANT LESIONS OF THE RECTUM, 290 Carcinoid, 290 Clinical Presentation, 291 Carcinoid, 290 Investigation, 291 Clinical Presentation, 291 Pathology, 291291 Investigation, Treatment, 291 Pathology, Results, 291291 Treatment, Results, 291 Lymphoma, 292 Clinical Presentation, 292 Lymphoma, 292 Treatment and Results,292 292 Clinical Presentation, Treatment and Results, 292 Sarcoma, 292 Sarcoma, 292 Gastrointestinal Stromal Tumor (GIST), 293 Gastrointestinal Stromal Tumor (GIST), 293 Secondary Carcinoma, 294 Secondary Carcinoma, 294 Miscellaneous Neoplasms, 294 Miscellaneous Neoplasms, 294 References, 294 References, 294

208

& PART I: COLORECTAL DISORDERS

ADENOCARCINOMA It has become customary in textbooks on colon and rectal surgery to divide malignant neoplasms of the colon and rectum into separate chapters. We have continued this tradition but recognize that the division is purely arbitrary. Consequently, this chapter will omit topics that the two subjects have in common—epidemiology, etiology, pathogenesis, and much of the pathology except those features that are characteristic of the rectum. These topics are discussed in detail in Chapter 23. General assessment of the patient is similar, but certain items will be highlighted. Therapeutic options in specific circumstances will differ and consequently will be discussed in detail.

& MECHANISMS OF SPREAD OF RECTAL CARCINOMA The general mechanisms of spread for carcinoma of the rectum are similar to those of the colon, but because of the rectum’s location within the pelvis, it is appropriate to elaborate on these mechanisms, especially as they may pertain to treatment.

& DIRECT EXTENSION Carcinoma of the rectum originates in the mucosa. Eventually it penetrates the thickness of the rectal wall rather than growing in the longitudinal axis. Untreated, it finally involves the full thickness of the rectum and may invade adjacent tissues. & TRANSPERITONEAL SPREAD Peritoneal involvement by carcinoma of the rectum probably starts from local extension, continues through the peritoneum, and is disseminated within the peritoneal cavity. Once the carcinoma has spread throughout the peritoneal cavity, it is beyond hope of operative cure. & IMPLANTATION It has been postulated that desquamated malignant cells from carcinoma of the colon or rectum may implant on anal wounds after hemorrhoidectomy, fistulectomy, fissurectomy, and the cut ends of bowel. This subject is discussed in Chapter 23. & LYMPHATIC SPREAD Miles (1) believed that the draining lymphatic pathways of carcinoma of the rectum followed three directions: upward along the superior rectal glands, laterally along the middle rectal nodes, and downward to the inguinal nodes. His belief was based on an autopsy study in which most patients died from advanced disease with proximal blockade of the lymph nodes. Later studies revealed that lesions of the upper and middle thirds of the rectum drain upward along the superior rectal vessels; lesions in the lower third drain both upward along the superior mesenteric vessels and laterally along the middle rectal vessels and frequently along the internal iliac glands (2). More recent studies, employing lymphoscintigraphy, have demonstrated lymphatic drainage proximally along the

inferior mesenteric vessels (3,4). Retrograde drainage of carcinoma is unusual unless the proximal lymph nodes are blocked or the carcinoma is poorly differentiated (5). Spread to inguinal nodes generally does not occur unless the carcinoma has invaded the dentate line.

Retrograde Intramural Metastases With the present trend of performing a low anterior resection for carcinoma of the middle and upper rectum, the surgeon must select a site that compromises neither the adequacy of the resection nor the ability to perform a sphincter-saving procedure. It therefore becomes necessary to define an adequate distal margin. A comprehensive prospective study of this subject was conducted by Quer et al. (6). Examinations included 91 specimens obtained directly from the operating room. To obviate shrinkage, the specimens were stretched to conform to the preoperative or operative length measurements and were then immediately fixed for 48 hours in 10% formalin. If the two patients who had clinically obvious metastatic deposits were excluded, only one of 89 patients had retrograde spread beyond 1.5 cm, and 86 had no retrograde intramural spread. To decrease the chance of error, they recommended a margin of 2.5 cm below the lowest grossly palpable or visible edge of the carcinoma. This margin is adequate from the standpoint of retrograde intramural spread when the lesions were of low-grade malignancy. For high-grade malignancy, they advised against performing low anterior resections unless a margin of 6 cm or more could be obtained. Grinnell (7) carried out a similar study. Of 76 patients with carcinoma of the rectum who had resection for cure, 67 patients had no retrograde intramural spread. Of the nine patients with retrograde spread, only three had spread beyond 1 cm from the lower margin of the carcinoma. Two had a poorly differentiated carcinoma, whereas the other one had retrograde subserosal venous spread. More recent studies have confirmed the limited nature of distal intramural spread in that it is within 2 cm in 95% of cases (8–11). It has been reported that even poorly differentiated carcinomas derive no significant benefit from a longer distal margin (12). Retrograde Extramural Metastases Spread of malignant cells in lymphatics is normally embolic from the primary lesion to regional nodes and then from node to node, with occasional bypassing of a node or a group of nodes. However, if lymph node metastases have developed, lymph flow may become blocked and be forced to seek alternate routes. As lymph pressure increases, lymphatic valves become incompetent, and retrograde flow and metastases may occur. Metastases retrograde to pararectal nodes occurred in only 1.6% of 309 cleared specimens of rectal carcinoma. All appeared to be the result of proximal lymphatic blockage from metastases (13). & VENOUS SPREAD The development of distant metastases from primary carcinoma can result only from dissemination of malignant cells into the bloodstream. Circulating malignant cells are found infrequently in the peripheral blood of patients with carcinoma of the colon and rectum. However, the incidence rises

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

during induction of anesthesia (28%), and such cells have been identified in the peripheral blood of 50% of patients during the operation (14). Surprisingly, follow-up studies of patients who had malignant cells in the peripheral blood during the operation have not shown any adverse effect on the ultimate prognosis (14). Unfortunately, little is known about the factors that favor the implantation of circulating malignant cells and subsequent metastases. The incidence of vascular spread correlates with the depth of invasion and histologic grade (15). The most common site of distant metastases for carcinoma of the rectum is the liver, followed by the lungs. The significance of venous involvement is described in Chapter 23.

& CLINICAL FEATURES & SYMPTOMS Many carcinomas of the rectum produce no symptoms initially and are discovered only as part of a routine proctosigmoidoscopy. Bleeding is the most common symptom of rectal carcinoma and is all too often incorrectly attributed by the patients to hemorrhoids. Profuse bleeding is unusual, and anemia is found only in the late stage. Occasionally considerable mucus may appear in the stool. As the disease progresses, the patient will notice a change in caliber of the stool because of partial obstruction. Complete obstruction is rare with carcinoma of the rectum because of the organ’s large lumen and frequent sloughing of the lesion. If located low in the rectum, a carcinoma may cause a feeling of incomplete evacuation after a bowel movement or an urge to defecate, which is also known as tenesmus. Mild abdominal symptoms such as bloating or cramps may occur. Severe rectal or low back pain occurs only when local fixation is extensive and the major nerve trunks are involved by pressure or by invasion. If the carcinoma has invaded the bladder, signs of cystitis or a fistula may become apparent. In the report of the commission on cancer (16) in which 5696 patients with carcinoma of the rectum were diagnosed, the presenting symptoms in order of frequency were rectal bleeding (60.4%), change in bowel habits (43.3%), occult bleeding (25.8%), abdominal pain (20.9%), malaise (9.1%), bowel obstruction (9%), pelvic pain (5%), emergency presentation (3.4%), and jaundice (0.8%). & GENERAL AND ABDOMINAL EXAMINATION The general physical examination is essential for determining the extent of local disease, disclosing distant metastases, and appraising the operative risk of the patient with regard to nutritional, cardiovascular, pulmonary, and renal status. Particular attention should be directed to the liver, inguinal and supraclavicular lymph nodes, and the presence of jaundice. The development of inguinal lymph node metastases augurs poorly for the patient as median survival following discovery is approximately one year if the diagnosis is made at the same time or within one year of the primary and about 16 months if made more than one year after the primary lesion (17). Digital Rectal Examination Despite the sophisticated imaging technology available, useful and important information can be obtained from

209

the digital rectal examination. A hard protuberant mass with a central ulceration is characteristic of a rectal carcinoma. The lesion may occupy a varying degree of the circumference of the rectum or may even be annular. Evaluation of the carcinoma should be made with respect to location and extent, mobility or fixity, size and gross configuration, and involvement in relation to extension to adjacent viscera or fixation to the sacrum. These are all helpful in establishing the true nature of the problem and hence direct appropriate therapy. The all-important digital rectal examination is such an age-old examination that studies have not been considered necessary. One exception is the report by Nicholls et al. (18) who assessed the depth and invasion of rectal carcinomas prior to operation and found colorectal specialists had an accuracy of 67% to 83% whereas the accuracy for trainees for 44% to 78% thus highlighting the need for experience. Others have reported similar accuracy for T-staging (67–84%) (19,20).

& INVESTIGATIONS & ENDOSCOPY The crucial nature of the endoscopic assessment, including sigmoidoscopy and colonoscopy, is fully explained in Chapter 23. This is especially true for rigid sigmoidoscopy, during which it is of utmost importance to carefully record the lesion’s size, lower margin distance from the anal verge, and gross appearance. The flexible sigmoidoscope does not provide the examiner with an accurate measurement of the distance from the distal margin of the carcinoma to the anal verge. Multiple biopsies always should be done to confirm the diagnosis and establish the grade of the carcinoma. All patients with carcinoma of the rectum should have the remaining colon assessed for the presence of synchronous neoplastic lesions. The preferred method of examination is colonoscopy performed preoperatively if possible or at least at some reasonable time within the first year of operation. & ROUTINE LABORATORY BLOOD WORK A complete blood count may reveal the presence of anemia. Abnormal liver function tests may suggest the presence of metastatic disease. A carcino-embryonic antigen determination is obtained as a baseline against which subsequent values may be compared. In addition, urinalysis, coagulation studies, and renal function tests are obtained. & RADIOLOGY Chest X-Ray Examination A chest X-ray examination is required for all patients with carcinoma of the rectum. It is the most practical and noninvasive method to detect pulmonary metastases. It is also a rough guide to evaluate pulmonary status. Barium Enema For patients in whom a total colonoscopy has not been performed or is not available, a barium enema, preferably an air-contrast study, should be obtained to rule out a synchronous carcinoma or polyp unless a high-grade obstruction is present. In a review of 545 patients with carcinoma of the rectum, of the 118 patients who underwent

210

& PART I: COLORECTAL DISORDERS

barium enema, synchronous carcinomas were detected in 9.3% and polyps in 14.7% (21). The investigators noted that 29% of the synchronous carcinomas and 11% of the polyps were beyond the reach of the flexible sigmoidoscope.

Intravenous Pyelography and Cystoscopy An intravenous pyelogram (IVP) is still advised by some surgeons in patients with carcinoma of the rectum to outline the anatomy of the ureters, detect anomalies, evaluate renal function, and reveal obstructive uropathy. Cystoscopy is indicated when bladder symptoms suggest invasion by the rectal carcinoma or when the IVP suggests bladder involvement. Many surgeons do not obtain an IVP unless the lesion is large and fixed or the computed tomography (CT) scan demonstrates invasion into adjacent organs. Imaging Techniques The role of the various imaging techniques is discussed in Chapters 3 and 23. Of specific relevance to carcinoma of the rectum is the use of the CT scan in assessing the extent of pelvic disease. A recent review of the literature cited a meta analysis of 78 studies including 4897 patients that found an accuracy for T staging of 73% (22). Cance et al. (23) reported that a negative CT scan will fail to detect 10% of patients with small liver metastases or positive periaortic lymph nodes.

Lupo et al. (24) compared water enema CT in 57 patients with standard CT in 64 patients and found the former to be more accurate in the preoperative staging of rectal carcinoma (84.2% vs. 62.5%). The diagnostic gain was mainly evident in the identification of rectal wall invasion within or beyond the muscle layer (94.7% vs. 61%). Tada and Endo (25) used abdominal ultrasonography to detect lateral pelvic lymph node involvement and reported an accuracy rate of 92.5%. Ultrasonographic images of the normal colon and rectum identify five distinct layers: (i) mucosa, (ii) mucosa and muscularis mucosae, (iii) submucosa plus interface between submucosa and muscularis propria, (iv) muscularis propria minus the interface between the submucosa and muscularis propria, and (v) serosa and perirectal fascia (26). The first, third, and fifth layers are hyperechoic, whereas the second and fourth layers are hypo-echoic (Fig. 1). A Tl lesion is confined to the first three layers, a T2 lesion infiltrates the fourth layer, a T3 lesion involves all layers into perirectal tissues, and a T4 lesion extends into adjacent organs (Fig. 2). The development of intrarectal ultrasonography has introduced a new dimension in the treatment of patients with rectal carcinoma. This diagnostic modality has proven to be a useful tool in the preoperative assessment of patients who might be candidates for local forms of therapy. The layers of the rectal wall can be identified and

FIGURE 1 & (A) Five ultrasonographic layers of rectum. Layers of rectal wall are seen as rings, which represent interfaces between tissues of different density. Critical ring is outer dark ring, which represents muscularis propria. Penetration of this ring by carcinoma denotes Dukes’ stage B. (B) Carcinoma of rectum. Outer dark interface (arrow) is intact adjacent to thicker dark layer that represents carcinoma. (C) Carcinoma of rectum with regional lymph node involvement. White arrow marks outer dark interface that disappears into adjacent thickening of carcinoma. Irregular surface of carcinoma where muscularis propria ring is not discernible denotes full bowel wall penetration. Black arrow points out large lymph node. Source: Courtesy of Lee E. Smith, M.D., Washington, D.C., U.S.A.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

211

FIGURE 2 & (A) Tl carcinoma. Thickening of mucosa and submucosa layer. (B) T2 carcinoma. Expansion of mucosa and submucosa into muscularis propria. (C) T3 carcinoma. Extension through full thickness of rectal wall. Source: Courtesy of Michael Hill, M.D., George Washington University, Washington, D.C., U.S.A.

the depth of penetration determined. The status of lymph nodes also can be assessed. Marohn (27) reviewed the literature on the subject of endorectal ultrasonographic staging for rectal carcinoma. From 26 reports, which included 1966 patients, the average accuracy for staging was 88% (range, 75–96%). Sensitivity ranged from 55% to 100%, specificity from 24% to 100%, positive predictive value from 55% to 100%, negative predictive value from 85% to 100%, percent overstaging from 10% to 40%, and percent understaging from 1% to 30%. Values tend to vary with depth of penetration. Garcia-Aguilar et al. (28) evaluated the accuracy of endorectal ultrasonography in preoperative staging of 1184 patients with rectal carcinoma or villous adenoma. They compared the endorectal ultrasonographic staging with the pathology findings based on the surgical specimens in 545 patients who had surgery (307 by transanal excision, 238 by radical proctectomy) without adjuvant preoperative chemoradiation. Overall accuracy in assessing the level of rectal wall invasion was 69%, with 18% of the neoplasms overstaged and 13% understaged. Accuracy depended on the stage and on the ultrasonographer. Overall accuracy in assessing nodal involvement in the 238 patients treated with radical surgery was 64%, with 25% overstaged and 11% understaged. In this study, the accuracy of endorectal ultrasonography in assessing the depth of invasion, particularly for early carcinomas was lower than previously reported. Differences in imaging interpretation may in part explain the discrepancies in accuracy between studies.

Kauer et al. (29) determined the accuracy of endosonography in the local staging of 458 rectal carcinomas. The overall rate for correctly classified patients was 69% with respect to the T category and 68% with respect to the N category. There was no difference between the 7.5-MHz and the 10-MHz scanners. In terms of accuracy, the T3 category carcinomas were the most (86%) and the T4 carcinomas the least (36%) accurately classified. Overstaging of carcinomas (19%) was much more frequent than understaging (12%). A high interobserver variability of 61% to 77% was noted. For pT1 carcinomas, the 10-MHz scanner was almost two times more accurate than the 7.5-MHz scanner (71% vs. 36%). The accuracy of endosonographic staging of rectal carcinoma very much depends on the T category. A review of this subject found almost identical values with accuracy for T-staging varying between 69% and 97% (22). The latter review found higher accuracy for superficial lesions but more limited for deeply penetrating carcinomas. Yet, another review of 16 publications cited similar data with an overall accuracy rate of 83% for T-stage and 75% for N-stage (30). Approximately 14% of rectal carcinomas cannot be staged accurately because stenosis prevents insertion of the ultrasound probe (30). Endorectal ultrasonography’s ability to detect metastatic lymph nodes lags behind its ability to assess depth of invasion. Lymph node status can be predicted with an accuracy ranging from 73% to 85% (31–35). It is feasible and safe to perform biopsy of pararectal lymph nodes under direct ultrasonographic guidance. Ultrasonography-directed

212

& PART I: COLORECTAL DISORDERS

lymph node biopsy has a sensitivity of 71%, specificity of 89%, a positive predictive value of 92%, and a negative predictive value of 62%. When endorectal ultrasonography staging is incorrect for depth of penetration, it is usually because of overstaging rather than understaging. Overstaging is presumably due to the related inflammatory cell infiltration around the carcinoma, which can have the same ultrasonographic characteristic as malignancy. Other potential explanations include an inflammatory response to a previous biopsy or compression of the rectal wall by the water-filled balloon leading to obliteration of the five-layer echo structure. It may also occur when a carcinoma is located on a haustral fold or sharp angulation that can result in tangential imaging. Overstaging most commonly occurs with T2 lesions that appear as T3 lesion (30). Understaging can be caused by microscopic infiltration, which is not detectable by the resolution obtained by the current ultrasonography instruments. Preoperative radiation therapy can influence the accuracy of staging by causing thickening of the rectal wall with loss of its five sonographic layers secondary to inflammation and/or fibrosis. Another variable is operator experience. Williamson et al. (36) assessed the ability and accuracy of endorectal ultrasonography to predict changes in the stage of rectal carcinoma after preoperative chemoradiation. Of 15 patients studied, sonographic evidence of level of invasion and nodal status were each downstaged in 38% of patients. Pathologic evaluation revealed that the level of invasion was downstaged in 47% and nodal status in 88% compared with the initial endorectal ultrasonography staging. The authors concluded that although the method demonstrated shrinkage, it did not closely predict pathologic results. Bernini et al. (37) found downstaging in 53% with wall invasion and in 72% with lymph node involvement. The authors found endorectal ultrasonography of limited value in this setting because of overstaging. Fleshman et al. (38) reported that preoperative radiation therapy makes transrectal ultrasonography and CT scan less effective as staging techniques. A recent review suggested that the T-stage accuracy after radiation is 50% with a 40% overstaging rate (30). Lymph node staging accuracy is also decreased. The introduction of magnetic resonance imaging (MRI) providing a high contrast soft tissue resolution without the need for imaging radiation was promising. However, in a recent review, the overall accuracy reported for body coil MRI ranges from 59% to 88% and is not better than CT (22). With the introduction of endoluminal coils reported T-stage accuracies varied from 71% to 91% (22). Further improvement was expected with the phased array coil but most reports record T-stage accuracy rates from 65% to 86% (22). The latter technique may be accurate and reliable for the prediction of circumferential resection margin with accuracy rates reported as high as 93% to 97% (22). Preoperative MRI may be of prognostic value in rectal carcinoma and may be helpful in selecting patients for neoadjuvant radio/chemotherapy. This idea is supported by the study of Martling et al. (39) who found five-year overall survival to be 77% when radial margins were not involved but only 43% when they were involved. Bisset et al. (40) reported the MRI detected penetration of carcinoma through the fascia propria with a sensitivity of

67%, specificity of 100%, and accuracy of 95%. This method of assessment may offer a new way to select those patients who require preoperative neoadjuvant therapy. A number of comparative studies of different imaging techniques have been conducted. Leite et al. (41) compared water enema CT and transrectal ultrasonography in 40 patients with rectal carcinoma. The authors believe the former improves local staging without the difficulties associated with transrectal ultrasonography as in cases with stenoses and lesions located at a high level. Endorectal ultrasonography has been used in the evaluation of recurrence. CT is reported to be more accurate than intrarectal ultrasonography in the detection of recurrent carcinoma (42). Tempero et al. (26) collated four series comprising 258 patients in which endorectal ultrasonography was compared to CT. Endorectal ultrasonography was superior in T stage (82% vs. 68%) and in N stage (83% vs. 66%). With respect to lymph node evaluation: (i) the likelihood of lymph node metastases is very low if lymph nodes are not identified, (ii) hyperechoic lymph nodes suggest nonspecific inflammatory changes, (iii) hypoechoic lymph nodes suggest metastases, and (iv) lymph nodes with mixed echoic patterns cannot be accurately classified and should be considered malignant. Overall accuracy of lymph node assessment in seven combined series comprising 566 patients was 80% (range, 70–83%) (26). A more recent review of comparative studies supported these findings of superiority of US over CT with accuracy of 87% versus 76% for T stage, and 78% versus 62% for N-stage (30). Nevertheless, CT scanning provides excellent definition of distant and locoregional spread. Meyenberger et al. (43) compared endoscopic ultrasonography to MRI with an endorectal coil in 21 patients. The results of preoperative examination were compared to histopathologic findings with a special focus on transmural invasion. Endoscopic ultrasonography identified all carcinomas, whereas MRI missed one carcinoma. Endoscopic ultrasonography was superior to MRI (accuracy, 83% vs. 40%). The MRI could not differentiate between T1 and T2 lesions. The accuracy of MRI in assessing perirectal infiltration was 80% compared to 100% with ultrasonography. Local recurrence was found in six patients all detected by endorectal ultrasonography and one missed by the MRI. In this study, endoscopic ultrasonography was more useful. MRI was found to be more operator-dependent (44). A more recent review suggested similar accuracy for MRI and US for staging rectal carcinoma (30). Katsura et al. (45) using a radial scanner for endo-rectal ultrasonography in 120 patients with rectal carcinoma, found an accuracy rate of 92% in assessing wall penetration. No swollen lymph nodes were found in 35 of 98 cases but metastatic disease was found in five of these 35 cases (14.3%). Metastatic disease was observed more frequently in lymph nodes with a diameter greater than 5 mm (53.8%) and with an uneven and markedly hypoechoic pattern (72.3%). In a study of 20 patients with rectal carcinoma, McNicholas et al. (46) evaluated MRI and found it accurate in all but one case with respect to transmural invasion and it diagnosed metastatic deposits in lymph nodes in 12 patients. It overstaged one patient with enlarged lymph nodes but the nodes were found to be histologically negative.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

Kim et al. (47) studied the accuracy rates and clinical usefulness of MRI in preoperative staging of rectal carcinoma in 217 patients with histologically proven rectal carcinoma who had surgical resections performed. The MRI criteria for determining the depth of invasion was the degree of disruption of the rectal wall. Metastatic perirectal lymph nodes were considered to be present if they showed heterogeneous texture, irregular margin, and enlargement (greater than 10 mm). The accuracy of the MRI for determining depth of invasion was 81% and regional lymph node invasion was 63%. In the T-stage, accuracy rate of T1 was 75%, T2 was 54%, T3 was 87%, and T4 was 86%. The specificity of lymph node invasion was 41% and the sensitivity was 85%. The accuracy rate of regional lymph node involvement was 63%. T1 and T2 were overstaged in 25% and 46%, respectively, and T3 was understaged in 9.2%. The accuracy rate to detect metastatic lateral pelvic lymph node was 29% after lateral pelvic lymph node dissection was done. The accuracy rate in assessing levator ani malignant involvement was 72%. The accuracy of three dimensional endorectal ultrasound and endorectal MRI in the assessment of the infiltration depth of rectal carcinoma is comparable to conventional endorectal ultrasound (44). Brown et al. (48) determined the accuracy of preoperative MRI in the evaluation of pathological prognostic factors that influence local recurrence and survival in 98 patients undergoing total mesorectal excision (TME) for biopsy-proven rectal carcinoma. There was 94% agreement between MRI and pathology assessment of T-stage. Agreement between MRI and histological assessment of nodal status was 85%. Although involvement of small veins by carcinoma was not discernible using MRI, large (caliber greater than 3 mm) extramural venous invasion was identified correctly in 15 of 18 patients. MRI predicted circumferential resection margin involvement with 92% agreement. Seven of nine patients with peritoneal perforation of the carcinoma (stage T4) were identified correctly using MRI. It may allow both better selection and assessment of patients undergoing preoperative therapy. Branagan et al. (49) assessed whether preoperative MRI scans were able to predict (i) pathologic carcinoma and node stage, and (ii) those patients with pathologically clear circumferential resection margin. From a total of 40 patients, MRI correctly staged the carcinoma in 20 patients, understaged in 12, and overstaged in eight. Statistically, there was poor correlation between pathologic and radiologic staging of the carcinoma. MRI correctly staged node status in 27 patients, overstaged in nine, and understaged in four. Statistically, there was poor correlation between pathologic and radiologic node staging. MRI correctly reported the status of circumferential resection margin in 39 patients and understaged one patient. Statistically, there was good correlation between pathologic and radiologic reporting the circumferential resection margin involvement. In a comparative study, Thaler et al. (50) found preoperative endoluminal ultrasonography to be accurate in 60% and MRI in 48% for preoperative staging of rectal carcinoma, but the differences were not significant. Others have found an MRI sensitivity of 84% and specificity of 93% in deter mining extramural extension (51). Billingham (52) reviewed seven series from the literature that encompassed 306

213

patients and found the accuracy of CT and MRI for detecting depth of invasion to be 76% and 83%, respectively. Staging of colonic neoplasms by endoscopic ultrasonography has not gained importance because to date preoperative staging is without any clinical consequences. This may change with the introduction of minimally invasive surgical procedures and endoscopic resection techniques as an alternative to conventional open operation. Stergiou et al. (53) performed endorectal ultrasound with a miniprobe in 54 consecutive patients with colonic neoplasms who had been referred for endoscopic resection or for laparoscopic resection of their lesions. A sufficient endorectal ultrasound evaluation of the colonic lesion was possible in 93% of patients. The infiltration depth was correctly classified in 17 adenomas, 16 T1, 8 T2, 5 T3, and 1 T4 carcinoma (accuracy for T-staging 94%). Two T2 and one T3 carcinomas were overstaged by endorectal ultrasound while no understaging was recorded. The lymph node status was correctly classified in 84% of patients and a false-negative lymph node status was found in 8%. The overall accuracy of endorectal ultrasound was 80%. Miniprobe endorectal ultrasound is not optimal accuracy for staging of colonic neoplasms but with refinement may find a role in the future. Heriot et al. (54) prospectively assessed the impact of PET scanning on the management of 46 primary rectal carcinomas. The operative management of 78% was unchanged as a result of the PET scan even though PET scan upstaged disease in 8% and downstaged disease in 14%. In 17%, management was altered because of the PET scan findings including 13% in which operation was cancelled and 4% in which the radiotherapy field was changed. Where available, follow-up confirmed the appropriateness of the PET scan induced management change in each case. Overall stage of disease was changed following PET scanning in 39% of patients. In view of this, they suggest that PET scanning be considered part of the standard investigation for such patients.

& PREOPERATIVE PREPARATION With growing fiscal constraints, same-day admission has come into vogue and indeed is being mandated by a growing number of health centers and insuring agencies unless the patient suffers from severe comorbid disease— in which case patients should be admitted to the hospital in sufficient time to allow measures designed to establish the optimum physical condition. In addition to laboratory studies, an electrocardiogram, especially in patients over 40 years old, always should be obtained and pulmonary function tests performed if indicated. A vigorous mechanical and antibiotic bowel preparation such as one of those described in Chapters 4 and 23 should be undertaken. For otherwise healthy patients, outpatient preparation has become increasingly utilized with patients arriving at the hospital on the morning of operation. Potential complications such as impotence in males should be discussed with patients preoperatively. If abdominoperineal resection is anticipated, patients ideally should be visited by an enterostomal therapist who shall

214

& PART I: COLORECTAL DISORDERS

discuss and give information regarding life with a colostomy. Successful stomal management usually begins with preoperative education, and this greatly facilitates adjustment to the colostomy, which most patients find difficult. The ideal site for the colostomy is marked preoperatively.

& RADICAL EXTIRPATIVE OPERATIONS & ASSESSMENT OF RESECTABILITY Cases classified as operable are those in which the surgeon considers the conditions favorable for removal of the disease by operation. Inoperability is determined either by an unsatisfactory general condition that renders the patient unsuitable for operation or by an advanced disease state that is beyond hope for cure. In most cases it is only at exploratory laparotomy that an accurate assessment of the resectability of a rectal carcinoma can be made. Clinical examination alone gives no reliable information as to venous or lymphatic spread unless the spread is extensive or obvious distant metastases are present. Fixation of the primary growth as determined by rectal examination is unreliable in estimating the resectability. Preoperative CT scanning of the pelvis adds further information, but the ultimate decision is made at the time of laparotomy. The decision to resect is determined by the degree of fixation of the growth, the presence and extent of hepatic metastases, and the presence and extent of other metastases or peritoneal seedings. Although fixation of a rectal carcinoma to an adjacent organ or the pelvis does not necessarily indicate contiguous spread, such appears to be the case more often with rectal carcinoma than with colon carcinoma. In a study of 625 patients who had undergone rectal excision, Durdey and Williams (55) noted that 27% of the patients had fixation—by malignant invasion (20%) and by inflammatory tissue (7%). Inflammatory attachment does not increase the risk of recurrence or decrease survival. Even for an incurable situation, palliative resection may relieve the patient of symptoms such as bleeding, tenesmus, and obstruction. In an unresectable case, the patient may benefit by an operation in which an obstructing lesion may be ‘‘bypassed’’ with a ‘‘hidden’’ colostomy (56). & SELECTION OF APPROPRIATE OPERATION A host of factors influence the choice of operation for a given patient. Probably the most important factor is the level of the lesion. For practical purposes the rectum is divided into thirds (Fig. 3). The lower third extends from the anorectal ring (3–4 cm from the anal verge) to 7 cm from the anal verge, the middle third is 7 to 11 cm from the anal verge, and the upper third is 11 to 15 cm from the anal verge. With the present knowledge of the usual lymphatic pathways at different levels of the rectum, it is generally accepted that low anterior resection is the treatment of choice for carcinoma of the upper third of the rectum. Abdominoperineal resection is the treatment of choice for many patients with carcinoma of the lower third of the rectum. However, with growing expertise, surgeons have become increasingly confident in offering an extended low anterior resection to a select number of patients whose

FIGURE 3 & Three levels of rectum.

carcinoma lie in the lower third of the rectum. The move toward sphincter-preserving operations began when technical expertise improved and anorectal physiology studies demonstrated that the distal 1 to 2 cm of rectum and upper internal anal sphincter were not absolutely necessary for continence. It has been suggested that it is now technically possible to remove rectal carcinoma that is extending into the anal canal with preservation of the anal sphincter mechanism and with a satisfactory oncologic outcome (57). However, enthusiasm for sphincter saving operations must be tempered by concern over incontinence and recurrence in suboptimally selected patients. Ultra-low colorectal and coloanal anastomosis together with a colonic pouch or coloplasty, produces acceptable function in many patients. There is still controversy about the risk of implantation of malignant cells, the place of downsizing neoadjuvant therapy, and true long-term functional outcome. Despite these concerns, surgeons should strive to perform rectal resection with sphincter preservation for low-lying rectal carcinoma whenever possible. The rationale is the same as noted in the following discussion on middle third lesions. Controversy exists on the best curative resection for carcinoma of the middle third of the rectum. The technical difficulties in performing the low anastomosis and the high morbidity from anastomotic leaks and sepsis are well known. Another concern to the surgeon is what constitutes an adequate distal margin for resection. Studies of operative specimens have shown that a distal margin of 2 cm from the lower border of the carcinoma is adequate to contain both intramural and retrograde lymphatic spread (6,7,13). Indeed, one study showed that three quarters of rectal carcinomas exhibited no intramural spread (58). The exception to this generalization is the poorly differentiated carcinoma; for this lesion a distal margin of at least 6 cm has been suggested (6). The clinical studies by Wilson and Beahrs (59) noted no difference in suture line recurrence, pelvic recurrence, or five-year survival rates whether the distal margin was less than 2 cm or more than

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

5 cm from the lesion. Anastomoses as low as the levator ani muscle have been performed with complete maintenance of anal continence. Other factors influencing the choice between a sphincter-saving operation and abdominoperineal resection for carcinoma of the middle third of the rectum include body build, sex, obesity, lesion level, local spread, perforation or abscess, size and fixation of lesion, histologic grade of carcinoma (particularly undifferentiated carcinoma), presence of obstruction, adequacy of bowel preparation, and general medical condition. However, most patients with carcinoma of the middle third of the rectum can undergo a sphincter-saving operation. From a Nationwide Inpatient Sample database, Purves et al. (60) reported that rectal carcinoma patients treated by high-volume surgeons are five times more likely to undergo sphincter-sparing procedures than those treated by low-volume surgeons. This has significant implications for those seeking a sphincter-preserving option for the treatment of their rectal carcinoma. A comparison of the amount of tissue excised with a low anterior resection and an abdominoperineal resection is best described pictorially (Fig. 4). From the point of view of the proximal extent of dissection, the two operations are identical. The amount of tissue removed laterally is the same in both procedures. With respect to the distal line of resection, a margin as long as a 2 cm can be obtained, and there appears to be no benefit from sacrifice of the sphincter. Distal dissection of the vessels, mesorectum, and lymphatics are otherwise identical. Consequently, over the past 20 years, abdominoperineal resection has increasingly been displaced by low anterior resection (61). Numerous reports have shown that sphincter-saving operations for carcinoma of the middle third of the rectum give comparable morbidity, mortality, and five-year survival

215

rates to abdominOperineal resection (62–68). Although these reports were not controlled trials and are therefore not comparable, they do support the concept that in properly selected cases, a sphincter-saving operation is a good alternative for carcinoma of the middle third of the rectum. Sphincter-saving procedures currently in use include low anterior resection and the pull-through operations. To determine the procedure of choice for rectal carcinoma, particularly low rectal carcinoma, Di Betta et al. (69) completed a search according to evidence based methods of comparative studies and national surveys published since 1990. Comparative studies between abdominoperineal resection and sphincter saving operations with a minimum of 50 patients presenting with carcinoma in the lower onethird of the rectum were analyzed including 6570 patients. Postoperative morbidity after abdominoperineal resection and sphincter saving operation is comparable and postoperative mortality decreased to 2% or less. The type of operation was not identified as a prognostic factor in terms of local disease control and survival. Quality of life is significantly inferior after abdominoperineal resection. National data reveal an abdominoperineal resection rate for carcinoma of the whole rectum (up to 16 cm) at 50% or higher and sphincter saving operations still would represent only 32% of the radical resections for low rectal carcinoma. All available evidence indicates that sphincter saving operations should be the procedure of choice for rectal carcinoma, even in the lower one-third. An abdominoperineal resection should only be performed when the carcinoma invades the anal sphincters and negative resection margins cannot be achieved by a sphincter saving operation. Still other considerations may enter the decision in selecting the appropriate operation. Preexisting anal incontinence may impede against a sphincter-saving operation

FIGURE 4 & Comparison of amount of tissue excised with low anterior resection (A) and with abdominoperineal resection (B). Red lines indicate boundaries of excision.

216

& PART I: COLORECTAL DISORDERS

because rectal resection removes the reservoir function, and both the frequency of bowel movements and incontinence may be increased. An allied consideration is advanced age where decreased continence may prove problematic. However, this should not be an overriding consideration. The presence of extensive pelvic disease, which may ultimately obstruct the bowel, may be a relative contraindication to restoring intestinal continuity. By the same token, patients with disseminated disease may not live long enough to benefit from establishment of intestinal continuity such as a coloanal anastomosis and later closure of an ileostomy. These patients would benefit from a permanent stoma. It should be noted that in the past operations for rectal carcinoma have focused mainly on avoidance of local recurrence and preservation of the anal sphincter. These goals were and still are very important. However, a patient who suffers from stool incontinence, straining and frequent bowel movements after sphincter saving rectal resection might have done better with a stoma. The time has now come to take more account of the postoperative quality of life of the patient and to modify our therapeutic options accordingly.

& OPERATIVE PROCEDURES Low Anterior Resection Precise definitions of extent of resection have varied from author to author. The term ‘‘anterior resection’’ is a recollection from the days when posterior proctectomy was in vogue and the term anterior resection differentiated the abdominal approach from the posterior approach. The operation, which entails removal of the sigmoid colon and anastomosis of the proximal sigmoid or descending colon to the proximal rectum, is probably better-called a sigmoid resection. The term ‘‘low anterior resection’’ is applied when the operation necessitates full mobilization of the rectum and transection of both lateral ligaments. The anastomosis is performed below the anterior peritoneal reflection. The term ‘‘extended low anterior resection’’ refers to a resection in which the anastomosis is constructed at or just above the levator ani muscles. Because of the lateral curves of the rectum, it is possible that a carcinoma at the 7 cm level may be at the 12 cm level after full rectal mobilization. This is especially true with a posterior lesion. The feasibility of low anterior resection is decided on only after the rectum has been completely mobilized posteriorly, anteriorly, and laterally. If low anterior resection is to be done, the distal clearance should be at least 2 cm from the lower margin of the carcinoma. The splenic flexure must be taken down if there is any question of tension at the anastomosis. At the completion of the anastomosis the pelvic peritoneum is left open so that the hollow of the sacrum freely communicates with the peritoneal cavity to allow drainage of the accumulated fluid. Drains are generally not used. Closed suction drainage may be used in selected cases when the pelvis is not completely dry. A proximal ileostomy or colostomy is rarely performed. However, when the integrity of the anastomosis is in question, the surgeon should not hesitate to perform a complementary diverting stoma, either a transverse colostomy or an ileostomy.

FIGURE 5 & Allen stirrups. With simple padding, stirrups minimize risk of pressure on calf and eliminate pressure on peroneal nerve. Thighs are slightly abducted and hips and knees slightly flexed. Source: Courtesy Allen Medical Systems, Bedford Heights, Ohio, U.S.A.

Position The patient is placed supine and in a slight Trendelenburg position for better exposure of pelvic structures and to aid in venous return from the lower extremities. The lower extremities are placed in Yellow Fin or Allen stirrups, with the buttocks slightly elevated and near the edge of the table. With the knees flexed and the hips minimally flexed and abducted simultaneously, access to the abdomen and perineum is obtained (Fig. 5). Care should be taken to avoid excess abduction lest excessive traction be placed on the sciatic nerve. The lower extremities are best placed in a neutral position with only enough abduction to permit access to the perineum. Pressure over the fibular head has the potential to cause peroneal nerve palsy, and undue calf pressure may result in a compartment syndrome (70). Therefore adequate padding should be employed. The second assistant can be positioned between the patient’s legs. A No. 16 Foley catheter is inserted into the bladder, and the urine output is monitored continuously throughout the operation. Incision Depending on the body habitus, both a transverse incision at a level between the pubis and umbilicus and a lower midline incision provide satisfactory exposure (Fig. 6A). If the proposed transverse incision interferes with the ideal location of a possible stoma, a midline incision should be used. A self-retaining retractor aids exposure. Mobilization of Sigmoid Colon After packing the small bowel into the upper abdomen, the sigmoid colon is mobilized by incising the lateral peritoneal reflection (white line of Toldt). The incision is carried cephalad to the distal descending colon and caudad parallel to the rectum until the cul-de-sac is reached. The extent of proximal mobilization depends on the redundancy of the sigmoid colon. If the sigmoid colon is short, the splenic flexure may require mobilization. This can be accomplished conveniently with a cautery or a scissors. The intersigmoid fossa acts as a useful guide to the ureter located just behind it (Fig. 6B). The retroperitoneal areolar tissue is pushed aside with a stick sponge so that a fan-shaped flap of

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

217

FIGURE 6 & Low anterior resection. (A) Incision selection. (B) Intersigmoid fossa is a marker of the underlying ureter. (C) Mobilization of sigmoid colon. Ureter is just medial to left spermatic or left ovarian vein. (D) T-shaped incision from colon toward inferior mesenteric artery and then toward the cul-de-sac. (E to M on next pages)

sigmoid mesentery is created. The left spermatic or left ovarian vein can be identified. At the level of the iliac crest, the ureter is just medial to this vein (Fig. 6C). Next the peritoneum on the right side of the sigmoid and rectum is incised, starting near the sigmoid vessels and continuing toward the origin of the inferior mesenteric artery. A T-shaped incision is made in the peritoneum, again parallel to the rectum, and carried down to the pelvis until the culde-sac is reached (Fig. 6D). Unless invasion or inflammation of the pelvis is encountered, the right ureter is not routinely identified although some surgeons believe it should be. The inferior mesenteric artery is identified and doubly clamped, divided, and doubly ligated just distal to the take-off of the left colic artery (Fig. 6E,F). Before the inferior mesenteric artery is clamped, care should be

taken to ensure that the ureter is out of harm’s way. The inferior mesenteric vein is ligated at the corresponding level. Posterior Mobilization of Rectum By drawing the rectum taut, a plane of areolar tissue behind the rectum at the level just above the promontory of the sacrum can be identified. With sharp dissection, the retrorectal space can usually be entered easily with minimal bleeding. Care must be taken at this point to develop the plane anterior to the presacral nerves. At the level just below the promontory, these nerves bifurcate (Fig. 6G,H). There are four key zones where nerve damage may occur: (i) the inferior mesenteric artery origin (sympathetic

218

& PART I: COLORECTAL DISORDERS

FIGURE 6 & (E) Ligation of inferior mesenteric vessels distal to left colic artery. (F) Ligation of inferior mesenteric vessels at origin. (G) Identification of presacral nerve at level of sacral promontory. (Continued)

hypogastric nerves), (ii) posterior rectal dissection (sympathetic hypogastric nerves), (iii) lateral dissection (mixed sympathetic and parasympathetic), and (iv) anterior dissection (cavernous nerves) (57). Further cautery dissection in a semicircular manner readily mobilizes the rectum until the lateral stalks become

prominent (Fig. 6I). The plane is developed by sharp dissection. At the S3 or S4 level, the rectosacral fascia, which varies from a thin fibrous band to a thick ligament, is encountered. The fascia is cut with a long heavy scissors or electrocautery (Fig. 6J). Failure to do so risks tearing the presacral venous plexus, which may then bleed

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

219

FIGURE 6 & (H) Presacral nerves (arrows) passing over sacral promontory into the pelvis. (I) At level of sacral promontory, presacral space is entered and developed. Cautery mobilization of posterior rectal wall and development of lateral stalks are done. (J) Division of rectosacral fascia by cautery or scissors. (K) Anterior mobilization of rectum by cautery or scissors. (L) Division of lateral stalks. (M) Right-angle bowel clamp placed distal to carcinoma.

220

& PART I: COLORECTAL DISORDERS

profusely and can be difficult to control. Once this fascia is cut, the level of the coccyx is reached. Prior to posterior mobilization, some surgeons have found it easier to transect the proximal colon initially and reflect it anteriorly to facilitate exposure of the mesorectum. The use of a lighted pelvic retractor may dramatically facilitate pelvic exposure both posteriorly and anteriorly. Although massive presacral bleeding during rectal mobilization is uncommon, it can rapidly destabilize a patient. Harrison et al. (71) described the anatomic basis of injury and summarized their experience with muscle fragment welding for control of massive presacral bleeding during rectal mobilization. Rectus abdominis muscle welding for the control of presacral bleeding was first described by Xu and Lin (72) in 1994. Harrison et al. (71) found this method to be an easy, rapid, extremely effective means of controlling a potentially life-threatening complication. The technique involves direct pressure held over the bleeding site with a finger or sponge stick while the patient is resuscitated. Once the patient is hemodynamically stable, a 1.5 to 2 cm2 segment of rectus abdominis muscle is harvested from the incision and held in place with a forceps over the bleeding area while vigorous suctioning is implemented to expose the presacral operative field. Electrocautery at a high setting (100 Hz) is then applied to the forceps and transmitted to the muscle fragment, literally welding the bleeding site (Fig. 7). The muscle fragment or coagulum may actually fall free from the site, but the source of bleeding is welded closed. Advantages of the muscle welding method include its efficacy (all eight patients in whom it was used by Harrison et al. (71) had prompt cessation of their hemorrhage); applicability with multiple bleeding sites; ready availability of the equipment required is already on the operative field; and the lack of foreign bodies left in the patient that require removal with the risk of secondary hemorrhage or anastomotic disruption. The anatomic basis for bleeding is the sacral-basivertebral veins that traverse the body of the sacrum to connect the internal vertebral venous system with the presacral venous plexus in at least 16% of patients (73). Because the adventitia of these veins

FIGURE 7 & Muscle welding with patch of rectus muscle on the bleeding site.

blends with the sacral periosteum at the edge of the foramen, lifting or disrupting the presacral fascia during pelvic dissection can lacerate the veins, causing the ruptured ends to retract into the sacral foramina (73). The presacral venous plexus is composed of contributions from the two lateral sacral veins and the middle sacral vein (Fig. 8). Because this entire pelvic venous system lacks valves, it has been estimated that with the patient in the lithotomy position, the hydrostatic pressure in the presacral plexus is up the three times that of the inferior vena cava. Bleeding from small vessels in this area can be torrential and extremely difficult to control. Surgeons have traditionally relied upon direct pressure or electro-coagulation to contain the hemorrhage. Nivatvongs and Fang (74) have described the use of a titanium thumbtack for bleeding arising from bone. In very rare circumstances, prolonged packing for several days can be employed (75,76). Another described procedure involves the use of cyanoacrylate adhesive-coated hemostatic gelatin sponge placed over the bleeding vessel. Anterior Mobilization of Rectum In men the peritoneum at the rectovesical reflection is incised; mobilization is continued in the plane between

FIGURE 8 & Presacral venous plexus with communication with basivertebral veins.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

the seminal vesicles and Denonvilliers’ fascia. For posterior lesions, the plane posterior to Denonvilliers’ fascia is pursued to diminish the risk of nerve damage. Most parasympathetic nerve damage probably occurs with anterior dissection. The cavernous nerves lie anteriorly to Denonvilliers’ fascia, at the posterolateral border of the apex and base of the prostate. At this point, they lie closely related to the rectum. Controversy exists about the best anterior plane of excision; some recommend taking Denonvilliers’ fascia whereas others believe it important to stay posterior to this fascia. The actual dissection performed should probably be tailored to the specific surgical situation, with emphasis on the position of the carcinoma (57). Dissection is extended distally until the rectum is separated from the seminal vesicles and is continued distal to the prostate (Fig. 6K). Bleeding is controlled with electrocoagulation. Junginger et al. (77) conducted a study to determine the frequency of identification and preservation of the pelvic autonomic nerves and to identify a possible link between postoperative micturition disturbances and the extent of the radical resection in 150 patients with carcinoma of the rectum. The pelvis autonomic nerves were identified completely in 72%, partially in 10.7%, and not at all in 17.3%. Multivariate analysis showed gender, T stage, blood loss, curative operation, previous operation, learning curve, and depth of penetration of the rectal wall (T1/T2 vs. T3/T4) exerted an independent influence on achievement of complete pelvic nerve identification. Discharge from hospital with a urinary catheter occurred in 10.7% of patients. Identification and preservation of pelvic autonomic nerves was associated with low bladder dysfunction rates (4.5% vs. 38.5%). In women the rectovaginal reflection is incised and the plane is developed between the rectum and the vagina until the pubis can be felt anteriorly. Exposure may be facilitated by passage of a suture around each fallopian tube, with elevation and retraction of the uterus anteriorly by attaching the sutures to the self-retaining retractor. Dividing Lateral Ligaments (Stalks) At this stage the rectum has been mobilized posteriorly to the tip of the coccyx and anteriorly to the level of the pubis. The rectum remains attached laterally on each side by the pelvic fascia or lateral ligaments (containing the accessory middle rectal vessels). The rectum is pulled taut with the left hand, and the right hand is placed behind the rectum and swept laterally on each side of the ligament, with care taken to avoid the ureters. The ligaments then can be divided by electrocautery. Alternatively, the ligaments are clamped, divided, and ligated (Fig. 6L). Based on the recommendation by Heald et al. (78), an effort is made to excise the mesorectum rather than following the tendency to cone down through the mesorectum to the proposed distal line of resection. The latter technique leaves mesorectum in the pelvis, and Heald cautions that this is the main reason for local recurrence (78,79). For lesions located higher in the rectum, the distal portion of mesorectum may be left undisturbed but given an adequate distal margin; transection of the mesorectum is performed perpendicular to the rectal wall and not in a cone fashion. When adequate mobilization has been achieved, a right-

221

angle bowel clamp is placed across the rectum distal to the carcinoma (Fig. 6M). At this point some surgeons irrigate the rectal stump with a cancericidal agent. Greater emphasis had been placed on this measure by European surgeons than by North American surgeons. This maneuver may disqualify patients who are otherwise eligible for clinical trials of adjuvant therapy. Agaba (80) evaluated the effectiveness of cytocidal rectal washout in reducing the incidence of local recurrence. Of 141 patients who underwent curative anterior resection for carcinoma of the rectum and rectosigmoid, 90 patients underwent rectal washout using cetrimide before anastomosis while 51 patients did not have rectal washout before anastomosis. Among the washout group, the local recurrence rate was 4.4% compared with 5.9% among the no washout group. Because of the size of this study they were unable to demonstrate the benefit or lack thereof of this cytocidal agent in reducing local recurrence. Inadvertent Rectal Perforation An embarrassing but nevertheless real case scenario is the occasional inadvertent rectal perforation during the mobilization of a carcinoma of the rectum. It has been reported that the spillage of malignant cells in this situation results in a higher incidence of local recurrence (81). The question arises as to ‘‘what exactly should the surgeon do in such situations?’’ Several suggestions have been made. Irrigation with sterile water has been recommended with the hope that the hypo-osmolar nature of this solution will lyse any released malignant cells. Irrigation with povi-done-iodine (Betadine) has been used, also in the hope of killing malignant cells. If radiation has not been given preoperatively, a course of postoperative radiotherapy should be considered. Anastomosis with Circular Stapler The introduction of staplers into operative use in North America has markedly facilitated intestinal anastomoses. The advent of the circular stapler has extended the limits of low anterior resection by enabling surgeons to perform highly reliable anastomoses at a lower level than was technically possible with a traditional hand-sewn anastomosis, thus sparing a considerable number of patients from abdominoperineal resection and a permanent colostomy. The general principles of anastomoses must be maintained; generally, tissues not fit to sew should not be stapled. Conditions that are unfavorable for a hand-sewn anastomosis are also unfavorable for staples. Staples are only one method, albeit a convenient one, to establish intestinal continuity. Considerations of adequate blood supply, absence of tension, accurate apposition of tissue, and absence of sepsis apply equally to both stapled and hand-sutured anastomoses. A variety of staplers are commercially available (Fig. 9). The technique and pitfalls with the use of the circular stapler have been described previously in detail (82). Having determined that a low anterior resection with a circular stapler is feasible, the surgeon prepares the proximal bowel by clearing 1 to 1.5 cm from the proposed proximal resection margin. A pursestring suture may be applied using the specially designed fenestrated clamp (Fig. 10A). The

222

& PART I: COLORECTAL DISORDERS

FIGURE 9 & Circular staplers. (A) United States Surgical disposable EEA stapler. (B) Curved CEEA staplers. (C) Ethicon ILS disposable staplers. (D) Circular staplers which connect to a long flexible drivenshaft: 21, 25, 29, and 33mm size. (E) Transanal 29mm circular stapler with flexible wire trocar. Source: (A) and (B) Courtesy of United States Surgical Corp., Norwalk, Connecticut, U.S.A. (C) Courtesy of Ethicon, Inc., Somerville, New Jersey, U.S.A. (D and E) Courtesy of Power Medical Interventions, Langhorne, Pennsylvania, U.S.A.

anvil is detached from the instrument and inserted into the proximal bowel after its transection (Fig. 10B). The proximal pursestring suture is tied to avoid contamination from the proximal bowel. Attention is then directed to the distal stump. The rectum is prepared in a similar manner, except that for low anastomoses, the width of the pelvis is not adequate to permit application of the instrument and a Keith needle. A right-angle clamp is placed distal to the carcinoma and the bowel is transected. A whipstitch of 2–0 Prolene is placed on the rectal stump, with evenly spaced bites taken 4 to 5 mm from the cut edge (Fig. 10A). Sizers are used to determine the appropriate diameter of the stapler cartridge. Before insertion of the circular stapler, the operator must confirm the presence of the staples, circular knife, and Teflon ring. The appropriate size instrument is selected, lubricated, and inserted through the anus in the closed position with the handle up. The instrument is advanced until the cartridge is visualized through the rectal lumen, the central shaft is extruded, and the distal pursestring suture is tied (Fig. 10B). The anvil is then

engaged into the central shaft (Fig. 10C). Turning the wing nut clockwise closes the stapler, while the abdominal operator ensures that the gap is free of sponges, mesentery, bladder, and other tissues, especially vaginal (Fig. 10D). Care must be taken to ensure adequate dissection to separate the rectum from the vaginal wall. Failure to do so runs the risk of entrapment of a portion of the posterior vaginal wall, which will probably lead to the development of a rectovaginal fistula. As the stapler is being closed, the perineal operator can check with a digital examination of the vagina that the vaginal wall is not being incorporated. When the stapler is fully closed, the safety is released and the stapler is activated by squeezing the handle firmly. This action places a double, staggered, circular row of stainless steel staples that join the two ends of the bowel, while a circular knife simultaneously cuts two rings of tissue inside the staple line, thus creating an inverted end-to-end anastomosis (Fig. 10E). To remove the instrument, the stapler is opened by turning the wing-nut counterclockwise one and a half completed turns. The stapler is rotated in an arc and should

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

223

FIGURE 10 & Construction of anastomosis using the circular stapler. (A) Application of proximal pursestring suture using specially designed fenestrated clamp and application of distal pursestring suture with a whipstitch suture of 2–0 Prolene. (B) Distal pursestring suture is secured around central shaft. Proximal colon secured around anvil. (C) Anvil engaged into central shaft. (D) Manual exclusion of extraneous tissue during approximation of bowel ends. (E) Completed anastomosis with ‘‘rings of confidence.’’ Source: Adapted from Ref. 83.

move independently of the bowel. The instrument is then removed in the direction of the curve of the instrument by a simple, gentle, simultaneous withdrawing and back-and-forth rotational motion. When difficulty is encountered in extraction of the instrument, guide sutures are placed through the anastomosis to help lift the anastomosis over the anvil (Fig. 11). A check is made to ensure that the rings of tissue excised are intact. Anastomoses may be

inspected directly with a sigmoidoscope, looking for bleeding or obvious disruption. The integrity of the anastomosis is further tested by insufflating air into the bowel via the sigmoidoscope with saline in the pelvis. The abdominal operator checks for bubbles arising from the anastomosis. If an air leak is present, sutures can be placed to correct the defect. These sutures may be placed by the abdominal operator, but in the case of a very low anastomosis, per anal

224

& PART I: COLORECTAL DISORDERS

FIGURE 11 & Suture placement through anastomosis to help lift anastomosis over anvil. Source: Adapted from Ref. 83.

access may be more appropriate. A diverting stoma is not required unless the surgeon is unable to accomplish the closure or is in some way unhappy about the integrity of the anastomosis. If excess laxity exists in the bowel, the peritoneum on the right side of the pelvis is closed. One side is left open to freely communicate with the generalized peritoneal cavity

so that any fluid will be readily absorbed. More recently, the peritoneum has not been closed at all, and no adverse effects have appeared. It is unlikely that a loop of small bowel will pass around the colon that was led to the pelvis. A technical problem that is not infrequently encountered is the discrepancy in the diameter of the bowel ends to be anastomosed. Probably the simplest way to enlarge the bowel lumen is by insertion of progressively larger sizers (e.g., sizers by United States Surgical Corp., Norwalk, Connecticut, U.S.A.). A second option is the use of a sponge forceps to stretch the bowel. Yet another technique is the very slow expansion of a 30 mL Foley balloon catheter with saline after it is positioned in the bowel lumen (84). All these methods of dilation may result in tearing of the bowel wall. Simple techniques of enlarging the diameter of the bowel lumen for performing end-to-end anastomoses using the EEA stapler were described by Tchervenkov and Gordon (85). If the transected bowel end cannot be dilated to accept a staple cartridge of appropriate size and the pursestring suture has already been applied, an incision can be made along the anti-mesenteric border of the colon. Ideally, the purse-string suture will have been placed so that the free ends are at the antimesenteric position of the circumference of the bowel (Fig. 12A). It is a simple matter to incise the bowel between the two ends of the suture (Fig. 12B) and then continue the suture along the newly created border past the apex of the incision to meet the other end of the suture (Fig. 12C). The new configuration of the circumference will be egg shaped (Fig. 12D). If it is apparent from the outset that the bowel caliber is definitely too small and will not be successfully dilated by the previously described methods, the oblique application of the pursestring clamp will result in a larger diameter of the bowel end (Fig. 13). Double-Stapling Technique Knight and Griffin (86) introduced the double-stapling technique (Fig. 14). The cited advantage of this technique is the elimination of the need for a distal pursestring suture. A Roticulator or regular linear stapler is placed on the rectum distal to the carcinoma. The instrument is fired, a rightangle bowel clamp is applied distal to the carcinoma, and the rectum is transected just proximal to the stapler. Depending upon the variety of circular stapler used, the technique varies slightly. With the CEEA instrument

FIGURE 12 & (A) Application of pursestring clamp such that one end is on antimesenteric border. (B) Incision of bowel along taenia between sutures. (C) Continuation of suture along newly created border. (D) Completed application of pursestring suture. Source: Adapted from Ref. 85.

FIGURE 13 & Oblique application of pursestring clamp. Source: Adapted from Ref. 85.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

225

FIGURE 14 & Double-stapling technique. (A) Roticulator applied to distal rectum and pursestring clamp applied to proximal colon. (B) CEEA introduced with trocar piercing at level of anastomosis. (C) Trocar removed and detachable anvil inserted into proximal bowel. (D) Anvil engaged in central shaft. (E) Anastomosis completed.

the detachable anvil is removed and a trocar is placed in the central shaft, which is then retracted within the cartridge. The instrument is introduced through the anus and advanced to the staple line. The trocar is extruded through the staple line and removed; the anvil previously inserted into and secured on the proximal bowel is engaged into the central shaft. The anvil is approximated to the cartridge and the instrument is activated. The instrument is withdrawn and the anastomosis is inspected. The Ethicon circular stapler has the trocar built into the central shaft and hence one step is removed. When using a circular instrument care should be exercised to ensure that the trocar

penetrates the suture line or the area immediately adjacent to it. If a rim of tissue between the stapled rectum and the ring of tissue excised remains, it may become ischemic and result in a leak. Griffin et al. (87) reviewed 75 patients and found an anastomotic leak rate of 2.7% and a 2.7% incidence of stenosis that required treatment. No deaths were reported. Moran et al. (88) encountered a clinical leak rate of 9% and a stricture rate of 5% in a series of 55 patients. Laxamana et al. (89) reported on the long-term follow-up of 189 patients who underwent a double-stapling technique. The postoperative mortality was 1.6% and the clinical leak rate

226

& PART I: COLORECTAL DISORDERS

is made in the peritoneum on each side of the rectum. A flap of peritoneum is elevated on each side. Conserving the peritoneum facilitates closure at the end of the operation. Once the decision is made that an abdominoperineal resection is required, the perineal phase of the operation begins. COMPLETION OF ABDOMINAL PORTION IN SYNCHRONOUS COMBINED EXCISION OF RECTUM PREPARING COLOSTOMY AND RECONSTRUCTION OF PELVIC FLOOR & The site in the proximal sigmoid or distal des-

FIGURE 15 & Single-layer hand-sutured anastomosis. Alignment of posterior wall and placement of sutures before tying.

7.3%, significantly higher at 20% for the lower one-third of the rectum. Other surgeons have evaluated the safety of this technique (90). Hand-Sutured Anastomoses Many surgeons still prefer hand-sutured anastomoses. Although two-layered anastomoses were once considered standard technique, the trend is now toward single-layer anastomoses. The open technique, with no clamping of the bowel ends, prevents injury to the bowel wall. For approximation of the posterior wall, vertical mattress sutures are used. The first bite is full thickness on each side, but on the return bite only mucosa and submucosa are included (Fig. 15). Popularly used suture materials are 4–0 polypropylene, 4–0 polyglycolic acid, or poly-glactin. All knots are tied inside the lumen. An interrupted full-thickness suture with only mucosal inversion as described by Gambee is used for the anterior wall (Fig. 44 in Chapter 23). Many surgeons prefer a two-layer anastomosis: an outer seromuscular layer of 4–0 nonabsorbable or absorbable suture and a continuous inner layer of 4–0 absorbable suture. Details of suture techniques are described in Chapter 23 and illustrated in Figs. 46 and 47 of that chapter.

Abdominoperineal Resection Proctectomy for carcinoma has evolved from a procedure that originally consisted primarily of a perineal excision to the classic Miles procedure. In the latter procedure, the rectum is mobilized via the abdominal route and buried below the reconstructed pelvic peritoneum; it is removed via a perineal route. The two-team, or synchronous combined excision gained popularity and remains my preference, as well as the technique of choice of many surgeons today. The Miles method permits a simultaneous approach to the rectum and shortens the operating time. Abdominal Part of Operation The initial dissection in an abdominoperineal resection is almost identical to that described for a low anterior resection. The one exception is that a conservative incision

cending colon chosen for the colostomy is cleaned and transected, an act easily accomplished with a stapler (Fig. 16A). With blunt finger dissection, an extraperitoneal tunnel is created from the left flank toward the site of the colostomy (Fig. 16B). A 2 cm disk of skin and subcutaneous tissue is excised at the premarked colostomy site. The anterior fascia is incised in a cruciate manner, the rectus muscle is split, and the posterior fascia and peritoneum are incised in a cruciate manner. The opening should admit two fingers loosely. The colon is then brought through the extraperitoneal tunnel to the colostomy aperture (Fig. 16C). It should be matured at the completion of the abdominal part of the procedure. Details of colostomy construction are described in Chapter 32. With the synchronous combined excision, the rectum is passed to the perineal operator when that portion of the dissection is complete. If the patient is to be repositioned, the stapled end of the rectal stump is dropped into the pelvis. The pelvic peritoneum is closed with continuous 4–0 absorbable suture (Fig. 16D). It is at this point that preservation of the peritoneal flaps is appreciated. The abdominal cavity is copiously irrigated with saline and closed in layers. INTRAPERITONEAL COLOSTOMY & If the extraperitoneal colostomy cannot be performed conveniently, the colon is brought out through the colostomy aperture, and the projecting limb is attached to the lateral abdominal wall using interrupted 4–0 absorbable sutures to prevent herniation of the small bowel. Nevertheless, the extraperitoneal technique is much preferred because of fewer complications related to peristomal herniation, prolapse, retraction, and internal herniation.

The staples at the end of the colostomy are excised. The colostomy limb may be sutured to the anterior fascia. It is preferable to evert the stoma 1 cm rather than making it flush with the skin. The stoma is immediately matured with interrupted 3–0 or 4–0 chromic catgut (Fig. 16E). A temporary colostomy bag is then applied.

MATURATION OF COLOSTOMY &

PERINEAL PORTION WHEN SYNCHRONOUS COMBINED EXCISION OF RECTUM IS USED & During the original posi-

tioning of the patient, 2-in. tape is placed on each buttock to retract it laterally and permit better exposure of the perineum. In the male, the external genitalia also are taped to keep them out of the way of the perineal operator. The dissection is not begun until the abdominal operator has confirmed that the rectum is resectable. The perineal phase of the operation is begun by placing a heavy, silk pursestring suture around the anus. An

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

227

FIGURE 16 & Completion of abdominal portion of abdominoperineal resection. (A) Division of colon in preparation for colostomy. (B) Development of extraperitoneal tunnel. (C) Delivery of colon through tunnel. (D) Closure of pelvic peritoneum. (E) Maturation of colostomy. Source: From Ref. 91; Figs. C to E after M. Finch.

elliptical incision is made to encompass an adequate margin of tissue (Fig. 17A). A wider margin is obtained for low-lying lesions than for more proximal ones. The skin edges are then grasped with baby Kocher clamps. Spring retractors will aid exposure (Fig. 17B). Dissection is continued with a cautery or scissors to incise the fat in the

ischioanal fossa. Cautery dissection results in less bleeding. The inferior rectal vessels on each side may require ligation. Deeper dissection is started posteriorly where the anococcygeal ligament is divided. Care should be taken to remain anterior to the coccyx because of the tendency to migrate to a superficial plane. Anteriorly dissection is

228

& PART I: COLORECTAL DISORDERS

FIGURE 17 & Perineal portion of abdominoperineal resection for synchronous operation. (A) Elliptical incision. (B) Exposure with spring retractors. (C) Anterior deep dissection posterior to transverse perinei muscles. (D) Division of levator ani muscles. (E) Delivery of rectum into perineal wound. (F) Closure of perineum.

continued posterior to the transverse perinei muscles (Fig. 17C). Attention is once again directed posterior to the rectum, where a finger can pierce through the levator ani muscles just anterior to the coccyx to enter the presacral plane where the abdominal dissection has led (Fig. 17D). The muscle is divided between clamps and ligated or can often simply be divided with cautery. This is continued laterally on each side. At this point, the rectum can be delivered into the perineal wound posteriorly (Fig. 17E). This maneuver facilitates the anterior dissection. In the female, the rectum can easily be dissected from the vaginal wall. For lesions located on the anterior wall, it may be advisable to excise the posterior vaginal wall in continuity with the rectal wall. Reconstructing the vaginal wall and introitus is not a difficult task. In men, caution must be exercised to avoid injury to the urethra. This is the most

difficult part of the operation. Misjudgment of tissue incision might result in entry into the rectum or membranous urethra. Palpation of the Foley catheter will help ascertain the appropriate plane. As muscle is divided on each side of the midline, the plane between the prostate and rectum is developed. By alternating displacement of the rectum from one side to the other and using cautious scissor or cautery dissection, the rectum is eventually separated from the prostate and urethra. Once this dissection is complete, the diseased rectum can be removed. The pelvis and perineal wound then are copiously irrigated with saline. After hemostasis has been ensured, a suction catheter is placed in the pelvis and brought out through a separate stab wound in the perineum. Alternatively, some surgeons prefer transabdominal drainage in the belief that this is easier for the patient in the postoperative period. The tapes

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

229

placed on the buttocks at the beginning of the operation are now released. The ischioanal fat is approximated with absorbable sutures and the skin is closed with subcutaneous 3–0 Vicryl (Fig. 17F). In the extraordinary situation in which satisfactory control of bleeding has not been obtained, the pelvic wound can be packed with gauze rather than closed primarily. This may be accomplished via the perineum or through an abdominal route and brought out through the lower end of the wound (76). Wounds should be inspected regularly because of the possibility of perineal sepsis. If an abscess were to develop, it would be necessary to open the wound promptly.

and the rectal stump can be delivered (Fig. 18C). The levator ani muscles on each side of the rectum are clamped, cut, and tied. Alternatively, they can be divided by cautery. At this point the remaining rectum is only attached to the prostate in men or the vagina in women (Fig. 18D). An incision is made to include Denonvilliers’ fascia on the anterior wall of the rectum. Again for posterior lesions dissection posterior to Denonvillier’s fascia would appear appropriate. Through this plane the rectum is sharply dissected from the prostate gland or vagina. Usually a moderate amount of bleeding occurs over the prostatic capsule or vagina, requiring electrocoagulation or a stick-tie.

PERINEAL PORTION OF ABDOMINOPERINEAL RESECTION WHEN PATIENT IS REPOSITIONED & With the recognized

CLOSURE OF PERINEAL SPACE &

advantages of the synchronous combined excision of the rectum, there is seldom, if ever, a need for this technique. When employed, the patient is turned to the prone position with a 6-in. roll placed under the pubis (Fig. 18A). The buttocks are spread apart with 2-in. tape. The perineal area is prepared and draped. A purse-string suture of No. 1 silk is placed around the anus to avoid fecal spillage. An elliptical incision is made around the anus, just lateral to the boundary of the external sphincter muscle (Fig. 18B). Bleeding points are electrocoagulated. The incision is deepened circumferentially. Posteriorly, the anococcygeal ligament and the anococcygeal raphe are incised close to the tip of the coccyx. Once this is completed, the pelvic cavity is entered

The perineal space is copiously irrigated with saline. Complete hemostasis must be obtained. A closed suction drain is placed in the perineal space and brought out through a small stab wound more anteriorly so that the patient will not sit on it. The subcutaneous tissue is closed with interrupted absorbable sutures (Fig. 18E). The skin is closed with subcuticular absorbable sutures. The drain is connected to continuous suction and left in place until the drainage is less than 10 mL per eight-hour shift on two or three consecutive shifts, which usually takes about five days (Fig. 18F).

Closure of the perineal space should not be done if the wound continuously oozes or if gross fecal spillage has occurred. Instead, the wound should PACKING OF PERINEAL WOUND &

FIGURE 18 & Perineal portion when patient is repositioned. (A) Position. (B) Elliptical incision after placement of pursestring suture. (C) Delivery of rectal stump. (D) Remaining rectal attachment. (E) Closure of perineum. (F) Completed operation.

230

& PART I: COLORECTAL DISORDERS

FIGURE 19 & Placement of sling from sacral promontory to anterior abdominal wall.

be packed with a rubber dam over Kerlex gauze. This packing should be removed in 48 hours. The rubber dam permits removal without the gauze adhering to adjacent tissue. For patients in whom postoperative radiotherapy is anticipated, elimination of the small bowel from the pelvis diminishes the chance of radiation enteritis. To this end it has been recommended that a mesh be sutured to the pelvic brim. Devereux et al. (92) have successfully adopted the use of an absorbable polyglycolic acid mesh sling to protect against radiation enteritis (Fig. 19). In addition, a silicone rubber prosthesis can be used to fill the pelvic cavity (93). The Silastic material is of water density and does not interfere with radiation therapy.

Restoration of Continuity after Sphincter Preserving Operations Reconstructive options following a sphincter-sparing resection for rectal carcinoma include straight colorectal anastomosis, straight coloanal anatomosis, colonic pouchanal anastomosis, and coloplasty. Several factors must be considered in the selection of an appropriate technique for restoration of continuity if the postoperative neo-rectal function is to be optimized. Very low anterior resection and anastomosis can result in less than perfect function, the so-called anterior resection syndrome which is characterized by frequency, urgency, and soiling (94) and is thought to be due to the loss of the reservoir function as well as diminished compliance (95). The lower the level of the anastomosis, the more adverse the functional outcome. This is supported by reports that patients undergoing very low anterior resection have poorer quality of life scores than those undergoing high anterior resection (96) or even abdominoperineal resection (97). Postoperative function is also impaired in patients who have a straight colorectal anastomosis and receive postoperative radiotherapy or in those who have preoperative radiotherapy followed by anastomosis of colon to an irradiated rectal remnant (98). Direct coloanal anastomosis may be associated with patients experiencing increased frequency of defecation, increased nocturnal defecation, fecal urgency, and incontinence (95).

Coloanal Anastomosis Parks (99) described a technique in which the rectum is transected just above the anorectal ring (Fig. 20A). The anal mucosa above the dentate line is removed through the anal orifice (Fig. 20B). The proximal colon is drawn down into the anorectal remnant and sutured to the level of the dentate line (Fig. 20C,D). A proximal ileostomy or colostomy is constructed to protect the anastomosis. Coloanal anastomoses have historically been associated with a price to be paid for the restoration of intestinal continuity. Apart from the anastomotic leakage, functional outcome has been affected with varying degrees of urgency and alteration in continence. Miller et al. (100) assessed the anorectal factors with manometric studies of 30 patents who underwent rectal excision and stapled coloanal anastomosis. Only 11 patients experienced perfect continence. Those with poor continence had lower resting and squeeze pressures. Poor function was more common in women. The authors raised the possibility that occult damage may have arisen before low anterior resection and such damage might be detected with preoperative manometry and endoanal ultrasonography. Patey et al. (101) reported on their experience in 140 patients with resection and coloanal anastomosis. There were no deaths and the five-year actuarial survival was 73%. Pelvic recurrence was documented at an actuarial rate of 11% at five years. Braun et al. (102) compared 63 patients who underwent intersphincteric resection with direct coloanal anastomosis and 77 who had an abdominoperineal resection. During the mean follow-up period of 6.7 years, of those patients with curative resection, 11% presented with pelvic recurrence, and 33% with distant metastases after coloanal anastomosis; the rates of recurrence and distant metastases after abdominoperineal resection were 17% and 35%, respectively. The corrected five-year survival rates were 62% following coloanal anastomosis and 53% following abdominoperineal resection. Eighty-five percent of the patients with coloanal anastomosis reported good functional results regarding anal continence. This study demonstrates that the intersphincteric resection with coloanal anastomosis is a valuable surgical technique for rectal carcinoma with the benefit of preservation of continence. Leo et al. (103) reported on 141 consecutive patients treated for a primary carcinoma of the distal rectum from 3.5 to 8 cm from the anal verge. Patient stratification, included 31 Dukes’ stage A (T2N0), 44 stage B (T3N0), and 66 stage C (T2Nþ to T3Nþ). Overall recurrence rate was 9.2%. Perfect continence was documented in 61% of cases. The only pathological factor related to local recurrence was lymphocytic reaction inside and around the carcinoma. Saito et al. (104) extended their indication for sphincter saving operation. They investigated the curability and functional results of intersphincteric resection and additional partial external sphincter resection for carcinoma of the anorectal junction. Thirty-five patients with carcinoma located between 0 and 2 cm above the dentate line underwent abdominotransanal rectal resection with TME. All patients underwent diverting colostomy. Twenty patients received preoperative radio/chemotherapy. All patients had curative intent with microscopic safety margins. No postoperative mortality was encountered. Morbidity was identified in 13 patients and included peri-anastomotic

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

231

FIGURE 20 & Coloanal anastomosis. (A) Transection of rectum at level of anorectal ring. (B) Stripping of mucosa from anorectal stump. (C) and (D) Anastomosis of proximal colon to dentate line.

abscess, anastomotic leakage and fistula; postoperative bleeding, infection, peri-anastomotic stenosis, and anovaginal fistula. One of these patients required a permanent colostomy. Five patients developed recurrence during the median observation period of 23 months. Two of these patients underwent curative resection of liver or lung metastases. Twenty-one patients had undergone stoma closure and although continence was satisfactory in all, five displayed occasional minor soiling 12 months after stoma closure. Anal canal manometry demonstrated significant reduction in maximum resting pressure but acceptable functional results were obtained. They believe these procedures

can be recommended for low rectal carcinoma patients who are candidates for abdominoperineal resection. Gamagami et al. (105) assessed the influence of partial excision of the superior portion of the anal canal when necessary for margin clearance in distant rectal carcinoma on fecal continence after coloanal anastomosis. The 209 patients were categorized into three groups according to their level of anastomosis from the anal verge: 43 group 1 patients with anastomosis 0.5 to less than 2 cm from the anal verge, 75 group 2 patients with anastomosis 2 cm to less than 3 cm from the anal verge, and 73 group 3 patients with anastomosis 3 to 2.5 cm from the anal verge. In the

232

& PART I: COLORECTAL DISORDERS

first year, there was progressive improvement in anal continence in all three groups. At two years, 50% in group 1, 73% in group 2, and 62% in group 3 were fully continent. The proportion of all patients fully continent in group 1 remained unchanged as compared to continued improvement in groups 2 and 3 following the first year. At four years, 50% in group 1, 80% in group 2, 68% in group 3 were completely continent. Nathanson et al. (106) reported that straight (nonresevoir) coloanal anastomoses with postoperative pelvic radiotherapy had significant adverse effects on anorectal function, with higher rates of clustering and frequency of defecation than with preoperative radiotherapy. No differences in continence rates were demonstrated. They attribute the adverse effects of postoperative radiotherapy to irradiation of the neorectum, which is spared when treatment is given preoperatively. Olagne et al. (107) assessed the functional outcome of patients who had delayed coloanal anastomosis for a lower third rectal carcinoma after preoperative radiotherapy in 35 patients. Colorectal resection was performed about 32 days after the end of the radiotherapy. The distal colon stump was pulled through the anal canal. On postoperative day 5, the colonic stump was resected and a direct coloanal anastomosis performed without colostomy diversion. There was no mortality. There was no leakage. One patient had a pelvic abscess. One patient had necrosis of the left colon requiring reoperation. Another delayed coloanal anastomosis could be performed. Median follow-up was 43 months. Function was considered good in 59% and 70% at one and two years, respectively. They concluded this procedure is a safe and effective sphincter preserving operation that avoids a diverting stoma for patients with rectal carcinoma of the lower third of the rectum. Most recently, Rullier et al. (108) assessed the oncologic outcome of patients treated by conservative radical surgery for carcinomas below 5 cm from the anal verge. Ninety-two patients with a nonfixed rectal carcinoma at 1.5 to 4.5 cm from the anal verge and without external sphincter infiltration underwent TME with intersphincteric resection, that is, removal of the internal sphincter to achieve adequate distal margin. Patients with T3 disease or internal sphincter infiltration received preoperative radiotherapy. There was no mortality and morbidity was 27%. The rate of complete microscopic resection (R0) was 89% with 98% negative distal margin and 89% negative circumferential margin. In 58 patients with a follow-up of more than 24 months, the rate of local recurrence was 2% and the five-year overall and disease free survival were 81% and 70%, respectively. They concluded, the technique of intersphincter resection permits us to achieve conservative surgery in patients with a carcinoma close to or in the anal canal without compromising local control and survival. J Pouch Establishment of a colonic reservoir analogous to the J-shaped ileal pouch was originally described by Lazorthes et al. (109,110) and has gained favor with a growing number of surgeons (111–114). Advocates of the procedure believe that adding capacity in the form of a colonic reser-

voir can substantially reduce dysfunction and greatly improve quality of life. Urgency is diminished and continence is improved. Technically, it is simple to make the reservoir in many cases, but in patients with obesity or a foreshortened mesentery, construction may prove difficult or impossible. The anastomosis itself is no more difficult than that of the straight coloanal reconstruction. Some patients may develop difficulty with evacuation and it is important to use no more than 12 to 16 cm of colon for the two-loop construction (115). The operative technique of dissection follows the general oncologic principles previously enunciated. The splenic flexure will most likely require mobilization unless there is considerable redundancy of the colon. The proximal sigmoid or distal descending colon is used for reconstruction. An anastomosing instrument is applied at the line of transection to both divide the bowel and close the end (Fig. 21A). A colonic J pouch is made with 7 to 8 cm of bowel by folding the colon and creating a side-to-side anastomosis with a stapler introduced through the apex of the pouch (Fig. 21B,C). A double-stapling technique can be used to establish intestinal continuity. A temporary ileostomy is created. A sutured anastomosis can be constructed (Fig. 22). Hallbo¨o¨k et al. (116) measured blood flow by laser Doppler fiowmetry before construction of a straight or J-pouch anastomosis. In the straight group (end-to-end anastomosis), blood flow levels at the site intended for anastomosis were significantly decreased following dissection of bowel. In the pouch group (side to end), blood levels at the site of the anastomosis were similar following dissection of the bowel and pouch construction. They concluded unaffected blood flow at the side of the anastomosis of the pouch might be a favorable factor for anastomotic healing. Mortensen et al. (117) reported on a consecutive series of 23 patients with colonic J pouch-anal anastomosis for low rectal carcinoma. The mean distance from the pouchanal anastomosis to the anal verge was 3.5 cm (range, 2.0– 4.5 cm). In 19 surviving patients a mean of seven months after ileostomy closure, mean bowel frequency was 2.1 (range, one to four) per day, five patients had urgency, four had mild fecal seepage up to three times per week, and seven patients had some degree of incomplete evacuation. In 13 patients, there were no manometric differences before and after operation with respect to maximum tolerated volume or maximum resting pressure, but maximum squeeze pressure was significantly lower after operation (mean, 189 vs. 132 cm H2O before and after operation, respectively). The authors believe that colonic pouch reconstruction should be considered as an alternative to straight coloanal anastomosis in patients undergoing very low anterior resection. The physiologic deficiency in patients undergoing a low anterior resection for rectal carcinoma was highlighted in a study by Lewis et al. (118). In a series of 34 patients that were divided into three groups on the basis of residual rectum (none, less than 4 cm, greater than 4 cm) and a control group of 10 patients in whom the rectum was left intact, resting anal pressure was lower and the capacity of the neorectum was less after coloanal than after colorectal anastomosis. The greater the length of the residual rectum, the greater was the capacity of the neorectum. Their findings support the use of a colonic pouch when the entire rectum or almost the entire rectum has been removed.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

233

FIGURE 21 & (A) Transected proximal bowel. (B) Completed colonic J pouch with anvil secured by pursestring suture in apex of pouch. (C) Double-stapling technique used to construct side-to-end coloanal anastomosis.

A remarkable case of a patient developing a carcinoma in a colonic J pouch after a low anterior resection for villous adenoma was reported (119). In a review of the literature, McNamara and Parc (95) found that up to 20% of patients suffer evacuation problems and require the use of enemas or suppositories to evacuate the pouch. This has been attributed to the use of excessively large reservoirs with pouches as large as 12 cm. They compared the number of bowel movements for 24 hours in eight series that compared coloanal anastomosis and the colonic J-pouch and with a follow-up of six weeks to two years found with the former a range of 2.0 to 6.4 (range 6–20) and with the latter 1.4 to 3.1 (range 4– 10), statistically significant in most series. Hida et al. (120) conducted a prospective randomized study to determine optimum pouch size. They found that a 5-cm pouch was significantly superior to a 10-cm pouch with respect to evacuation function. Dehni and Parc (121) reviewed the different aspects of colonic J-pouch reconstruction with special focus on functional results and complications. They cited five randomized controlled trials comparing colon pouch anal anastomosis and straight coloanal anastomosis. According to the current literature, local recurrence rate and survival are comparable after either a low anterior resection, a coloanal anastomosis or an abdominoperineal resection for rectal carcinoma. In terms of functional outcome, there is mounting evidence supporting the superiority of the colonic J-pouch reconstruction over a straight anastomosis. The advantage of the pouch is most marked in the early postoperative period but may persist up to two years. Comparative studies have shown better results with pouch reconstruction in terms of stool frequency, urgency, nocturnal movements, and continence one year after operation. However, problems with evacuation of stools from the pouch are reported by 20% to 30% of patients in the long-term follow-up. Smaller pouches may reduce this difficulty. Overall, the use of a colonic J-pouch is compatible with curative operation and has the

additive benefit of optimizing the postoperative quality of life of patients after total rectal resection for carcinoma. Laurent et al. (122) compared handsewn or stapled colonic J-pouch-anal anastomosis performed after complete proctectomy and TME for carcinoma of the rectum. They found stapled coloanal anastomosis to be significantly faster than handsewn coloanal anastomosis and it had similar functional results. Machado et al. (123) randomized 150 patients with rectal carcinoma undergoing TME and coloanal anastomosis to receive either a colonic pouch or a side-to-side anastomosis using the descending colon. A large proportion of the patients received short-term preoperative radiotherapy (78%). There was no significant difference in surgical outcome between the two techniques with respect to anastomotic height (4 cm), perioperative blood loss (500 mL), hospital stay (11 days), postoperative complications, reoperations, or pelvic sepsis rates. Comparing

FIGURE 22 & Hand-sutured coloanal anastomosis for colonic J pouch.

234

& PART I: COLORECTAL DISORDERS

functional results in the two study groups, only the ability to evacuate the bowel in less than 15 minutes at 16 months reached a significant difference in favor of the pouch. They concluded that either a colonic J-pouch or a side-to-end anastomosis performed on the descending colon in low anterior resection with total mesorecal excision are methods that can be used with similar expected functional and surgical results. Machado et al. (124) compared the functional and physiologic outcome two years after low anterior resection and TME with a colonic J-pouch or a side-to-end anastomosis in 35 patients. There was no statistical difference in functional outcome between groups at two years. Maximum neorectal volume increased in both groups but was approximately 40% greater at two years in pouches compared with the side-to-side anastomosis. Anal sphincter pressure volumes were halved postoperatively and did not recover during follow-up of two years. Male gender, low anastomotic level, pelvic sepsis, and the postoperative decrease of sphincter pressures were independent factors for more incontinence symptoms. Dehni et al. (125) compared long-term bowel habits in 28 patients receiving preoperative radiation for rectal carcinoma followed by colonic J-pouch-anal anastomosis with those of 97 patients having similar rectal carcinoma surgery without radiation. The number of bowel movements per 24 hours in patients with or without radiation was 1.8. In the irradiated group, diarrhea (39% vs. 13%) and nocturnal defecation (36% vs. 15%) were more frequent than in the nonirradiated group. No other significant difference existed between groups with regard to stool clustering, use of protective pads, ability to defer evacuation more than 15 minutes, ability to evacuate the bowel within 30 minutes, incontinence score, use of medications, or dietary restriction. Sailer et al. (126) reported that patients with a pouch reconstruction had a significantly better quality of life, particularly in the early postoperative period. Patients undergoing low anterior rectal resection and coloanal J-pouch reconstruction may expect not only better functional results but also an improved quality of life in the early months after operation compared with patients who receive a straight coloanal anastomosis. In a detailed study Heah et al. (127) reported that pouches made from sigmoid or descending colon give similar bowel function after ultra-low anterior resection for rectal carcinoma. Ho et al. (128) conducted a randomized controlled study to compare functional outcome following the colonic J-pouch and the straight coloanal anastomosis immediately after ultra-low anterior resection. Of 42 consecutive patients recruited, 19 of the straight group completed the study. At six months, the pouch patients had significantly less frequent stools (32.9 vs. 49 per week) and less soiling at passing flatus (38% vs. 73.7%). At two years, both groups had improved with no longer any differences in stool frequency (7.3 vs. 8 per week) and soiling and passing flatus (38% vs. 53%). Anal squeeze pressures were significantly impaired in both groups up to two years. The rectal maximum tolerable volume and compliance were not different between groups. Rectal sensory testing on the barostat phasic program showed impairment at six months and recovery at two years suggesting that postoperative recovery of residual afferent sympathetic nerves may play a role in functional recovery.

Wang et al. (129) reported the outcome of 30 consecutive patients with colonic J-pouch-anal anastomosis without a diverting stoma. All patients had carcinoma of the lower two-thirds of the rectum. Functional results were compared with those of 21 rectal carcinoma patients with straight coloanal anastomosis who underwent operation in the same period and 20 normal patients. There were two anastomotic leakages and one postoperative death. After one year, patients with pouch anastomosis had significantly less frequency of defecation and rectal urgency compared with those with straight anastomosis; 48% of patients with straight anastomosis had more than five bowel movements per day whereas all patients with pouch anastomosis had five or fewer bowel movements per day. Manometric studies showed that the maximum tolerable volume was significantly higher in patients with pouch anastomosis (81 vs. 152 mL). If construction of a colonic J-pouch is not possible due to lack of colonic length, especially after prior colonic resections, the ileocecal interpositional reservoir may offer an alternative to rectal replacement. Hamel et al. (130) studied the functional results of the long-term follow-up of patients after TME and ileocecal interposition as rectal replacement. At five years, 78% of the patients were continent; mean stool frequency was 2.5 per day. Da Silva et al. (131) assessed the impact of diverticular disease on function and on postoperative complications of the colonic J-pouch with pouch-anal anastomosis. The presence of diverticular disease in the colonic J-pouch was assessed on pouchogram prior to ileostomy closure. The median follow-up period was 22 months. Twenty-four patients comprised the diverticular group and 42 were in the nondiverticular group. The total evacuation score and total incontinence score did not significantly differ between the two groups. Furthermore, there was no significant difference in the total incidence of pouch complications between the two groups. The presence of diverticular disease in a colonic J-pouch does not seem to have an impact on pouch function or the postoperative complication rate. Good function of the colonic-J-pouch reconstruction in the elderly would obviate the need for colostomy that is sometimes performed because of concern about fecal incontinence which increases with age. Hida et al. (132) compared functional outcome in 20 patients age 75 years or older (older group), and 27 patients age 60 to 74 years (old group) and 60 patients age 59 or younger (young group), three years after colonic J-pouch reconstruction, using a functional scoring system with a 17-item questionnaire [score range, 0 (overall good) to 26 (overall poor)]. The functional scores in the three age groups were satisfactory and similar. Among patients with anastomoses 1 to 4 cm from the anal verge, all 17 categories on the questionnaire in the three age groups were similar. Among patients with anastomoses 5 to 8 cm from the anal verge, only the use of laxatives or glycerine enemas was more common in the older group than in the old and young group (90% vs. 38.5% and 43.3%). They concluded that low anterior resection with colonic J-pouch reconstruction provides excellent functional outcome, including continence, for elderly patients. Gervaz et al. (133) evaluated the impact of adjuvant chemoradiation therapy on pouch function in 74 patients with midrectal or low rectal carcinomas (less than 10 cm from the anal verge) who underwent a proctectomy with coloanal anastomosis with colonic J-pouch reconstruction.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

235

The mean age of patients was 68.9 years and the mean duration of follow-up was 28.8 months. There were 28 patients in the surgery alone group and 17 patients who received either preoperative (13) or postoperative (4) adjuvant chemoradiation therapy. Patients in the surgery along group had a significantly better continence score: 18.1 versus 13.3 and were less likely to experience evacuatory problems (evacuation score 21.3 vs. 16.4). Use of pad was more frequent in the chemoradiation therapy than in the surgery alone group (53% vs. 18%). The incidence of functional disorders was also more frequent in the irradiated group of patients: incontinence to gas (76% vs. 43%) to liquid stool (64% vs. 25%) and to solid stool (47% vs. 11%). Moreover, irradiated patients reported more frequent pouch-related specific problems such as clustering (82% vs. 32%) and sensation of incomplete evacuation (82% vs. 32%). Finally, regression analysis demonstrated that radiation-induced sphincter dysfunction was progressive over time. Both preoperative and postoperative chemoradiation adversely affects continence and evacuation in patients with colonic J-pouch. Harris et al. (134) identified seven reasons for failure to be able to construct a colonic J-pouch. These reasons were: (i) technical (narrow pelvis, bulky anal sphincters or need for mucosectomy, diverticulosis, insufficient colon length or pregnancy), and (ii) nontechnical (complex surgery or distant metastases present). Failure to construct a neorectal reservoir occurred in 30.7% of patients. This was reduced to 5.3% of patients in the latter period of the study. Coloplasty To overcome poor bowel function after resection of the distal rectum with a straight reanastomosis, a neo-reservoir using a colonic J-pouch has been advocated. However, difficulties in reach, inability to fit the pouch into a narrow pelvis, and postoperative evacuation problems can make the colonic J-pouch problematic. Furst et al. (135) randomized 40 consecutive patients with distal rectal carcinoma (less than 12 cm from the anal verge) into the J-pouch or coloplasty group. A low rectal resection and coloanal anastomosis was performed in all patients. The construction of a coloplasty pouch was feasible in all cases for the coloplasty group but not in 25% of patients of the J-pouch group because of colonic adipose tissue (Fig. 23). Six months after operation or stoma closure the stool frequency was 2.75 þ/– 1 per day in the J-pouch group and 2 þ/– 2 per day in the coloplasty group, respectively. There was no significant difference in resting and squeeze pressure and neo-rectal volume between the two groups. They found similar functional results in the coloplasty group compared to the J-pouch group. The neo-rectal sensitivity was increased in the coloplasty group. Therefore the colonic coloplasty seems to be an attractive pouch design because of its feasibility, simplicity, and effectiveness. They speculate that the advantage of the colonic J-pouch is not in creating a larger neo-rectal reservoir but is rather related to decreased motility. Mantyh et al. (136) compared the functional results after a low colorectal anastomosis among patients receiving a coloplasty (20), colonic J-pouch (16), or straight anastomosis (17). Maximum tolerated volume was significantly favorable in the coloplasty and colonic J-pouch group versus the straight anastomosis group. The compliance was also significantly favorable for the coloplasty and the colonic J-pouch group versus the straight

FIGURE 23 & Coloplasty construction. (A) An 8 to 10 cm longitudinal incision is made between the tenia 4 to 6 cm from the cut edge of the colon. (B) The longitudinal opening is closed in a transverse manner with interrupted 2-0 polyglycolic sutures starting with stay sutures in the middle and at each end. (C) Completed coloplasty and withdrawal of circular stapler after completion of the coloanal anastomosis.

anastmosis group. The coloplasty and colonic J-pouch had significantly fewer bowel movements per day than the straight anastomosis group. Similar complications rates were noted in the three groups. Patients with a coloplasty and low colorectal anastomosis seem to have similar functional outcome along with similar pouch compliance compared with patients with colonic J-pouch and low colorectal anastomosis. However, the coloplasty may provide an alternative method to the colonic J-pouch for a neo-rectal reservoir construction when reach or narrow pelvis prohibits its formation. Technically, it may also be easier to construct. Z’graggen et al. (137) reported on 41 patients who underwent low anterior rectal resection with continuity restored by a transverse coloplasty pouch anastomosis and the colon was defunctionalized for three months. Intraoperative complications occurred in 7%, none related to the transverse coloplasty pouch. There were no hospital deaths and the total complication rate was 27%; an anastomotic leakage rate of 7% was recorded. The stool frequency was 3.4 per 24-hours at two months follow-up and

236

& PART I: COLORECTAL DISORDERS

gradually decreased to 2.1 per 24-hours at eight months. Stool dysfunctions such as stool urgency, fragmentation, and incontinence were regularly observed until six months; the incidence significantly decreased thereafter. None of the patients had difficulty in pouch evacuation. The most recent report by Remzi et al. (138) compared the quality of life, functional outcome, and complications among 162 patients undergoing coloplasty (69) colonic J-pouch (43) or straight anastomosis (50). Usually coloplasty or straight anastomosis was favored in male patients with a narrow pelvis or when a handsewn anatomosis was used. Quality of life assessment with the short form-36 questionnaire revealed better scores in coloplasty and colonic J-pouch groups. The coloplasty (1.0) and colonic J-pouch (1.0) groups had fewer night bowel movements than the straight anastomosis (1.5). The coloplasty group also had fewer bowel movements per day than the straight anastomosis group (3.8 vs. 4.8); also, less clustering and less anti-diarrheal medication use were observed than in the straight anastomosis group. Colonic J-pouch patients with hand-sewn anastomosis had a higher anastomotic leak rate (44%) than the patients in the coloplasty with hand-sewn anastomosis group (3.6%). They concluded, coloplasty seems to be a safe, effective technique for improving the outcome of low colorectal or coloanal anastomosis. It is especially applicable when a colonic J-pouch anastomosis is technically difficult.

Other Procedures A variety of pull-through operations have been described. Their operative morbidity and functional outcome have often left a lot to be desired. Because other operations have supplanted them, none of the pull-through operations will be described here. Kraske’s approach, or the parasacral approach to rectal resection, commanded some popularity in the early part of the century but because of the associated problems of sepsis, anastomotic leaks, and fecal fistulas, it fell into disrepute. Kraske’s operation is seldom used for the treatment of rectal carcinoma. The abdominosacral resection championed by Localio et al. (139) enjoyed the favor of some surgeons, but enthusiasm has waned with the introduction of the circular stapler. In an effort to create a sutureless anastomosis, analogous to Murphy’s button, Hardy et al. (140) developed a biofragmentable ring composed of polyglycolic acid and barium sulfate. A multicenter trial compared the biofragmentable ring anastomosis to the stapled and sutured anastomoses. No statistical differences emerged with respect to the complication rate, in particular, of the anastomotic leaks (141). The biofragmentable ring has not been adapted for low anastomoses and indeed has not garnered great enthusiasm among surgeons. Extended resections will be described in more detail under the operative treatment of recurrent disease but may also be applied to patients who initially present with locally advanced disease. Moriya et al. (142) described en bloc excision of the lower ureter and internal iliac vessels for locally advanced upper rectal and rectosigmoid carcinomas. Excellent prognoses were noted with respect to local control and functional results in eight patients.

Double Dynamic Graciloplasty In patients with a very low rectal carcinoma, an abdominoperineal resection with creation of a permanent colostomy is the operative treatment of choice. Creation of a colostomy can be avoided without compromise of oncologic principles by delivering the distal colon to the perineum and wrapping it with both gracilis muscles, creating a new sphincter and pelvic floor. These muscles are electrically stimulated by an implanted neurostimulator. The details of the dynamic graciloplasty were described in Chapter 15. In a series of 11 patients so treated by Geerdes et al. (143) with a follow-up of 1.3 years, continence was achieved in seven patients, and two patients were awaiting completion of therapy. In two patients, necrosis of the distal colon led to failure of the technique. There were no local recurrences but two patients had distant metastases. & POSTOPERATIVE CARE Postoperative management of the patient is discussed at length in Chapter 4. & RESULTS The same difficulty encountered in determining the survival, morbidity, and mortality rates of colon carcinoma has been encountered with rectal carcinoma. Martling et al. (144) analyzed the prognostic value of surgeons’ and pathologists’ assessment of clearance in 1550 patients with primary rectal carcinoma who underwent resection. In patients assessed as having a complete surgical clearance, recurrence developed in 33.3%. For patients with an uncertain or incomplete clearance, recurrence rate was 59.5% and 61%, respectively. The relative risk of recurrence was twice as high when the surgeon or pathologist disagreed than when they both agreed on the complete clearance. Survival in patients with a complete, uncertain, or incomplete surgical clearance was 55.3%, 23.0%, and 10%, respectively. An effort was made to present representative data in Tables 1–4. A frequently debated point is the virtue of sphinctersaving operations. Proponents of abdominoperineal resection believe anything less than the radical operation might jeopardize the patient’s chance of cure. However, evidence suggests that this is not the case. From a multicenter trial, Wolmark and Fisher (68) found no survival advantage of abdominoperineal resection over low anterior resection. A review of several series from the literature concurred (63,66,67,178–182). The general acceptance of the circular stapler method mandates a detailed discussion of the technique, with its inherent advantages and disadvantages as they relate to results. We reviewed a series of 215 anastomoses that I constructed with the circular stapler following resection (183). The operative mortality was 0.47%, with the one death being unrelated to the anastomosis. Intraoperative complications included bleeding, difficult extraction, instrument failure, incomplete ‘‘doughnuts,’’ and deficient anastomoses. Early postoperative complications included one leak and a number of complications unrelated to the anastomoses. Anastomotic stenosis developed in 27 patients,

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

237

TABLE 1 & Operative Mortality, Five-Year Survival Rate, and Local Recurrence of Carcinoma of Rectum Five-Year Survival Rate: No. of Potentially Curable Patients

Author(s) Pa¨hlman et al. (145) (1985) Davis et al. (146) (1987) Enblad et al. (147) (1988) Michelassi et al. (148) (1988) Amato et al. (149) (1991) Tagliacozzo and Accordino (150) (1992) MacFarlane et al. (79) (1993) Fandrich et al. (151) (1994) Adam et al. (152) (1994) Clemmesen and Sprechler (153) (1994) Enker et al. (154) (1995) Hermanek et al. (155) (1995) Laxamana et al. (89) (1995) Singh et al. (156) (1995) Zaheer et al. (157) (1998) Killingback et al. (158) (2001) Staib et al. (159) (2002) Croxford et al. (160) (2004) Wibe et al. (161) (2004)a Law et al. (162) (2005) Gordon et al. (163) (2006) (unpublished data)

161 235 23,603 154 147 274 135 331 141 196 246 887 189 251 514 549 1099 683 2136

Operative Mortality (%)

Crude (%)

5 3

60 42 42 53

3 4–2a 3 2

1 2 2 3 1

80, 83, 78c 73

3

63 overall 69, 62, 59d 71 64

2 2

Local Recurrence Rate (%)

55 53

59 78 80, 40, 30b 62 50 74 55 78

6

224e 184

Corrected (%)

– 81

12 11–12a 17 4 14 23 6, 11, 22b 7 9 8 4, 6, 7c 7 16 4 12 overall, 9, 13, 15a 6 18

a

Abdominoperineal plus anterior resection. According to Dukes’ classification. c Abdominoperineal resection, coloanal anastomosis, low anterior resection. d Upper middle lower third of rectum. e Stage II. b

but only eight were permanent and only three of these were symptomatic. Two of these patients were treated with balloon dilatation. Anastomotic recurrences developed in 13.1% of patients. The time to recurrence ranged from 5 to 39 months. The overall follow-up time ranged from 4 to 124 months. Our experience and that of many other surgeons has shown the circular stapling device to be a reliable method for performing anastomoses to the rectum in a safe and expeditious manner.

One concern with the circular stapler is the high incidence of narrowing of the anastomoses. Others who have recorded complications after use of the EEA stapler in rectal anastomoses have reported narrowings ranging from 0% to 30% (Table 5). Few authors have given their definition of stenosis. Leff et al. (191) and Fazio (199) have defined a stricture as narrowing that does not allow passage of a 15 mm sigmoidoscope. I have considered as stenotic any anastomosis that did not accept the 19 mm sigmoidoscope.

TABLE 2 & Results of Curative Low Anterior Resection for Carcinoma of Rectum

Author(s) Fedorov and Odarjuk (164) (1983) Zhou et al. (165) (1983) Williams et al. (66) (1985) Malmberg et al. (166) (1986) Davis et al. (146) (1987) Heberer et al. (167) (1987) Belli et al. (168) (1988) Amato et al. (149) (1991) Jatzko et al. (169) (1992) Isenberg et al. (170) (1995) Singh et al. (156) (1995) Enker et al. (154) (1995) Laxamana et al. (89) (1995) Zaheer et al. (157) (1998) Killingback (158) (2001) Wibe et al. (171) (2004) Law and Chu (164) (2004) Law et al. (162) (2005) Gordon et al. (163) (2006) (unpublished data) a

Stage II.

No. of Patients

Operative Mortality (%)

Five-Year Survival Rate: Crude (%)

356 158 100 83 111 354 74 78 175 89 135 170 186 272 468 1315 419 224a 129

4 1 7 5 5 4 3 2 3

62 63 74 71 39 76 67

4 1 2 2 3 3 2 2 2

77 45 81 78 78 — 68 74 (disease specific) 71 63

238

& PART I: COLORECTAL DISORDERS

TABLE 3 & Results of Curative Abdominoperineal Resection for Carcinoma of Rectum

Author(s) Fedorov and Odarjuk (165) (1983) Zhou et al. (166) (1983) Elliot et al. (172) (1984) Williams et al. (67) (1985) Davis et al. (146) (1987) Heberer et al. (168) (1987) Huguier et al. (173) (1990) Amato et al. (149) (1991) Jatzko et al. (170) (1992) Isenberg et al. (171) (1995) Singh et al. (156) (1995) Enker et al. (154) (1995) Zaheer et al. (157) (1998) Killingback et al. (158) (2001) Dehni et al. (174) (2003) Law and Chu (164) (2004) Wibe et al. (161) (2004) Gordon et al. (163) (2006) (unpublished data)

No. of Patients

Operative Mortality (%)

Five-Year Survival Rate: Crude (%)

829 251 196 33 104 317 100 69 72 53 45 76 169 58 91a 69 821 21

8 1 2 4 1 7 1 4 1

53 48 39 68 56 73 45

4 1 2 3 1 2 3 0

49 56 60 80 76 60 disease specific 55 57

a

Some patients received preoperative radiology, some postoperative radiotherapy, some chemotherapy.

The problem of stenosis is probably an ischemic one, and in the enthusiasm to clear the bowel for application of the pursestring instrument, excess blood supply is stripped from the edge. By decreasing the amount of bowel that is prepared for the anastomosis, I have noted a decrease in the incidence of stenosis from 20% in my early experience to 13% in subsequent patients. In the last 72 cases, the incidence of stenosis was further reduced to 4.2%. From the point of view of anastomotic security, reported clinical rates of leakage have ranged from 0% to 16% (Table 5). Those who have studied patients with meglumine diatrizoate (Gastrografin) enemas have found radiologic leakage rates ranging from 3% to 36% and even higher if only very low anterior resections were considered (211). Several authors have compared stapled with hand-sewn anastomoses (212–214). In two of these series stapled anastomoses fared better than hand-sewn anastomoses. In the report by Beart and Kelly (212) the

two techniques were equal, but it must be noted that in 12% of the patients in that series, the rectum was preserved as a result of the stapling procedure. Other authors have reported that the use of a circular stapling device has diminished the need for abdominoperineal resection (63,187,188,215–218), and I certainly support this view. In the report by McGinn et al. (203) stapled anastomoses suffered a 12% clinical leak, whereas hand-sutured anastomoses experienced only a 3% clinical leak. These results are clearly at variance with other reports. Concern has been expressed that the introduction of the circular stapling devices might result in a higher incidence of anastomotic or local recurrence (Table 6) (194,219– 221). The concern is that surgeons might compromise on the distal margin of resection. It might also be argued that since a lower anastomosis can be created with a circular stapling device, a greater distal margin might be achieved in a given case while still preserving the anal sphincter.

TABLE 4 & Five-Year Survival Rate According to Stage Following Curative Resection of Carcinoma of Rectum Dukes’ A Author(s) Rosen et al. (175) (1982) Hojo (176) (1986) Malmberg et al. (167) (1986) Davis et al. (146) (1987) Michelassi et al. (148) (1988) Fandrich et al. (151) (1994) Hermanek et al. (155) (1995) Enker et al. (154) (1995) Sugihara et al. (177) (1996) Staib et al. (159) (2002) Dehni et al. (174) (2003) Gordon et al. (163) (2006) (unpublished data) a

No. of Patients 180 273 83 235 154 371 864 246 214 91 186

No. 45 26

Crude (%) 86 93 96 75

257

70–80a 74

55 259 34 59

96 91 88 63

Dukes’ B Corrected (%)

No.

Crude (%)

75 32 98

257 99 72 269 26 63

Submucosa or muscularis propria. Some patients received preoperative radiotherapy, some postoperative radiotherapy, some chemotherapy.

b

65 73 72 55 60 45 62 87 84 73 84 73

Dukes’ C Corrected (%)

No. 60 25

72 61 350 147 87 215 31 54

Crude (%) 33 49 45 31 43 30 40 64 67 38 54 46

Corrected (%)

40 43

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

239

TABLE 5 & Complications with Circular Stapler Leaks (%)a Author(s) Cutait et al. (184) (1981) Killingback (185) (1981) Dorricott et al. (186) (1982) Friis, Hjortrup, and Nielson (187) (1982) Goligher (188) (1982) Hamelmann et al. (189) (1982) Helm and Rowe (190) (1982) Leff et al. (191) (1982) Polglase et al. (192) (1982) Vezeridis et al. (193) (1982) Anderberg et al. (194) (1983) Fegiz, Angelini, and Bezzi (195) (1983) Isbister, Beasley, and Dowle (196) (1983) Kennedy et al. (197) (1983) Resnick, Burstein, and Viner (198) (1983) Fazio (199) (1984) Hedberg and Helmy (200) (1984) Steichen and Ravitch (201) (1984) Fazio (202) (1984) McGinn et al. (203) (1985) Gillen and Peel (180) (1986) Malmberg et al. (166) (1986) Antonsen and Kronberg (204) (1987) Zannini et al. (205) (1987) Belli et al. (168) (1988) Dehong et al. (206) (1991) Steegmuller and Brown (207) (1991) Kyzer and Gordon (183) (1992) Moran et al. (88) (1992) Karanjia et al. (208) (1994) Fingerhut et al. (209) (1995)c Fingerhut et al. (209) (1995)d Detry et al. (210) (1995)

No. of Cases

Clinical

49 64 50 38 101 54 78 106 19 58 34 134 88 236 61 183 63 33 85 58 55 96 178 209 74 84 133 215 55 219 54 85 1000

12 9 6 11 3 11 9 8 16 0 12 16 15 3 3 3 3 6 1 12 6 14 15 9 4 5 3 0.4 9 11 4 0 4

Radiologic b b

Stenosis (%) 6 14

20

b

b

b

9 28

5 17 5 11 30 3

b b

36 b b

b

30

b

b

1

b

b

b

b

6 b

1 10

b

b

4 24 24

b

b

b

b

8 9 3

b b

b b

b

b

b

b

b

13 5

b

6 7 b b

b

16 5 0.5

a

Percentages have been rounded to the closest whole number. Not addressed. c Infraperitoneal anastomoses. d Supraperitoneal anastomoses. b

Several reports have shown that five-year survival rates for patients with carcinoma of the middle third of the rectum are at least as good or better with the use of a low anterior resection than with abdominoperineal resection (62,63,66– 68) It also has been reported that local recurrences are higher after abdominoperineal resections than after low anterior resections (66,227). Since anastomoses are being created at a lower level than possible before the introduction of the stapler, if the incidence of local recurrence after the use of the EEA stapler is shown to be higher than with the conventional hand-sutured anastomoses, it might be appropriate to compare at least some of these recurrences to recurrences in patients who have had abdominoperineal resection. In an effort to shed light on this controversy, Wolmark et al. (226) compared the incidence of local recurrence in patients undergoing stapled or hand-sewn anastomoses following curative resection of Dukes’ B and C colorectal carcinoma. There were 99 patients in the sutured group and 82 in the stapled group. The patients’ average time in the study was 41 months. Analyses of the distal resection margins were made in the two groups. For anterior resections the length of distal margin was 2.7
0.2 cm for hand-sewn anastomoses and 2.8
0.2 cm for those effected by the EEA instruments. No

significant difference emerged in the development of local recurrences as a first site of treatment failure when handsewn and stapled anastomoses were compared. If any trend was in evidence, it was in favor of the patient cohort with stapled anastomoses, in which the local recurrence was 12% compared with 19% for the hand-sewn group. The authors concluded that the use of stapled anastomoses does not compromise the patient. In their review of the literature (18 reports), Enker et al. (154) reported that conventional operations are associated with a worldwide incidence of pelvic recurrence of 30% and disseminated disease in 60% to 65%. Local recurrence has been reported to be related to depth of penetration of the bowel wall. Willett et al. (230) reviewed the clinical course of 64 such patients undergoing abdominoperineal resection. Follow-up revealed that all 12 patients with disease limited to the submucosa were free of recurrence. Recurrence developed in eight of the 52 patients with muscularis propria involvement. The six-year actuarial disease-free survival, local control, and freedom from distant metastases rates for these patients were 80%, 84%, and 85%, respectively. Croxford et al. (160) reported that the recurrence rate following TME is influenced by the level of the carcinoma.

240

& PART I: COLORECTAL DISORDERS

TABLE 6 & Local Recurrence with Circular Stapler

Author(s) Heald and Leicester (222) (1981) Hurst et al. (219) (1982) Anderberg et al. (194) (1983) Fegiz et al. (195) (1983) Isbister et al. (196) (1983) Luke et al. (223) (1983) Kennedy et al. (224) (1985) Leff et al. (225) (1985) Rosen et al. (221) (1985) Gillen and Peel (180) (1986) Malmberg et al. (166) (1986) Wolmark et al. (226) (1986) Carlsson et al. (227) (1987) Colombo et al. (179) (1987) Neville et al. (228) (1987) Zannini et al. (205) (1987) Belli et al. (168) (1988) Akyol et al. (229) (1991) Amato et al. (149) (1991) Dehong et al. (206) (1991) Kyzer and Gordon (183) (1992) Moran et al. (88) (1992)

No. of Cases

Follow-Up Mean or Range (mo)

Local Recurrence Rate (%)

40 32 34 102 63 79 63 70 76 55 96 82 40 61 76 108 74 152 40 84 215 55

16 7–25 5 9–38 ? 24 44 36 24 > 24 65 41 78 6–52 43 > 24 37 24 24 ? 4–124 24–84

3 32 21 9 14 22 36 11 21 15 18 12 35 10 24 18 4 19 13 4 13 7

Overall recurrence rate was 4% in a series of 480 patients with upper, middle, and lower third rates being 0%, 2%, and 7%, respectively. With the growing number of publications reporting results of TME or partial mesorectal excision, it might be appropriate to report their results separately and these are listed in Table 7. A caveat in interpreting these results is that in many of these publications there are a large number of exclusions which would make their survival rates artificially enhanced and local recurrence diminished over potentially previous reports of survival and local recurrence rates so that results are not necessarily comparable to the historical controls. For example, in the report by Heald et al. (232) exclusions resulted in a 50% decrease in local recurrence (6 to 3%) and a five-year survival improvement from 68% to 80%. With the usual emphasis on technical expertise, Cecil et al. (239) reported that following TME the local recurrence rate was 2% for Dukes’ A cases,

4% for Dukes’ B, and 7.5% for Dukes’ C. The systemic recurrence rate was 8% for Dukes’ A, 18% for Dukes’ B, and 37% for Dukes’ C. They believe if surgical priority is given to the difficult task of excision of the whole mesorectum, anterior resection with TME in node positive rectal carcinoma, recurrence rates of less than 10% can be achieved. Once again, case exclusion may have favorably biased results. Kockerling et al. (240), monitored 1581 consecutive patients who underwent curative resection (R0) for rectal carcinoma, for recurrence and survival. No patient received adjuvant radiotherapy or chemotherapy. The median follow-up time was greater than 13 years. The local recurrence rate decreased from 39.4% to 9.8% during the study period. The observed five-year survival rate improved from 50% to 71%. Three hundred and six patients with local recurrence had a significantly lower observed five-year survival rate. A total of 1285 patients had no local recurrence but 275 of them developed distant metastases

TABLE 7 & Operative Mortality, Survival and Local Recurrence Following Curative Mesorectal Excision for Carcinoma of Rectum

Author(s) Arbman et al. (231) (1996) Heald et al. (232) (1998) Hall et al. (233) (1998) Dahlberg et al. (234) (1999) Martling et al. (235) (2000) Kapiteijn et al. (236) (2001) Nesbakken et al. (237) (2002) Bulow et al. (238) (2003) Wibe et al. (161) (2004) a

Carcinoma-specific survival. Received preoperative radiotherapy.

b

No. of Potentially Curative Patients

Operative Mortality

Overall Survival (%)

Local Recurrence (%)

129 405 152 83 119b 381 937 924b 161 311 2136

2 3 3 2 – 3 3

68 (4 yr) 80 (5 yr) 68 (5 yr) 75 (5 yr)a – 91 (2 yr) 82 (2 yr) 82 (2 yr) 66 77 (3 yr) 63

4 (4 yr) 3 (5 yr) 11 (41 months) 12 3b 6 (2 yr) 8 (2 yr) 3 11 (5 yr) 11 (3 yr) 12 (5 yr)

3 7 3

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

(stage I 8%; stage II 16%; stage III 40%). Better quality operation had no effect on the incidence of initial distant metastases that remained constant. Local recurrence continues to be a major problem following surgical treatment for rectal carcinoma and proposed ways of reducing this remain controversial. McCall et al. (241), reviewed published surgical series in which adjuvant therapies were not used and follow-up on at least 50 patients was available. Fifty-one papers reported follow-up on 10,465 patients with a median local recurrence rate of 18.5%. Local recurrence was 8.5%, 16.3%, and 28.6% in Dukes’ A, B, and C patients, respectively, 16.2% following anterior resection, and 19.3% following abdomino-perineal resection. Nine papers (1176 patients) reported local recurrence rates of 10% or less. Local recurrence was 7.1% in 1333 patients having TME and 12.4% in 476 patients having extended pelvic lymphadenopathy. In 52% of cases, local recurrence was reported to have occurred with no evidence of disseminated disease. Surgical technique is an important determinant of local recurrence risk. Some reports include all patients with carcinoma of the rectum including those treated by abdomino-perineal resection or low anterior resection. Further difficulty in comparison of local recurrence rates is the very definition of local recurrence—local recurrence alone, or local recurrence combined with distant spread. The varying observation times also plays a role. In some series, the rate is calculated on the basis of all patients and others report curative resection only. Also, the stage of the disease should be considered as pointed out as in the study of Bulow where a higher proportion of patients with Dukes’ C carcinomas were in the control group (238). Another point to note is that leak rates of 8% to 16% are higher than should be expected following such operations. In the Hall report 73% of patients had a diverting stoma and mortality rates up to 7% were reported (233). What these reports do clearly emphasize is that careful oncologic operations can and should be performed by those specifically trained in the conduct of these procedures in a specialized unit and this should result in a reduction in the frequency in abdominoperineal resection, postoperative complications, mortality, local recurrence, and improved survival (233,234). The importance was further emphasized by the report of Nagtegaal (242). They reviewed the pathology reports of all patients entered into a Dutch multicenter randomized trial. A three-tiered classification was applied to assess completeness of the TME. Prognostic value of this classification was tested using the data of all patients who did not receive any adjuvant treatment. In 24% of 180 patients, the mesorectum was incomplete. Patients in this group had an increased risk of local and distant recurrence 36.1% versus 20.3% recurrence in the group with complete mesorectum. In their very comprehensive study, Wibe et al. (161) examined the outcome of carcinoma of the lower rectum, particularly the rates of local recurrence and survival for carcinoma located in this area that have been treated by anterior or abdominoperineal resections. The prospective observational national cohort study that is part of the Norwegian Rectal Cancer Project, included all patients undergoing TME in 47 hospitals. A total of 2136 patients

241

with rectal carcinoma within 12 cm of the anal verge were analyzed; 62% anterior resection and 38% abdomino-perineal resections. The lower edge of the carcinoma was located 0 to 5 cm from the anal verge in 791 patients, 6 to 8 cm in 558 patients, and 9 to 12 cm in 787 patients. According to the TNM classification, there were 33% stage I, 35% stage II, and 32% stage III. The five-year local recurrence rate was 15% in the lower level, 13% in the intermediate level, and 9% in the upper level. It was 10% local recurrence after anterior resection and 15% after abdomino-perineal resection. Fifty percent of recurrences occurred within 19 months and 90% within 40 months. The five-year survival rate was 59% in the lower level, 62% in the intermediate level, and 69% in the upper level, respectively, and it was 68% in the anterior resection group and 55% in the abdominoperineal resection group. Overall survival declined with increasing age with mortality rate of those greater than 79 years, six times those less than 50 years. Females survived better than males. The level of the carcinoma influenced the risk of local recurrence but the operative procedure anterior resection versus abdomino perineal resection did not. On the contrary, operative procedure influenced survival but level of the carcinoma did not. In addition to patient and carcinoma characteristics (T4 lesions), intraoperative bowel perforation that occurred in 9% of patients (16% APR, 4% anterior resection) and positive circumferential margins that occurred in 8% of patients (12% after APR, 5% after anterior resection), fiveyear survival negative circumferential margins (66% vs. 31% with positive margin) were identified as significant prognostic factors which were more common in the lower rectum explaining the inferior prognosis for carcinoma in this region. At 44 months follow-up, the observed local recurrence rate was 23% after perforation and 9% in patients with no perforation. They concluded that T4, R1 resections and/or intraoperative perforation of the carcinoma of bowel wall are main features of low rectal carcinoma causing inferior oncologic outcomes for malignancies in this area. If the operation is optimized preventing intraoperative perforation and involvement of the circumferential margin, the prognosis for carcinoma of the lower rectum seems not to be different from that of carcinomas at higher levels. In that case, the level of the carcinoma or the type of resection will not be indicators for selecting patients for radiotherapy. Optimizing surgical techniques (TME) for rectal carcinoma can reduce the rate of local recurrence and increase overall survival. This was clearly exemplified by the results of the Norwegian Registry (243). In a national audit for the period of 1986 to 1988, 28% of patients developed local recurrence following treatment with a curative intent. Fiveyear overall survival was 55% for patients younger than 75 years. In 1994, the Norwegian Rectal Cancer Group was founded with the aim to improve the surgical standard by implementing TME on a national level. A rectal carcinoma registry recorded 5382 patients with a carcinoma located within 16 cm of the anal verge and 3432 patients underwent rectal resection with a curative intent. Of these, 9% had adjuvant radiotherapy and 2% were given chemotherapy. There was a rapid implementation of the new technique as 78% underwent TME in 1994 increasing to 96% in 1998. After 39 months mean follow-up, the rate of

242

& PART I: COLORECTAL DISORDERS

local recurrence was 8% and five-year survival was 71% for patients younger than 75 years. Specialization of surgeons, feedback of results, and a separate rectal carcinoma registry are thought to be major contributors to the improved treatment. Ike et al. (244) reviewed the medical records of 71 patients with T3 or T4 primary rectal carcinomas who underwent a curative pelvic exenteration. The postoperative mortality, hospital death, and morbidity rates were 1.4%, 4.2%, and 66.2%, respectively. The overall five-year survival rate after total pelvic exenteration was 54.1%. The five-year survival rate was 65.7% for patients with T3 lesions and 39% for patients with T4 lesions. Postoperative survival was affected by age, stage of disease, and lymph node metastases. They concluded total pelvic exenteration might enable long-term survival in young patients with stage T3 and T4 primary rectal carcinomas and little or no lymph node metastases. Manfredi et al. (245) determined factors influencing local recurrences from the cancer registry of the Cote d’Or (France). The five-year cumulative recurrence rate was 22.7%. The two variables significantly associated with local recurrence risk were stage at diagnosis and the macroscopic type of growth. There was a nonsignificant decrease in the local recurrence rate in patients treated by preoperative radiotherapy compared with that in the patients treated by operation alone. The proportion of patients re-resected for cure was 25.2%, an increase from 13% in 1976 to 1985 to 37.9% in 1986 to 1995. Killingback et al. (158) examined local recurrence after curative resection for carcinoma of the rectum in which operative technique of TME was not performed. Total excision of the distal mesorectum was not performed in the upper third or mid rectum. Curative resections were performed in 549 patients of whom 17 died postoperatively (3%). Sphincter saving resection was performed in 88% and abdominoperineal resection in 11%. The pathology stages (Dukes’) were A, 29.7%, B, 34.7%, C, 35.7%. Follow-up for a minimum of five years was recorded in 97.2% of patients. The median period of follow-up was 82 months. Local recurrence confined to the pelvis occurred in 3.2% of patients, and local recurrences associated with distant metastases occurred in 4.5% of patients. The total five-year local recurrence rate was 7.6%. Local recurrence was increased in stage C carcinomas. Diathermy dissection in the pelvis was associated with a decreased local recurrence rate. The five-year survival rate in curative resection was 72.5%. Bonadeo et al. (246) assessed the local recurrence rate and prognostic factors for local recurrence in patients undergoing curative anterior or abdomino perineal resections without radiotherapy in 514 consecutive patients. In 417 patients, postoperative chemotherapy was limited to patients with stage III lesions. The five-year local recurrence rate was 9.7% with a median time to diagnosis of 15 months. Local recurrence rates in stage I, II, and III were 3.1%, 4.1%, and 24.1%, respectively. In relation to node status, local recurrence rates were N0, 4.1%; N1, 12.6%; N2, 32.1%; and N3, 59.3%. Lower third carcinomas had a higher local recurrence rate than middle and upper third lesions 17.9%, 7.1%, 5.1%, respectively. Zaheer et al. (157) reported on the outcome of patients undergoing resection of rectal carcinoma achieved by

abdominoperineal resection, coloanal anastomosis, anterior resection without adjuvant therapy. Among 514 patients who underwent operation alone, abdominoperineal resection was performed in 169 patients, coloanal anastomosis in 19, anterior resection in 272, and other procedures in 54. Eighty-seven percent of patients were operated on with curative intent. The mean follow-up was 5.6 years; followup was complete in 92%. Abdominoperineal resection and coloanal anastomosis were performed excising the envelope of rectal mesentery posteriorly and the supporting tissues laterally from the sacral promontory to the pelvic floor. Anterior resection was performed using an appropriately wide rectal mesentery resection technique if the carcinoma was high; if the carcinoma was in the middle or low rectum, all mesentery was resected. The mean distal margin achieved by anterior resection was 3 þ/- 2 cm. Mortality was 2%. Anastomotic leaks after anterior resection occurred in 5% and overall transient urinary retention in 15%. Eleven percent of patients had a wound infection. The local recurrence and five-year disease free survival rates were 7% and 78%, respectively, after anterior resection; 6% and 83%, respectively, after coloanal anastomosis; 4% and 80%, respectively, after abdominoperineal resection. Patients with stage III disease had a 60% disease-free survival. The overall failure rate of 40% in stage III disease means that surgical approaches alone are not sufficient to achieve better long-term survival rates. Vironen et al. (247) conducted a study to find out whether TME technique alone or combined with preoperative radiotherapy reduces local recurrence rate and improves survival. A conventional surgical technique was used during the first period (144 patients) and TME alone or combined with preoperative radiotherapy during the second period (61). After anterior resection (5%) during the first period, and 9% during the second period developed anastomotic leaks. Operative 30-day mortality was 1% and 0%, respectively. Actuarial local recurrence rate was 17% in the first period and 9% in the second period. Actuarial crude five-year-survival improved from 55% to 78% and carcinoma-specific survival improved from 67% to 86% between the two study periods. Parks (248) reported on 76 patients who underwent a resection with coloanal anastomoses. Pelvic sepsis following colonic necrosis and anastomotic breakdown developed in two patients. Pelvic sepsis without anastomotic breakdown developed in another eight patients. Local recurrence developed in 8%. Survival at the three-year follow-up examination was 66% and at five years was 63%. Good functional results were reported in 69 of the 70 patients. In a report from St. Mark’s Hospital in London, Sweeney et al. (249) reviewed 84 patients who underwent this operation. They found a low mortality of 2.4%, but a high complication rate with pelvic sepsis in 40.5% and anastomotic dehiscence in 47.6%. The crude five-year survival rate was 56%; local recurrence developed in 9.2% and was associated with systemic recurrence in another 9.2%. Functional results were deemed satisfactory in 92%. The authors concluded that for patients in whom the only alternative was abdominoperineal resection, their results showed no disadvantage in terms of potential cure, and functional results are acceptable. Enker et al. (250) reported on 41 patients in whom this technique was used. At 31 months,

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

75% were disease free. Functional results were reported to be good in 80% of patients. Similar encouraging results have been reported by others (251–255). Kohler et al. (256) reported the results of 31 patients with a very low localization of the carcinoma (distal margin 1.3 cm above the dentate line) who underwent a low anterior rectal resection with coloanal anastomosis. If the function of the sphincter was acceptable and they could exclude infiltration of the carcinoma into the sphincter through endosonography, they relocated the resection plane distally into the intersphincteric region to obtain an acceptable margin of safety. After intersphincteric rectal dissection, the anastomosis was hand sewn, using interrupted sutures from the perineal approach, 2.5 to 3 cm above the anal verge, implementing Parks’ retractor. A protective stoma was performed in all cases. Postoperative mortality was 0%. The leakage rate was 48%. Only 16% later needed additional operation for anastomotic strictures or for rectovaginal fistulas. Long-term observation showed that the anastomosis healed well in 87.1%. During the follow-up period of 6.8 years, 19.4% developed progression of their carcinoma (9.7% local recurrence and 12.9% distant spread). The five-year survival rate was 79% (Dukes’ A 100%, Dukes’ B, 67%, and Dukes’ C, 44%). Anorectal incontinence for liquid stool developed in 29.6% and for solid stool in 3.7% of patients. Average stool frequency was 3.3 times per day. Resting pressure decreased significantly by 29% whereas squeeze pressure did not change. They concluded, in selected patients with carcinomas close to the dentate line, and intersphincteric resection of the rectum may help to avoid an abdominoperineal excision of the rectum with a terminal stoma, without any curtailment of oncologic standards. Ho et al. (257) investigated the functional outcome in elderly patients following low anterior resection for carcinoma of the rectum. The study included 87 patients with carcinoma of the middle and lower rectum who underwent curative low anterior resection with TME and remained alive without recurrence for at least six months following the resection or closure of stoma. Anorectal manometry and questionnaire survey of the patients’ bowel function were performed during follow-up (median 24.1 months). The median number of bowel motions was 2.5 per day in both elderly and young patients. Complete continence was achieved in 71.3% of patients, with both elderly and young patients performing similarly. The most common symptoms were clustering of bowel motions and urgency, which occurred in 30.3% and 34.9% of patients, respectively, regardless of age. Manometric findings were also similar between the elderly and the younger patients. An elderly patient undergoing anterior resection for rectal carcinoma has a reasonably good expectation of acceptable continence. Phillips et al. (258) reviewed functional results one year after restorative operation in patients older than age 75 years. A total of 133 patients who had restorative anterior resection were alive at one year. Significant problems with bowel function or continence were denied by 85% of patients. One patient had already reported severe difficulty and had been given a definitive stoma for incontinence. The remaining 14%, although experiencing some problems with continence, did not consider the situation serious enough to contemplate a stoma.

243

Rengan et al. (259) conducted a trial to determine whether preoperative external-beam radiation therapy can increase the rate of sphincter preservation for patients with distal cT2N0 carcinoma of the rectum who refused an abdominoperineal resection. There were 27 patients with distal rectal carcinomas stage T2 by clinical and/or endorectal ultrasound who were judged by the operating surgeon to require an abdominoperineal resection and were treated with preoperative pelvic radiation alone (50.4 Gy). Operation was performed four to seven weeks later. The median follow-up was 55 months. The pathologic complete response rate was 15% and 78% of patients who underwent a sphincter-sparing procedure. The crude incidence of local failure for patients undergoing a sphincter sparing procedure was 10% and the five-year actuarial incidence was 13%. The actuarial five-year survival for patients undergoing sphincter preservation was as follows: disease-free, 77%; colostomy-free, 100%; and overall, 85%. Their data suggest that for patients with cT2N0 distal rectal carcinoma who required abdominoperineal resection, preoperative pelvic radiation improves sphincter preservation without an apparent compromise in local control or survival. Engel et al. (260) assessed rectal carcinoma patients’ quality of life by using the European Organization for Research and Treatment of Cancer QLQ-30 and CR38 questionnaires in 329 patients. Overall, anterior resection patients had better quality of life scores than abdominoperineal extirpation patients. High-anterior resection patients had significantly better scores than both low-anterior resection and abdominoperineal extirpation patients. Lowanterior resection patients, however, overall had a better quality of life than abdominoperineal extirpation patients, especially after four years. Abdominoperineal extirpation patients’ quality of life scores did not improve over time. Stoma patients had significantly worse quality of life scores than nonstoma patients. Quality of life improved greatly for patients whose stoma was reversed. Anterior resection and nonstoma patients, despite suffering micturation and defecation problems, had better quality of life scores than abdominoperineal extirpation and stoma patients. Schmidt et al. (261) assessed the differences in perceived quality of life over time among patients treated with anterior resection or abdominoperineal resection for rectal carcinoma. In a prospective study, the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire C30 and a disease-specific module were administered to patients with rectal carcinoma before operation, at discharge, and 3, 6, and 12 months after the operation. Comparisons were made between patients receiving an anterior resection and those receiving an abdominoperineal resection. Data were available for 212 patients of whom 112 were female and 100 male. No differences in the distribution of age, sex, or disease stage were observed between groups. EORTC function scales showed no significant differences, including body image scales, between patients receiving an anterior resection and those receiving an abdominoperineal resection. In symptom scores, anterior resection patients had more difficulty with diarrhea and constipation, whereas patients with abdominoperineal resection experienced more impaired sexuality and pain in the

244

& PART I: COLORECTAL DISORDERS

anoperineal region. At discharge, patients receiving an anterior resection were more confident about their future. Ranbarger et al. (81) reported on the prognostic significance of intraoperative perforation of the rectum during abdominoperineal resection. In their review of 250 patients, they found that 25.6% sustained an intraoperative perforation. Perforation did not affect the long-term outcome in patients with Dukes’ A carcinoma or those with metastatic disease, but patients with Dukes’ B and Dukes’C carcinomas experienced an increased recurrence (57% vs. 34%), most dramatically with Dukes’ B lesions (25.9% vs. 8.1%), and a decreased five-year survival rate (31% vs. 46%). Porter et al. (262) reviewed 178 patients who underwent an abdominoperineal resection, 42 (24%) of whom had an inadvertent perforation. Local recurrence was significantly higher in the perforated group than the nonperforated group (54% vs. 17%). Similarly, five-year survival was significantly decreased with inadvertent perforation (29% vs. 59%). The detrimental implications of inadvertent perforation mandate meticulous avoidance. Malignant fixation of carcinoma of the rectum is associated with a poor prognosis. Durdey and Williams (55) studied 625 patients who had undergone rectal excision and found that fixation was present in 27% (20% by malignant extension and 7% by inflammatory attachment). Corrected five-year survival rates were 28.5% in patients with malignant fixation, 68.9% in those with mobile lesions, and 64.6% when the carcinoma was tethered by inflammation. The incidence of local recurrence in the three groups was 41.3%, 15.1%, and 20%, respectively. Phang et al. (263) conducted a retrospective population-based study to determine the influence of emergent presentation (obstruction, perforation, massive hemorrhage) on outcomes. There were 452 invasive carcinomas of the rectum of which 45 were emergent and 407 nonemergent. Disease-specific survival at four years for emergent and nonemergent stage II carcinomas were 66% versus 80%, respectively, and for stage III carcinomas 60% versus 73%, respectively. Local recurrence rates at four years for emergent and nonemergent stage II carcinomas were 20% versus 15%, respectively, and for stage III carcinomas 70% versus 20%, respectively. Resection more frequently involved a stoma for emergent (60%) than for nonemergent (35%) cases. Percent of patients having complete staging investigations were similar between emergent (42%) and nonemergent patients (39%). Adjuvant radiation was given in similar proportion to emergent (61%) and nonemergent (55%) patients. Adjuvant chemotherapy was given to a slightly higher proportion of emergent patients (63%) than nonemergent patients (43%). Vironen et al. (264) compared perioperative morbidity, mortality, and survival after operation for rectal carcinoma in patients younger than and aged 75 years or older. Of 294 patients, 32% were aged 75 or older and comprised the elderly group. Major curative operation was possible in 62% of patients in the elderly group and in 74% of patients in the younger group. Among those operated on with curative intent, 34% of the older age group and 27% in the younger age group had complications. Thirty-day mortality was 2% and 0%, respectively. Although five-year crude survival was significantly lower in the older age group (43% vs. 65%), five-year carcinoma-specific survival

(60% vs. 70%) and disease-free five-year survival (60% vs. 69%) were similar in both groups. The 17 patients treated with local excision had a carcinoma-specific survival of 81% and 83% in younger and older groups, respectively. After palliative resection, the two-year survival was similar (20% vs. 25%) in both groups. They concluded that major curative rectal carcinoma surgery in selected elderly patients can be performed with similar indications, perioperative morbidity, and mortality, as well as five-year carcinomaspecific and disease-free survival as in younger patients. A very important aspect relates to careful operative technique. Eriksen et al. (265) reviewed the outcome of inadvertent perforation of the bowel during resection of rectal carcinoma. The study included a prospective national cohort of 2873 patients undergoing major resection for rectal carcinoma at 54 Norwegian hospitals. The overall perforation rate was 8.1%. The risk of perforation was significantly greater in patients undergoing abdominoperineal resection (odds ratio 5.6) and in those aged 80 years or more (odds ratio 2.0). The five-year local recurrence rate was 28.8% following perforation compared with 9.9% in patients with no perforation. Survival rates were 41.5% and 67.1%, respectively. The high local recurrence rates and reduced survival following perforation called for increased attention to avoid this complication. A recent review by Leong (266) of multiple studies has shown that specialization in rectal carcinoma surgery results in lower postoperative morbidity and mortality, local and distant recurrence, and higher rates of sphincter saving resections. Kressner et al. (267) investigated whether an abdominal or perineal septic complication was associated with rectal carcinoma recurrence in 228 patients. There was no clear difference in the overall incidence of recurrence between the infection group (36%) and the noninfection group (26%). The incidence of local recurrence was higher in the infection group (23%) than the noninfection group (9%). This increased risk was restricted to patients with a perineal infection (33% vs. the noninfection group), whereas patients with an abdominal infection (13%) did not differ from the noninfection group. Sexual dysfunction is a recognized complication in men undergoing pelvic surgery for rectal carcinoma but there is little information on the influence of such surgery on sexual health in women. Platell et al. (268) evaluated sexual health in 50 women undergoing pelvic surgery for rectal carcinoma. A control group comprised women who had undergone operation for colonic carcinoma during the same interval. Of the 50 women in the study group who were contacted, 22 completed questionnaires. Sixtytwo women in the control group were contacted and 19 completed questionnaires. Women in the study group were significantly younger than those in the control group. Compared with those of the control group, women who had undergone pelvic surgery were significantly more likely to feel less attractive, feel that the vagina was either too short or less elastic during intercourse, experience superficial pain during intercourse, and complain of fecal soiling during intercourse. Women in the study group were concerned that these limitations would persist for the rest of their lives. There were no differences between the two groups in relationship to sexual arousal or libido. They concluded that

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

pelvic surgery for rectal carcinoma has a significant influence on sexual health of women. Schmidt et al. (269) investigated how sexuality and quality of life are affected by age, gender, and type of operation in 516 patients who had undergone operation for rectal carcinoma. Quality of life data was received from 261 patients. For patients receiving abdominoperineal resection, sexuality was most impaired. Females reported more distress from the medical treatment: insomnia, fatigue, and constipation. Both genders had impaired sexual life; however, males had significantly higher values and felt more distressed by this impairment. Younger females felt more distress through impaired sexuality. In males sexuality was impaired independent of age. Adjuvant therapy had no influence on sexuality but on quality of life one year after operation.

& LOCAL FORMS OF THERAPY & RATIONALE The advantage of a purely local form of treatment for an early lesion is self-evident. Elimination of the necessity to recommend a major resectional operation, with its attendant morbidity and mortality, is most attractive. The theoretical disadvantage of a local form of therapy for carcinoma of the rectum is the failure to remove nodes along the inferior mesenteric vessels. However, not all these patients have lymph node involvement. With the high morbidity of abdominoperineal resection and low anterior resection, clinicians continue the search for a lesser procedure that would still be adequate for long-term survival. The goal is to identify patients with carcinoma that is still confined to the bowel wall without lymph node metastases (Dukes’ A). This accounts for about 15% of carcinomas of the rectum. The advent of intrarectal ultrasonography has gone a long way to accurately identify these patients. In their review of the literature, Sharma et al. (270) noted that depth of invasion provides the best estimate of the probability of lymph node spread this occurring in 0% to 12% of T1, 12% to 28% of T2 and 36% to 79% of T3 lesions. Additional factors considered important predictors of lymph node spread are histologic grade, lymphovascular invasion of the rectal wall, mucinous histology, and ulceration of the carcinomas. Proponents of local forms of therapy agree that even when local therapy fails, salvage by abdominoperineal resection can be expected with the same hope for cure as if it had been the initial form of therapy. For patients who are deemed unfit to withstand a major rectal excision or for patients who refuse an operation that necessitates a colostomy, a number of purely local forms of treatment have been used. & PROCEDURES Local Excision For a select small group of patients, perhaps 3% to 5% of all patients with carcinoma of the rectum, per anal local excision of the carcinoma may prove to be the treatment of choice (271,272). Criteria for the selection of patients for a local excision are not well defined. Ideally lesions should be in the distal third of the rectum, mobile, not deeply penetrating the wall, less than a third of the circumference

245

of the rectal wall, preferably polypoid rather than ulcerated, and well or moderately well differentiated, with no suggestion of anal sphincter involvement. Furthermore, there should be no palpable or radiologic evidence of regional lymph node involvement. Consideration is given to a local form of treatment because it has been estimated that the risk of regional lymph node involvement is only 10% in selected cases (273). In an effort to determine selection criteria for local excision, Minsky et al. (274) studied 168 patients who had undergone curative resection for carcinoma of the rectum (low anterior resection or abdominoperineal resection). Patients with T1 lesions (submucosal invasion) had a zero incidence of positive lymph nodes regardless of size, grade, histology, and vessel or lymphatic invasion. All other categories had a 19% incidence of lymph node positivity. Huddy et al. (275) investigated the incidence of lymph node involvement in 454 rectal excision specimens. For lesions that were confined to the bowel wall, lymph node metastases were identified in 20% although two-thirds had only one or two involved nodes. For lesions confined to the submucosa, lymph node involvement occurred in 11%. The availability of intrarectal ultrasonography should help to define the status of these lesions and hence provide a comfort level that the disease is confined to the bowel wall. Billingham (52) reviewed five series from the literature that encompassed 761 patients. The percentages of lymph node involvement were related to depth of invasion and were as follows: submucosa (12%), muscularis propria (35%), subserosa (44%), and serosa (58%). To determine the frequency of lymph node metastases, Blumberg et al. (276) reviewed 318 patients with T1 and T2 rectal carcinomas who underwent radical resection. Of these, 159 (48 T1 and 111 T2) were potentially eligible for curative local excision (less than or equal to 4 cm in size, less than or equal to 10 cm from the anal verge, no synchronous metastases). The overall frequency of lymph node metastases was 15% (T1, 10%; T2, 17%). Differentiation (well-differentiated or moderately differentiated, 14%; and poorly differentiated, 30%), and lymphatic vessel invasion (lymphatic vessel invasion-negative, 14%; and lymphatic vessel invasion-positive, 33%) influenced the risk of lymph node metastases. However, only blood vessel invasion (blood vessel invasion-negative 13% and blood vessel invasion-positive 33%) reached statistical significance as a single predictive factor. Carcinomas with no adverse pathologic features (low risk group) had a lower overall frequency of lymph node metastases (11%) compared with the remaining carcinomas (high risk groups 31%). However, even in the most favorable group (T1 carcinomas with no adverse pathologic features), lymph node metastases were present in 7% of patients. Nascimbeni et al. (277) reviewed the clinical records of 7543 patients who underwent operative treatment for carcinoma of the colon to determine the risk factors for lymph node metastases. For the depth, the submucosa was divided into upper third (sm1), middle third (sm2), and lower third (sm3). The incidence of T1 lesions was 8.6% (353 patients). The lymph node metastasis rate was 13%. Significant predictors of lymph node metastasis were sm3, lymphovascular invasion, and lesions in the lower third of the rectum. They concluded T1 colorectal

246

& PART I: COLORECTAL DISORDERS

carcinomas with lymphovascular invasion, sm3 depth of invasion and location in the lower third of the rectum have a higher risk of lymph node metastases. These lesions should have an oncologic resection. In a case of the lesion in the lower third of the rectum, local excision plus adjuvant chemotherapy may be an alternative. Sakuragi et al. (278) retrospectively analyzed 278 consecutive cases of T1 stage colorectal carcinoma resected using endoscopic resection or bowel surgery to seek predictive markers of lymph node metastasis to assist patient management. Twenty-one had lymph node metastasis. Depth of submucosal invasion (greater than or equal to 2000 mm) and lymphatic channel invasion significantly predicted risk of lymph node metastasis. When these two factors were adopted for the prediction of lymph node metastasis, sensitivity, specificity, positive predictive value, and negative predictive value were 100%, 55.6%, 15.6%, and 100%, respectively. They concluded, these two factors could be used in selecting appropriate cases for operation after endoscopic resection. Okabe et al. (279) evaluated the clinical significance of location as a risk factor for lymph node metastasis in 428 T1 adenocarcinomas of the colon and rectum treated by radical resection. Location was assigned as right colon (cecum to transverse), left colon (splenic flexure to sigmoid), or rectum (0–18 cm from the anal verge). The overall rate of lymph node metastasis was 10%. On univariate analysis, lymph node metastasis was significantly more common in the rectum (15%) compared to the left colon (8%), or right colon (3%). However, on multivariate analysis, deep submucosal invasion and lymphovascular invasion were independent and significant risk factors, whereas location was not. T1 colorectal carcinomas have a progressively higher risk of lymph node metastasis as their location becomes more distal. However, the increasing rate of lymph node metastasis observed in carcinomas of the left colon and rectum is explained by a higher prevalence of high-risk pathologic features. In early colorectal carcinomas, morphology of the carcinoma is the strongest clinical predictor of metastatic behavior. In a review of the literature, an enormously wide range of the incidence of lymph node involvement was cited: T1, 6% to 11%; T2, 10% to 35%, and T3, 25% to 65% (30). T1 lesions are usually amenable to local excision if they satisfy the above selection criteria. T3 lesions are clearly unsuitable because of the high incidence of lymph node metastases.

There is considerable controversy about the suitability of T2 lesions and this will be elaborated upon below. The advantage of local excision over other forms of local treatment is that it provides the opportunity to assess the diseased tissue histologically. Local excision also may be used for palliation in patients with overt disseminated malignant disease or in patients who are too old or too ill to withstand a major operation or who are unwilling to accept a colostomy. The technique of local excision is similar to that described in Chapter 19 for benign lesions. The major difference, of course, is that the full thickness of the rectal wall is excised. The patient is placed in the prone or supine position, depending on the location of the lesion. Exposure is obtained through a Parks or Pratt retractor. A full-thickness excision is made around the lesion, leaving a 1 cm rim of normal rectal wall. The defect is closed transversely with full-thickness mattress sutures of 4–0 polyglycolic acid or polyglactin (Fig. 24). Some surgeons prefer to leave the wound open (280). The excised specimen is pinned on a board to facilitate the histologic interpretation by the pathologist. Gopaul et al. (281) reviewed 64 patients treated with local excision for rectal carcinoma. The median follow-up was 37 months. There were 15 local failures with a median time to local failure of 12 months. Seven patients were salvaged with further operation (four by repeat local excision, four by abdominoperineal resection, and one by low anterior resection). The incidence of local recurrence increased with advancing stage of the carcinoma (T1, 13%; T2, 24%; T3, 71%), histologic grade of differentiation (well, 12%; moderately, 24%; poorly, 44%), and margin status [negative, 16%; close (within 2 mm), 33%; positive, 50%]. Sixteen percent of carcinomas less than or equal to 3 cm failed compared with 47% for carcinomas greater than 3 cm. Nine percent of T2 patients treated with adjuvant radiotherapy recurred locally compared with 36% without radiation therapy. Three of four T3 patients who received radiation therapy failed locally compared with two of three who did not. The overall survival at five years was 71%, and disease-free survival was 83%. Actuarial local failure was 27% and freedom from distant metastasis was 86%. The sphincter preservation rate was 90% at five years. They concluded, local excision alone is an acceptable option for well differentiated, T1 carcinomas less than or equal to 3 cm. Adjuvant radiation is recommended for T2 lesions. The high local recurrence rate in patients after local

FIGURE 24 & Local excision of carcinoma of rectum. (A) Exposure of carcinoma. (B) Full-thickness disk excision with perirectal fat exposed. (C) Completed closure.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

excision of T3 lesions with or without adjuvant radiotherapy would mandate a radical resection. Nascimbeni et al. (282) identified 114 patients who had T1 sessile adenocarcinoma in the lower third or middle third of the rectum to compare the long-term recurrence rate, long-term survival and risk factors for patients treated with local excision or oncologic resection. Patients who received adjuvant therapy or who had pedunculated lesions were excluded. The mean follow-up was 9.2 years; median follow-up was 8.1 years. They compared 70 patients who underwent local excision with 74 patients who underwent oncologic resection. Among patients with lesions in the middle or lower third of the rectum (i) the five-year and 10-year outcomes were significantly better for overall survival (90.4% vs. 72%) and disease-free survival (83.6% vs. 69.8%) in the oncologic resection group compared with 72.4% and 44.3% overall survival and 66.6% and 39.6 disease-free survival for local excision. There were no significant differences in local recurrence or distant metastasis; (ii) multivariate risk factors for long-term carcinoma-free survival were invasion into the lower third of the submucosa, local excision, and older than age 68 years; and (iii) for lesions with invasion into the lower third of the submucosa, the oncologic resection group had lower rates of distant metastasis and better survival. Among patients with lesions in the lower third of the rectum (i) the five-year and ten-year outcome showed no significant differences in survival, local recurrence, or distant metastasis between the two groups; and (ii) for lesions with invasion into the lower third of the submucosa, the oncologic resection showed a trend of improved survival which was not statistically significant, possibly because of the lower statistical power from the small sample size. They concluded, patients who undergo local excision or oncologic resection for T1 carcinoma in the lower two-thirds of the rectum have a high incidence of local recurrence and distant metastasis. A meta-analysis indicates that local recurrence occurs in 9.7% of patients (range 0–24%) in patients with T1 carcinomas, 25% (range 0–67%) of those with T2 carcinomas and 38% (range 0–100%) of those with T3 carcinomas (283). The overall results of local excision for carcinoma of the rectum are presented in Table 8. It is once again stressed that these series are not necessarily comparable because patient selection and degree of bowel wall penetration will differ. A general idea of possible results can be gleaned from these reports, however. Graham et al. (299) reviewed 10 published series in which local excision was used as definitive treatment for patients with invasive rectal carcinoma located within 6 cm of the anal verge. The overall five-year crude survival rate was 69% (range, 55–100%) and the cancer-specific survival rate was 89% (range, 82–100%). Local recurrence developed in 19% (range, 0–27%), but 56% (range, 22–100%) of these patients were cured with additional operations. The pathologic features of positive surgical margins, poorly differentiated histology, and increased depth of invasion were associated with increased local recurrence and decreased survival. Lock et al. (287) assessed the outcome of 152 patients who underwent local excision of their carcinoma of the rectum at St. Mark’s Hospital, London, England, between 1948

247

and 1984. The authors continued their institutional policy of local excision being justified if the carcinoma is well or moderately well differentiated and the lesion is judged completely excised. Following local excision, if the lesion is deemed incompletely removed, a radical operative procedure is indicated. In cases with high-risk lesions (e.g., those with deep penetration, poor differentiation but completely removed, or lymphatic involvement) adjuvant radiotherapy can be considered. Of 56 patients with lowgrade malignancy, three underwent early reoperation (low anterior resection). There was one death from carcinoma. Of the 81 patients with average-grade malignancy, 10 died from carcinoma—two after early reoperation and eight following recurrence. Of the 15 patients with highgrade malignancy, six underwent early reoperation and three of these died from carcinoma. Among the nine patients who did not undergo early reoperation, there were three related deaths (two from carcinoma, one after rectal excision for recurrence). The authors concluded that local excision is a valid therapy for patients with low-grade lesions, not recommended for those with average-grade lesions, and they questioned the necessity for early reoperation in patients with high-grade lesions. Green et al. (289) reviewed 104 patients who underwent diskectomy with intent to cure, representing 16% of patients with carcinoma of the rectum. Of the nine patients whose operation was converted to a more radical procedure because of failure to meet criteria for cure, five had no residual disease and two died from carcinoma. The remaining 95 patients were followed up from 1 to 152 months (mean 64.1 months). The overall recurrence rate was 22.4% (Tl, 10.6%; T2, 38.9%; and T3, 100%; well differentiated, 0%; moderately differentiated, 21.1%; and moderately to poorly differentiated, 58.3%). The authors concluded that transanal diskectomy remains an option for cure in selected patients. Muto and Oya (300) reviewed colorectal T1 (submucosally invasive carcinoma) in terms of pathology, diagnosis, and treatment. The incidence of T1 carcinomas was approximately 10% and the percentage of node metastases was 10% to 11%, whereas hepatic metastases was less than 1%. Macroscopically, they were divided into three types: polypoid, ulcerating, and flat. The ulcerating type was divided into two subgroups: polypoid growth and nonpolypoid growth based on the histologic appearance of resected specimens. The tools for detecting T1 carcinomas were fecal occult blood test and colonoscopy, and magnifying colonoscopy and chromography facilitate more precise diagnosis. Ultrasonography also was useful for the correct diagnosis of the depth of carcinoma invasion. Most polypoid and some flat T1 carcinomas were safely treated by polypectomy or endoscopic mucosal resection alone; however, when resected specimens contained risk factors for metastasis such as deep invasion (sm-2, sm-3) and vessel invasion, additional surgery was necessary for cure. For rectal T1 carcinomas, the most appropriate procedure should be carefully selected from several therapeutic options to preserve anal function. Chambers et al. (297) conducted a study to determine whether the morphology of 91 rectal carcinomas predicts outcome following treatment by local excision. Morphology was divided into four types: polypoid, sessile, ulcerated, and flat raised. Survival and local recurrence

248

& PART I: COLORECTAL DISORDERS

TABLE 8 & Results of Local Excision of Carcinoma of the Rectum

Author(s) Biggers, Beart, and Ilstrup (284) (1986) Heberer et al. (168) (1987) Horn, Halvorsen, and Morild (285) (1989) Killingback (286) (1992) Billingham (52) (1992) Lock, Ritchie, and Hawley (287) (1993) Willett et al. (288) (1994)

No. of Patients

Mean Follow-Up (Yr)

141

5.0

42 17b 14c 40 33 126

Local Recurrence (%)

Salvage Procedure Performed (%)

65

27

ND

5.0 4.3

84 100

7 0

ND

4.8 5.0

83 72 88 93

43 25 27 14

66 50 56 61

66

20

ND

> 5.0

Green et al. (289) (1994)

32b 22c 104

5.0

Faivre et al. (290) (1996) Obrand and Gordon (291) (1996) Taylor et al. (20) (1998) Chakravarti et al. (292) (1999)

126 19 24 52

7.3 5.0 4.3 4.3

59 37

4.0 6.0

Mellgren et al. (295) (2000)

108

4.4

Paty et al. (296) (2002)

125

6.7

D56 &35 53 70

5.0 5.0 30 9.2

Steele et al. (293) (1999) Varma et al. (294) (1999)

Chambers et al. (297) (2004) Gopaul et al. (281) (2004) Nascimbeni et al. (282) (2004) Madbouly et al. (298) (2005)

52

46

Five-Year Survival Ratea(%)

Tl 93 T2 83 T3 0 85 82 52 T1þT2 (66)e (74)f T1 85 (78)d T1 95 T2 83 T3 57 T1 72 T2 65 T1 74 T2 72 91 54 71 (83%)d T1 72.4 (66.6)d T1 75

11 39 100 28 26 46 28e 18f 7 5 45 25 o 18f 47 o 17 26 9 40 27 7 29

ND

69 60 70 78 84 50 100

89 82 ND ND 67

93

Note: ND, no data;, overall survival; D, exophytic; &, nonexophytic. Survival and recurrence data are of 10 years. 25% regional radiotherapy one-half of which also received chemotherapy. a Not certain which values are crude or corrected. b Invasion of submucosa only. c Invasion of muscularis propria. d Survival (disease-free survival). e Operation alone. f Adjuvant chemoradiation.

were significantly better for patients with exophytic (polypoid and sessile) carcinomas than for those with nonexophytic (ulcerated and flat raised) lesions. The exophytic group included significantly more stage T1 and fewer T2 and T3 carcinomas, and a significantly smaller proportion of carcinomas that showed venous and lymphatic invasion than the nonexophytic. Nonexophytic carcinomas were associated with high-risk histopathological features that rendered carcinomas of this type unsuitable for local excision. Paty et al. (296) reported on 125 patients with T1 and T2 carcinomas of the rectum treated by local excision as definitive operation. Thirty-one patients were selected to receive adjuvant radiation therapy. Fifteen of these 31 patients received adjuvant radiation in combination with 5 fluorouracil chemotherapy. Median follow-up was 6.7 years. Ten-year local recurrence and survival rates were 17% and 74% for T1 rectal carcinomas and 26% and 72% for T2 carcinomas. Median time to relapse was 1.4 years for local recurrence and 2.5 years for distant recurrence.

In patients receiving radiotherapy, local recurrence was delayed (median 2.1 years vs. 1.1 years), but overall rates of local and overall recurrence and survival rates were similar to patients not receiving radiotherapy. Among 26 carcinoma deaths, 28% occurred more than five years after local excision. No clinical or pathologic features were predictive of local recurrence. Intra-lesional vascular invasion was the only significant predictor of survival. Among patients who developed recurrence, the pattern of first clinical recurrence was predominantly local: 50% local only, 18% local and distant, and 32% distant only. Among patients with isolated local recurrence, 82% underwent salvage resection. Actuarial survival among these salvaged patients was 30% at six years after salvage. They concluded the long-term risk of recurrence after local excision of T1 and T2 rectal carcinomas is substantial. Two-thirds of patients with recurrence have local failure implicating inadequate resection in treatment failure. In their study, neither adjuvant radiotherapy nor salvage operation was

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

reliable in preventing or controlling local recurrence. The postoperative interval to death was as long as 10 years, raising concern that mortality may be higher than is generally appreciated. Additional treatment strategies are needed to improve the outcome of local excision. To evaluate nationwide patterns of surgical treatment and survival of early stage rectal carcinoma using data from the National Cancer Database, Baxter et al. (301) examined rectal carcinoma cases reported to the National Cancer Database during three time periods; 1989 to 1991, 1994 to 1996, and 1999 to 2001. Patients had T1 or T2 carcinomas that were clinically and pathologically node negative and no neoadjuvant therapy was received. A total of 25,815 patients meeting selection criteria were identified; 28% in the first period, 36% in the second, and 36% in the third. T1 lesions were identified in 30%. Rates of local excision increased over time, both for T1 lesions (22–46%) and T2 lesions (6–17%). Local excision was more common in T1 lesions, the elderly (greater than 75 years), African Americans, and women. The five-year survival of patients with T1 lesions was similar to those treated with local excision and radical resection; 77% versus 80%, however, survival was worse with local excision for those with T2 lesions; 66% versus 73%. The use of local excision for stage I rectal carcinoma is increasing both for T1 and T2 lesions. This does not appear to have negative consequences for T1 lesions; however, local excisions for T2 lesions were associated with worse five-year survival. Although some of this effect may be due to inadequate staging for local excision, at the current time they believe local excision for T2 lesions should be used with great caution. Sengupta and Tjandra (283) conducted a literature review of 41 studies on curative local excision of rectal carcinoma. Local excision preserves anorectal function and seems to have limited morbidity (0–22%). Local excision alone was associated with local recurrences in 9.7% (range 0–24%) of T1, 25% (range 0–67%) of T2, and 38% (range 0–100%) of T3 carcinomas. The addition of adjuvant chemoradiotherapy after local excision yields local recurrence rates of 9.5% (range 0–50%) for T1, 13.6% (range 0– 24%) for T2, and 13.8% (range 0–50%) for T3 carcinomas. Factors other than T stage that lead to higher local recurrence rates after local excision include poor histologic grade, the presence of lymphovascular invasion, and positive margins. Local recurrences after local excision can be surgically salvaged 84 of 114 patients (74%) in 15 studies, with a disease-free survival rate between 40% and 100% at a follow-up of 0.1 to 13.5 years. Madbouly et al. (298) reviewed their results of T1 low rectal carcinoma undergoing local excision alone. Patients with poorly differentiated carcinomas, perineural or lymphovascular invasion, or with mucinous component were excluded. Fifty-two patients underwent transanal excision during the study period. Five-year recurrence was estimated to be 29.4%, five-year carcinoma-specific and overall survival rates were 89% and 75%, respectively. Fourteen of 15 patients with recurrence underwent salvage treatment with 56.2% five-year survival. Gender, preoperative staging by endorectal ultrasound, distance from the anal verge, size of the carcinoma, location, and T1 status discovered after transanal excision of a villous adenoma did not influence local recurrence or carcinoma-specific

249

survival. This high recurrence and low salvage rate raises the issue about the role of transanal excision alone for early rectal carcinoma and the possible need for adjuvant therapy or increased role of the resective procedures. In a series of 19 patients on whom I performed a per anal local excision, there were no intraoperative complications and postoperative complications included urinary retention in one patient and bleeding in one patient (291). Local recurrence developed in 26% of patients but salvage operation was successful in 60% of these patients. The fiveyear corrected survival rate was 82%. In this review, neither size nor grade of the carcinoma correlated with recurrence. The recurrence rate was higher in patients with inadequate margins of excision and ulcerative lesions. Minsky (302) reported an increase from 6% to 56%. Killingback (286) reported increases from 23% to 36% if margins were positive for malignant cells. Of the 19 patients in our series, resection margins were involved in one patient, who died of other causes at 13 months without evidence of recurrence. Margins of normal tissue were noted to be small in four patients and unassessable in one. Recurrence occurred in three of these five patients. Hahnloser et al. (303) determined the frequency and outcome of radical resection (within 30 days) after local excision of rectal carcinoma. Fifty-two locally excised rectal carcinomas (29 transanal and 23 polypectomies) were followed by radical operation (24 abdominoperineal resections and 28 low anterior resections) within seven days. Radical operation was performed because of a malignant polyp (n ¼ 42), positive margins (n ¼ 5), lymphovascular invasion (n¼3), and T3 stage carcinoma (n ¼ 2). Twentythree percent were found to have nodal involvement and 29% showed residual carcinoma in the resected specimen. The T2-3 N0-1 stage controls were well matched. No significant difference in location, size, adjuvant therapy, or length of follow-up was noted. Length of follow-up was not different. For T1, the controls were also comparable. Nodal involvement was 21% in T1 study cases and 15% in T1 primary radical-surgery controls, with a trend toward location in the lower third of the rectum in both groups (58% and 50%, respectively). Local recurrence rates were 3% in the study group, 5% for patients undergoing primary radical-operation, and 8% for local excision alone. Distant metastasis (11%, 12%, and 13%, respectively) and overall five-year survival were also not significantly different (78%, 89%, and 73%, respectively). They concluded, nodal involvement in attempted locally excised rectal carcinomas is not uncommon. Local excision of rectal carcinoma followed by radical operation within 30 days in carcinoma patients does not compromise outcome compared with primary radical-operation. Conflicting reports regarding the results of salvage operations for local recurrence have been published. Cuthbertson and Simpson (304) achieved good salvage results by radical resection. On the other hand, Killingback (286) found that five of 10 patients with local recurrence were suitable for radical resection, and only one of these survived the malignant disease (dying of cardiac disease three years after resection). Friel et al. (305) determined the outcome of 29 patients who underwent salvage radical operation for local recurrence after a full thickness local excision for stage I rectal

250

& PART I: COLORECTAL DISORDERS

carcinoma. Recurrence involved the rectal wall in 90% and was purely extrarectal in only 10%. Mean time between local excision and radical operation was 26 months. The resection was considered curative in 79%. At a mean follow-up of 39 (range 2–147) months after radical operation, 59% remained free of disease. The disease-free survival rate was 68% for patients with favorable histology versus 29% for patients with unfavorable histology. Weiser et al. (306) examined surgical salvage of locally recurrent rectal carcinoma in 50 patients following initial transanal excision of T1 or T2 rectal carcinomas. Eight patients had resectable synchronous distant disease. Salvage procedures included abdominoperineal resection (31), low anterior resection (11), total pelvic exenteration (4), and transanal excision (3). One patient had unresectable disease at exploration, requiring diverting ostomy. Of the 49 patients who underwent successful salvage, 55% required an extended pelvic dissection with en bloc resection of one or more of the following structures: pelvic sidewall and autonomic nerves (18); coccyx or portion of sacrum (6); prostate (5); seminal vesicle (5); bladder (4); portion of the vagina (3); ureter (2); ovary (1); and uterus (1). Complete pathologic resection (R0) was accomplished in 47 of 49 patients. Of the eight patients with distant and local recurrence, two underwent synchronous resection and six had delayed metastasectomy. Five-year disease-specific survival was 53%. Factors predictive of survival included evidence of any mucosal recurrence on endoscopy, low presalvage carcinoembryonic antigen, and absence of poor pathologic features (lymphovascular and perineural invasion). Patients who required an extended pelvic resection had a worse survival rate. Other salvage rates are reported to be between 25% and 100% (291). Baron et al. (307) compared the impact on survival of ‘‘immediate’’ resection for adverse features versus ‘‘salvage’’ resection for clinical recurrence. Of 155 patients who underwent some form of local therapy, 21 patients underwent abdominoperineal resection and/or low anterior resection, whereas another 21 patients underwent salvage operation for local recurrence. The disease-free survival rate was 94.1% for the former and 55.5% for the delayed group. The authors therefore recommended that immediate operation be performed when adverse pathologic features are present in the excised specimen. Adjuvant therapy has been used in an effort to decrease the incidence of recurrence after local excision (308). Bennett et al. (309) reported a recurrence rate of 16% after the administration of 45 Gy to 19 patients who underwent local excision. Bailey et al. (19) reported on 53 patients who underwent local excision; 24 received full-dose radiotherapy postoperatively. With a follow-up ranging from 12 to 130 months, there was only an 8% recurrence rate. Minsky (302) reviewed the results from seven institutions, with series numbers ranging from 12 to 46 patients who underwent margin-negative local excision and postoperative pelvic radiotherapy. Some also received adjuvant chemotherapy. They found a local recurrence rate ranging from 6% to 18%. Of the three institutions that reported failure by stage, local recurrence developed in 3% of patients with T1 lesions, 10% of patients with T2 lesions, and 24% of patients with T3 lesions. Minsky therefore recommended

that T3 lesions be treated by extended resection with preoperative or postoperative radiotherapy. Wagman et al. (310) reported on 39 patients who underwent a local excision followed by postoperative radiation therapy þ or – 5-FU based chemotherapy. The median follow-up was 41 months and 11 patients had positive margins. The five-year actuarial colostomy-free survival was 87% and overall survival was 70%. Crude local failure increase with T stage: 0% T1, 24% T2, and 25% T3. Of the eight patients (21%) who developed local failure, five underwent salvage abdominoperineal resection and were locally controlled. Actuarial local failure at five years was 31% for T2 disease and 27% for the total patient group. In the 32 patients with an intact sphincter, 94% had good to excellent sphincter function. They believe local excision and postoperative therapy remains a reasonable alternative to abdominoperineal resection in selected patients. Lamont et al. (311) reviewed 27 T1 lesions treated with local excision, 10 of whom received postoperative chemoradiation, and no local recurrences were identified. Of 17 T1 patients who did not receive adjuvant treatment, local recurrence occurred in 24%. In all cases of local recurrence, lesions had been excised to negative margins, none were poorly differentiated, none exhibited vascular or lymphatic invasion. Their data suggested a trend toward improved local control with adjuvant therapy after local excision of T1 rectal carcinomas. Coco et al. (312) reported that the addition of radiotherapy in the treatment of patients with T2 lesions provided equivalent survival to T1 who underwent local excision without adjuvant radiotherapy. In their series of 22 patients with T1 and 15 patients with T2 lesions overall survival was 75%, whereas the corrected survival rate was 90%. Graham et al. (313) reported on the use of postoperative chemotherapy and radiotherapy, using 5-fluorouracil (5-FU) and leucovorin, with radiation in doses of 45 Gy plus a perineal boost of 9 Gy. Three of 14 patients so treated experienced grade 3 to 4 toxicity manifested by cystitis, proctitis, or perineal skin desquamation, but there were no local recurrences and only one case of distant metastases. Russel et al. (314) assigned 65 eligible patients into one of three treatment groups following local excision: observation, or adjuvant treatment with 5-FU fluorouracil and one of two different dose levels of locoregional radiation. Minimum follow-up was five years and median follow-up was 6.1 years. Locoregional failure correlated with T-category revealed: T1, 4%; T2, 16%; T3, 23%. Locoregional failure escalated with percentage involvement of the rectal circumference; 6% among patients with carcinomas involving 20% or less of the rectal circumference, and 18% among patients with carcinomas involving 21% to 40% of the circumference. Distant dissemination rose with T-category with 4% T1, 12% T2, and 31% T3. Actuarial freedom from pelvic relapse at five years was 88% based on the entire study population and 86% for the less favorable patients treated with adjuvant radiation and 5-FU. Chakravarti et al. (292) reported on 99 patients with T1 or T2 carcinomas, 52 patients treated by local excision alone, and 47 patients treated by local excision plus adjuvant irradiation. Twenty-six of these 47 patients were treated by irradiation in combination with 5 fluorouracil

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

chemotherapy. The five-year actuarial local control and recurrence-free survival rates were 72% and 66%, respectively, for the local excision alone group and 90% and 74%, respectively for the adjuvant irradiation group. This improvement in outcome was evident despite the presence of a higher risk patient population in the adjuvant irradiation group. Adverse pathologic features such as poorly differentiated histology and lymphatic or blood vessel invasion decreased local control and recurrence free survival rates in the local excision only group. Adjuvant radiation significantly improved five-year outcomes in patients with high-risk pathologic features. Furthermore, adjuvant radiotherapy extended the time to recurrence from 13.5 months to 55 months. The authors recommend adjuvant chemoradiation for all patients undergoing local excision for T2 carcinomas and for T1 carcinomas with high-risk pathologic features. The four cases of late local failures beyond five years in the adjuvant irradiation group underlines the need for careful long-term follow-up in these patients. The authors reviewed five other studies with local excision plus postoperative radiotherapy that had reported local control from 76% to 86% and five-year survivals, 75% to 85%. The Cancer and Leukemia Group B (CALGB) and collaborators (293) reviewed 177 patients who had T1/T2 carcinomas less than or equal to 4 cm in diameter, which encompassed less than or equal to 40% of the bowel circumference, and were less than or equal to 10 cm from the dentate line. Of the 177 patients, 59 patients who were eligible for the study had T1 carcinomas and received no further treatments; 51 eligible T2 patients received external beam irradiation (5400 cGY/30 fractions five days/week) and 5-FU after local excision. At 48 months median follow-up, six-year survival and failure-free survival rates of the eligible patients were 85% and 78%, respectively. Eight patients died of disease, four with distant recurrence only. One T1 patient is alive with distant disease. Two T1 and seven T2 patients experienced isolated local recurrences; all underwent salvage abdominoperineal resection. After APR, one T1 and four of the seven T2 patients were free of disease at the time of last visit (two to seven years). One T1 patient died of local and distant disease. Three of seven T2 patients died with distant disease. The authors concluded that sphincter preservation could be achieved with excellent carcinoma control without initial sacrifice of anal function in most patients. Our own experience with 64 cases, 11 of which received adjuvant radiotherapy revealed the incidence of local recurrence increased with advancing stage: T1 13%, T2 24%, T3 71% (281). However, the addition of postoperative radiotherapy reduced the local recurrence rate of T2 lesions to 9%. Overall and disease-free survival at five years was 71% and 83%. Salvage operations were performed in 67% of patients with locoregional recurrence and the five-year sphincter preservation rate was 90%. Predictors of local failure in our study were size greater than 3 cm, transmural penetration (T3), and positive marginal status. Taylor et al. (20) reported that the addition of adjuvant radiotherapy reduced their recurrence after local excision from 46% to 13% and the five-year disease-free survival increased from 52% to 81%. Varma et al. (294) reported that

251

none of their 19 patients who received adjuvant therapy, either chemotherapy or radiotherapy developed a local recurrence. Steele et al. (293) conducted a prospective multi-institutional trial to determine whether survival of patients with T1 and T2 carcinomas of the rectum who were treated with local excision would be comparable to historical controls treated with radical operation (APR). T1 patients (n ¼ 59) had no further treatment while T2 patients (n ¼ 51) received postoperative radiation (5400 cGY/30 fractions) and 5-FU (500 mg/m2 days 1–3 and 29–31). For T1 lesions, the six-year overall survival was 87% and disease-free survival 83% and the corresponding values for T2 lesions 85% and 71%. Patients who developed only local recurrence (T1 and T2) all underwent salvage APR. At 48 months comparison of their experiences with historical data from the National Cancer Database (American Joint Commission on Cancer) showed no significant therapeutic disadvantage in the sphincter-sparing approach. The follow-up was too short to determine whether patients who failed locoregional disease can be cured by salvage radical operation. At the end of their study, they believed their data allowed the following conclusions: (i) sphincter preservation is not easy to accomplish in a uniformly defined manner; (ii) sphincter preservation approaches are significantly less morbid than radical operations; (iii) predictors of outcome are not yet assessable because of relatively few recurrences; and (iv) salvage therapy effectiveness is not yet determined. Schell et al. (315) believe that patients who have initially bulky (T3) lesions and experience significant downstaging after neoadjuvant chemoradiotherapy, peranal excision appears to be a safe and effective treatment preserving sphincter function and avoiding laparotomy. They reported 74 patients diagnosed with locally advanced T3 rectal carcinomas. After neoadjuvant therapy, 14.9% of patients who had significant downstaging of their carcinomas were selected to undergo peranal excision of their residual rectal carcinomas. Carcinomas were located between 1 and 7 cm from the anal verge and were located in lateral, anterior, and posterior positions. Mean followup was 55.2 months. There were no local recurrences, nodal metastases, or operative mortalities. Nine percent developed distant metastases (pulmonary nodules) and lived 30 months after peranal excision. Ruo et al. (316) evaluated the clinical outcome of 10 selected patients with invasive distal rectal carcinoma (six T2 and four T3) treated with preoperative radiotherapy with or without 5-FU based chemotherapy. A full thickness local excision was performed four to six weeks after completion of radiotherapy, primarily because of co-morbid diseases or patient refusal or permanent colostomy. Median follow-up was 28.5 months. Only one positive microscopic margin was detected. Among three cases of complete pathological response, two remain without evidence of disease. All patients retained sphincter function and avoided creation of a stoma. Two patients developed recurrence, one with widespread disease including pelvic recurrence 26 months after operation and the other with distant disease only at 23 months. There were four deaths, two unrelated to carcinoma, one of undetermined cause after seven years and one after widespread recurrence at 26 months with death four months later. Two-year actuarial

252

& PART I: COLORECTAL DISORDERS

survival was 78%. This pilot study demonstrated that preoperative radiotherapy and full thickness local excision avoids major abdominal operation yet facilitates sphincter preservation, excision with negative margins and short-term local control in selected patients with distal rectal carcinoma.

Electrocoagulation Many carcinomas of the lower and middle thirds of the rectum are treated locally for cure with electrocoagulation by some surgeons. Details and results of this technique are described in Chapter 19. The results have been encouraging. Intracavitary Radiotherapy In 1973, Papillon (317) of Lyons, France, introduced a method for delivering high-dose radiotherapy to a small area in a short period of time. The technique, known as the Papillon treatment of contact radiotherapy, or intracavitary radiotherapy, is attractive in that it requires no operation. It can be delivered in an outpatient setting and is reported to have good results in selected patients. The radiation is delivered through a special 3 cm-diameter proctoscope. At each visit 3000 rads can be delivered, and lesions usually respond after the delivery of 9 to 15 Gy. Selected lesions should be within 10 cm from the anal verge, preferably not larger than 3 cm and certainly not more than 3  5 cm, in which case overlapping fields can be delivered. The lesions should be polypoid, well differentiated, or moderately well differentiated and have no evidence of regional lymph node involvement. With these restrictions, less than 10% of rectal carcinomas are suitable for intracavitary radiotherapy (318). In his most recent report of 310 patients, Papillon (318) noted a five-year survival rate of 76%, a local recurrence rate of 8.3%, and a death rate from carcinoma of 7.7%. The results of other investigators reviewed by Papillon reveal a local recurrence of 18% to 30%. For elderly poor-risk patients with TNM classification T2 or T3 carcinomas of the lower third of the rectum, Papillon established a protocol of a short intensive course of external beam irradiation (30 Gy over 12 days) followed two months later by intracavitary radiation. In a series of 71 patients followed up for more than three years, the death rate from carcinoma was 11%, and the death rate from intercurrent disease was 27%. At five years, the death rate from carcinoma was 16%. Sischy et al. (319) applied intracavitary radiotherapy successfully for a selected group of 192 patients and reported local control in 95% and five-year survival rates of 94%. They believe this form of treatment may be applicable in 15% to 20% of all rectal carcinomas. Hull et al. (320) reviewed 126 patients who received endocavitary radiotherapy with curative intent. With a mean time to recurrence of 16.1 (range 1–56) months, 29% of patients had a recurrence. Following additional treatments, 14 additional patients were rendered free of disease. They concluded endocavitary irradiation initially rendered 71% of patients free of disease. With additional treatment another 11% were rendered free of disease. In the subgroup of patients followed more than five years, 68% had no evidence of disease at follow-up after endocavitary

irradiation, and 91% had no evidence of disease with additional treatment. The size of the lesion, differentiation, morphology, and distance from the anal verge did not influence recurrence. Debulking or operative excision before endocavitary irradiation did not increase recurrence. Horiot et al. (321) reported on 200 patients treated with intracavitary radiotherapy. Local failure occurred in 4.4% of patients with T1 lesions and 19.5% in those with T2 lesions and nodal failure in 0.9% of T1 lesions and 9.2% of those with T2 lesions. Ultimate local control after salvage of failures was 94.5%. A functional sphincter was preserved in 95% of patients with local control. Reed et al. (322) reported 32 patients who received 75 to 120 Gy of endocavitary radiation (2–4 doses of 30 Gy at three-week intervals). The carcinomas were polypoid in 22 patients, sessile in five, and ulcerated in five. After a mean followup of 43 months (range, 6–103 months) four of five patients with ulcerated lesions developed recurrence (80%) compared to only 15% for sessile or polypoid lesions. Furthermore, time to recurrence was shorter in the group with ulcerated lesions. The authors recommended that these patients should be considered for operation initially. Schild et al. (323) evaluated the results of 20 patients treated with curative intent. Radiation doses ranged from 20 to 155 Gy and one to four fractions. In follow-up from 5 to 84 months (median, 55 months), local control was achieved in 18 of 20 patients (89%) and the five-year survival rate was 76%. We reported the clinical outcome of 15 patients with low rectal adenocarcinoma treated with the long source– skin distance of endorectal irradiation technique (324). This method was designed at McGill University in 1986 as an alternative to the standard short source–skin distance rectal irradiation that was developed by Papillon. The only preparation required is two cleansing enemas prior to treatment. Fourteen patients were treated with curative intent and one for palliation. The median total dose was 85 Gy (range, 60–135 Gy) in a median of three fractions (range, 3–5) over a median treatment time of five weeks (range, 2–9.5 weeks). With a mean follow-up of 39 months and a median of 24 months (range, 3 months to 8.7 years), actuarial overall survival and disease-free survival rates were 50.8% and 71.4%, respectively, at 8.7 years. Subsequent to that report, one patient died of distant metastatic disease. One patient treated with curative intent required an abdominoperineal resection for progressive disease. Treatments were tolerated well by all patients. Four patients required steroid enemas for localized proctitis for a short period of time. They all responded well and had complete resolution of symptoms. Our results are comparable with those in other reports in the literature, and the complications are similar in type and frequency to other published series. The long source–skin distance technique may be an acceptable alternative to the standard short source–skin distance technique. Winslow et al. (325) examined outcomes after salvage abdominoperineal resection for recurrence in 38 patients after endocavitary radiation. The mean time to recurrence after completion of endocavitary radiation was 21 months with 29% having persistent disease, 63% recurrent disease, and 8% a second primary. At abdominoperineal resection, 47% had transection of the carcinoma, specimen perforation, or injury to the genitourinary or gynecologic tract

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

and 24% had positive radial margins. The mean time to perineal wound healing was 56 days postoperatively with 36.8% taking more than 60 days. Re-recurrence developed in 45% of patients at a mean of 21 months after salvage, with a local control rate of 26% at 45 months of followup. Median disease-specific survival from completion of endocavitary radiation was 115.5 months with a five-year disease-specific survival rate of 66%. Patients with recurrent disease after endocavitary radiation had significantly better disease-specific survival than those with persistent disease (median survival 115 vs. 25 months). Although technically difficult and associated with a high morbidity, abdominoperineal resection salvage was possible in 55% of patients failing endocavitary radiation in this study.

Cryotherapy Cryotherapy is another form of local tissue destruction. However, the technique is associated with considerable malodorous discharge and post-treatment bleeding (326). It has not found general acceptance. A noted exception is the report by Heberer et al. (168), who treated 268 patients by cryotherapy as a primary procedure and 20 because of locoregional recurrence. Indications for palliative treatment were high operative risk in 67% and advanced incurable and unresectable disease in 21%. Fourteen patients refused a colostomy or radical operation, and 19 patients had multiple factors. Three deaths occurred during the treatment courses— all due to perforation. Two of the deaths resulted from necrosis of the carcinoma related to cryotherapy, and one was caused by perforation of the descending colon due to chronic obstruction by uncontrolled intrapelvic disease. Predictably, results were favorable for the early carcinomas, 77% of which disappeared with no evidence of disease. With respect to quality of life, cryotherapy enabled 80% of all patients to avoid a colostomy. For the 162 patients without evidence of metastatic disease, the five-year survival was about 35%. Laser Therapy Mathus-Vliegan (327) reported on the use of laser therapy for the management of carcinomas with endosonographic evidence of infiltration no deeper than the submucosa without evidence of lymph node involvement. Of the 15 patients who fulfilled these requirements, together with 15 patients treated in the presonographic era, complete eradication was achieved in 78 patients. No recurrence was seen in a median follow-up period of three years. Since the authors believe eradication is possible in greater than 90% of cases, they suggest that curative laser treatment be considered in highly selected cases. Its role in palliation was discussed in Chapter 23. External Beam Radiotherapy External beam radiation has been used as primary definitive therapy for selected cases of carcinoma of the rectum. Cummings (328) reported the use of 45 to 50 Gy in 144 patients. He reported an uncorrected five-year survival rate of 22%. The rates were 36% for patients with mobile lesions and 11% for those with fixed lesions. He suggested it is possible to offer initial high-dose radiation followed by a period of observation of up to six months (provided the

253

carcinoma is regressing) before recommending operative intervention. Sischy (329) has reported a similar experience. Most physicians, however, do not consider external beam radiotherapy the first-line recommendation.

& SPECIAL CONSIDERATIONS & DISTAL MARGINS Controversy has long raged over what should be the ideal, necessary, or even acceptable distal margin of resection. The distal margin of resection is determined by the likelihood of downward lymphatic spread and intramural spread of the carcinoma. Based on a better understanding of distal intramural spread, it would appear unnecessary to demand the traditional 5 cm distal margin, especially when it may mean sacrifice of the anal sphincter. Wilson and Beahrs (59) demonstrated that a 2 cm distal margin is probably adequate to encompass any distal intramural lymphatic spread. From the same institution Beart (330) reported that no significant differences appeared in 5- and 10-year survival rates in patients whose length of distal margin ranged from less than or equal to 2 cm to greater than or equal to 5 cm for those who underwent a low anterior resection. Several reports have suggested that the 5 cm distal margin is too stringent a requirement (10,58,254,331). Pollet and Nicholls (10) found no significant difference in either local recurrence or survival when distal margins of less than 2 cm were compared to distal margins greater than 2 cm. Vernava et al. (254) found that a distal margin of even 0.8 cm in the fresh unpinned specimen provided adequate clearance for most rectal carcinomas. In a review of 273 patients with sphincter-saving operations, Hojo (176) concluded that a 2 cm distal margin is adequate. Madsen and Christiansen (9) found in their series of 43 specimens that all potentially curable carcinomas would have been adequately resected with a margin of only 1.5 cm. Heimann et al. (332) found that patients having anterior resection with distal margins of less than or equal to 1 cm had an extremely high recurrence rate (36%) but that the rates of pelvic recurrence did not continue to improve when the distal margins were greater than 2 cm. In a labor-intensive study, Kirwan et al. (333) examined 7626 sections with a mean number of 381 per ‘‘doughnut’’ for 20 patients who underwent a low anterior resection. In 19 of the 20 patients, no malignant tissue was encountered. In the patient in whom carcinoma was identified, the perirectal lymphatics were involved, which the authors believed represented lymphatic rather than contiguous spread. The authors concluded that the classic 5 cm margin of resection is not necessary in low anterior resection for carcinoma. To ascertain the optimal distal margin of resection of sphincter-saving surgery, Shirouzu et al. (334) examined 610 consecutive patients who underwent resection for carcinoma. Distal spread was found in 10% overall, but in only 3.8% of patients who underwent curative operation. Distal spread was not seen in stage I; however, it was seen 1.2% in stage II, and 5.1% in stage III, but this was confined to within a 1 cm length. Most patients with distal spread

254

& PART I: COLORECTAL DISORDERS

had a lower survival rate and died of distant metastases rather than local recurrence. The authors concluded that a distal margin of resection of 1 cm might be appropriate clearance for most rectal carcinomas. Vernava et al. (335) also reported that distal margins as short as 0.8 cm do not prejudice cure nor do they increase the risk of local recurrence. Measurements were made in the fresh, not pinned, resected specimens. In a series of 243 patients, anastomotic recurrence rates for margins less than 0.8 cm were 30% vs. 10.5% for greater than 0.8 cm. The five-year survival rate in patients with distal margins less than 0.8 cm was 49.3% vs. 67.5% in those with greater distal margins. In a study of 55 patients, Kwok et al. (336) found distal intramural spread of greater than 10 mm in only three patients, all of whom had advanced disease. Although most authors agree that the 5 cm distal margin is not necessary, dissenting opinions exist. Enker et al. (337) reported that for patients with Dukes’ B carcinoma with a resection margin of less than 5 cm, the recurrence rate was 20% compared to 7% if the margin was greater than 5 cm. For Dukes’ C lesions with margins less than10 cm, the recurrence rate of 37% compared to 7% if the margin was greater than10 cm. Perhaps the patients with distal margins greater than10 cm had upper rectal lesions and the survival differences were due to greater mesorectal excision. Tonak et al. (338) noted a local recurrence rate of 33% for margins less than 3 cm and only 13% for margins greater than 3 cm. Despite these reports, for patients with a well-differentiated or moderately distinct carcinoma, a 2 cm distal margin seems adequate. Review of previous studies reveals that 5% to 19% of rectal carcinomas will have intramural spread of greater than 1 cm (339). Moore et al. (340) determined the adequacy of a distal margin of less than or equal to 1 cm in 94 patients with locally advanced rectal carcinoma requiring preoperative combined modality therapy. Distal margin length ranged from 0.1 to 9.5 cm (median 2.0 cm) and did not correlate with local recurrence or recurrence free survival. Estimates of recurrence free survival and local recurrence at three years for the less than or equal to 1 cm versus greater than 1 cm and the less than or equal to 2 cm versus greater than 2 cm groups were not significantly different. Their data suggest that for patients with locally advanced rectal carcinoma undergoing resection and preoperative combined modality therapy distal margins less than or equal to 1 cm do not seem to compromise oncologic outcome. Nakagoe et al. (341) examined microscopic distal intramural spread in 134 consecutive specimens of resected rectal carcinoma. Distal intramural spread was noted in 24.6% of patients. Patients with distal intramural spread had a shorter disease-specific or disease-free survival time after curative operation than those without distal intramural spread. Most patients with distal intramural spread developed distant reccurrence. Distal intramural spread was an independent risk factor for poor prognosis in patients with rectal carcinoma. It has generally been reported that patients with poorly differentiated carcinoma exhibit a greater extent of distal intramural spread; hence a longer distal margin is required even if the resection entails an abdominoperineal resection. However, Elliot et al. (12) reviewed 42 patients with poorly differentiated carcinoma of the rectum and

found no significant difference in the five-year survival rate or in local recurrence between those patients who underwent abdominoperineal resection and those who underwent low anterior resection, despite the fact that the average distal margin of those who had a sphincter-saving operation was only 2.7 cm.

& CIRCUMFERENTIAL MARGINS A long overlooked factor in the development of local recurrence following resection for rectal carcinoma is the circumferential margin. Although the literature is replete with discussion and heated debate as to appropriate distal margins, Quirke et al. (342) emphasized the importance of obtaining an adequate circumferential margin in order to secure a cure. In the most recent report from that institution, Hall et al. (233) examined the prognostic significance of circumferential margin involvement in resected specimens after potentially curative operations. Of 152 patients having curative resection, 13% had carcinomas within 1 mm of the circumferential margin. After followup until death or a median period of 41 months, recurrent disease was seen in 24% of patients with a negative margin and 50% with a positive margin. Local recurrence, however, was not significantly different in the two groups (11% and 15%, respectively). They concluded that when mesorectal excision is performed, circumferential margin involvement is more an indicator of advanced disease than inadequate local surgery. Patients with an involved margin may die from distant disease before local recurrence becomes apparent. Wibe et al. (343) examined the prognostic impact of the circumferential resection margin on local recurrence, distant metastasis, and survival rates. From the Norwegian national population-based rectal carcinoma registry, 686 patients underwent TME with a known circumferential margin. None of the patients had adjuvant radiotherapy. Following potential curative resection and after a median follow-up of 29 (range 14–60) months, the overall local recurrence rate was 7%: 22% among patients with a positive resection margin and 5% in those with a negative margin (greater than 1 mm). Forty percent of patients with a positive margin developed distant metastases, compared with 12% of those with a negative margin. Therefore, what is the value of extending the distal margin for 2 to 5 cm if the lateral margins of resection are involved? Information on circumferential margin is important in the selection of patients for postoperative adjuvant therapy. & TOTAL MESORECTAL EXCISION In recent years the subject of TME has become increasingly discussed and promoted as the ‘‘gold standard’’ in operative technique when operating for cure upon patients with carcinoma of the rectum. A growing number of publications have raved as to the efficacy and indeed the necessity for religiously adhering to the principle of TME in each and every patient who is to undergo a curative resection. Many of these publications can be classified as testimonials and hence it may be appropriate to review the recommendation and at least try to place it in proper perspective. To broach the subject it is important to consider the definition of what TME is, to review the mechanism of spread of carcinoma of the rectum with particular focus on how it pertains to TME,

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

255

to consider why TME has been deemed necessary in particular on how it relates to the incidence of local recurrence after radical resection, and finally, to weigh the available evidence to decide whether TME is really required in the operative management of each and every patient who suffers from carcinoma of the rectum (344). TME may be defined as complete excision of all mesorectal tissue enveloped in an intact visceral layer of the pelvic fascia that has no carcinoma at the lateral or circumferential margins. The procedure has been popularized by Heald since he and his colleagues first advocated the technique in 1982 (78). In fact, details of the operation (not labeled TME at the time) were described a half century earlier by Abel in his address entitled ‘‘The Modern Treatment of Cancer of the Rectum’’ to the Interstate Postgraduate Medical Assembly of North America in Milwaukee, Wisconsin October 21, 1931 (345). The salient components of the operation as recommended by Heald are as follows (Fig. 25). Meticulous dissection of the avascular plane between the mesorectum and parietes is completed under direct vision. Precise, sharp dissection is undertaken around the integral mesentery of the hindgut that envelops the entire midrectum. The excised specimen therefore includes the entire posterior, distal and lateral mesorectum out to the plane of the inferior hypogastric plexuses that have been carefully preserved. Anteriorly the specimen includes intact Denonvilliers’ fascia and the peritoneal reflection. In their most recent publication, Heald et al. (346) stressed the importance of dissecting anterior to Denonvilliers’ fascia until dissection is beyond the carcinoma in order to optimize outcomes in the technically challenging but rewarding operation. They describe the fascia as intimately adherent to the anterior mesorectal envelope. They describe a loose areolar tissue between Denonvilliers’ fascia and the seminal vesicles. The desire to dissect posterior to the fascia is because it is perilously close to the neurovascular bundles

and other authors believe this may be appropriate for posterior lesions (Fig. 26). The characteristic smooth, bilobed, encapsulated appearance posteriorly and distally reflects the contours of the pelvic floor and the midline anococcygeal raphe. This procedure adds to operative time and complications but is has been claimed that it eliminates virtually all locally recurrent disease after ‘‘curative’’ resection. By adopting this technique, the amount of the rectum with adequate blood supply above the levator is limited, and only 3 to 4 cm can safely be used for anastomosis. This may explain the high leak rate reported by Heald et al. (78,79) and their recommendation for diversion. Hermanek et al. (347) stressed that the pathological examination of resected rectal carcinoma should always include a visual assessment of the mesorectal excision to ensure oncologic adequacy. The clinical practice guidelines of the German Cancer Society recommends reporting the distal extent of mesorectal excision (total or partial without coning) and excision in an inviolate fascia envelope. Macroscopic evaluation of the resection specimen is supplemented by stain marking after postoperative filling the inferior mesenteric or superior rectal artery with ink or methylene blue solution. They believe the pathological assessment of adequacy of mesorectal excision should be taken into account in selection for adjuvant radiotherapy. Sterk et al. (348) performed specimen angiography to confirm completeness of the removed mesorectum. The rectal blood supply comes almost exclusively through the superior rectal vessels. Thus the fascia covering the mesorectum forms, as far as the rectal vascularization is concerned, a closed compartment. The mesorectal vessels are enclosed in the fibrous avascular mesorectal fascia. They run close above the fascia. In the case of an incomplete mesorectal excision, the specimen angiography shows a stain leaking from the mesorectal fascia. The rationale for proposing this operation is based on the observation that one of the mechanisms of spread of rectal carcinoma is retrograde extramural metastases. Pathologic assessment of resected specimens has been

FIGURE 25 & Total excision of mesorectum.

FIGURE 26 & The extent of resection for total mesorectal excision.

256

& PART I: COLORECTAL DISORDERS

conducted by a number of investigators. In the study of 100 specimens, Heald et al. (78), found distal spread in five patients the furthest being 4 cm distal to the mural lesion. Cawthorn et al. (349), also related the extent of mesorectal spread as a prognostic factor after operation for rectal carcinoma. From a study of 167 consecutive patients, the authors noted that the five-year survival rate was significantly greater in patients with slight mesorectal spread (less than 4 mm) than in those with more extensive mesorectal spread (55% vs. 25%). Scott et al. (350), noted that distal mesorectal spread often exceeds intramural spread and found distal mesorectal spread in 4 of 20 patients who underwent TME. Patients with deposits in the distal mesorectum had a worse outcome at four-year follow-up, a greater risk of local recurrence, and an increased frequency of distant metastasis. Choi (351) found distal spread in 11 of 53 specimens studied but the furthest deposit was only 2 cm away. Reynolds et al. (352) conducted a detailed pathologic evaluation of the mesorectum from 50 consecutive patients who underwent curative resection for rectal carcinoma. Of 21 patients with Duke’s B lesions, five had discrete foci of adenocarcinoma with no evidence of lymph node metastases, as did 12 of 23 patients with Duke’s C lesions. In 12 patients, the distal mesorectum was involved, and in one a mesorectal deposit 5 cm distal of the lower mural level of the carcinoma was retrieved. In five of these 12 cases, mesorectal involvement was greater than 2 cm below the mural level of the lesion. In five cases, overt mesorectal deposits were identified in the absence of mesorectal node involvement. It can thus be seen distal extramural retrograde spread has not been reported greater than 5 cm and rarely greater than 3 cm. Prabhudesai et al. (353) studied the impact of extranodal deposits on the outcome of rectal carcinoma in 55 patients whose resected specimens for rectal carcinoma were staged as Dukes C or Dukes B. Distant metastases were diagnosed earlier in patients with extranodal deposits (mean 14 months) compared with controls (mean 37 months). On follow-up, 31% from the extranodal deposit group developed liver metastases compared with 11.5% of the control group. Local recurrence was seen in 17.2% of patients from the extranodal deposit group and 3.8% of the control group. Carcinoma-related mortality was higher in the extranodal deposit group (16 vs. 7 patients). The three-year actuarial survival was 48.3% in patients with extranodal deposits and 65.4% in those without. A significant association was noted between the number of extranodal deposits and intramural vascular invasion, extramural vascular invasion, perineural invasion, and lymph node involvement. These data suggest that extranodal deposit is a distinct form of metastatic disease in patients with rectal carcinoma. The association with vascular invasion and earlier development of metastases probably infers that a significant proportion of extranodal deposits may represent blood born spread. These foci of malignant cells should be considered as indicators of poor prognosis. The rationale for advocating TME is that Heald believes local recurrence is the result of spread of carcinoma into the mesorectum. A very large range of local recurrences have been reported after radical resection for carcinoma of the rectum (3–36%) (Table 1, Table 6). McCall

et al. (354) in a review of 52 articles in the literature in which at least 50 patients were included (10,465 patients) found a median local recurrence rate of 18.5% for operation alone. The true rates are probably higher because in their review, patients up to 20 cm from the anal verge were included and hence these would in most series be considered colon carcinoma. Local recurrence was 8.5%, 16.3%, and 28.6% in Dukes’ A, B, and C patients, respectively, 16.2% following anterior resection and 19.3% following abdomino-perineal resection. Nine papers (1176 patients) reported local recurrence rates of 10% or less. Routine cytocidal stump washout in 1364 patients was associated with 12.2% local recurrence. Although not a comparative value in light of the other numbers, it did not appear to be very helpful. Others also support the adoption of TME (355,356). Bokey et al. (357) have cited numerous reasons for the wide range of reported local recurrence rates. These include case mix, surgical expertise, definition of recurrence (pelvic only vs. pelvic and distant metastases), incomplete histopathology reporting, lack in clarity in definition of curative operation, variable length of follow-up, variability in data collection, and variability in calculation of local recurrence rate. Notwithstanding these difficulties, it is important to seek ways of decreasing the incidence of local recurrence. In a 1993 report of patients of MacFarlane et al. (79) an independent analyst reviewed 135 patients with Dukes’ B and C carcinomas of the rectum who underwent curative resection (126 anterior resections and nine abdominoperineal resection). With follow-up over a 13-year period, the actuarial recurrence rate at five years was 4% and the overall recurrence rate was 18%. Ten-year figures were 4% and 19%. The operative mortality was 2%. The five-year disease-free survival was 78% (Dukes’ B, 85%; Dukes’ C, 68%). Of nagging concern in the tabulation of their data was the 25% exclusion rate of what was considered to be incurable or believed to harbor residual disease on the sidewall of the pelvis. The authors further reported that the results from TME alone were substantially superior to the best results reported by the North Central Cancer Treatment Group (358) (NCCTG) with conventional operation plus radiotherapy or combination chemoradiotherapy: 5% local recurrence at five years compared with NCCTG 25% (radiotherapy) and 13% (chemoradiotherapy), respectively; and 22% overall recurrence compared with NCCTG 62.7% (radiotherapy) and 41.5% (chemoradiotherapy), respectively. These large differences are somewhat hard to reconcile and may imply that residual disease remained more often in the NCCTG patients than those who underwent TME. Possibly some undefined reason may have contributed to the difference. I mention this because the pathologic studies conducted to determine the incidence of metastatic disease in the mesorectum distal to the mural lesion (i.e., the extramural retrograde metastases) reveal a relatively low incidence that would not account for the difference between the Heald and the NCCTG reported incidences of local recurrence. TME results are also superior to the National Surgical Adjuvant Breast and Bowel Project (NSABP), in which a five-year recurrence rate of 70% was seen in a similar group treated by operation alone (359).

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

Enker et al. (154) reported on 246 patients who underwent resection of rectal carcinoma according to the principles of TME (low anterior resection, 69%; abdominoperineal resection, 31%). The operative mortality rate was 0.8%. The overall five-year survival rate was 74.2% (86.7% for patients with Dukes’ B and 64% for patients with Dukes’ C lesions). Adjuvant radiotherapy with or without chemotherapy was administered to 70 patients. Pelvic recurrences were observed in 7.3%. They believe TME is compatible with both autonomic nerve and sphincter preservation, thus preserving sexual and urinary function in most patients. Furthermore, statistics revealed that adjuvant radiation therapy was of no benefit in preventing local recurrence. The investigators’ results led them to question the current role of combined-modality adjuvant therapy in patients who have undergone resection in accordance with TME. They noted that various trials of adjuvant radiotherapy have reduced the incidence of local failure to approximately 15% but that such adjuvant treatment has no impact on survival. The results of chemoradiation following conventional resection do not approach the rates achieved by TME. Furthermore, the morbidity and mortality of adjuvant treatment are frequently ignored. Situations these authors believe appropriate for adjuvant treatment include (i) patients with marginally resectable or unresectable disease; (ii) patients with N2 disease, perineural invasion, or multiple adverse pathologic features in the mesorectum; and (iii) patients with T3, Nl, M0 disease. A summary of local recurrence of other authors reporting on TME have ranged from 0% to 13% with most in the 6% to 9% range (344). Wibe (360) who reviewed 3516 patients in the Norwegian Cancer Registry, reported on the largest series of patients undergoing TME. After a mean follow-up period of 36 months, local recurrence rates were 6.4% for low anterior resection and 8.2% for abdominoperineal resection compared to 11% and 14.3%, respectively, after conventional operation. Noteworthy was the fact that local recurrence rates in 25 departments operating on more than 10 patients per year varied from 1.8% to 20.6%. Ridgway and Darzi (361) reviewed local recurrence rates following TME from selected series with more than 50 patients combining 13 studies with 6058 patients followed two to 10 years. The average local recurrence rate was 6.6%. In a recent review, the five-year survivals have ranged from 64% to 83% (95). In an effort to even further decrease the incidence of local recurrence, several investigators have added adjuvant therapy to TME. Local recurrence rates in these reports have ranged from 0% to 4.0% (344), but the adjuvant therapy has varied from series to series and from patient to patient within any given series and follow-up has been relatively short. In their comprehensive most recent review of the literature, Colquhoun et al. (362) analyzed seven case series that employed TME and found that with a follow-up of one to seven years local recurrence rates varied from 5% to 9%. In one study in which postoperative radiotherapy was employed, the local recurrence rate was reduced to 1.5% with a follow-up of two years. No meaningful conclusion can be drawn from these reports only to say that there may be advantage to add adjuvant therapy to TME. The cost of improved survival is not inconsequential. Procedures require more operating room time, transfusion

257

requirements have increased, and anastomosotic leak rates have increased to 17.4% (11%, clinical: 6.4%, radiologic) to the point that MacFarlane et al. (79) now routinely fashion a protective colostomy. In their detailed analysis of these later complications, there were 11% of major anastomotic leaks associated with peritonitis or a pelvic collection and 6.4% minor leaks that were asymptomatic and detected by contrast enema. All major leaks occurred at an anastomotic height of less than 6 cm (208). Major leaks were associated with failure to function in 11 of 62 patients with a defunctioning colostomy in 13 of 157 patients. Of the 24 patients with major leaks, three died and only nine patients with major leakage and a temporary stoma have had these closed. Nesbakken et al. (363) evaluated the long-term functional outcome after anastomotic leakage in the treatment of rectal carcinoma in 92 patients who underwent low anterior resection with TME. Eighteen percent developed clinical anastomotic leakage. The functional outcome of 11 of 12 patients in whom a stoma was subsequently closed and bowel continuity was restored without stricture was compared with that of 11 matched patients who had undergone low anterior resection without leakage. Nine patients made an uneventful recovery after the initial treatment of anastomotic leakage. Eight developed serious septic complications, of whom, four had a pelvic abscess. Five patients had chronic complications that precluded closure of the stoma. Patients who had experienced leakage showed reduced neo-rectal capacity, more evacuation problems and a trend to more fecal urgency and incontinence than control patients. Nesbakken et al. (364) examined the frequency of urinary and sexual dysfunction following rectal excision for carcinoma. Spontaneous flowmetry, residual volume of urine measurement and urodynamic examinations including cystometry and simultaneous detrusor pressure and urinary flow recording, was carried out before and three months after curative rectal excision. Forty-nine consecutive patients, 39 of whom had a TME and 10 of partial mesorectal excision were examined before operation at 35 again after operation. In two patients, a weak detrusor was detected before operation. Two patients developed signs of bladder denervation after operation. Transitory moderate urinary incontinence appeared in four other women. Six of 24 men reported some reduction in erectile function and one became impotent. Two men reported retrograde ejaculation. All the complications were seen in the TME group. Kim et al. (365) assessed the safety of TME with pelvic autonomic nerve preservation in terms of voiding and sexual function in males with rectal carcinoma. They performed urine flowmetry using Urodyn and a standard questionnaire using the International Index of Erectile Function and the International Prostate Symptom Score before and after surgery in 68 males with rectal carcinoma. Significant differences in mean maximal urinary flow rate and voided volume were seen before and after operation but no differences in residual volume before and after operation were apparent. The total International Prostate Symptom Score was increased after operation. Five International Index of Erectile Function domain scores (erectile function, intercourse satisfaction, orgasmic function, sexual desire, and overall satisfaction) were significantly decreased after operation. Erection was possible in 80.9%; penetration

258

& PART I: COLORECTAL DISORDERS

ability was possible in 75%. Complete inability for erection and intercourse was observed in 5.5%. Retrograde ejaculation was noted in 13.2%. International Index of Erectile Function domains such as sexual desire and overall satisfaction were greatly decreased in 57.4% and 63.2%, respectively. Urinary dysfunction may occur after mesorectal excision and pelvic autonomic nerve preservation in patients with rectal carcinoma. Kneist et al. (366) identified factors predictive of long-term urinary catheterization in 210 patients without significant urologic problems who underwent resection of rectal carcinomas with mesorectal excision. Eight patients (3.8%) required long-term urinary catheterization: two after complete pelvic autonomic nerve preservation (2 of 168) and six in whom a pelvic autonomic nerve preservation was incomplete (6 of 42). Incomplete pelvic autonomic preservation (odds ratio 13.8) was a predictive factor for major urinary dysfunction. Peeters et al. (367) investigated risk factors associated with symptomatic anastomotic leakage after TME. Patients with operable rectal carcinoma were randomized to receive short-term radiotherapy followed by TME or to undergo TME alone. Eligible Dutch patients who underwent an anterior resection (924 patients) were studied retrospectively. Symptomatic anastomotic leakage occurred in 11.6%. Pelvic drainage and the use of defunctioning stoma were significantly associated with a lower anastomotic failure rate. A significant correlation between the absence of a stoma and anastomotic dehiscence was observed in both men and women, for both distal and proximal rectal carcinomas. In patients with anastomotic failure, the presence of pelvic drains and a covering stoma were both related to a lower requirement for a surgical reintervention. They concluded, placement of one or more pelvic drains after TME might limit the consequences of anastomotic failure. The clinical decision to construct a defunctioning stoma is supported by this study. A novel solution to the problem of increased stool frequency, urgency, and incontinence following TME was described by von Flu¨e et al. (368). An interpositional ileocecal graft based on its original blood supply was rotated counterclockwise and placed between the sigmoid colon and the anal canal. In 20 consecutive patients, at six-month follow-up, 16 patients showed excellent and four patients showed good defecation quality with maximal tolerable volumes, compliance, and mean colonic transit time comparable to age and gender matched healthy volunteers. Before fully endorsing TME other information must be considered. Quirke (342) emphasized the importance of adequate circumferential margins to secure cure. In a review of 141 patients undergoing curative resection 25% of patients were found to have involved margins. The hazard of local recurrence was 12.2 and for survival 3.2. For patients with a positive margin at five-year follow-up local recurrence was 78% compared with 10% for patients with negative margins. In a more recent report (233) the same group reviewed 152 curative resections and at a median follow-up of 41 months found a local recurrence of 11% if margins were negative compared with 15% if margins were positive. Overall recurrence was 24% for those with negative margins and 50% for those with positive margins. The authors now concluded that circumferential involvement

was more an indicator of advanced disease than inadequate local surgery. Patients with involved margins may die from distant disease before local recurrence becomes apparent. Lopez-Kostner (369) and colleagues compared the outcomes of the treatment of upper rectal carcinoma (n ¼ 229) in which TME was not performed with outcomes of sigmoid colon carcinoma (n ¼ 225) and lower rectal carcinomas (n ¼ 437). The risk of local recurrence alone, local and distant recurrence, death as a result of carcinoma or any recurrence or death as a result of carcinoma was 3.5, 2.7, 2.1, and 1.9 times higher for patients with lower rectal carcinoma than for patients with upper rectal carcinoma or sigmoid colon carcinoma. The outcome for upper rectal and sigmoid carcinomas was the same. Another important contribution was that of Bokey and colleagues (357) who reviewed a series of 596 patients who had no adjuvant therapy and had a five-year actuarial recurrence rate of 11%. In their study independent predictive factors for local recurrence were: (i) positive nodes (hazard ratio 5.5); (ii) distal margin less than 1 cm (3.8); (iii) venous invasion (2.6); and (iv) total anatomic dissection (2.0). Pertinent to this discussion these authors noted no difference in the local recurrence between those patients in whom the mesorectum was divided or fully excised. So in this morass of opinionated authors and plethora of often confusing data, what conclusions can the operating surgeon draw? It seems clear that from pathologic studies cited and the operative results obtained, that TME for the upper rectum and possibly even upper part of mid rectum is not necessary. For patients with carcinoma of the distal rectum, a ‘‘standard’’ dissection in the appropriate plane ‘‘automatically’’ includes a TME. The key would seem to be careful not to cone down through the mesentery but rather to proceed in the areolar plane outside the investing fascia of the rectum and when the appropriate distal margin is obtained a sharp incision is made perpendicular to the bowel thus including in the mesorectum any potential extramural retrograde spread. Indeed upon review of Heald’s original publication on the subject (78), it can be observed that he had stated that ‘‘this technique was not necessary for some upper rectal carcinomas which mobilized up so well that they could be resected like sigmoid lesions, i.e., the mesorectum was transected at least 5 cm below the carcinoma.’’ I certainly support Heald’s contention that precise operative technique is fundamental to a sound oncologic resection, but do not concur that TME is necessary for all patients undergoing operation for a rectal carcinoma. This dissertation in no way is to be misinterpreted as casting any disparaging remarks on Heald’s enthusiasm for the procedure. On the contrary, he is to be complimented for drawing our attention to the serious problem of local recurrence following operation for rectal carcinoma and suggesting a plan of action to diminish the problem.

& RADICAL LYMPHADENECTOMY The role of extended proximal lymphadenectomy will be considered in the discussion on high ligation of the inferior mesenteric artery. The role, if any, of radical pelvic lymphadenectomy is unclear. Advocates of the extended procedure claim improved survival. A report by Enker et al. (154),

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

in which TME was adopted, resulted in an improved survival rate of 74% and so they no longer adopt extended lymphadenectomy as a standard operative approach. A report by Michelassi et al. (148) found no statistical reduction in pelvic recurrence by the addition of pelvic lymphadenectomy. Hojo et al. (370) reported on 437 patients who underwent curative resection for carcinoma of the middle and lower rectum. Abdominoperineal resections were performed in 262 patients and low anterior resections in 175 patients. The overall pelvic recurrence rate was 12.4%. In a comparison of wide ileopelvic and conventional lymphadenectomy, local recurrence rates according to Dukes’ stage were 0% and 5.2% for Dukes’ A, 21.9% and 6.3% for Dukes’ B, and 32.8% and 23.6% for Dukes’ C, respectively. Corresponding five-year survival rates were 94% and 91% for Dukes’ A, 88% and 74% for Dukes’ B, and 61% and 43% for Dukes’ C patients. The operation, however, is associated with significant morbidity, which may include impotence, urinary bladder difficulties, possible vascular injury with subsequent deep venous thrombosis, and sciatic nerve injury. Hojo et al. (370) noted that the extent of operation increased intraoperative bleeding from 1500 to 1900 ML, the anastomotic leak rate from 16.9% to 22.8%, the incidence of long-term neurogenic bladder from 8.8% to 39.4%, and the incidence of impotence from 37.5% to 76%. In order to decrease the urinary and sexual morbidity that follows radical pelvic lymphadenectomy for rectal carcinoma, Hojo et al. (371) began selective preservation of the pelvic autonomic nerves. One hundred and thirty-four patients with rectal carcinoma underwent a curative resection (52 abdominoperineal resections, 82 sphincter-saving resections) with extended pelvic lymphadenectomy and selective pelvic autonomic nerve preservation. Pelvic autonomic nerve preservation was classified into five degrees depending on the extent of pelvic dissection. First-degree indicated complete preservation of the nerves; second-degree indicated destruction of the hypogastric plexus; third-degree indicated partial preservation of the pelvic autonomic plexus; fourth-degree indicated bilateral or unilateral preservation of only the fourth pelvic parasympathetic nerve; and fifth-degree indicated complete destruction of the pelvic autonomic nerves. Most patients with firstdegree preservation were able to void spontaneously 7 to 10 days following the operation. However, 28 of 36 patients (78%) with fifth-degree preservation had not regained bladder sensation by the third postoperative week and were discharged with an indwelling catheter; 21 of 36 (58%) had not regained bladder sensation by the sixtieth postoperative day. The cystometric data indicate a progressive decline in bladder sensation and function with increasingly extensive pelvic dissection. However, preservation of only the fourth parasympathetic nerve (fourth-degree preservation) resulted in partial sparing of bladder sensation and voiding function. Evaluation of sexual function in males below 60 years of age revealed that only 12 of 39 patients (31%) recovered erectile function and only six of 39 (19%) recovered normal ejaculatory function in the first postoperative year. In most of these patients, the pelvic autonomic plexus was preserved (i.e., first-degree preservation). Four patients with partial preservation recovered erectile function. Complete pelvic

259

autonomic nerve preservation is the best way to prevent urinary and sexual morbidity after rectal resection. The opposing goals of maximizing the chance for cure and minimizing morbidity must be individualized and balanced in each patient. By contrast, Glass et al. (372) analyzed the results of 75 patients who underwent extended pelvic lymphadenectomy in an attempt to improve the prognosis but found no improvement in survival. By comparing the results of current series to those of patients previously operated on in the hospital, the five-year survival rates of extended resection and conventional techniques by Dukes’ stage were Dukes’ A, 100% vs. 82.4%; Dukes’ B, 78.3% vs. 65.6%; Dukes’ C1, 29.2% vs. 40.4%; and Dukes’ C2, 25% vs. 22.4%. The authors concluded that extension of the conventional operation is unlikely to improve results. Moreira et al. (373) also failed to demonstrate a benefit in terms of local recurrence and survival rates. In contrast, Moriya et al. (374) reported promising results using nerve-sparing operations with lateral node dissection for advanced lower rectal carcinoma. With respect to technique, nerve-sparing procedures were distinguished as three types—total autonomic nerve preservation, complete preservation, and partial preservation of the pelvic nerve with lateral dissection. In the authors’ opinion, total autonomic nerve preservation should be applied to patients with Dukes’ A carcinoma, complete preservation should be used for patients with Dukes’ B, and partial preservation of the pelvic nerve with lateral dissection is appropriate for patients with Dukes’ C, on the basis of findings from endorectal ultrasonography. Cutting the lateral ligament of the rectum, which consists primarily of autonomic nerves, is the most intense phase during this nerve-sparing technique. The dissection should start along the internal iliac vessels and advance downward to the middle rectal artery, while removing the lymphatic tissue covering them. By meticulous sharp cuts of fascia on the piriform muscle, the surgeon can expose the root of the pelvic nerves. In 133 patients who underwent this nerve-sparing operation with lateral dissection for lower rectal carcinoma, the authors analyzed survival and functional results, operative burdens, and modes of recurrence. In 84% of patients, an acceptable level of urinary function was preserved. The five-year survival rate was 67% in all patients, 88% for Dukes’ A, 74% for Dukes’ B, and 59% for Dukes’ C. According to the number of positive nodes, the five-year survival rate comprised 83% of patients with up to three involved nodes, and 34% of those with more than four nodes. Local recurrence rates were 2.7% in patients with Dukes’ B and 13% with Dukes’ C. The authors believe at present, a pelvic nerve-sparing procedure with lateral dissection is the most promising operation, guaranteeing both adequate lymphadenectomy and preservation of urinary function. In the most recent review of the controversy, Morita et al. (375) continue to support the use of lateral node dissection and pelvic autonomic nerve preservation, but in a commentary following the article Moriya now believes that the number of patients who have suffered loss of function (impotence and bladder dysfunction) because of lateral node dissection is no match for the very rare patient who benefits from it. Hocht et al. (376) analyzed data to evaluate

260

& PART I: COLORECTAL DISORDERS

the pattern of recurrence in rectal carcinoma with special emphasis on lateral extension. Initially 54% of the evaluated patients were N0, and the others were distributed evenly between N1 and N2; initial T stage was T1 in 2%, T2 in 24%, T3 in 60%, and T4 in 13%. Recurrent carcinomas were situated mainly in the posterior part of the bony pelvis. The pelvic sidewall was a rare site of recurrence and involved in fewer than 5%. Because most recurrences arise in the central pelvis extending surgery to include dissecting the iliac vessels would probably offer only a moderate benefit, which must be balanced against potential side effects. Fujita et al. (377) investigated the oncologic outcome of patients who underwent curative operation for lower rectal carcinoma to clarify whether lateral pelvic node dissection conferred any benefit. A total of 246 patients who underwent curative operation for stage II and III lower rectal carcinoma were reviewed. Forty-two of these patients did not undergo lateral pelvic lymph node dissection. There was no difference in survival among patients with stage II and III disease between the two groups. However, in patients with pathological N1 lymph node metastases, the five-year disease free survival was 73% in patients who had lateral pelvic lymph node dissection compared with 35.3% among those who did not. They believe a randomized trial is needed to verify the benefit of lateral pelvic lymph node dissection. Nagawa et al. (378) evaluated the effectiveness of preoperative radiation therapy for advanced lower rectal carcinoma to preserve the function of pelvic organs and reduce local recurrences in a prospective randomized, controlled study. The patients were randomly allocated to complete autonomic nerve preserving surgery without lateral node dissection, or surgery with dissection of the lateral lymph nodes including autonomic nerves followed by oral administration of carmofur for one year. No difference was observed in either survival or disease-free survival between groups. There was no difference between the two groups in terms of recurrence rate. A significant difference was observed in urinary and sexual function one year after surgery. Their study suggests that lateral node dissection is not necessary in terms of curability for patients with advanced carcinoma of the lower rectum who undergo preoperative radiotherapy. Koda et al. (379) examined whether lateral lymph node dissection with or without preoperative chemoradiotherapy benefits patients with rectal carcinoma. A total of 452 consecutive cases of curatively resected pT2, pT3, and pT4 middle to lower rectal carcinomas were retrospectively analyzed. Of these, 265 patients underwent curative lateral lymph node dissection and 155 chemoradiotherapy. Lateral lymph node metastases were identified in 7.7% of patients. Of the pT3/pT4 extraperitoneal carcinoma patients, 13.5/18.8% had lateral lymph node metastases. In the treatment of middle rectal carcinomas and pT2 extraperitoneal carcinomas, lateral lymph node dissection either with or without chemoradiotherapy did not improve survival rate. Other surgeons have conceded that when lymph nodes on the lateral pelvic wall are implicated, the condition is essentially incurable, since removal of all involved tissue will probably not be complete. Hida et al. (380) investigated the therapeutic efficacy of lateral lymph node dissection. They studied 198 patients with rectal

carcinoma who underwent lateral lymph node dissection. The rate of metastases to lateral lymph nodes was 11.1% and metastases to the lateral lymph nodes occurred more frequently with lower rectal carcinoma classified as pT3 or pT4 in the TNM system. The rate of local recurrence was 12.5% and the five-year survival rate after curative resection was 70.1%. The five-year survival rate in patients with metastases to the lateral lymph nodes was 25.1% and this rate was significantly lower than the five-year survival of 74.3% in patients without metastases to the lateral lymph nodes. Urinary dysfunction was observed in 67.5% of patients, and male sexual dysfunction was found in 97.4% of men younger than 60 years of age with prior sexual ability. They concluded the prognosis for patients with metastases to the lateral lymph nodes is poor and the improvement in survival rate from lateral lymph node dissection is minimal. Furthermore, the operation will be prolonged, and the extended dissection will undoubtedly be associated with more bleeding and ultimately a greater morbidity and mortality. No convincing evidence exists to support a significant survival advantage with the radical operation to compensate for the increased morbidity.

& CONCOMITANT PELVIC ORGAN EXCISION The incidence of occult ovarian metastases from carcinoma of rectum and colon is about 6%. For this reason bilateral salpingo-oophorectomy should be considered, especially when the carcinoma is in the lower part of the rectum where the lymphatics communicate directly to the ovaries and fallopian tubes. The risk of this added surgical procedure is small and the benefit to the patient may be significant. Most of these patients have passed menopause, thus minimizing the adverse physiologic effect. Lymphatic drainage from the lower part of the rectum also communicates with the uterus, cervix, and broad ligament. Theoretically a total hysterectomy should be performed, along with abdominoperineal resection for carcinoma of the lower rectum. However, because a hysterectomy may add to the morbidity and mortality, such practice is usually limited to only those patients in whom associated pathology such as direct extension of the carcinoma or large fibroids exists. A rich lymphatic system communicates between the lower rectum and the posterior vaginal wall. A concomitant partial vaginectomy may be required in women with locally advanced rectal carcinoma. Ruo et al. (381) identified 64 patients requiring a partial vaginectomy during resection of primary rectal carcinoma. Locally advanced disease was reflected by presentation with malignant rectovaginal fistula (n ¼ 6) or carcinomas described as bulky or adherent/tethered to the rectovaginal septum (n ¼ 32). Thirty-five patients received adjuvant radiation with or without chemotherapy. At a median follow-up of 22 months, 42% of patients developed recurrent disease, with most of these occurring at distant sites. The five-year overall survival was 46% with a median survival of 44 months. The two-year local recurrence-free survival was 84%. The crude local failure rate was 16% and local recurrence was more common in patients with a positive as opposed to a negative microscopic margin (50% vs. 13%, respectively).

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

Positive nodal status had a significant effect on overall survival. Excision of the posterior vaginal wall if attached should be included in abdominoperineal resection for carcinoma on the anterior wall of the lower rectum. This should not add significantly to the risk of the procedure. The seminal vesicles and prostate usually are not removed along with abdominoperineal resection of the rectum unless there is direct extension of the carcinoma into them. Bladder wall not involving the trigone can be excised in continuity. In the rare situation in which the bladder and prostate are involved by the disease, they can be removed if the patient is an acceptable medical risk and agreeable to the operation. Consideration can be given to a cystoprostatectomy and ileal bladder resection in addition to the abdominoperineal resection, but it must be remembered that this is a formidable undertaking and is associated with significant morbidity and mortality. The role for a procedure of this magnitude is clearly limited. If this procedure is being considered, it would be wise to seek the counsel of a urologist. Sokmen et al. (382) reported fixation of the locally advanced rectal carcinoma at the time of operation as an important prognostic variable. It may be difficult to determine whether fixation is caused by inflammatory adhesions or by direct extension tethering the carcinoma to the surrounding pelvic structures. Of 83 patients with rectal carcinoma, 24% had locally advanced lesions. Perioperative mortality was 5%. Only 24% showed histopathological confirmed carcinoma adhesion into adjacent structures. There was no significant difference between the patients with positive and negative histopathological confirmation of malignant spread in terms of survival rates. The presence of local extension does not necessarily mean incurability and sound surgical judgment should dictate that in the face of a tethered lesion, must one extend the surgical intervention radically to resect any malignancy en bloc. Orkin et al. (383) reviewed 65 patients who underwent extended resection for locally advanced carcinoma of the rectum. Their study included 37 abdominoperineal resections, 20 low anterior resections, and eight Hartmann’s procedures. Attached viscera most frequently resected were the uterus, 32; ovaries, 37; partial vagina, 25; and bladder, 26. Examination of these organs revealed carcinomatous involvement in 57%. This extended resection represented only 1.5% of patients who underwent operation for carcinoma of the rectum. No deaths occurred; the incidence of morbidity was 20%, and the five-year survival rate was 52%. A similar experience was reported by Federov et al. (384). When a carcinoma is fixed to the sacrum, Sugarbaker (385) has advocated partial sacrectomy and en bloc resection. He reported a survival time of more than three years in four of six patients so treated.

& PALLIATIVE THERAPY FOR ADVANCED RECTAL CARCINOMA Palliative therapy for advanced rectal carcinoma constitutes a major health care issue. A workshop was held on this challenging problem and much of the following information was derived from the summary of that undertaking (386). About three-quarters of the patients with rectal carcinoma are treated with curative intent. The remaining 25% of patients are treated for palliation and half of these will

261

undergo abdominal operation as part of their palliative treatment. Of the patients treated with curative intent, approximately 40% will develop recurrences that cannot be treated with curative intent. It can thus be estimated that in the United States 20,000 to 30,000 patients per year will require palliative care for rectal carcinoma. The primary goal of palliative therapy is to maximize the quality of remaining life. Palliative care is a complex undertaking and requires collaboration of a multidisciplinary team. The treatment plan should focus on pain and symptom management and treatment should be commensurate with the expected improved quality of life. Most patients with incurable rectal carcinoma fear the development of severe pain. Reassurance and good communication about all aspects of the disease process in combination with optimal pain control by modern pain management methods should play a pivotal role in the palliative treatment regimen. Investigations, especially when clinical findings include malnutrition, ascites, extensive lymphadenopathy and palpable metastases should be minimized. If operative palliation appears feasible, resectability and extent of abdominal metastases should be assessed with CT of the abdomen and pelvis. Endorectal ultrasound and MRI should be used only if resectability is uncertain. If the primary lesion is resectable and abdominal metastases are absent, further imaging studies, e.g., PET scan may be indicated. Diagnostic laparoscopy is occasionally useful to determine if palliative resection would be beneficial. Rectal carcinoma may be deemed incurable for several reasons. The patient may have advanced locoregional disease or distant metastases. There may be significant co-morbidities rendering the patient unfit for operation or the patient may simply decline an operation that consequently results in the construction of a permanent colostomy. For the most part, operative palliative therapy is generally indicated if the patient will tolerate the procedure and if the operation has a high likelihood of relieving symptoms related to the carcinoma and maintaining normal functions to maximize the quality of life. Operative intervention is only one of several options for palliation. The indication for surgical palliation depends on symptoms, extent of local disease, expected duration of life, and perioperative morbidity and mortality. Indications forcing the surgeon’s hand include bowel obstruction, rectal perforation with localized sepsis or fistula formation, rectovesical, rectoprostatic, or rectovaginal fistula formation, or fecal incontinence. Resection should not be considered if there is extensive pelvic disease, lower extremity lymphedema, invasion of ileofemoral vessels, extensive lymphatic involvement, sciatic nerve pain, bilateral ureteral obstruction, neural or bony involvement at a higher than S1/S2 level, DVT, multiple peritoneal metastases, nonresectable distant metastases (liver, lung, etc.), or if life expectancy is less than three to six months. Anterior resection with anastomosis is considered if clearance allows the rectal reminant to be longer than 3 to 4 cm. Increased risk of anastomotic breakdown due to preoperative irradiation, the disadvantage of a temporary proximal stoma, and the time to achieve good anal function should be taken into account and indeed may negate the creation of an anastomosis. A Hartmann operation is often a better alternative if the patient is willing

262

& PART I: COLORECTAL DISORDERS

to accept a permanent colostomy when anal function is poor. Perineal wound healing is obviated. Whereas posterior vaginectomy and hysterectomy are not a deterrent to palliative rectal resection, more extensive pelvic exenterations are rarely performed because of their high morbidity and diminished quality of life. Surgical treatment for patients with unresectable carcinomas is frequently limited to construction of a sigmoid colostomy for fecal diversion; this is now often performed laparoscopically. Construction of an end stoma with an adjacent small mucous fistula (end loop) avoids the disadvantages of a loop colostomy. Loop stomas are useful in patients with short life expectancy and with significant obesity. With extensive local disease but not complete obstruction a hidden colostomy may be created and the patient may succumb from metastatic disease before complete obstruction occurs. In a number of patients, colostomies are being replaced by endorectal debulking and endorectal stenting particulary for mid and low rectal lesions. Seldom may these lesions be amenable to peranal excision. In the minds of some surgeons, it seems hard to justify a low anterior resection or abdominoperineal resection in patients who obviously have no chance for long-term survival. Because of the high morbidity and mortality, some authors have advocated lesser procedures (387). On the other hand, most of these patients are symptomatic. If untreated, the symptoms invariably become worse with progression of the disease. A colostomy without excision of the primary lesion affords little palliation and only adds to the inconvenience of stoma care. Although a colostomy may overcome obstructive symptoms, it does nothing to relieve the incessant diarrhea with bleeding and mucus discharge or the sacral and sciatic nerve pain. The most reliable way to diminish these symptoms is excision of the carcinoma-bearing rectum. In a series of 125 patients who underwent palliative operative treatment, Moran et al. (387) found a median survival of 6.4 months for those who underwent a diverting colostomy, 14.8 months for abdominal resection, and 14.7 months for per anal excision. Longo et al. (388) reviewed a series of 103 patients, 68 of whom underwent palliative resection and 55 of whom were treated without resection (colostomy or nonoperative treatment). The postoperative mortality was similar in both groups. The survival time was significantly better in patients who underwent resection, but this is probably a reflection of the more extensive disease in patients who did not undergo resection. Resection may also improve the quality of life. The optimal use of radical surgery to palliate primary rectal carcinomas presenting with synchronous distant metastases is poorly defined. Nash et al. (389) reviewed 80 stage IV rectal carcinoma patients to evaluate the effectiveness of radical operation without radiation as local therapy. Seventy-six percent of patients received chemotherapy; response information was available for 34 patients. Radical resection was accomplished by low anterior resection (n ¼ 65), abdominoperineal resection (n ¼ 11), and Hartmann’s resection (n ¼ 4). Surgical complications were seen in 15% with one death and four reoperations. The local recurrence rate was 6% with a median time to local recurrence of 14 months. Only one patient received pelvic radiotherapy as salvage treatment. One patient required subsequent

diverting colostomy. Median survival was 25 months. The extent of metastases and response to chemotherapy were determinants of prolonged survival. They concluded that radical surgery without radiotherapy would be able to provide durable local control with acceptable morbidity. The results of the largest series of palliative operations were reported by Johnson et al. (390). In a total of 338 patients, the operative mortality for patients undergoing resection was 11.7%; colostomy, 5.3%; and diagnostic laparotomy, 6.8%. The carcinoma-specific survival rate was longer in patients undergoing resection, but this only reflects a bias of patient selection toward more favorable cases. The five-year survival rate for the palliative operation was 4.5% with a median survival of 10 months. In a symposium on advanced carcinoma of the rectum, Gordon et al. (391) discussed the subject of palliative resection in patients who have established metastatic disease. It was believed that palliative resection of the primary lesion is very important because it eliminates the constant urge for stool, the bleeding problem, and the pelvic pressure symptoms. In these patients, an anastomosis can be done and the usual margin for a curative resection is not required. A low anterior resection is an option, even with liver metastases. There is also a role for palliative abdominoperineal resection, and patients who survive the procedure may have palliation for up to two to three years. In the presence of ascites, there is concern about the ability of the anastomosis to heal. If there is extensive pelvic disease and concern that there might be residual carcinoma in the pelvis, perhaps an anatomosis will not work well, and the patient will be better served by a colostomy. If an extensive portion of the patient’s liver has been replaced by malignancy, perhaps nothing should be done. Endocavitary radiation, local excision, fulguration, or some primary control of the rectal lesion can be considered, knowing that the patient has a limited life expectancy of less than six months. Electrocoagulation and/or laser or intraluminal stenting may be used to open a channel through an obstructing rectal malignancy for palliation. These patients have a limited lifespan, and it is preferable to make them comfortable without a major operation, if possible. Nevertheless, there are certain situations when Hartmann’s procedure is appropriate for palliation. If the carcinoma is large and bulky, removal of the lesion can prevent tenesmus, pain, bleeding, and impending obstruction. At the same time, if the carcinoma is located so low that an anastomosis would result in a perineal colostomy, Hartmann’s procedure is appropriate. It alleviates the patient’s symptoms and offers better quality of life for the time remaining. Heah et al. (392) analyzed the outcomes of 28 patients treated by Hartmann’s procedure versus 26 patients by abdominoperineal resection. Postoperatively, Hartmann’s procedure group started oral intake at a mean of 2.3 days, and stomas were functioning at a mean of 3.1 days compared with 2.6 days for oral intake and three days for stoma functioning in the abdominoperinreal resection group. Hartmann’s procedure group was ambulant after a mean of 2.4 days versus a mean of 3.2 days in the abdominoperineal resection group. Postoperative abdominal wound infection occurred in 18% and 19%, respectively, in the Hartmann and abdominoperineal groups. Forty-six percent of patients had perineal wound sepsis, and 38% had

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

perineal wound pain in the abdominoperineal resection group. These complications were absent in Hartmann’s procedure group. Postoperative stay was similar in both groups. They concluded that Hartmann’s procedure offers superior palliation compared with abdominoperineal resection because it provides good symptomatic control without any perineal wound complications and pain. In patients with a large resectable carcinoma, significant palliation can be achieved in terms of pain, tenesmus, and obstruction with a palliative abdominoperineal resection recognizing that the operation will not necessarily allow the patient to live any longer but will certainly make the patient more comfortable for the time remaining. If the patient is otherwise a good risk, abdominoperineal resection may be the best palliative procedure. The prognosis and role of operation in patients with incurable rectal carcinoma was studied by Mahteme et al. (393). The authors noted that resection of the primary lesion in patients with incurable carcinoma was followed by a median survival of 7.5 months, approximately four months longer than that after laparotomy without resection. Massive liver involvement, abnormal liver function tests, peritoneal growth, or abnormal lymph nodes correlated with a short survival. Local symptoms demanded active surgical treatment in approximately 15% of patients with a retained primary carcinoma. A colostomy was necessary for intestinal obstruction in 12% of patients. The authors believe their results support a selective approach to patients with incurable rectal carcinoma. In light of these findings, I tend to favor resection of the rectum or a limited low anterior resection for patients with tenesmus, excessive bleeding, or perineal pain. Exceptions are patients whose general condition is considered too poor to withstand the operation or the presence of malignant disease that has disseminated extensively. Self-expanding metal stents have become a useful addition to the palliative armamentorium in the past decade. Stent placement is indicated in patients with obstructing rectal carcinoma who have extensive disease, who are poor surgical candidates, and who have incurable recurrent disease after resection. Stent placement does not palliate rectal bleeding. In experienced centers, stents are successfully placed in approximately 90% of cases. Stenting has been discussed in detail in Chapter 23. Laser ablation is still a useful therapy for some patients, particularly when the predominant symptom is rectal bleeding. In patients with obstructing rectal carcinoma, several repeated treatment sessions may be necessary to achieve initial luminal patency and further sessions will become necessary every few months or as symptoms recur. Palliation of obstructive symptoms is achieved after two to five laser sessions in 80% to 90% patients and lasts up to six months. Complications occur in 5% of 15% of patients that are mostly minor, although perforation, sepsis, and death have been reported. Laser therapy is an important adjunct in patients with recurrent obstruction after self-expanding metal stent placement. Laser ablation is not effective in treating painful infiltration of pelvic nerves by the malignancy. Argon plasma coagulation is a cost-effective alternative to laser treatment for control of bleeding but it is less useful for treating rectal obstruction. Injection therapy of rectal carcinoma using

263

alcohol or sclerosing agents has a great advantage of being low cost and simple. Photo-dynamic therapy, endoscopic electrocoagulation, and cryotherapy are less suitable because of side effects or complications. Photo-dynamic therapy is limited by the cutaneous photo-toxicity of the systemically administered hematoporphyrin. Chemotherapy and external beam radiotherapy play an important role in the palliation of the incurable carcinoma. Lesions that are asymptomatic or minimally symptomatic can be managed with chemotherapy or with chemotherapy and radiation. Combination regimens (especially with newer agents Irinotecan and oxaliplatin) may achieve responses in excess of 75% for chemotherapy-na€
ve rectal carcinoma and offer a considerable chance for palliation and may offer a survival benefit. The addition of pelvic radiation therapy should be made dependent on the extent of extrapelvic metastases and the size of the rectal carcinoma. A large rectal carcinoma might best be offered pelvic radiation. Radiation alone has definite benefits in relieving symptoms. Pain and bleeding can be treated with success in 75% of patients with low doses of radiation. However, symptom relief is relatively short-lived and can be expected to last for three to nine months. It is most useful in patients with advanced disease and a short life expectancy. Radiation therapy does not offer a survival benefit. Addition of chemotherapy will generally increase survival. Attempts have been made to improve palliation in patients with locally recurrent carcinoma by combining surgery with external beam and intraoperative radiotherapy if available. Nakfoor et al. (394) reported on 145 patients with locally advanced rectal carcinoma who underwent moderate to high dose preoperative irradiation followed by resection, 93 of whom received 5-FU. At operation, intraoperative electron beam irradiation was administered to the surgical bed of 73 patients with persistent malignancy or residual disease in the pelvis. No differences in sphincter preservation, pathological downstaging or resectability rates were observed by 5-FU use. However, there were statistically significant improvements in five-year actuarial local control and disease specific survival in patients receiving 5-FU during irradiation compared with patients undergoing irradiation without 5-FU. For the 73 patients selected to receive intraoperative electron beam irradiation, local control and disease-specific survival correlated with resection extent. For the 45 patients undergoing complete resection and intraoperative electron beam irradiation, the five-year actuarial local control and disease specific survival were 89% and 63%, respectively. These figures were 65%, 32% for the 28 patients undergoing intraoperative radiotherapy for residual disease. Treatment strategies using 5-FU during irradiation and intraoperative radiotherapy for patients with locally advanced rectal carcinoma are beneficial and well tolerated. Patients should not be considered for operation, chemotherapy, or radiation therapy if they have significant pre-existing medical conditions, are unable to maintain alimentation because of metastatic disease, or are so debilitated in their performance status that they are limited to bed-to-chair existence. Parenteral administration of fluids may provide some additional comfort, however, total parenteral nutrition or tube feedings have not been shown to be of benefit in the nonsurgical carcinoma patient.

264

& PART I: COLORECTAL DISORDERS

& HARTMANN’S PROCEDURE Occasionally one encounters a carcinoma of the middle third of the rectum that can be resected with an adequate lower margin of clearance, but it is technically difficult to safely perform a low anastomosis, or the likelihood of local recurrence is high. In such situations, closure of the distal rectal stump with establishment of an end sigmoid colostomy is an alternative to abdominoperineal resection. Likewise, in the presence of an obstructing carcinoma of the rectum, where one is faced with poorly prepared bowel, Hartmann’s procedure might be considered. However, this procedure should be performed with caution since the complication rate of pelvic abscess as the principal cause of death is high. ReMine and Dozois (395) reviewed 107 patients who underwent Hartmann’s procedure for carcinoma of the rectum. In approximately half of these patients the operation was considered palliative. In only 10% of patients was intestinal continuity reestablished. Doci et al. (396) reported a consecutive series of 50 patients who underwent Hartmann’s resection for palliation because of advanced disease (62%), poor risk conditions and advanced age (24%), and intraoperative complications or a difficult primary anastomosis (14%). The overall operative morbidity and mortality were 80% and 8%, respectively, with pelvic sepsis accounting for 37.5% of the complications. The overall five-year survival rate for patients undergoing operation for cure was 46%. An extended Hartmann’s procedure is occasionally useful in rectal resections, because anastomotic, perineal, and functional problems are eliminated. Tottrup and Frost (397) reported 163 patients undergoing rectal resection with colostomy and closure of the rectal remnant. Pelvic sepsis developed in 18.6%. When the rectum had been transected less than 2 cm above the pelvic floor, 32.9% developed an abscess in contrast to 7.8% after higher transection. Other risk factors were male gender and missing foot pulses. Only 61% of pelvic abscesses healed after a median of 59 days, leaving 39% unhealed after an observation period of 277 days. & UNRESECTABLE CARCINOMA OF THE RECTUM When confronted with a carcinoma that is unresectable, the surgeon has a difficult decision. The spectre of a future colonic obstruction may motivate many surgeons to perform an immediate diverting colostomy despite the fact that the patient does not have an obstruction and the surgeon knows many patients will die of metastatic disease before they develop an obstruction. The ‘‘hidden’’ colostomy, in combination with nonoperative therapy, such as radiotherapy, electrocoagulation, and, more recently, the neodymium-yttrium-aluminum-garnet (Nd:YAG) laser, provides these patients a better quality of life. The ‘‘hidden’’ colostomy is not a new procedure. It was initially described in 1967, but is a procedure that is seldom considered or remembered (398). The technique of construction of the hidden colostomy is not difficult (56,398). A short transverse incision is made in the right upper quadrant over the rectus muscle and deepened through the subcutaneous tissue, anterior rectus sheath, rectus muscle, posterior rectus sheath, and peritoneum.

FIGURE 27 & Hidden colostomy.

The proximal transverse colon is mobile and generally easily delivered into the incision. The omentum is gently dissected off the right side of the transverse colon to facilitate delivery of the intestine. A small window, approximately 3 cm, is created adjacent to the portion of the intestine to be delivered. The purpose of the window is to allow the creation of a fascial bridge. The intestine is then delivered into the wound and three sutures of 0 Vicryl (polyglactin 910) are used to approximate the fascia through the mesenteric window (Fig. 27). Care must be taken to ensure adequate width of the bridge because, if a narrow bridge is created, the intestine may be acutely angulated and the patient may experience an obstruction at the level of the operative site during the early postoperative period (Fig. 28). A small subcutaneous pocket is created above and below the incision to allow better seating of the transposed colon. A few catgut sutures are used to approximate the subcutaneous tissue superficial to the colon and the skin is closed with 3–0 subcuticular Vicryl. In the event that the distal part of the intestine becomes obstructed, it is a simple matter to open the colostomy. Using local infiltrative anesthesia, a short incision is made superficial to the transposed intestine. The colon is opened and immediately matured. The wound should be quarantined with drapes during the opening of the intestine and after aspiration of the initial flood of intestinal contents, a gauze can be inserted into each of the distal and proximal intestinal

FIGURE 28 & Collapsed hidden colostomy due to inadequate length of fascial bridge.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

loops while the colostomy is matured. The gauze is then removed and an appliance promptly applied. Of six patients on whom I performed this procedure, two died without signs of obstruction. One patient with carcinoma of the ovary responded to chemotherapy and had no subsequent symptoms of obstruction. Two patients underwent opening of their colostomy eight and nine months later. After courses of chemotherapy and radiotherapy, one patient underwent a pelvic exenteration.

& HIGH LIGATION OF INFERIOR MESENTERIC ARTERY In performing abdominoperineal resections for carcinoma of the rectum, if the inferior mesenteric artery is ligated below the take-off of the left colic artery, a significant number of patients (7–11%) will have lymph node metastases at the origin of the inferior mesenteric artery. These metastases would not have been removed unless a high ligation had been performed (399). Based on this finding, many surgeons advocated ligation of the inferior mesenteric artery at the origin of the aorta, hoping to improve long-term survival rates. In a study of 179 specimens with high ligation by Grinnell (400), 19 (10.6%) had lymph node metastases between the point of ligation of the artery at the aorta and the level of its left colic branch. These were nodes that would not have been removed if ligation had been carried out at a point distal to the left colic artery. However, follow-up records of these 19 patients showed that none were salvaged by the radical procedure. This suggests that still higher aortocaval nodes were involved and were beyond the reach of operation. A study by Busuttil et al. (401) of treatment for carcinoma of the sigmoid colon and upper rectum also failed to show any benefit of ligating the inferior mesenteric artery at its origin. Indeed, the complication rate after segmental colectomy was a third as great as that for the radical left hemicolectomy. The hospital mortality rate was 1% after the former and 6.2% after the latter, whereas the five-year survival rate was 70.3% after the former and 56.3% after the latter. Surtees et al. (402) reviewed the results of a series of 250 patients who underwent resection of carcinoma of the rectum at St. Mark’s Hospital in London. For patients with Dukes’ C carcinomas, survival rate did not improve if ligation of inferior mesenteric artery at its origin was performed instead of ligation distal to the left colic artery. Other authors have also reported the failure of high ligation to confer a survival benefit (403,404). In a series of 198 patients with rectal carcinoma who underwent resection with high ligation of the inferior mesenteric artery, Hida et al. (405) found the incidence of metastases to lymph nodes surrounding the origin of the inferior mesenteric artery (root nodes) was 8.6%. Inferior mesenteric artery root nodal metastases occurred more frequently with pT3 and pT4 carcinomas. The five-year survival rate in patients with inferior mesenteric artery root nodal metastases was 38.5%; this rate was significantly lower than in those without inferior mesenteric root nodal metastases (73.4%), clearly an indicator of more advanced disease. The only contrary publication on the subject was by Slanetz and Grimson (406). They reviewed a series of 2409 consecutive patients undergoing curative resection

265

with detailed descriptions of the operative procedure and lymphatic drainage in the surgical specimens that provided a unique database to provide meaningful comparisons between high and intermediate level ligation. The probability of five-year survival rate increased with high ligation from 73.9% to 84% in patients with Dukes’ B colon carcinomas and from 49% to 58% in patients with Dukes’ C1 colon carcinomas. In patients with Dukes’ A C (lymph node involvement but not penetration of the full thickness of the muscularis propria) carcinomas, high ligation increased five-year survival rate from 64.9% to 80.4%. In patients with Dukes’ C carcinomas with involved middle level lymph nodes, the five-year survival rate increased from 20.5% to 33% and the death rate from recurrent carcinoma fell from 77% to 59% with high ligation. In patients with Dukes’ A C carcinomas with four or fewer involved nodes, the five-year survival rate was increased by high ligation from 50% to 78.6% in the colon and from 40% to 71.4% in the rectum. When more than four lymph nodes were involved, the survival rate had been unaffected by the level of ligation. Although high ligation reduced distant recurrences, its greatest effect was observed in the incidence of local and suture line recurrence. The five-year local recurrence rate in patients with Dukes’ B who were managed by high ligation was 11.4% compared with 18.7% with intermediate ligation. In patients with Dukes’ C carcinoma the local recurrence rate was 20.8% five years following high ligation compared with 30.7% for intermediate ligation. In patients with Dukes’ B carcinoma who were undergoing curative resections, the incidence of suture line recurrence was 3.9% following high ligation compared with 5.5% following intermediate ligation. In patients with Dukes’ C carcinoma, the incidence of suture line recurrence was 6.9% with high ligation and 11.4% with intermediate ligation. They concluded in certain stages of colorectal carcinoma, the more extensive resection of mesenteric lymphatic drainage associated with high ligation appears to increase the survival rate and reduce the recurrence rate following curative resections. High ligation is of no value in the treatment of Dukes’ A and B lesions, nor is it of any value if metastases have occurred (407). High ligation is of potential benefit to the patient with a Dukes’ C lesion only at the precise time when nodal metastases have spread to a level proximal to the left colic artery but have not spread to the origin of the inferior mesenteric artery (Fig. 29). This clearly constitutes only a small number of patients. The additional operative morbidity and mortality would probably negate any potential benefit.

& MARKING THE RECTUM When a polyp has been removed from the rectum and the pathologist reports the excised tissue to contain invasive adenocarcinoma, a low anterior resection may be recommended. At the time of operation it will be impossible to determine the level at which the lesion was located and the level at which the rectum is to be transected. To mark the site, the rectum can be injected with methylene blue at the commencement of the procedure. Injection of 0.25 ML usually turns the external surface blue at the time of laparotomy. If the injection is insufficient, the rectum will not be marked,

266

& PART I: COLORECTAL DISORDERS

FIGURE 29 & Potential value of high ligation of inferior mesenteric artery. (A) Conventional low ligation would be sufficient. (B) High ligation provides potential benefit. (C) Proximal lymphatic spread is beyond confines of even high ligation. Source: Adapted from Ref. 407.

but if excess methylene blue is injected, the entire pelvis may be stained. Injecting the appropriate amount permits identification of the correct level for transection.

& ADJUVANT THERAPY FOR CARCINOMA OF THE RECTUM & RADIOTHERAPY Although the resectability of carcinoma of the rectum has remarkably increased during the past five decades, the five-year survival rate has not substantially improved. Radical operation remains the principal treatment for rectal carcinoma, but unfortunately failure, particularly local recurrence, is a common event (30–50%) (408). The symptoms of locally recurrent disease such as pelvic pain and altered bowel habit are often difficult to palliate and many patients die from locally recurrent disease. The search for additional treatment modalities to employ in conjunction with operative excision has been widespread during the past several decades. It has long been known that carcinoma of the rectum is relatively sensitive to irradiation. Both preoperative and postoperative adjuvant radiation therapy has been evaluated in randomized trials involving thousands of patients. Most studies have reported that both pre- and postoperative radiotherapy significantly reduce local recurrence but do not improve survival compared with results obtained from resection alone. McLean and Duncan (408) published an excellent review of randomized trials of adjuvant radiotherapy and

some of the following information has been derived from their article.

Preoperative Irradiation Interpretation of results in the literature is difficult. Many of the reports are not controlled trials, the modes of treatment were different, and the dosages varied widely from a mere single dose of 500 cGy in the immediate preoperative period to multiple doses totaling 5000 cGy or more. Preoperative adjuvant radiotherapy offers three potential advantages (408,409): (i) an increase in resectability and the probability of achieving a curative resection; (ii) ‘‘downstaging’’ of the resected carcinoma; and (iii) a reduction in the number of viable malignant cells, which may then reduce the probability of both local and distant relapses. Other potential advantages cited include decreased seeding of malignant cells, less acute toxicity, increased radiosensitivity because of more oxygenated cells, and enhanced sphincter preservation (410). The possible disadvantages include (408): (i) the loss of accurate surgical staging, which may confound the selection of patients for adjuvant chemotherapy; (ii) delay in performing the definitive curative procedure; and (iii) a potential increase in the risk of postoperative mortality and morbidity. Furthermore, routine use of preoperative radiation will subject many patients to unnecessary therapy, because information from randomized trials suggests that 20% to 30% of patients have Tl or T2 lesions. These patients risk bowel, bladder, and systemic toxicity. Patients with unresectable disease or those found to have extensive liver

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

metastases may be excluded from being given inappropriate additional therapy. Women in their childbearing years should be informed of the sacrifice of fertility and induction of premature menopause. The increasing availability of sophisticated imaging techniques such as the PET scan may decrease the number of patients who are overtreated. The results of the European Organisation for Research and Treatment of Cancer (EORTC) were reported by Gerard et al. (411). The report included 466 patients who were randomized to groups of operation only or operation and 34.5 Gy preoperatively. High morbidity and mortality were noted in the irradiated group, although the values did not reach statistical significance. Both perineal wound healing and hospitalization was prolonged in irradiated patients. During a mean follow-up of 75 months, no improvement in the survival rate of irradiated patients was noted, but the rate of local recurrence decreased. The Medical Research Council in the United Kingdom conducted a trial in which patients were randomly assigned into groups for no adjuvant treatment, a course of 2000 cGy in 10 fractions, or a single exposure of 500 cGy (412). With a minimum follow-up of four years neither form of adjuvant radiotherapy yielded a survival benefit, nor was the incidence of local recurrence reduced. Other authors support the use of preoperative radiotherapy (413–416). Mendenhall et al. (417) reviewed a series of 132 patients with rectal carcinoma who underwent preoperative radiotherapy with a variety of doses. Compared with historical controls treated by operation alone, the local recurrence rate at five years was 8% versus 33%, and the five-year absolute survival rate was 66% versus 40%. Some authors have reported improved survival rates (418). Preoperative radiotherapy has reportedly been beneficial in the treatment of unresectable carcinoma (415,419). It also has been suggested that no increase in morbidity or mortality occurs with preoperative radiotherapy (420). A report on preoperative short-term radiation therapy in operable rectal carcinoma was provided by the Stockholm Rectal Cancer Study Group (421). The investigators randomized 849 patients in a controlled clinical trial (Stockholm I Trial) of radiation therapy (2500 cGy over five to seven days) before operation versus operation alone. At a median follow-up time of 107 months (range, 62–144 months) the incidence of pelvic recurrence was significantly lower (422). A reduction was observed in all Dukes’ stages, but no differences with respect to frequency of distant metastases or overall survival were observed. The time to local recurrence or distant metastases and survival was significantly prolonged in the irradiation group. However, the postoperative morbidity was significantly higher among irradiated patients, and the postoperative mortality was 8% in the radiation therapy group compared to 2% in the operation alone group. Pa¨hlman and Glimelius (423) reported on a multicenter trial in which patients with Astler-Goller stages B2, C1, C2 were randomized to treatments of 25.5 Gy in five to seven days preoperatively (236 patients) and 60 Gy postoperatively (235 patients). No radiation-related complications or increased mortality were noted. The incidence of perineal wound sepsis was increased in the abdominoperineal resection group that received radiotherapy. Although

267

the local recurrence was statistically lower after preoperative than postoperative radiotherapy (12% vs. 21%), with a minimum follow-up of three years and a mean followup of 6.3 years, the two groups had equal survival rates. From their review of randomized controlled radiotherapy trials, Pa¨hlman and Glimelius (424) believed that the reduction in local recurrence is higher if radiotherapy is given preoperatively rather than postoperatively. The effect of radiation is also dependent on the dose of each fraction and the total treatment time. In an initial report from a Swedish multicenter study examining the role of preoperative irradiation in the treatment of patients with resectable rectal carcinoma, 1168 patients were recruited into a nationwide Swedish trial (425). Patients were randomly allocated to receive either preoperative irradiation (25 Gy in five fractions over one week) followed by operation within one week or operation alone. In each group, 454 patients underwent curative operation. The postoperative mortality rates did not differ, 4% after irradiation versus 3% after operation alone, but the postoperative mortality rate was significantly higher in patients treated with a two- rather than three- or four-portal technique (15% vs. 3%). Perineal wound infections were found more frequently in patients receiving irradiation (20% vs. 10%). No difference in the incidence of anastomotic dehiscence or other postoperative complication was found between the groups. In the most recent report on the Swedish Rectal Cancer Trial (426) it was stated that irradiation did not increase postoperative mortality. After five years follow-up, the rate of local recurrence was 11% in the group receiving preoperative radiotherapy and 27% in the group undergoing operation alone. The difference was found in all subgroups defined according to Dukes’ stage. The overall five-year survival rate was 58% for those receiving preoperative radiotherapy and 48% for those undergoing operation alone. The cancer-specific survival rates at nine years among patients treated with curative resection were 74% and 65%, respectively. The Stockholm Rectal Cancer Study Group initiated a second trial in March 1987 (Stockholm II Trial). The protocol was identical with that of the first trial, except that a four-portal technique was used, the target volume was reduced, and patients older than 80 years old were excluded. In both trials, the patients were treated with a fractionation dose of 5 Gy over five or seven days (treatment interruption occurred during the weekend). The Stockholm I Trial used a two-field technique to a large volume, including the anus, rectum, perirectal tissues, perineum, and regional lymph node stations (the inguinal lymph nodes, the obturator foramina, and the paravertebral lymph nodes) up to the level of the second lumbar vertebra. In the Stockholm II Trial, a four-field box technique was used and the treatment volume was reduced cranially. The latter technique is believed to reduce risk of radiationrelated complications. The Stockholm II Trial closed when 557 patients had been included. In a preliminary report from this trial (427), the local recurrence rate was reduced, the length of survival improved after preoperative radiotherapy, and there was no significant increase in postoperative mortality in irradiated patients compared with nonirradiated patients (2% vs. 1%).

268

& PART I: COLORECTAL DISORDERS

Holm et al. (428) assessed the outcome of abdominoperineal resection and anterior resection in the treatment of rectal carcinoma in relation to adjuvant preoperative radiotherapy in 1292 patients. Patients in the radiotherapy group received 25 Gy over five to seven days [776 patients (60%) had been treated with abdominoperineal resection and 516 (40%) with anterior resection]. The overall postoperative complication rate was 33% in the group who underwent abdominoperineal resection and 29% in those who had undergone anterior resection. In both groups, complications occurred more frequently in irradiated patients. The anastomotic leakage rate after anterior resection was similar in irradiated and nonirradiated patients (13% vs. 12%). The postoperative mortality rate was similar after abdominoperineal resection and anterior resection (4% vs. 3%). Postoperative mortality rate was increased in irradiated patients compared with nonirradiated and this was higher after abdominoperineal resection (7% vs. 1%) than after anterior resection (4% vs. 2%). There was no statistically significant difference in survival between the abdominoperineal resection and the anterior resection. Radiotherapy did not significantly influence the overall survival in either of the two surgical groups. There was no statistical difference in local recurrence between abdominoperineal resection and anterior resection groups, but radiotherapy significantly reduced local recurrence, regardless of the operative method. After abdominoperineal resection the local recurrence rate was 16% in irradiated patients and 28% in nonirradiated patients. Corresponding figures after anterior resection were 11% and 24%. The cumulative incidence of distant metastases was 27% after abdominoperineal resection and 26% after anterior resection. These investigators reviewed the causes of death after operation in an attempt to identify risk factors for postoperative mortality in patients treated with high doses of preoperative radiotherapy (429). The majority of deaths were from cardiovascular disease or infection. The authors noted that the risk of death was related to the preoperative radiotherapy technique, that is, patients irradiated with a two-portal technique and a relatively large volume compared with those radiated with a four-portal technique and a limited volume. Holm et al. (430) reported on 1027 curatively operated patients included in two prospective randomized trials of preoperative radiotherapy for rectal carcinoma patients (Stockholm I and II trials). The goal was to assess whether long-term intercurrent morbidity and mortality were increased in patients allocated to the preoperative treatment. Preoperative radiotherapy significantly increased the incidence of venous thromboembolism (7.5% vs. 3.6%), femoral neck and pelvic fractures (5.3% vs. 2.4%), intestinal obstruction (13.5% vs. 8.5%), and postoperative fistulas (4.8% vs. 1.9%). For arterial disease and genitourinary tract diseases, no difference in risk was found between irradiated and nonirradiated patients. Radiotherapy significantly reduced rectal carcinoma deaths in both trials and also improved overall survival in the Stockholm II Trial. Pelvic radiation-related complications are a direct function of the volume of the radiation-field, the overall treatment time, fraction size, radiation energy, total dose, and technique (431). Small bowel-related complications, a dreaded consequence, are directly proportional to the

volume of small bowel in the radiation field. Surgical techniques employed to minimize toxicity to the small bowel from radiation therapy include reperitonealization of the pelvic floor, retroversion of the uterus, construction of an omental sling, placement of an absorbable mesh, placement of pelvic clips to delineate high-risk areas, and insertion of tissue expanders (431). Some radiation therapy techniques include using the prone position, multiple-field techniques (three-field preferred), computerized dosimetry, highenergy linear accelerator (greater than or equal to 6 mv) doses that do not exceed 5040 cGy to the small bowel, a standard fraction size (180 to 200 cGy/day), and bladder distention (431). The importance of technique was highlighted in the report from the Swedish Rectal Cancer Trial in which in-hospital mortality was similar in patients receiving preoperative therapy (4%) when compared with those who underwent operation alone (3%). However, in the group of patients irradiated with two beams, the rate of in-hospital mortality was considerably higher (15%) than in those irradiated with three or four beams (3%) (426). Sphincter function has also been a concern following pelvic radiation. A preliminary analysis of response to a questionnaire about anal function that was sent to all recurrencefree patients who had undergone a sphincter-saving procedure and were alive at least five years following treatment in the Swedish Rectal Cancer Trial indicated that patients who underwent irradiation had more problems with the number of bowel movements, incontinence, urgency, and soiling than those who underwent operation alone (426). Dahlberg et al. (432) investigated the effect of preoperative high dose radiotherapy in long-term bowel function in patients treated with anterior resection in the Swedish Rectal Cancer Trial. Median bowel frequency per week was 20 in the irradiated group and 10 in the surgery alone group. Incontinence for loose stools, urgency, and emptying difficulties were all more common after irradiation. Sensory function such as discrimination between gas and stool and ability to safely release flatus did not however differ between groups. Thirty percent of the irradiated group stated that they had an impaired social life because of bowel dysfunction compared with 10% of the surgery alone group. The study indicates that high dose radiotherapy influences long-term bowel function thus emphasizing the need for finding predictive factors for local recurrence to exclude patients with very high probability for cure with surgery alone and to use optimized radiation techniques. The investigators now suggest that to minimize disturbance of bowel function, the routine inclusion of the anal canal in the irradiated volume in proximal carcinomas should probably be avoided. Preoperative radiotherapy followed by proctectomy and colonic J-pouch anastomosis significantly increases nocturnal defecation frequency (36% vs. 15%) and diarrhea (39% vs. 13%) compared with similar nonirradiated patients but had no influence on the other bowel-habit parameters studied (98). Goldberg et al. (433) reported a prospective randomized multicenter trial that compared preoperative radiotherapy followed by operation with operation alone for rectal carcinoma less than or equal to 12 cm from the anal verge (Imperial Cancer Research Fund). Of 468

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

patients, 228 were randomized to radiotherapy (3–5 Gy over five days within two days of operation) followed by operation and 239 to operation alone. Follow-up until either the patient’s death or five years after the procedure was achieved in 97% of patients. The operative mortality rate for radiotherapy and operation was 9% compared with 4% for operation alone. Cardiovascular and thromboembolic complications were more common after radiotherapy and operation (13%) than after operation alone (3%). Of the 280 patients who had curative operation, 52% of those who had radiotherapy and operation and 56% of those who had operation alone survived five years. Local treatment failure was identified during follow-up in 17% of patients with radiotherapy and operation and 24% for operation alone. It occurred in 33 of 258 patients who had a curative resection (radiotherapy and operation, 9%; operation alone, 16%). Long-term survival was unaffected but long-term local recurrence was reduced by the addition of low-dose radiotherapy to the operation. Perioperative mortality, however, was increased. Kerman et al. (409) reported an analysis of a 14-year experience of moderately high-dose (4500–5000 cGy) preoperative radiation as an adjuvant to low anterior resection of 95 cases of adenocarcinoma of the rectum. The treatment was well tolerated, without treatment-related mortality, and with a low incidence (5.2%) of severe complications. The local recurrence rate was 4.2% and the distant failure rate was 10.5%. At five years, the actuarial survival rate was 66% and the disease-free survival rate was 64%. At 10 years, the actuarial survival rate and disease-free survival rate were 52%. The authors concluded that moderately high-dose (4500–5000 cGy) neoadjuvant radiation in clinically resectable adenocarcinoma of the rectum in which one segment of the anastomosis is in the preoperative radiation field is a safe, effective adjuvant to low anterior resection and that it offers patients excellent local control, long-term survival, and sphincter preservation. In an uncontrolled study, Bannon et al. (434) reported on sphincter-sparing procedures for patients with carcinoma of the distal 3 cm of the rectum following high-dose radiation. Following administration of 4500 to 7000 cGy, 65 patients underwent transanal abdominal radical proctosigmoidectomy and 44 patients underwent full-thickness local excision with a mean follow-up of 40 months. Local recurrence rates were 9% and 14%, respectively, and fiveyear actuarial survival rates were 85% and 90%, respectively. Minsky et al. (435) reported 30 patients (two with T2 and 28 with T3) with rectal carcinoma 3 to 7 cm from the anal verge who underwent preoperative radiotherapy (46.8 to the pelvis and a 3.6 Gy boost) followed by a low anterior resection and coloanal anastomosis. The incidence of local failure was a crude rate of 12% and a four-year actuarial rate of 23%. The four-year actuarial survival rate was 75%. Function was good or excellent in 77% of patients, with median number of two bowel movements. The authors thought that this technique was an alternative to an abdominoperineal resection. Myerson et al. (436) reported on 251 patients with carcinoma of the rectum who received 40 to 50 Gy followed by operation six to seven weeks later. The five-year rates for local control and freedom from disease were 90% and 73%, respectively.

269

Izar et al. (437) reported on 241 patients with rectal carcinoma who underwent preoperative radiotherapy with 36 Gy. The postoperative mortality rate was 2.9%. The most frequent complications were delayed healing of abdominal wounds, 18%; perineal wounds, 14%; and intestinal obstruction, 5%. Local failure occurred in 12% (Dukes’ A, 10%; B, 12%; and C, 23%). Five- and 10-year actuarial survival rates were 70% and 50%, respectively. Pacelli et al. (438) compared neoadjuvant treatment of preoperative radiotherapy (38 Gy) plus intraoperative radiotherapy (10 Gy: n ¼ 69) to no preoperative treatment in a series of 113 patients with middle and lower T3 rectal carcinomas consecutively submitted to TME. Overall, 68.2% of patients were downstaged by the preoperative regimens. Postoperative complications were comparable in the two groups. Five-year disease specific survival was 81.4% and 58.1% in preoperative radiotherapy plus intraoperative radiotherapy group and TME group, respectively. Corresponding figures for disease-free survival were 73.1% and 57.2% in the two groups, respectively. The rates of local recurrence at five years were 6.6% and 23.2% in preoperative radiotherapy plus intraoperative radiotherapy and TME groups, respectively. Minsky reviewed 12 modern randomized trials of preoperative radiation therapy for clinically resectable rectal carcinomas. All used low to moderate doses of radiation. Overall, most of the trials showed a decrease in local recurrence and this difference reached statistical significance in six of the trials. Although in some trials, a subset analysis has shown a significant improvement in survival, the Swedish Rectal Cancer Trial is the only one that reported a survival advantage for the total treatment group (58% vs. 48%) (426). Given that the other 11 randomized trials of preoperative radiation therapy did not report a survival benefit, these data clearly need to be confirmed by additional studies (439). The most recent trial reported was the Dutch CKVO 95–04 study in which 1805 patients with clinically resectable (T1 to T3) disease were randomized to operation alone (with a TME or to extensive short course preoperative radiation therapy followed by operation with TME) (236). Although radiation significantly decreased the local recurrence rate (8% vs. 2%), there was no difference in two-year survival (82%). With longer follow-up, the fiveyear local recurrence rate was higher with operation alone (12%); however, it was significantly decreased (6%) with preoperative radiation followed by operation with TME (440). Second, even if future trials confirm a survival benefit, there are other equally important endpoints in rectal carcinoma that need to be addressed. These include acute toxicity, sphincter preservation and function, and quality of life. For example, acute toxicity in the Dutch CKVO 95–04 trial included 10% neurotoxicity, 29% perineal wound complications, and 12% postoperative leaks (441). In the patients who developed postoperative leaks, 80% required operation resulting in an 11% mortality rate. Hartley et al. (442) reported a 6% mortality with 5 Gy  5 followed by TME. Two meta-analyses reported conflicting results. While both reported a decrease in local recurrence, the analysis by Camma et al. (443) showed a survival advantage, whereas the analysis by the Colorectal Cancer Collaborative Group (439) did not. The benefit of adjuvant radiotherapy for resectable rectal carcinoma has been extensively studied, but data

270

& PART I: COLORECTAL DISORDERS

on survival are still equivocal despite a reduction in the rate of local recurrence. At least 28 randomized controlled trials have compared outcomes of operation for rectal carcinoma combined with preoperative or postoperative radiotherapy with those of operation alone. The Colorectal Cancer Collaborative Group conducted a collaborative meta-analysis of these results to give a more balanced view of the evidence and to increase statistical precision. They analyzed individual patient data from 22 randomized comparisons between preoperative (6350 patients in 14 trials) or postoperative (2157 in eight trials) radiotherapy and no radiotherapy for rectal carcinoma. Overall survival was only marginally better in patients who were allocated to radiotherapy than those allocated to operation alone (45% vs. 42.4% at five years and 26.9% vs. 25.3% at 10 years). Rates of apparently curative resection were not improved by preoperative radiotherapy (85% radiotherapy vs. 86% control). Yearly risk of local recurrence was 46% lower in those who had preoperative radiotherapy than those who had operation alone and 37% lower in those who had postoperative treatment than those who had operation alone. Fewer patients who had preoperative radiotherapy died from rectal carcinoma than those who had operation alone (45% vs. 50%, respectively) but early (less than or equal to 1 year after treatment) deaths from other causes increased (8% vs. 4% died). Preoperative radiotherapy (at biologically effective doses greater than or equal to 30 Gy) reduces the risk of local recurrence and death from rectal carcinoma. Postoperative radiotherapy also reduces local recurrence but short preoperative radiation schedules seem to be at least as effective as longer schedules. In this systematic review of data from 8507 patients in 22 trials of adjuvant radiotherapy for rectal carcinoma, radiotherapy both before and after operation substantially reduced the risk of local recurrence in apparently curatively resected patients and moderately reduced deaths from rectal carcinoma. The largest reductions were in studies of preoperative radiotherapy with biologically effective doses of 30 Gy or more; no significant reductions were recorded in studies of radiotherapy schedules with low biologically effective doses. One of the most important controversies with preoperative therapy is whether the degree of downstaging is adequate to enhance sphincter preservation. An analysis of 1316 patients treated on two previously published Scandinavian trials of intensive short course radiation showed the downstaging was most pronounced when the interval between the completion of radiation and operation was at least 10 days (444). In the Dutch CKVO 95–04 trial, where the interval was one week, there was no downstaging (445). Data from the Lyon R90–1 trial of preoperative radiation suggest that an interval of more than two weeks following the completion of radiation increases the chance of downstaging (446). Whether increasing interval between the end of intensive short-course radiation and surgery to greater than or equal to four weeks will increase downstaging is not known. This question is being addressed in an ongoing randomized trial from Sweden (Stockholm III trial). Even if preoperative radiation therapy were effective, I would not perform an anastomosis through a portion of bowel that several weeks earlier harboured a malignancy. Furthermore, sphincter preservation without good function

is of questionable benefit. In a series of 73 patients who underwent operation, Grumann et al. (97) reported that 23 patients who underwent an abdominoperineal resection had a more favorable quality of life compared with the 50 who underwent a low anterior resection. Guillem et al. (447) conducted a prospective analysis to determine the operating surgeon’s ability to assess response to combined modality therapy (CMT) in 94 prospectively accrued patients with locally advanced rectal carcinoma. Clinical assessment using digital rectal examination (DRE) underestimated pathologic response in 78%. In addition, DRE was able to identify only 21% with a pathologic complete response. There were no clinical overestimates of response. None of the clinicopathologic characteristics of the carcinoma examined had a significant impact on DRE estimation of response. Given the inaccuracy of DRE following preoperative CMT, it should not be used a sole means of assessing efficacy of therapy nor for selecting patients following CMT for local therapies. The timing of operation following neoadjuvant therapy for rectal carcinoma is not well defined. An interval of six to eight weeks between completion of preoperative chemo-radiation therapy and operative resection of advanced rectal carcinoma has been described. Stein et al. (448) studied whether a longer time interval between completion of therapy and resection increases disease downstaging and affects perioperative mortality. They enrolled 40 patients with advanced carcinoma of the rectum who underwent preoperative chemoradiation on a prospective trial with irinotecan (50 mg/m2), 5-FU (225 mg/m2), and concomitant external-beam radiation (45–54 Gy) followed by complete resection of the carcinoma with TME. The patients were divided into two groups with 33 eligible patients: Group A (four- to eight-week time interval) and Group B (10–14-week interval). There were no statistical differences in perioperative morbidity with three anastomotic leaks in Group A. The carcinomas were downstaged in 58% of patients in Group A and 43% of those in Group B. Nodal down staging occurred in 78% of Group A and 67% of Group B. The pathologic complete response rate was 21% in Group A and 14% in Group B and a residual microfocus of carcinoma was found in 33% of patients in Group A and 42% of those in Group B. These differences were not statistically significant. They concluded a longer interval between completion of neoadjuvant chemoradiation and operative resection did not increase the response of advanced rectal carcinoma in this cohort. Francois et al. (446) presented contrary data. They suggested that a long interval between preoperative irradiation and operation provides increased downstaging with no detrimental effect on toxicity and early clinical results. They randomized 20 patients with rectal carcinoma accessible to digital examination, staged T2 to T3, NX, MO before radiotherapy (39 Gy in 13 fractions) into two groups: in the short interval group, operation had to be performed within two weeks after completion of radiation therapy, compared with six to eight weeks in the long interval group. A long interval between preoperative radiotherapy and operation was associated with a significantly better clinical response (53.1% in the short group and 71.7% in the long group), and pathologic downstaging (10.3% in the short group and 26% in the long group). At a median follow-up of 33

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

271

addition to distant dissemination, may be decreased by altering the viability of shed malignant cells. Currently, no unanimous agreement on the dose of radiation to be delivered exists. A frequently administered dosage is 40 to 45 Gy delivered over four to six weeks with a six-week interval before operation. For patients who have received radiotherapy, it has been customary to establish a protective ileostomy or colostomy, but some reports suggest that it is safe to perform an anastomosis if the radiation dose does not exceed 45 Gy (451,452). A summary of recent published randomized trials is provided in Table 9. Caution should be exercised when data from different sources are cited, because each trial has different patient selection criteria, different radiotherapy schedules and fields, and different end points. As seen from the table, only the Swedish trials observed significant improvement in overall survival with preoperative radiotherapy. There is some overlap between the Swedish Rectal Cancer Trial and the so-called Stockholm II Trial in that 316 patients who were enrolled in Stockholm were also included in the Swedish Rectal Cancer Trial (426). The Sao Paulo study also demonstrated improvement, but only 68 patients were entered (454).

months, there were no differences in morbidity, local relapse, and short-term survival between the two groups. Sphincter preserving operation was performed in 76% of cases in the long group and 68% in the short group. They believe that when sphincter preservation is questionable, a long interval may increase the chance of successful sphincter-saving operation. This issue continues to remain controversial. Marijnen et al. (449) reported on the heath related quality of life and sexual functioning of 990 patients who underwent TME and were randomly assigned to short-term preoperative radiotherapy (5  5 Gy). Patients without a recurrence the first two years were analyzed (n ¼ 990). Daily activities were significantly fewer for preoperative radiotherapy patients three months postoperatively. Irradiated patients recovered slower from defecation problems than TME-only patients. Preoperative radiotherapy had a negative effect on sexual functioning in males and females. Irradiated males had more ejaculation disorders and erectile functioning deteriorated over time. Preoperative radiotherapy had similar effects in patients who underwent a low anterior resection versus an abdominoperineal resection. Patients with an abdominoperineal resection scored better on the physical and psychologic dimension than low anterior resection patients, but worse on voiding. With the information available at this time, the effect of preoperative irradiation on carcinoma of the rectum can be summarized as follows: the resectability may be improved in those patients with a fixed rectal carcinoma (450); malignant cells in the regional lymph nodes may be destroyed and thus may convert the disease to a more favorable stage; and local seeding of malignant cells, in

Postoperative Radiotherapy Since most recurrent carcinoma of the rectum is localized in the pelvis and usually occurs in those patients with Dukes’ B or C lesions, some authors believe that postoperative irradiation is more appropriate when details of the pathologic staging have been established. Gunderson and Sosin (457) cited the following advantages and disadvantages of postoperative irradiation.

TABLE 9 & Results of Randomized Trials of Preoperative Radiotherapy

Trial (Authors)

No. of Patients

MRC 1 (412) (1984)

824

VASOG 2 (453) (Higgins, Humphrey, and Dweight, 1986) EORTC (411) (Gerard et al., 1990)

361

Sao Paulo (454) (Resis Neto, Guilici, and Reis Neto, 1989) Norway (455) (Dahl et al., 1990)

68 309

ICRF (433) (Goldberg et al., 1994)

468

Northwest Region (456) (Marsh, James, and Schofield, 1994) Stockholm I (422) (Cedermark et al., 1995) Stockholm II (427) (Cedermark et al., 1996) SRCT (426) (1997)

284

Dutch (Kapiteijn et al., 2001) (236) a

466

849 557 1168 937 924

Dose (Gy/Fractions)

Dukes’ Stage C (%)

Local Recurrence (%)

Control 5/1 20/10 Control 34.5/18

46 No difference 45 36a 41 35

No difference

Control 34.5/15 Control 40/20 Control 32.5/18 Control 15/3 Control 20/4 Control 25/5-7 Control 25/5 Control 25/5 Preop 5/5 TME alone

59 55 47 26a 28 18a No difference

28 28

33 36

30 15a 47 15a 21 15 24 17a 37 13a 28 14a 21 10a 27 11 2 (2 yr) 8

Distant Metastases (%)

Overall Years Survival (%) 38 41 40 42 43

39 39 32 15a 21 23

36 43 37 30 26 19a

15 (2 yr) 17

59 69 34 80a 58 57 40 39 70 70 36 36 56 70a 48 58a 82 (2 yr) 82

Statistically significant. Abbreviations: MRC, Medical Research Council; VASOG, Veterans’ Administration Surgical Oncology Group Trial II; EORTC, European Organisation for Research and Treatment of Cancer; ICRF, Imperial Cancer Research Foundation; SRCT, Swedish Rectal Cancer Trial.

272

& PART I: COLORECTAL DISORDERS

Advantages

1. The total extent of the carcinoma is known and thus irradiation is unnecessary, particularly in those patients who have Dukes’ A lesions. 2. Postoperative irradiation could be used in patients with either anterior resection or abdominoperineal resection and therefore would potentially benefit a larger number of patients. This treatment does not delay operation or alter pathologic staging. Disadvantages

1. Postoperative irradiation has no effect on cells that may be spread at the time of operation. 2. Residual malignant cells in tissues rendered hypoxic as a result of operation may be more resistant to ionizing irradiation than those in a normal oxygenated environment. 3. A long delay in initiating postoperative irradiation could ensue if postoperative complications occur or perineal wound healing is delayed. Hoskins et al. (458) reported on 97 patients who received postoperative radiotherapy. When compared with a historical control group of 103 patients, a statistically significant decrease in local recurrence at 3 years was noted in patients with Dukes’ B2 or C lesions. The Gastrointestinal Tumor Study Group (GITSG) trial consisted of a four-arm study in which patients were randomized to the following treatment groups: operation alone, chemotherapy, radiotherapy, or combination chemotherapy (459). After a median follow-up of more than 5 years, recurrence was seen in 55% of patients in the control group, in 46% who received chemotherapy or radiotherapy, and in only 30% of those who received combination treatment. Fisher et al. (359) reported on 555 patients entered into the multicenter NSABP Protocol R-01. Patients were randomized to receive operation alone; postoperative adjuvant chemotherapy with 5-FU, semustine, and vincristine; or postoperative radiotherapy (46 to 47 Gy). The postoperative radiotherapy group demonstrated an overall reduction in local and regional recurrence from 25% to 16%, but no significant benefit in overall disease-free survival or the survival rate

was found. In summary, postoperative radiotherapy appears to decrease the risk of local recurrence but results in no significant improvement in survival rate. Martijn et al. (460) reported a retrospective analysis of 178 patients receiving adjuvant postoperative radiotherapy (median total dosage, 50 Gy) after curative operation for adenocarcinoma of the rectum and rectosigmoid. The overall 5-year survival rate was 42% and the 5year disease-free survival rate was 37%. The respective rates for the Gunderson-Sosin stage B2 patients were 59% and 53% and 25% and 25% for stage C2 patients. Five-year local relapse rates were 27% for stage B and 40% for stage C2 carcinomas. De Neve et al. (461) reviewed the results of postoperative radiotherapy given to 40 patients with gross or microscopically proven residual disease after resection of rectal or rectosigmoid carcinoma. Mean radiation dosage was 50 Gy. Survival for patients with microscopic residual disease was 40% at 5 years compared to 12% for those with gross residual disease. Mak et al. (462) reported on the late complications of postoperative radiation therapy used to treat 224 patients with rectal carcinoma. The median dose was 54 Gy. Fortyseven patients received concomitant 5-FU. The small bowel obstruction rate was 30% for extended field radiotherapy, 21% for single pelvic field, and 9% for multiple pelvic fields with median time to obstruction of 7 months. Small bowel obstruction was correlated with post-surgical adhesions prior to radiotherapy and absence of reperitonealization at time of the initial operation. Tang et al. (463) compared 127 patients with AstlerColler stage B2 or C rectal carcinoma who underwent operation and postoperative radiotherapy to 122 patients undergoing operation alone. Severe or life-threatening radiation-related complications were encountered in 8% of patients. Postoperative radiotherapy did not improve survival. Five-year survival rates for patients with Dukes’ B2 for operation alone and operation and radiotherapy were 80% and 64%, respectively, and for Dukes’ C, 52% and 41%, respectively. A summary of randomized trials of postoperative radiotherapy is presented in Table 10.

TABLE 10 & Results of Randomized Trials of Postoperative Radiotherapy No. of Patients

Dose (Gy/Fractions)

Dukes’ Stage

GITSG (459) (1985)

227

B2 and C

Denmark (464) (Balsev et al., 1986)

494

NSABP (359) (Fisher et al., 1988)

555

Netherlands (465) (Treurniet-Donker et al., 1991)

172

Operation alone CT 40–48 Gy/22–27 CT þ 40–44 Gy/22–24 Operation alone 50 Gy/25 Operation alone CT 46–47 Gy/26–27 Operation alone

MRC3 (466) (Gates et al., 1995)

469

Trial (Authors)

*

50 Gy/25 Operation alone 40 Gy/20

B/C B and C

B2/C

B/C

Local Recurrence (%)

Distant Metastases (%)

Overall 5-Year Survival (%)

24 27 20 11 18 16 25 21 16 33

34 27 30 26 14 19 27 24 31 26

43 57 50 59 Similar

20 34 213*

36 35 31

45 38 41

43 53* 50 57

Statistically significant. Abbreviations: CT, Chemotherapy; GITSIG, Gastrointestinal Tumor Study Group; MRC, Medical Research Council; NSABP, National Surgical Adjuvant Breast and Bowel Project.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

Preoperative vs. Postoperative Radiotherapy Pa¨hlman and Glimelius (423) reported on a study that compared patients who received 2550 cGy preoperatively to those who received 6000 cGy postoperatively. The authors found no significant difference in the survival rate, but a local recurrence rate of 12% after preoperative radiotherapy was better than the 21% noted after postoperative radiotherapy. Preoperative and Postoperative Irradiation (Sandwich Therapy) The search continues to improve the survival rate in patients with carcinoma of the rectum and to reduce morbidity and unnecessary treatment. Using 500 cGy preoperatively is attractive since this amount of irradiation does not interface with wound healing or alter pathologic staging. The ‘‘sandwich technique’’ uses a single dose of 500 cGy preoperatively and 4500 cGy over 5 weeks postoperatively. The postoperative dose can be eliminated if the resected specimen does not warrant further radiotherapy. Sause et al. (467) used this protocol in 353 patients with carcinoma of the ascending and descending colon and rectum. Five-year follow-up estimates of no preoperative therapy vs. preoperative therapy were as follows: local recurrence, 29% vs. 26%; metastasis, 41% vs. 43%; and survival, 54% vs. 54%. No benefit was derived from preoperative treatment. Bayer et al. (468) proposed a different treatment regimen, with 3000 cGy administered preoperatively, and only patients with Dukes’ B or C lesions received another 3000 cGy postoperatively. However, the follow-up time was too short to permit any conclusions. From a nonrandomized retrospective analysis Botti et al. (469) compared 124 patients who received preoperative and postoperative radiotherapy (40), postoperative radiotherapy (30), or operation alone (54). Operative mortality was 2% in the sandwich radiotherapy group vs. 7% in the operation alone group. After a median follow-up of 60 months, the actuarial locoregional recurrence rate at 5 years was 3% for the sandwich radiotherapy group compared with 18% and 30% for the postoperative radiotherapy and surgery group alone, respectively. The actuarial 5-year survival rates were 86%, 50%, and 28% in the sandwich radiotherapy group, postoperative radiotherapy group, and operation alone group, respectively. Intraoperative Irradiation For selected patients in whom a fixed carcinoma has responded to external beam radiation, Tepper et al. (470) suggest that the addition of intraoperative radiation therapy increases the regional and probably the overall disease-free survival rate beyond that found when external beam radiation alone is used. Encouraging preliminary results also have been reported by Sischy (471). Intraoperative radiotherapy has been recommended for patients with fixed, unresectable carcinomas. It is believed that direct delivery of the high-dose radiotherapy may enhance control of the disease. Valentini et al. (472) reported on 15 studies at the fifth IORT International Congress held in 1994; 700 patients with rectal carcinoma had undergone intraoperative radiation therapy. In primary carcinoma unresectable for cure and in local recurrence it is evidenced that external beam radiotherapy plus operation plus intraoperative radiation therapy enable an improvement in local

273

control and survival compared to external radiotherapy alone. In the Mayo Clinic experience 50 patients treated with external beam radiotherapy (45–50 Gy) plus operation showed a 24% local control and 24% three-year survival, while 20 patients in the intraoperative radiation therapy group (10–20 Gy) showed 80% local control and 50% three-year survival. At the Massachusetts General Hospital 103 patients with rectal carcinoma unresectable for cure were given preoperative radiotherapy (50.4 Gy) and, if at operation frozen sections revealed residual disease or a disease-free margin less than 5 mm, intraoperative radiation therapy (15 Gy) was performed. Local control in the former was 67%, while it was 82% in the latter group. Five-year disease-free survival was similar (55% vs. 54%). Investigators of the Fox Chase Cancer Center in Philadelphia stated that in the presence of residual carcinoma intraoperative radiation therapy is ineffective. Significant intraoperative radiation therapy-induced perioperative and late toxicities have been reported—irradiation of ureters resulting in a 44% rate of obstruction requiring stents, peripheral neuropathies (32%), bone necrosis, fatal bleeding from a Hartmann pouch, and enteritis. Gunderson et al. (473) evaluated 123 patients with previously unirradiated locally recurrent colorectal carcinomas who received intraoperative electron radiation therapy usually as a supplement to external beam irradiation and maximum resection. All received external beam radiation therapy with or without concomitant 5-FU-based chemotherapy (45 Gy in 25 fractions and a boost of 5.4 to 9 Gy in three to five fractions). Maximum resection was performed before or after external beam radiation. Intraoperative electron radiation therapy doses ranged from 10 to 20 Gy in 119 of 123 patients, Central failure (within the intraoperative electron field) was documented in 11%, with a five-year actuarial. rate of 26%. Local relapse (in the external beam field) occurred in 20%, with a five-year rate of 37%. Distant metastases occurred in 54%, with a five-year rate of 72%. Median survival was 28 months, with overall survival at two, three, and five years of 62%, 39%, and 20%, respectively. Although there was a trend for reduction in local relapse rates with gross total versus partial resection, this neither achieved statistical significance nor translated into improved survival. Disease control within the intraoperative electron and external fields is decreased when the surgeon is unable to accomplish a gross total resection. Therefore it is reasonable to consistently add 5-FU or other dose modifiers during external beam radiation therapy and to evaluate the use of dose modifiers in conjunction with intraoperative electron radiation therapy (sensitizers and hyperthermia). Even with locally recurrent lesions, the aggressive multimodality approaches, including intraoperative electron radiation therapy, have resulted in improved local control and long-term survival rates of 20% versus an expectecd 5% with conventional techniques. Sadahiro et al. (474) reported the efficacy of intraoperative radiotherapy for curatively resected rectal carcinoma in 62 patients who received preoperative radiotherapy with 20 Gy. Retrospective comparisons were made with 248 patients treated by operation alone. Survival, disease-free survival, and local recurrence-free survival in the intraoperative radiotherapy group were significantly more favorable than in the nonintraoperative radiotherapy

274

& PART I: COLORECTAL DISORDERS

group. Differences in survival were observed in stage II patients but not in stage I or stage III patients. The local failure rate was 2.6% in the intraoperative radiotherapy group and 11.3% in the nonintraoperative radiotherapy group. The distant metastasis rate was 18% in the intraoperative radiotherapy group and 19.5% in the nonintraoperative radiotherapy group. There was a significantly higher rate of wound infection in the intraoperative radiotherapy group. However, in reviewing a series of 71 patients who received intraoperative radiotherapy for locally advanced carcinoma, Fuchs and Bleday (475) noted high rates of complications, 78.9% averaging 2.83 complications per patient. One of the most severe complications observed was massive iliac artery bleeding or blowout. Two such cases were successfully treated with angiographic embolization, both patients presented to the hospital emergently with considerable blood loss requiring intensive care unit resuscitation in the range of 6 to 19 units of blood. The necessity for specialized expensive equipment, as well as the cooperation of the radiation oncologist, anesthesiologist, and surgeon, will limit the acceptance of this regimen unless dramatic results can be obtained.

& CHEMOTHERAPY Adjuvant chemotherapy involves the postoperative use of cancericidal drugs to eradicate microfoci of malignant cells. The most useful single agents for colorectal carcinoma have been 5-FU and 20 -deoxy-5-fluorouridine (5-FUDR). Most investigators use 5-FU because of its identical effectiveness to 5-FUDR and lower cost. 5-FU blocks the formation of thymidylic acid and therefore the biosynthesis of DNA. The NSABP Protocol R-01 (359) used 5-FU, semustine, and vincristine and found an improved disease-free survival rate and overall survival rate. However, the observed advantage was restricted to males and was more advantageous in those younger than 65 years of age. The advantage continues to be significant at eight years (476). The GITSG trial reported an overall survival benefit for patients treated with adjuvant postoperative radiation (4000–4800 cGy) and chemotherapy (5-FU and semustine) following low anterior resection or abdominoperineal resection for Dukes-Kirklin B2, C1, and C2 carcinomas (459). The recurrence rate was 55% in control patients and 33% among patients receiving a combination of radiation therapy and chemotherapy. O’Connell et al. (477) studied 660 patients with TNM stage II and III rectal carcinoma to determine whether the efficacy of chemotherapy could be improved by administering 5-FU by protracted infusion throughout the course of radiotherapy and whether the omission of semustine would reduce the toxicity and delayed complications of chemotherapy without decreasing its efficacy. At a median follow-up of 46 months, patients who received a protracted infusion had a significantly increased time to relapse, 53% to 63% (a decrease of 27%), and improved survival, 60% to 70% (a decrease of 30% in their death rate). The addition of semustine provided no benefit. In a study of 141 patients with curative resection for colorectal carcinoma, approximately one-third of which were carcinoma of the rectum, patients were randomized to a control group (operation alone); resection and 5-FU group; or resection, 5-FU, and levamisole group (478). With a five-year follow-up, patients who had received

levamisole had a survival advantage. Patients who died of recurrence of the carcinoma in the control group comprised 52%; in the group who underwent resection and 5-FU, 44%; and in the group receiving 5-FU and levamisole, 32%.

& COMBINATION CHEMORADIOTHERAPY Frustration with inadequate results has led to the recommendation of combined chemotherapy and radiotherapy. GITSG published encouraging results (459). In a four-arm study that compared operation alone, chemotherapy (5-FU and semustine), radiotherapy (4000–4800 cGy), and combined radiation (4400 cGy) and chemotherapy (5-FU and semustine), the results showed an advantage for combined treatment for time to recurrence and survival rate. It has been suggested that the preoperative irradiation combined with 5-FU suppositories can reduce the incidence of postoperative recurrence (479). In an effort to improve resectability and possibly survival, Sischy et al. (480) employed chemosensitizers (5-FU and mitomycin) combined with moderate-dose radiation for carcinomas larger than 5 cm requiring abdominoperineal resection. In a series of 60 patients, the size of the lesions at operation decreased by more than 50% in over 75% of patients. No residual carcinoma was found in 10%, and only microscopic foci were present in another 21%. Only 24% of resected specimens contained positive lymph nodes. Of the 33 patients followed up for five years, 60% were alive and well. As in the treatment of anal carcinoma, chemosensitizers may play a role in the treatment of rectal carcinoma in the future. Krook et al. (358) reported on the NCCTG 794751 trial. Patients with T3, T4, N1, N2 (equivalent to Dukes’ B and C) lesions were randomized to treatment with postoperative radiation alone (45 Gy delivered in 180 cGy fractions five days a week over five weeks plus a 5.4 Gy boost delivered in 180 cGy fractions to the carcinoma bed after acute radiation tolerance was demonstrated) or to radiation (similar dosage) plus concurrent 5-FU. The latter treatment was both preceded and followed by a cycle of systemic therapy with 5-FU plus methyl-CCNU (semustine). For the combined modality, chemotherapy was initiated with a single dose of methyl-CCNU at 130 mg/m2 of body surface area. 5-FU was administered by rapid intravenous injection at 300 mg/m2 on days 1 through 5. On day 36, a course of 5-FU was given at 400 mg/m2 daily for five days. Four weeks later (on day 64), therapy of 5-FU and radiation was initiated. 5-FU was given by rapid intravenous injection at a dose of 500 mg/m2 during the first three days of radiation and repeated during the first three days of the fifth week of radiation. One month after radiation, a single dose of methyl-CCNU was given (100 mg/m2) plus 5-FU (300 mg/m2) for five days. One month later, 5-FU was given at 400 mg/m2 daily for five days. After a median follow-up time of more than seven years, the combined modality showed a relative reduction in recurrence of 34% (63% vs. 42%), with an initial 46% reduction in pelvic recurrence and a 37% reduction in distant metastases. In addition, the overall patient death rate was reduced by 29%, and carcinoma-related deaths dropped by 36%. The five-year disease-free survival rate increased from 42% to 63%, and the overall five-year survival rate increased from 47% to 58%.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

In the NSABP R-02 trial, 694 eligible patients with Dukes’ B or C carcinoma of the rectum were randomly assigned to receive either postoperative adjuvant chemotherapy alone (n ¼ 348) or chemotherapy with postoperative radiotherapy (n ¼ 346) (481). All female patients (n ¼ 287) received 5-FU plus LV chemotherapy; male patients received either MOF (n ¼ 207) or 5-FU plus LV (n ¼ 200). The average time on study for surviving patients was 93 months. Postoperative radiotherapy resulted in no beneficial effect on disease-free survival or overall survival regardless of which chemotherapy was utilized, although it reduced the cumulative incidence of locoregional relapse from 13% to 8% at five-year follow-up. Male patients who received 5-FU plus LV demonstrated a statistically significant benefit in disease-free survival at five-years compared with those who received MOF (55% vs. 47%) but not in five-year overall survival (65% vs. 62%). The addition of postoperative radiation therapy to chemotherapy in Dukes’ B and C rectal carcinoma did not alter the subsequent incidence of distant disease, although there was a reduction in locoregional relapse when compared with chemotherapy alone. While a logical argument may be made for the elimination of radiotherapy in the postoperative setting on the basis of the outcome from the study described here, enthusiasm for the approach must be tempered by the confirmed demonstration that radiotherapy is effective in reducing the incidence of locoregional recurrence, an event that can be associated with substantial morbidity and an attenuation in quality of life. Whether the 5% absolute decrease in the cumulative incidence of locoregional relapse is sufficient to justify the routine use of postoperative radiotherapy is a decision that must be made by the clinician. In a thoughtful and provocative editorial that followed this publication, Haller (482) noted that although it is accepted that a beneficial outcome of adjuvant treatment of rectal carcinoma is prevention of clinically relevant local recurrences, most clinicians and patients would agree that overall survival and quality of life are more pertinent end points of treatment. Since it is likely that most of the long-term morbidity of postoperative chemotherapy and radiation therapy is due to the late effects of the radiation therapy, it is reasonable to ask whether all patients should be exposed to this treatment and whether similar clinical outcomes could be achieved with improved surgical techniques or with systemic chemotherapy alone. If radiation therapy reduces local recurrence, then perhaps its use should be limited to those patients with particularly highrisk anatomic or biologic determinants for this pattern of failure, i.e., those with greater local extension (macroscopic T3 or T4) or multiple lymph node involvement. Theodoropoulos et al. (483) evaluated the impact of response to preoperative and, specifically, of T-level downstaging, nodal downstaging, and complete pathologic response after chemoradiation therapy on oncologic outcome of 88 patients with locally advanced rectal carcinoma. T-level downstaging after neoadjuvant treatment was demonstrated in 41% and complete pathologic response was observed in 18%. Of the 42 patients with ultrasound-positive nodes, 27 had no evidence of nodal involvement on pathologic evaluation (64%). The overall response rate (T-level downstaging or nodal downstaging)

275

was 51%. At a median follow-up of 33 months, 86.4% of patients were alive. The overall recurrence rate was 10.2% (three patients had local and six had metastatic recurrences). Patients with T-level downstaging and complete pathologic response were characterized by significantly better five-year survival and better overall survival. None of the patients with complete pathologic response developed recurrence or died during the follow-up period. Burmeister et al. (484) conducted a phase 2 study to collect data prospectively on the toxicity of postoperative combined chemoradiation therapy. The prescribed radiation dose was 50.4 Gy in 28 fractions, and the 5-FU chemotherapy was 450 mg/m2 given with fractions 1 to 3 and 26 to 28 (n ¼ 80). On completion of the radiation therapy, the patient was given a further four cycles of bolus 5-FU at monthly intervals. Acute toxicity of the therapy was significant with 16% of patients experiencing severe bowel morbidity. The other major side effects of the therapy were skin reactions, neutropenia and bladder problems. Late bowel toxicity was also severe. The local in field relapse rate was 10%. The majority of relapses were at distant sites, mostly in the liver and lungs. The actuarial survival at five years was 55%. They concluded that the combined adjuvant postoperative chemoradiation therapy using their protocol was effective but had significant acute and late morbidity. The optimum regimen for those patients requiring postoperative adjuvant therapy is yet to be determined. Tepper et al. (485) published the final report of the gastrointestinal Intergroup 0114. All patients had a potentially curative resection and were treated with two cycles of chemotherapy followed by chemoradiation therapy and two additional cycles of chemotherapy. Chemotherapy regimens were bolus 5-FU, 5-FU and leucovorin, 5-FU and levamisole, and 5-FU leucovorin and levamisole. Pelvic irradiation was given to a dose of 45 Gy to the whole pelvis and a boost to 50.4 to 54 Gy. One thousand, six hundred and ninety-five patients were entered and fully assessable with a median follow-up of 7.4 years. There was no difference in overall survival or disease-free survival by drug regimen. Disease free survival and overall survival decreased between years five and seven (from 54% to 50% and 64% to 56%, respectively), although recurrence free rates had only a small decrease. The local recurrence rate was 14% [9% in low-risk (T1 to N2þ) and 18% in high-risk patients (T3Nþ, T4N)]. Overall, sevenyear survival rates were 70% and 45% for the low-risk and high-risk groups, respectively. Males had a poorer overall survival rate than females. There is no advantage to leucovorin or levamisole containing regimens over bolus 5-FU alone in the adjuvant treatment for rectal carcinoma when combined with irradiation. Local and distant recurrence rates are still high especially in T3Nþ and T4 patients even with full adjuvant chemoradiation therapy. The EORTC trial compared preoperative radiotherapy (34.5 Gy) to the same treatment and chemotherapy (5-FU) and found no difference in local recurrence, distant metastases, disease-free survival, or overall survival (486). In an ongoing effort to improve survival, preoperative chemoradiation has been tried by several investigators. In a study by Meade et al. (487) patients underwent preoperative endorectal ultrasonography staging followed by

276

& PART I: COLORECTAL DISORDERS

high-dose radiotherapy combined with 5-FU. Following resection, the downstaging was recorded in 14 of 20 patients and no residual disease was present in eight patients. Local recurrence developed in two patients. Disease-free survival was noted in 17 patients with 9 to 51 month follow-up. Others have also reported encouraging results for preoperative chemoradiation (488). In a nonrandomized trial, Stryker et al. (489) compared the results of chemoradiation (45–50 Gy plus 5-FU plus mitomycin, four to eight weeks prior to operation) in 30 patients with stage II and III rectal carcinoma with the results of treatment in 56 patients who did not undergo preoperative chemoradiation of which 24 patients received postoperative adjuvant chemoradiation. Five-year actuarial control rates were 96%, 83%, and 88%, respectively. Diseasefree survival rates were 80%, 57%, and 47%, respectively. Overall survival rates were 85%, 48%, and 78%, respectively. Kollmorgan et al. (490) studied the adverse effects of long-term bowel function following postoperative chemoradiotherapy for rectal carcinoma. The authors compared two similar groups of patients who underwent anterior resection. One group received postoperative radiation and chemotherapy, while the other did not. The chemoradiotherapy patients had more bowel movements per day (median seven vs. median two) and more of these patients had clustered bowel movements (42% vs. 3%), nighttime movements (46% vs. 14%), incontinence (occasional, 39% vs. 7%; frequent, 17% vs. 0%), and inability to differentiate stool from gas (39% vs. 15%). More patients in the chemoradiation group had to wear a pad (41% vs. 10%), were unable to defer defecation longer than 15 minutes (78% vs. 19%), had liquid stools sometimes or always (29% vs. 5%), regularly used Lomotil and/or Imodium (58% vs. 5%), and had perianal skin irritation (41% vs. 12%). Overall, 93% of the patients receiving chemoradiation reported that their bowel function was different than before operation versus 61% of the nonradiation group who so reported. Although it may improve survival, adjuvant postoperative chemoradiotherapy also results in significant long-term detrimental effects on bowel function. Physicians should inform patients of these effects before the treatment is initiated. Picciocchi et al. (491) entered 64 patients with stage II or III rectal carcinoma into a study to receive preoperative radiotherapy, 37.8 Gy combined with continuous 5-FU and mitomycin C. Toxicity was recorded in 27% of patients with one patient dying from the effects of chemotherapy. Of the 61 patients operated on, 46 underwent low anterior resections and 15 underwent abdominoperineal resection. There were no postoperative mortalities but morbidity was recorded at 28%, 8.7% sustaining anastomotic dehiscences. Definitive histologic staging was downgraded from initial clinical staging (stage 0, 0–5; stage I, 0–19; stage II, 13–21; stage III, 48–15; and stage IV, 0–1). With a median follow-up of 23 months, the incidence of local recurrence was 5% and that of distant metastases was 8%. Multidisciplinary treatment efforts continue to select patients who would benefit most from perioperative treatment while minimizing toxicity. Suggested criteria for the selection of neoadjuvant therapy include patients with carcinomas of the distal rectum that are locally advanced by either clinical or imaging criteria, poorly differentiated

carcinomas, circumferential lesions, or carcinomas that are obstructing or perforated (492). Garcia-Aguilar et al. (493) studied the prognostic value of pathologic complete response to preoperative chemoradiation in rectal carcinoma patients. They prospectively followed up 168 consecutive patients with ultrasound stages II (57) and III (122) rectal carcinoma treated by preoperative chemoradiation followed by radical resection with mesorectal excision; 161 had a curative resection. Average follow-up was 37 months. Downstaging occurred in 58% of patients, including 13% patients who had a pathologic complete response. The estimated five-year rate of local recurrence was 5%; of distant metastases, 14%. None of the patients with pathologic complete response has developed disease recurrence. They found no difference in survival among patients with pathologic stage I, stage II, or stage III carcinomas. They concluded a pathologic complete response to preoperative chemoradiation is associated with improved local control and patient survival. Ruo et al. (494) found that a marked response to preoperative radiotherapy þ/ chemotherapy may be associated with good long-term outcome but was not predictive of recurrence free survival. The presence of poor histopathologic features and positive nodal status are the most important prognostic indicators after neo-adjuvant therapy. Stipa et al. (495) evaluated the impact of preoperative radiation and chemotherapy on primary rectal carcinoma and mesorectal lymph nodes in 187 consecutive patients who underwent abdominoperineal resection or low anterior resection for locally advanced stage T3-4. Comparison of pre-combined modality therapy, endorectal ultrasound stage with pathologic stage revealed a decrease in T stage in 49% as well as a decrease in the percentage of individuals with positive mesorectal lymph nodes, from 54% to 27%. The incidence of residual mesorectal lymph node involvement remains significant and parallels increasing stage. They believe that locally advanced distal rectal carcinoma should continue to include formal resection. Rullier et al. (496) reported 43 patients who underwent preoperative radiochemotherapy (50 Gy range 40–54) and concomitant chemotherapy with 5-FU continuous infusion (n ¼ 36) or bolus (n ¼ 7) for rectal carcinoma located 2 to 6 cm from the anal verge. There were 40 T3 lesions, and three T4 lesions. Sphincter saving resection was performed six weeks after treatment in 25 patients by using intersphincteric resection. Coloanal anastomoses were associated with a colonic pouch in 86% of the patients, and all patients had a protecting stoma. There were no deaths related to the preoperative radiochemotherapy and operation. Acute toxicity was mainly due to diarrhea with 54% of grade 1 to 2. Four anastomotic fistulas and two pelvic hematomas occurred; all patients but one had closure of the stoma. Downstaging was observed in 42% of the patients and was associated with a greater radial margin. After a median follow-up of 30 months, the rate of local recurrence was 2%, and 10% had distal metastases. Overall and disease-free survival rates were both 85% at three years. Functional results were good (Kirwin continence I, II) in 79% of the available patients (n ¼ 37). They were slightly altered by intersphincteric resection (57% vs. 75%) for perfect continence; but were significantly

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

improved by a colonic pouch (74% vs. 16%). They concluded these results suggested preoperative radiochemotherapy allowed sphincter saving resection to be performed with good local control and good functional results in patients with T3 low rectal carcinomas. Read et al. (497) reviewed 191 consecutive patients undergoing abdominal surgical procedures for primary rectal carcinoma, 89% of whom were treated with preoperative external beam radiotherapy. Curative resection was performed in 80% including low anterior resection with coloproctostomy or coloanal anastomosis (n ¼ 103), abdominoperineal resection (n ¼ 44), Hartmann’s procedure (n ¼ 4), and pelvic exenteration (n ¼ 1). Mean follow-up of patients undergoing curative resection was 96 months. Palliative procedures were performed in 20%. Perioperative mortality was 0.5%. Complications occurred in 34%. The anastomotic leak rate was 4%. Disease-free five-year survival rate by pathologic stage was as follows: stage I, 90%; stage II, 85%; stage III, 54%; stage IV, 0% and no residual carcinoma, 90%. Of the 152 patients treated with curative resection, disease-free survival rate was 80% at five years. The carcinoma recurred in 21% treated with curative resection. The predominant pattern of recurrence was distant failure only. Overall, local recurrence (local and local plus distant) at five years was 6.6%. The local recurrence rate paralleled stage: stage I, 0%; stage II, 6%; stage III, 20%; and no residual carcinoma, 0%. The review of retrospective data by Minsky (410) suggests that preoperative combined modality therapy increases pathologic downstaging compared with preoperative radiation without chemotherapy, and is associated with a lower incidence of acute toxicity compared with postoperative combined-modality therapy. In general, the incidence of grade 3/4 acute toxicity during the combined-modality segment is 15% to 25%, the complete response rates are 10% to 30% pathologic, and 10% to 20% clinical, and the incidence of local recurrence is 2% to 10%. In his review of 5 oxaliplatin-based combined modality therapy studies in which 45 to 50.4 Gy were used in combination with 5-FU, leucovorin, and oxaliplatin (one study substituted ralitrexal for 5-FU) pathologic complete response rates varied from 14% to 37% (410). Volter et al. (498) combined irinotecan with hyperfractionated radiation (1.6 Gy twice daily to 41.6 Gy) and encountered a high incidence of anastomotic leak and/or abscess of 30%. Despite a lack of randomized data demonstrating clinical benefit, preoperative chemoradiation has been increasingly used in patients with T3 disease in North America (499). Luna-Perez et al. (500) evaluated the feasibility, morbidity, and functional results of anal sphincter preservation after preoperative chemoradiation therapy and coloanal anastomosis in 32 patients with rectal carcinoma located between 3 and 6 cm above the anal verge. Twenty-two patients underwent coloanal anastmosis with the J-pouch; ten underwent straight anastomosis. The mean distal surgical margin was 1.3 cm. Major complications included coloanal anastomotic leakage (3); pelvic abscess (3); and coloanal stenosis (2). Mean follow-up was 25 months. Recurrences occurred in four patients and were local and distant (1) and distant (3). Anal sphincter function was perfect (20), incontinent to gas (3), occasional minor leak (2), frequent major soiling (3), and colostomy (n ¼ 2).

277

They concluded that in patients with locally advanced rectal carcinoma located 3 to 6 cm from the anal verge who are traditionally treated with abdominoperineal resection, preservation of anal sphincter after preoperative chemoradiation therapy plus complete rectal excision with coloanal anastomosis is feasible and is associated with acceptable morbidity and no mortality. To determine whether patients downsized with preoperative chemoradiotherapy may be potential candidates for local excision, Bedrosian et al. (501) investigated residual disease patterns in 219 patients after neoadjuvant treatment. Preoperatively, 88% were staged as T3 and 47% had clinical N1 disease. The pathologic complete response rate was 20%. T stage was downsized in 64% of the patients and 69% of patients with clinical N1 disease were rendered node negative. Seventeen percent of patients downsized to less than or equal to T2 had residual disease in the mesentery. With a median follow-up of 40 months, 83% of patients remain alive and free of disease. Local recurrence developed in 4.1% of patients. Although response rates to preoperative chemoradiotherpay within the bowel wall and lymph node basin are similar, one in six patients with pT0-2 carcinomas will have residual disease in the rectal mesentery and nodes. Stipa et al. (495) evaluated the impact of preoperative radiation and chemotherapy on primary mid and distal rectal carcinomas and mesorectal lymph nodes in 187 consecutive patients who underwent abdominoperineal resection or low anterior resection for locally advanced (T3-4) carcinoma. Comparison of pre-combined modality therapy, endorectal ultrasound stage with pathologic stage revealed a decrease in stage in 49% as well as a decrease in the percentage of individuals with positive mesorectal lymph nodes from 54% to 27%. They concluded following preoperative combined modality therapy, the incidence of residual mesorectal lymph node involvement remains significant and care for locally advanced distal rectal carcinoma should continue to include formal rectal resection. Enker (502) noted that despite recent enthusiasm, neoadjuvant therapy in previous studies have indicated that from 10% to 30% of patients demonstrate objective regression of disease (i.e., a complete pathologic response) after neoadjuvant therapy. The corollary is that 70% to 90% of patients harbour residual mesorectal carcinoma despite their preoperative treatment requiring formal curative rectal and mesorectal resection. Furthermore, no proven correlation exists between the regression of the primary carcinoma and the regression of regional, mesorectal node disease. Whereas shrinkage of a bulky primary carcinoma situated within a narrow pelvis can offer the surgeon an easier opportunity to perform an adequate resection, shrinkage alone is not a sign of clinical downstaging in a disease that is overwhelmingly regional in its presentation (greater than or equal to T3, or N1–N2 in 65% to 80% of patients). Designing patient care around the aftereffects of response to chemoradiotherapy as opposed to determining treatment of the original stage of presentation has many pitfalls and hazards. Therefore, shrinkage of the primary carcinoma is not a sufficient foundation upon which to alter treatment from one’s original intentional to perform a regional resection in such patients despite regression of the primary lesion after radiation and chemotherapy.

278

& PART I: COLORECTAL DISORDERS

Onaitis et al. (503) reported up to 30% of patients with locally advanced rectal carcinoma have a complete clinical or pathologic response to neoadjuvant chemoradiation. They analyzed complete clinical and pathologic responders among 141 rectal carcinoma patients treated with neoadjuvant chemoradiation. Clinical restaging after treatment consisted of proctoscopic examination and often CT scan. Clinical complete responders had no advantage in local recurrence, disease-free survival, or overall survival rates when compared with clinical partial responders. Pathologic complete responders also had no recurrence or survival advantage when compared with pathologic partial responders. Of the 34 pathologic T0 lesions, 13% had lymph node metastases. They concluded clinical assessment of complete response to neoadjuvant chemoradiation is unreliable. Micrometastatic disease persists in a proportion of patients despite pathologic complete response. Hofheinz et al. (504) evaluated the feasibility and efficacy of capecitabine in combination with weekly irinotecan (CAPIRI) with concurrent pelvic radiotherapy in 19 patients with locally advanced rectal carcinoma. All patients underwent operation and R0 resection was achieved in all patients. Pathologic complete remission was observed in four patients and another five patients had only microfoci of residual disease. They concluded preoperative chemoradioatherapy with CAPIRI is feasible and well tolerated. It must be remembered that chemoradiation is not without its complications. Chessin et al. (505) reported on complications of 297 consecutive patients with locally advanced rectal carcinoma treated with preoperative combined modality therapy (radiation: 5040 cGy; chemotherapy: 5-FU-based) and then operation. Major complications were defined as those requiring medical or surgical treatment. Median follow-up was 43.9 months. There were no postoperative mortalities, but there were 145 major complications in 98 patients (33% of the study population). The most common complications were small bowel obstruction (11%) and wound infection (10%), anastomotic leaks (4%) and pelvic abscesses (4%) in patients treated with low anterior resection. Postoperative complications had no significant impact on oncologic outcomes. Although postoperative mortalities are rare, complications requiring treatment can be anticipated in one-third of patients undergoing preoperative combined modality therapy and TME. They recommend a policy of selective fecal diversion after preoperative combined modality therapy and TME for locally advanced rectal carcinoma to achieve low rates of pelvic sepsis but this may lead to an increased incidence of small bowel obstruction. Most recently, Habr-Gama et al. (506) conducted a study to determine the correlation between final stage and survival in patients receiving neoadjuvant chemoradiation treatment regardless of initial disease stage. Two hundred and sixty patients with distal (0–7 cm from anal verge) rectal carcinoma considered resectable were treated by neoadjuvant chemoradiotheapy with 5-FU and leucovorin plus 5040 cGy. Patients with incomplete clinical response eight weeks after chemoradiotherapy completion were treated by radical resection. Patients with complete clinical response were managed by observation alone. Seventy-one patients (28%) showed complete clinical

response (clinical stage 0). One hundred and sixty-nine patients showed incomplete clinical response and were treated with operation. In 22 of these patients (9%), pathologic examination revealed pT0 N0 M0 (stage p0), 59 patients (22%) had stage I, 68 patients (26%) had stage II, and 40 patients (50%) had stage III disease. Overall survival rates were significantly higher in stage c0 compared with stage p0. Disease-free survival rate showed better results in stage c0, but the results were not significant. Five year overall and disease-free survivals were 97.7% and 84% (stage 0); 94% and 74% (stage I); 83% and 50% (stage II); and 56% and 26% (stage III), respectively. Carcinoma-related overall and disease-free survival may be correlated to final pathologic staging following neoadjuvant chemoradiotherapy for distal carcinoma. Also stage 0 is significantly associated with improved outcome. Moore et al. (507) conducted a review to determine whether prolongation of the interval between preoperative combined modality therapy and operation resulted in an increased pathologic complete response rate. They identified 155 rectal carcinoma patients undergoing preoperative pelvic external beam radiation and 5-FU chemo based therapy followed by rectal resection. A pathologic complete response occurred in 15% of patients. A pathologic complete response occurred in 19% of patients with an interval greater than 44 days versus 12% in those with an interval less than or equal to 44 days. The benefit of a prolonged interval between completion of preoperative combined modality therapy and operation remains unclear. The German Rectal Cancer Group addressed the controversy of preoperative versus postoperative chemoradiotherapy (508). Patients with clinically staged T3/T4 Nþ rectal carcinoma were randomly assigned to preoperative or postoperative chemoradiotherapy: 50.4 Gy in 28 fractions were applied to the carcinomas and the pelvic lymph nodes. 5-FU (1 g/m2/day) was administered concomitantly during the first and fifth week of radiotherapy as 120 hours continuous infusion. Four additional cycles of 5-FU adjuvant chemotherapy (500 mg/m2/day  5 day, q four weeks) were delivered. Chemoradiotherapy was identical in both arms except for a small-volume boost of 5.4 Gy in the postop arm. The interval between chemoradiotherapy and operation was four to six weeks in both arms. Techniques of operation were standardized and included TME. Median follow-up was 43 months (range 4– 89 months). Of 823 patients randomized in 26 participating institutions, 392 and 405 were evaluable in the postop and preop chemoradiotherapy arms, respectively. The five-year pelvic and distant recurrence rates were 11% versus 7% and 34% versus 30%, respectively. Fewer patients experienced chronic anastomotic stenosis following preoperative chemoradiotherapy versus postoperative chemoradiotherapy (2.7% vs. 8.5%). Following preoperative chemoradiotherapy, there was significant down staging of the carcinoma with a 8% pathologic complete response rate. The UICC-stages were I to IV: 18%, 28%, 39%, 7%, missing 8% in the postoperative chemoradiotherapy arm versus 24%, 28%, 27%, 6%, missing 7% in the preoperative chemoradiotherpy arm. In the subgroup of 188 patients with low lying carcinomas who were declared by the surgeon prior to randomization to require an abdominoperineal resection, 19% underwent a sphincter saving procedure in the postoperative

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

chemotherapy arm. This was significantly increased to 39% following preoperative chemoradiotherapy. In the most recent update of their data, Sauer et al. (509) reported 421 patients were randomly assigned to receive preoperative chemoradiotherapy and 402 patients to receive postoperative chemoradiotherapy. The overall survival rates were 76% and 74%, respectively. The fiveyear cumulative incidence of local relapse was 6% for patients assigned to preoperative chemoradiotherapy and13% to the postoperative treatment group. Grade 3 and 4 toxicity effects occurred in 27% of patients in the preoperative treatment group as compared with 40% of patients in the postoperative treatment group. The corresponding rates of long-term toxic effects were 14% and 24%, respectively. They concluded preoperative chemoradiotherapy as compared to postoperative chemoradiotherapy improved local control and was associated with reduced toxicity but did not improve overall survival. Ravasco et al. (510) investigated the impact of dietary counseling or nutritional supplements on outcomes in carcinoma patients: nutritional, morbidity, and quality of life during and three months after radiotherapy. A total of 111 colorectal carcinoma patients referred for radiotherapy, stratified by stage, were randomly assigned: group 1 (n ¼ 37), dietary counseling (regular foods); group 2 (n ¼ 37), protein supplements; group 3 (n ¼ 37), ad libitum intake. Nutritional intake (dietary history), status (Ottery’s Subjective Global Assessment), and quality of life (European Organization for Research and Treatment of Cancer Quality of Life Questionnaire version 3.0) were evaluated at baseline, at the end, and three months after radiotherapy. At radiotherapy completion, energy intake increased in groups 1 and 2, group 1 more than group 2, and decreased in group 3. Protein intake increased in groups 1 and 2, group 1 less than group 2, and decreased in group 3.

279

At three months, group 1 maintained nutritional intake and group 2 and 3 returned to baseline. After radiotherapy and at three months, rates of anorexia, nausea, vomiting, and diarrhea were higher in group 3. At radiotherapy completion, in group 1, all quality of life function scores improved proportionally to adequate intake or nutritional status; whereas in group 2 only three of six function scores improved proportionally to protein intake, and in group 3 all scores worsened. At three months, group 1 patients maintained/improved function, symptoms, and singleitem scores; in group 2, only few function and symptoms scores improved; in group 3, quality of life remained as poor as after radiotherapy. The results of the randomized trials of combined chemoradiation is presented in Table 11.

& IMMUNOTHERAPY Malignant cells have repeatedly been found in the systemic venous blood in a high percentage of patients undergoing colonic or rectal excision for carcinoma (14). Their presence or absence in the peripheral blood during the operation in no way correlates with survival, suggesting that host factors are important in preventing the growth of these circulating cells (513). Many other observations support the premise of a close relationship between immunologic competency and the growth of human carcinomas (513). The rationale of using immunotherapy for carcinoma assumes patients have potential antineoplastic immunity that is either blocked or at a low level, but that can be effectively stimulated to destroy malignant cells (514) A variety of modalities have been used to increase immunocompetence. Specific methods include the use of living carrier cells, neuraminidase, and purified tumor antigen (515). Nonspecific immunologic adjuvants have been shown to

TABLE 11 & Results of Randomized Trials of Combined Chemoradiation

Trial (Author)

No. of Patients

Dose (Gy/Fraction)

EORTC (486) (Boulis-Wassif et al., 1984) GITSG (459) (1985)

247

NCCTG (358) (Krook et al., 1991) GITSG (511) (1992)

209

Intergroup (477) (O’Connell et al., 1994) NARCPG (512) (Tveit et al., 1997) NSABP R-02 (481)

660

Preop 34.5 Gy/15.0 Preop 34.5 Gy/15.0 þ 5-FU Operation alone CT Postop 40–48 Gy/22–27 CT þ 40–44 Gy/22–21 Postop 45 Gy/25 þ 5.4 Gy boost Postop 45 Gy/25 þ 5-FU Postop 44.4 Gy/23 þ 5-FU Postop 41.4 Gy/23 þ 5-FU þ Semustine 50.4/28 Gy þ CT bolus 54 Gy/30 þ CT continuous Operation alone Postop 46 Gy þ 5-FU MOF 5-FU-leucovorin 5-FU þ postop 46 Gy (26/5) 5-FU-leucovorin þ postop 46 Gy Preop 50.4/28 Postop 50.4/28 þ 5.4 Gy

227

210

144 694

(Wolmark et al., 2000) CAO/ARO/A10–94 Sauer (508) (2003)

405 392

Local Recurrence (%) 15 15 24 27 20 11 23 14a 15 11 11 8 30 12a 13

g g8

a

7%a 11%

Distant Metastases (%)

Overall Five-year Survival (%)

30 30 34 27 30 26 46 29a 25 33 40 31

59 46 43 57 50 59 38 53a 44 46 60 70a 46 64

-

g29 g31 30% 34%

g 66 g 68 78% 73%

a Statistically significant. Abbreviations: CT, chemotherapy; EORTC, European Organisation for Research and Treatment of Cancer; GITSG, Gastrointestinal Tumor Study Group; MOF, methyl CCNU, vincristine sulfate (Oncovin), 5-FU; NCCTG, North Central Cancer Treatment Group; NSABP, National Surgical Adjuvant Breast and Bowel Project. NARCPG Norwegian Adjuvant Rectal Cancer Project Group.

280

& PART I: COLORECTAL DISORDERS

be effective in eliciting an immune response against a wide range of neoplasms, including colorectal carcinoma. The most widely used immunoadjuvants are bacillus Calmette. Guerin (BGG), Corynebacterium parvum (C. parvam), and methanol-extracted residue of BCG (MER). Only nonspecific adjuvants have been used in colorectal carcinoma. Another approach is combination immunotherapy and chemotherapy Chemoimmunotherapy aims at combining the ‘‘debulking’’ capacity of chemotherapy with the potential of immunotherapy for controlling microscopic disease and therefore producing long-term disease-free survival. Reed et al. (516) noted that patients given C. parvum had less immunosuppression secondary to chemotherapy than the control group and were able to receive chemotherapy twice as frequently. It may be that C. parvum exhibited no immunotherapeutic value in these patients but merely allowed the more aggressive utilization of chemotherapy. Immunotherapy is most effective when the bulk of the carcinoma is small, either from surgical removal or in response to chemotherapy or radiation therapy. Immunotherapy can only be effective in an immunocompetent host. Immunotherapy at the M.D. Anderson Hospital in Houston using BCG showed improvement in survival rates (517). Unfortunately, the results were not compared to a control group, making interpretation difficult. In the NSABP study, BCG was used but did not result in improved survival rates (359).

& SUMMARY Despite a lack of randomized data demonstrating clinical benefit, preoperative chemoradiation has been increasingly used in patients with T3 disease in North America (499). The phlethora of information from the published data and the controversies amongst the various experts makes it very difficult to mandate a course of action for each clinical situation. Even the magnitude of the different problems encountered in the treatment of rectal carcinoma varies tremendously. In their comprehensive review of the literature, Colquhoun et al. (362) analyzed nine randomized clinical trials designed to determine the effects of neoadjuvant therapy in the treatment of rectal carcinoma with a follow-up that varied from two to seven years. The recurrence rates following operation alone ranged from 8% to 30%. With the addition of adjuvant therapy, rates ranged from 2.4% to 27%. The death rate was reduced by 25% to 29% in two studies in which postoperative chemoradiotherapy was given and by 21% in one study in which preoperative chemoradiotherapy was given. In the remaining six trials, no survival benefit was derived from adjuvant therapy. In a meta-analysis of 36 randomized clinical trials using neoadjuvant and adjuvant therapy, Ooi et al. (518) demonstrated the morbidity of such treatments to be frequent but tolerable. Short-term (acute) complications of preoperative radiotherapy include lethargy, nausea, diarrhea, and skin erythema or desquamation. These acute effects develop to some degree in most patients during treatment but are usually self-limiting. With preoperative radiotherapy the incidence of perineal wound infection increases from 10% to 20%. The acute toxicities after postoperative radiotherapy for rectal carcinoma occur in 4% to 48% of cases, and serious toxicities requiring

hospitalization or operative intervention occur in 3% to 10% of cases. The main problems with postoperative radiotherapy are small bowel obstruction (5–10%), delay in starting radiotherapy caused by delayed wound healing (6%), postoperative fatigue (14%), and toxicities precluding completion of adjuvant therapy (49–97%). The morbidity and mortality of both preoperative and postoperative radiotherapy are higher in elderly patients and when two-portal rather than three-portal or four-portal radiation technique is used. After combined adjuvant chemotherapy and radiotherapy, acute hematologic and gastrointestinal toxic effects are frequent (5–50%). Delayed radiation toxicities include radiation enteritis (4%), small bowel obstruction (5%), and rectal stricture (5%). Nevertheless, Colquhoun et al. (362) concluded in their review that adjuvant treatment remains indicated for patients with T3-4N0 and T1-4N1-2 carcinomas. Careful consideration should be given to those patients greater than 80 years who are less likely to experience recurrence based on age alone and who appear less tolerant of the toxicity associated with adjuvant therapy of any sort. Despite many randomized trials of adjuvant therapy for rectal carcinoma, no consensus currently exists. Although the early trials of postoperative radiation and/ or chemotherapy led to the 1991 NIH Consensus Statement, subsequent protocols of preoperative therapy challenged the consensus. Given the variation in timing and amount of radiation therapy used in existing trials, and given the lack of consensus on this topic, decision analysis provides a means to determine optimal adjuvant therapy. Kent et al. (519) conducted a literature search. Baseline values and ranges were determined from reference sources restricted to randomized controlled trials. Model variables included adverse effects and overall survival. They found postoperative combination therapy (expected value ¼ 0.68) is the preferred treatment in terms of overall survival, in comparison to postoperative radiation alone, preoperative combination therapy, and preoperative radiation alone. Their decision analysis demonstrated a preference for postoperative combination therapy for stage II and III rectal carcinoma in terms of overall survival.

& POSTOPERATIVE COMPLICATIONS The complications associated with rectal surgery are discussed in detail in Chapter 36.

& RECURRENT DISEASE & FOLLOW-UP The rationale and policy for the follow-up of patients with carcinoma of the large bowel is detailed in Chapter 25. Most anastomotic recurrences will be detected within two years of operation. & INCIDENCE A wide range of recurrence rates has been reported. As might be expected, the incidence of recurrence increases with the duration of follow-up. Long-term follow-up and

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

an intensive search for recurrence, including a high autopsy rate, are factors that yield an increase in both total and local recurrence rates (227). One series of 101 patients followed up for five years turned up a recurrence rate of 39%, whereas a series of 231 patients followed up for 18 years yielded an incidence of 54% (227). In a thorough review of the literature, Sagar and Pemberton (520) found that the incidence of local recurrence after curative resection of rectal carcinoma ranged from 3% to 32%. Values varied, depending on the Dukes’ stage of the disease and the portion of the rectum involved, that is, the upper, middle, or lower third. Lower values were seen with Dukes’ A and proximal rectal lesions, whereas high values were seen with Dukes’ C and distal rectal lesions. Abdominoperineal resections often have been recommended over low anterior resections because of the fear of local recurrence. However, when compared for the same degree of differentiation and stage of the carcinoma, local recurrence rates following low anterior resection are reportedly the same as rates for abdominoperineal resection (66,68,520,521). Other prognostic discriminants were fully discussed in Chapter 23. In a review of 1008 patients undergoing potentially curative resection, McDermott et al. (331) reported that local recurrence developed in 14% of patients after resection of the upper third of the rectum compared with 21% for the middle third and 26% for the lower third. No relationship between local recurrence and the type of curative resection performed was documented. In a study that supported an intensive follow-up program to detect early local recurrence of colorectal carcinoma, Schiessel et al. (522) found that 22% of patients developed recurrence. Most of the recurrences (76%) developed after operation for rectal carcinoma. The rate of local recurrence was similar after a sphincter-saving operation (14.4%) as for an abdominoperineal resection (16.7%). An overall recurrence rate for rectal carcinoma was 14.7% as opposed to 4.4% after colonic carcinoma. Local recurrence only developed in 74.6% of patients, whereas disseminated disease was manifest in the other 25.4%. Castro-Sousa et al. (523) reported on 84 patients with carcinoma of the rectum who underwent radical excision. There was no significant difference in the incidence of recurrence between the group with sphincter preservation (17%) and abdominoperineal resection (13%). Rullier et al. (524) also found no difference in local recurrence between sphincter preservation (27.8%) and abdominoperineal resection (30.8%), but these rates are appreciably higher. Following low anterior resection, Pilipshen (525) found that local recurrence rates varied from 14% to 43%. Local recurrence rates following low anterior resection with the use of the circular stapler are noted in Table 6 (3–36%). Quirke et al. (342) reported that in different hands, the incidence of local recurrence after resection of rectal carcinoma varies from 4% to 40%. In a dramatic departure from the generally high rates of recurrence, Heald (526) noted a remarkably low local recurrence rate of only 2.6%. He attributed this favorable outcome to complete excision of the mesorectum. It has been reported that wide pelvic lymphadenectomy can reduce the incidence of recurrence to 6% to 8%. Pihl et al. (527) also reported the amazing low anastomotic recurrence rate of 3% after resection of carcinoma of the rectum.

281

& FACTORS CONTRIBUTING TO RECURRENCE A number of local factors may contribute to the development of local recurrence following a low anterior resection of the rectum: (i) incomplete excision of the primary lesion, either inadequate distal or lateral margins of resection, or incomplete removal of the mesorectum; (ii) implantation of exfoliated neoplastic cells on the anastomosis or other raw surface; and (iii) development of metachronous lesions at the site of the anastomosis. General prognostic discriminants were described in great detail in Chapter 23. In their local recurrence rate of 20%, Feil et al. (528) reported that local recurrence depended on Dukes’ stage, grading, gross appearance, lymphatic stroma reaction, venous invasion, perineural invasion, and margin of clearance. In a review of factors that influence local recurrence, Twomey et al. (529) conducted a meta-analysis of the English language literature through 1988 and found that adjuvant radiotherapy with doses of 3000 cGy or more resulted in a reduction of up to 40% in local recurrence in all but the lowest risk patient. & PATTERNS OF RECURRENCE Pilipshen (525) has classified local recurrence with a view toward possible further treatment. He defined the five categories as follows: & & &

& &

Anastomotic. Recurrence arising in and contained to bowel wall. Perianastomotic. Recurrence in proximity or involving the anastomosis by extensive disease with inward invasion. Perineal. Recurrence in the perineal scar following abdominoperineal resection. If minimal, the scar may be amenable to resection. Pelvic wall. Recurrence fixed to bone, major blood vessels, and/or nerves, usually precluding resection. Anterior genitourinary. Recurrence that, if amenable to resection, involves pelvic exenteration.

In most instances, however, recurrences overlap these categories at the time of diagnosis. Gunderson and Sosin (457) reported that in patients who develop recurrent disease, local recurrence alone was noted in 50% of patients and in 92% with distant metastases. In an autopsy series, Welch and Donaldson (530) found that 25% of the patients died of only local disease, 25% of distant disease, and 50% of combined regional and distant metastases. Among patients who develop recurrent disease, Pilipshen et al. (531) found that the pelvis was the single site of recurrence in 53% or was combined with other sites in 63%. They suggested that resection of local recurrent disease would be feasible in 20% of patients who underwent low anterior resection and in 10% of patients who underwent abdominoperineal resection. In a review of the patterns of recurrence following curative abdominoperineal resection, Rosen et al. (175) found that 43% of patients developed recurrence, 10% developed local recurrence, and the other 33% manifest with distant metastases. They noted that local recurrence appears much earlier in patients with Dukes’ C lesions than in those with Dukes’ B lesions (6 months vs. 21.5 months). However, once recurrence appears, differences in survival

282

& PART I: COLORECTAL DISORDERS

time from recurrence to death were not significant (10.5 months vs. 17 months), regardless of whether the initial presentation of recurrence was local or distant. In a review of 1008 patients who underwent curative resection, McDermott et al. (331) found that 11% of patients developed local recurrence without evidence of systemic spread, whereas 9% developed both local and systemic recurrence. Of the patients who developed metastases, the recurrence was evident within two years in 60%. Of the patients who died from recurrence, 27% had evidence of local recurrence only, 24% had combined local and systemic recurrence, and 48% had evidence of systemic spread only. The corresponding median survival times were 35, 34, and 39 months, respectively. Luna-Perez et al. (532) reported on the patterns of recurrence in a group of 49 patients treated with pelvic exenteration and radiotherapy. Thirty-one received preoperative radiotherapy, 4500 cGy. Six weeks later they performed posterior pelvic exenteration in 21 patients and total pelvic exenteration in 10. Nine patients received postoperative radiotherapy, 5000 cGy, after a posterior pelvic exenteration. Of nine patients who had surgery, only seven had posterior pelvic exenteration and two had total pelvic exenteration. Surgical mortality occurred in 16% of those patients who received preoperative radiotherapy. The median follow-up was 52 months. Recurrences occurred in 23% of those patients who received preoperative radiotherapy (local, one patient; local/distant, one; distant, four); in 88% of those patients treated with surgery only (local/distant, four; distant, four); and in 11% of those treated with postoperative radiotherapy (distant, one). The five-year survival rate for patients who received radiotherapy was 66% vs. 44% for those treated with surgery only. The authors concluded that local control of locally advanced primary rectal adenocarcinoma requiring a pelvic exenteration is improved by the addition of radiotherapy. When recurrences do occur, they are predominantly at extrapelvic sites. The extensive review of the patterns of recurrence reported by Obrand and Gordon was described in Chapter 23.

& CLINICAL FEATURES With anastomotic recurrence, the patient may be asymptomatic, but an irregularity may be found upon follow-up rectal, digital, or vaginal examination, or recurrence may be seen during sigmoidoscopy. Symptoms, when present, may include bleeding, narrowed stools, or pain. A biopsy confirms the diagnosis. Following abdominoperineal resection with perineal recurrence, a mass may be seen. Pelvic recurrence may be asymptomatic or may manifest with pain or pressure in the abdomen, pelvis, or perineum (Fig. 30). The pain may also radiate to the back, buttocks, or lower extremities. Urinary tract symptoms may develop, or vaginal bleeding may occur. As many as half the patients with recurrence are symptom free at the time of diagnosis (522). & INVESTIGATIONS A barium enema is not necessary for diagnosing an anastomotic recurrence following low anterior resection but may show a narrow-caliber lumen and rule out a synchronous

FIGURE 30 & Perineal recurrence with ulcerating masses and numerous metastatic nodules along vulva and groin.

neoplastic lesion if a colonoscopy is not planned or unable to to performed. Investigations should include a search for the extent of disease using an IVP, which may show ureteral displacement or compression although this information can usually be obtained from a CT scan. The carcinoembryonic antigen (CEA) level may or may not be elevated in patients with a pelvic recurrence. A search for disseminated disease also should be made. CT scanning is probably the most useful test to determine a recurrence. Postoperative changes, especially after abdominoperineal excision of the rectum, mimic recurrent disease and serial scans may be of benefit especially if a postoperative baseline scan had been obtained. Following abdominoperineal resection, percutaneous biopsy under CT guidance will confirm the diagnosis (533). MRI may also demonstrate the extent of a pelvic recurrence and/or guide percutaneous biopsy. Intrarectal ultrasonography may detect early intramural or extrarectal recurrences before there is evidence of intraluminal recurrence (534). Lohnert et al. (535) conducted a prospective study to assess the diagnostic potential of endorectal and endovaginal ultrasound to detect asymptomatic resectable local recurrence. In 338 patients, 721 endoluminal ultrasound examinations were added to routine follow-up of rectal and left colonic carcinoma. A total of 116 patients (34.3%) were shown to have local recurrence, which was suggested by endoluminal ultrasound and proven by endoluminal ultrasound guided needle biopsy in all cases of unclear pararectal structures that could not be verified by endoscopic biopsy. Digital examination failed to detect local recurrence in 91 patients, endoscopy failed to detect local recurrence in 80 patients, and the levels of markers for carcinoma were normal in 25 patients with confirmed local recurrence. In 33 cases of local recurrence, both digital examination and endoscopy

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

results were normal. Twenty-five patients in whom carcinoembryonic antigen levels, digital examinations, and endoscopy results were normal, underwent potential curative reoperation, with total resection of the local recurrence. All 25 patients were still alive at the end of the study period, and 21 were free from disease. On the other hand, only 6 of 67 patients with local recurrence detectable by conventional follow-up could be operated on with curative intention. Detecting radiolabeled monoclonal antibodies with a hand-held gamma-counter probe appears to be of some help in intraoperative decision making at the time of second-look operation. Unnecessary operation can be avoided and the extent of potentially curative resections may be expanded (536). Moore et al. (537) found that PET scanning is an accurate modality for detecting pelvic recurrence of rectal carcinoma after full dose external beam radiotherapy. It correctly identified 16 of 19 recurrences for a sensitivity of 84% and specificity of 88%. Overall accuracy was 87%.

& TREATMENT OF RECURRENT DISEASE Effective therapy for pelvic or perineal recurrence is limited. It appears, however, that to adopt a totally therapeutic nihilistic attitude is not entirely warranted. Resectional therapy has proven curative in some, albeit a limited number, patients [estimated by some to be up to one-third of patients (520)] and palliative care can lessen the discomfort of many of these patients. Yamada et al. (538) studied 83 patients with locally recurrent rectal carcinoma for survival benefit by re-resection. Sixty patients underwent resection for recurrent carcinoma including total pelvic exenteration in 30 patients and sacrectomy in 23 patients. The extent of locally recurrent carcinoma classified by the pattern of pelvic invasion was: localized, sacral invasion, and lateral invasion. This simple classification proved to be helpful to the surgeon making a recommendation for treatment as the pattern of pelvic invasion was a significant prognostic factor that independently influenced survival after resection of recurrent carcinoma. The five-year survival rates were 38% in the localized type (recurrent carcinoma localized to adjacent pelvic organs or connective tissue), 10% in the sacral invasion type (recurrent carcinoma invading the lower sacrum S3-4–5, coccyx, or periostium) and 0% in the lateral invasive type (recurrent carcinoma invading the sciatic nerve, greater sciatic foramen, lateral pelvic wall or upper sacrum S1/2). Hence resection for locally recurrent rectal carcinoma is potentially curative in patients with localized or sacral invasion patterns of recurrence, but alternatives should be explored in patients with recurrence involving the lateral pelvic wall. Operative Treatment In selected circumstances when technically possible, the best chance for cure following a recurrence is re-resection. Failing that, radical surgery may offer good palliation and a better quality of life (520). More than 50% of these patients have disease amenable to operative resection (520). In a review of seven reports from the literature, Herefarth et al. (539) deemed that 31% of patients with locoregional recurrence may be suitable for curative reintervention. This was a paltry increase of only 6% over reports two and three

283

decades earlier, despite the advanced technology used in the postoperative follow-up. In the case of a previous low anterior resection, occasionally it may be feasible to effect a re-resection with an extended low anterior resection, but more likely an abdominoperineal resection will be necessary. In preparation for operation, insertion of ureteral catheters may be helpful to identify the ureters. Segall et al. (540) reviewed 12 patients who underwent abdominoperineal resection for recurrence subsequent to low anterior resection. These procedures were technically more difficult, but no death occurred. Although no long-term cures resulted, significant palliation was achieved, and one patient survived more than eight years. In the series reviewed by Schiessel et al. (522) the treatment of local recurrence was operative in 86.5% of patients (48.6% of the patients underwent radical resection and 51.4% underwent palliative operation). The type of operative intervention depended on the localization of the recurrence and included abdominoperineal resection, sacral excision, or pelvic exenteration. Hojo (176) performed re-resection in 22 of 30 patients with anastomotic recurrence; 11 had curative resections. Suzuki et al. (541) reported on 65 patients who underwent re-resection for recurrent rectal carcinoma. Procedures included abdominoperineal resection, wide local resection, Hartmann’s procedure, low anterior resection, and abdominosacral resection along with an en bloc resection of adherent organs, including vagina, uterus, bladder, prostate, seminal vesicles, and sacrum when appropriate. There were no deaths, but 17 severe complications occurred. Three-year, five-year, and median survivals were 57, 34, and 45 months, respectively. The cumulative probability of failure was 24%, 41%, and 47% at one, three, and five years, respectively. Cumulative risk of distant metastases was 30%, 51%, and 62% at one, three, and five years, respectively. The authors concluded that complete excision of locally recurrent carcinoma would be able to provide a significant number of patients with long-term survival. Wiggers et al. (542) reviewed 163 patients with local recurrence of rectal carcinoma after previous ‘‘curative’’ resection and of the 35 patients who underwent exploratory laparotomy, 27 were amenable to re-resection. There was no operative mortality but, with a median follow-up of 42 months, local re-recurrence developed in 59% of patients. The estimated five-year survival rate was 20%. Saito et al. (543) reviewed 58 consecutive patients with local recurrence of rectal carcinoma after previous ‘‘curative’’ resection. Of the 58 patients, 27 underwent re-resection, nine had palliative resection, and 22 were treated by conservative therapy. Among the 27 patients with curative resections, 17 received preoperative radiotherapy (40 Gy) plus operation, and 10 operation alone. No patients were lost to follow-up; median follow-up time was 36.3 months. The overall rate of curative resection was 46.6%. With regard to surgical procedure, abdominoperineal resection with or without sacral resection was standard following previous low anterior resection and total pelvic exenteration with or without sacral resection was common after abdominoperineal resection. There was a high incidence of morbidity (71.4%) after total pelvic exenteration. Re-recurrence was observed in 44.4% after curative

284

& PART I: COLORECTAL DISORDERS

re-resection. There was local re-recurrence in 22.2%. The local re-recurrence rate was 11.8% with radiotherapy plus operation and 40% with operation alone. The estimated five-year survival following curative re-resection was 45.6% (61.2% with radiotherapy plus operation, 29.6% with operation alone). Huguier et al. (544) reported that of 80 patients having local recurrence, 48% underwent a re-resection combined in 10 cases with resection of metastases. The incidence of asymptomatic detected recurrence was higher after anterior resection (39%) than after abdominoperineal resection (18%). Re-resection was performed more often in the past two decades after anterior or Hartmann first procedure than after abdominoperineal resection (67% vs. 21%) and more often in asymptomatic patients than in symptomatic patients (71% vs. 38%). The actuarial five-year survival rate after re-resection was 20%. They concluded early detection of local recurrence with PET scan leads to an improved re-resection rate. Boyle et al. (545) assessed the outcome of a series of 64 patients who underwent resection of locally recurrent rectal carcinoma with curative intent. The median time interval between resection of primary carcinoma and surgery for locally recurrent disease was 31 months. Twenty-three patients had central disease, 10 patients had sacral involvement, 21 patients had pelvic sidewall involvement, and 10 patients had both sacral and sidewall involvement. Fiftyseven patients underwent resection of the carcinoma. Thirty-nine out of 57 patients underwent wide resection (abdominoperineal excision of rectum, anterior resection, or Hartmann’s procedure) whereas 18 patients (31.6%) required radical resection, pelvic exenteration, or sacrectomy. Curative, negative resection margins were obtained in 36.8% of patients who had excision. Perioperative mortality was 1.6%. Significant postoperative morbidity occurred in 40% of patients. Vermaas et al. (546), compared the results of preoperative radiotherapy followed by operation with operation alone for recurrent rectal carcinoma. With 92 patients suitable for resection with curative intent, preoperative radiation with a median dosage of 50 Gy was performed in 59 patients; 33 patients did not receive preoperative radiotherapy. The median follow-up of patients alive for the total group was 16 months. Complete resections were performed in 64% of the patients who received preoperative radiation and 45% of the nonirradiated patients. A complete response with radiotherapy was found in 10% of the preoperative irradiated patients. There were no differences in morbidity and reintervention rate between the two groups. Local control after preoperative radiotherapy was statistically significantly higher after three and five years. Overall survival and metastasis-free survival were not different in both groups. Complete response to preoperative radiotherapy was predictive for an improved survival. Garcia-Aguilar et al. (547) analyzed the outcome of patients with isolated local recurrence after radical treatment for rectal carcinoma in 87 patients. Symptomatic treatment alone or chemotherapy and/or radiation therapy was provided in 23 patients (26%) and surgical exploration was performed in 64 patients. In 22 patients (25%), the carcinoma was considered unresectable at operation (n ¼ 13) or was resected for palliation with gross or

microscopic-positive margins (n ¼ 9). In 48% curative intent resection was performed. The only independent predictors of resectability were younger age at diagnosis, earlier stage of the primary carcinoma, and initial treatment by sphincter-saving procedure. There was no difference in survival between patients who had no operation and those who had palliative operation. The estimated five-year survival rate for patients who had curative intent resection was better than for those who had no operation or palliative operation (35% vs. 7%). Of the 42 patients who underwent curative intent resection, 33% developed a second recurrence at a mean of 15 months after reoperation. Twentyfive percent of patients developed major complications. Hahnloser et al. (548) assessed the results of multimodality therapy for patients with recurrent rectal carcinoma and analyzed factors predictive of curative resection and prognostic for overall survival. A total of 394 patients underwent surgical exploration for recurrent rectal carcinoma. Ninety were found to have unresectable local or extrapelvic disease and 304 underwent resection of the recurrence. Overall five-year survival was 25%. Curative, negative resection margins were obtained in 45% of patients; in these patients a five-year survival of 37% was achieved compared to 16% in patients with either microscopic or gross residual disease. Overall survival was significantly decreased for symptomatic pain and more than one fixation. Survival following extended resection of adjacent organs was not different from limited resection (28% vs. 21%). Perioperative mortality was only 0.3% but significant morbidity occurred in 26% of patients with pelvic abscess being the most common complication. They concluded long-term survival can be achieved especially for patients with no symptoms and minimal fixation of the recurrence in the pelvis provided no gross residual disease remains. Lopez-Kostner et al. (549) determined whether salvage operation in appropriate selected patients could significantly lengthen disease-free survival after a previously curative resection for rectal carcinoma. Of 937 patients who underwent operation with curative intent, 8.6% experienced local recurrence. Thirty-six patients with locally current rectal carcinoma were referred from other institutions. Of 117 patients with locally recurrent rectal carcinoma, 36.7% underwent salvage operation. Factors associated with higher chance of receiving salvage operation were female gender, the first operation performed at outside institutions, and transanal local excision as the initial operation. For 43 patients who underwent salvage operation, five-year carcinoma-specific and disease-free survival rates were 49.7% and 32.2%, respectively. A trend for poor prognosis was observed in patients with recurrence diameter greater than 3 cm and fixation of the carcinoma. Mannaerts et al. (550) compared three treatment modalities for patients with locally recurrent rectal carcinoma. Ninety-four patients were treated with electron-beam radiation therapy only, 19 with combined preoperative electron-beam radiation therapy and operation and 33 with intraoperative radiation therapy-multimodality treatment. The three-year survival, disease-free survival and local control rates were 14%, 8%, and 10%, respectively, in the electron-beam radiation therapy only group and 11%, 0%, and 14%, respectively, in the combined electron-beam radiation therapy surgery group. The overall intraoperative

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

radiation therapy multimodality treatment group showed significant better three-year survival, disease-free survival, and local control rates of 60%, 43%, and 73%, respectively compared with the historical control groups. Isolated perineal recurrence after abdominoperineal resection of the rectum may be suitable for wide local excision. Cure is rare as perineal deposits are usually associated with disease deeper in the pelvis. Local excision results in some palliation but may also result in nonhealing wounds and even enterocutaneous fistulas. To determine the value of total pelvic exenteration, Lopez et al. (551) reviewed the results of 24 patients undergoing the procedure, which involved removal of the distal colon and rectum along with the lower ureters, bladder, internal reproductive organs, perineum, draining lymph nodes, and pelvic peritoneum. The operative mortality was 20% (9% during the past decade) with an overall survival rate of 42%. Falk et al. (552) reported on 45 patients who underwent pelvic exenteration for advanced primary or recurrent adenocarcinoma. Many patients had previous radiotherapy. The operative mortality was 15.5%. Of the 16 patients who had palliative exenteration, nine died, seven of progressive malignancy (median survival, five months). Exenteration was deemed curative in 22 patients, but six died of the disease (median survival, 15.5 months). At the time of reporting, four patients were alive at 11,17, 21, and 23 months. After operation the remaining 12 patients were disease free at 4 to 21 months postoperatively (median, 10.5 months). Sugarbaker (385) reported on six patients who underwent en bloc resection for rectal carcinoma with sacrectomy for lesions fixed posteriorly. Voiding occurred spontaneously, but the men were impotent. Four of six such patients survived more than three years. Temple and Ketcham (553) performed five en bloc resections and one sacral resection only. The operative mortality was 9% with an 18% five-year survival. Pearlman et al. (554) reported on 19 patients. Four underwent extended proctectomy, four underwent a standard pelvic exenteration, and 12 had sacropelvic exenterations. One operative death occurred, two died free of disease, four died of disease, two are alive with disease, and eight are living free of disease at the time of publication. The authors believe that radiation is not as effective as radical operation. Wanebo et al. (555) have used abdominosacral resection with or without pelvic exenteration for pelvic recurrence in 53 patients, 47 with curative intent. Almost all the patients had been irradiated previously (4000– 5900 cGy). Organs resected included the rectum, 18; bladder, 25; partial bladder, two; prostate and/or seminal vesicles, 19; vagina, two; total abdominal hysterectomy, eight; segmental bowel resection, eight; and other, eight. The level of sacral resection was L5-S1, one; high SI or Sl-2, 26; mid-S2 or mid-S3, 14; and low S4-5, six. Added resection included sidewall pelvic vessels in two patients and pelvic lymph node dissection in 46. The operative mortality rate was 8.5% and postoperative morbidity was encountered in most patients. Survivors had relief of sacral root pain and good motor function. The division of SI, S2, S3 roots compromises bladder function, resulting in bladder denervation. Although this can be managed by Crede’s

285

method of voiding and the use of a-agonists or selfcatheterization, it requires careful attention and care by the attending physician. The actuarial five-year survival rate was 33% and the median survival was 39 months. Temple and Ketcham (556) described a unique palliative approach of radical debridement for recurrent carcinoma ulcerating through the perineum that cannot be controlled by other means. Seven patients were treated with resection of the recurrence along with portions of the sacrum. Coverage was obtained with myocutaneous flaps. Temple and Ketcham believe that the duration and quality of life are improved with this method. Tepper et al. (557) reported on the prevalence and influence of salvage therapy among patients with recurrent disease following curative resection of T3, T4 rectal carcinomas in the intergroup study 0114. A total of 1792 patients were entered into the study and 1696 were assessable. After a median of 8.9 years of follow-up, 715 patients (42%) had disease recurrence, and an additional 10% died without evidence of disease. Five hundred patients with follow-up information available had a single organ or single site of first recurrence (73.5% of all recurrences). A total of 171 patients (34% of those with a single organ or single site of recurrence) had a potentially curative resection of the metastatic or locally recurrent disease. Single-site first recurrences in the liver, lung, or pelvis occurred in 448 patients (90% of the single-site recurrences), with 159 (35%) of these undergoing surgical resection for attempted cure. Overall survival differed significantly between the resected and nonresected groups with overall five-year probabilities of 0.27 and 0.06, respectively. From their data, the chance of a long-term cure for surgical salvage of rectal carcinoma recurrence is approximately 27%. Jimenez et al. (558) reviewed 55 patients undergoing total pelvic exenteration for locally advanced or recurrent colorectal carcinoma. Indications for operation were recurrent colorectal carcinoma in 71% and primary rectal carcinoma in 29%. Of 39 patients with recurrent colorectal carcinoma, 85% had previous radiotherapy and 64% had previous abdominoperineal resection. At the time of pelvic exenteration, 49% of patients receiving intraoperative radiation and 20% required sacrectomy. Complete resection with negative margins was achieved in 73%. Perioperative mortality after pelvic exenteration was 5.5% and complications including perineal wound infection (40%), pelvic abscess (20%), abdominal wound infection (18%), and cardiopulmonary events (18%). Median disease-specific survival for all patients was 48.9 (range, 3.2–105.6) months. Less satisfactory outcomes are observed in patients whose indication for pelvic exenteration is recurrent colorectal carcinoma after abdomoniperineal resection. Ike et al. (559) reviewed charts of 45 patients with rectal carcinoma who underwent curative total pelvic exenteration for local recurrence. Post-operative morbidity was 77.8% and in-hospital death occurred in 13.3% of patients. The overall five-year survival rate was 14.1%. The five-year survival rates stratified according to the expectation of curability were 31.6% for absolutely curative resection, 7.8% for relatively curative resection, and 0% for noncurative resection. The disease-free interval was the only independent prognostic factor. There was no benefit from perioperative radiation or intraoperative continuous

286

& PART I: COLORECTAL DISORDERS

pelvic peritoneal perfusion of the pelvis. They concluded total pelvic exenteration for local recurrence of rectal carcinoma would be able to achieve long-term survival when curative resection is possible and the disease-free interval would be long. Bakx et al. (560) reported on the results of 26 patients who underwent sacral resection for both primary and recurrent rectal carcinoma as well as recurrent anal carcinoma. Sacral resection alone was performed in five patients, with abdominoperineal resection in six patients and exenteration in 13 patients. Median operating time was six hours (range 2.5–10 hours) with median blood loss of 3600 ML (range 420–11,500 ML). A transpelvic rectus abdominis musculocutaneous flap was used in 50% and an omental transposition in 65% of procedures. There was only one death. Estimated two and five-year survivals were 82% and 51%, respectively. There were 27 major complications (five requiring re-laparotomy). They believe their results support the use of sacral resection in patients with nonmetastatic malignant disease. Mannaerts et al. (561) reported their results in patients with locally advanced primary (13) or locally recurrent rectal (37) carcinoma with dorsolateral fixation who underwent abdominosacral resection as part of a multimodality treatment, i.e., preoperative irradiation, operation, and intraoperative irradiation. Margins were microscopically negative in 52%, microscopically positive in 36%, and positive with gross residual disease in 12%. Operation time ranged from 210 to 590 minutes and blood loss ranged from 400 to 10,000 ML. No operative or hospital deaths occurred. Postoperative complications occurred in 82%; most notably were perineal wound infections or dehiscence (48%). Other complications were postoperative urinary retention or incontinence (18%), peritonitis, grade II neuropathy, and fistula formation. The three-year overall survival, disease-free survival, and local control rates were respectively 41%, 31%, and 61%. Completeness of the resection (negative vs. positive margins was a significant factor influencing survival, disease-free survival, and local control. Moriya et al. (562) evaluated the effectiveness or total pelvic exenteration with distal sacrectomy for fixed recurrent carcinoma that developed from primary rectal carcinoma in 57 patients. Negative margins were present in 84%. Two hospital deaths were observed in the early period and none in the later period. The most common sacral amputation level was the S3 superior margin, followed by the S4 inferior margin and the S2 margin. The most frequent complication was sacral wound dehiscence in 51%, followed by pelvic sepsis in 39%. The incidence of pelvic sepsis in the latter period was significantly decreased to 23% compared with 72% in the earlier period. Negative margins and negative CEA predicted improved survival. In 48 patients with negative margins, three-year and five-year disease-specific survivals were 62% and 42%, respectively. Shirouzu et al. (563) reported on 26 patients who underwent total pelvic exenteration for locally advanced colorectal carcinoma with an 8% mortality rate. In patients with stage II primary disease the five year survival rate was 71% (mean survival, 58 months). In patients with stage III, the mean survival was 14 months and in stage IV, the mean survival was five months. In patients who underwent cura-

tive operation for recurrent disease, the five-year survival rate was 25% (mean survival, 33 months). In their review of the literature, Sagar and Pemberton (520) noted that if patients are selected carefully, pelvic exenteration is associated with a median survival of 21 to 30 months and a five-year survival rate of 50%. In their review of pelvic exenteration for recurrent colorectal carcinoma, Yeung et al. (564) noted that after they combined a series of 13 reports in the literature, the incidence of pelvic recurrence by Dukes’ staging was 11%, 23%, and 35% for Dukes’ A, B, and C, respectively, with an overall recurrence rate of 25%. It is very clear that the incidence of recurrence is dependent on the length of follow-up, but approximately 70% of failures occur within two years of treatment. A significant portion of patients with local recurrence have disease confined to the pelvis without distant metastases. Faced with the problem of isolated pelvic recurrence, complete extirpation may have a chance of cure or at least good palliation. Elements of the procedure consist of a composite that in women are radical hysterectomy, total cystectomy, and abdominoperineal resection; and in men total cystectomy, radical prostatectomy, and abdominoperineal resection. More radical resection may include sacrectomy. In their series of 43 patients ranging in age from 31 to 77 years, the median duration from the time of the initial operation to exenteration was 39.7 months. Of the patients studied, 60% had received radiotherapy. Internal iliac vessels were ligated, but involvement of common and external iliac vessels implied inoperability. The most popular form of urinary diversion was the ileal conduit. In their series of 43 patients, 26 had ileal conduits, six had colonic conduits, and 11 had wet colostomies. Among the different constructions of a continent reservoir, the ileal colonic (Indiana) pouch was attractive because it utilized the least amount of ileum along with the ascending colon, which has the greatest likelihood of not being exposed to pelvic radiation. The learning curve for this procedure is steep with high postoperative morbidity and mortality rates. Cited operative mortality rates fell from 13% to 20% to from 2% to 7% as experience accumulated. Complication rates range from 30% to 75%. Types of complications included enteric fistula, conduit leak and/or fistula, bowel obstruction, pelvic hemorrhage/abscess, urosepsis, renal failure, wound infection, deep venous thromboses/pulmonary embolus, myocardial infarction/cardiac arrest, cerebrovascular accident, sepsis (nonpelvic), gastrointestinal hemorrhage, prolonged ileus, and hernia. The reoperation rate was high among this group of patients in whom the complication rate was even higher. These sobering statistics mandate that a careful patient selection process was undertaken prior to embarking on this procedure. Survival rates following this procedure have not been overly encouraging. Reported rates of recurrence range from 40% to 70% and overall five-year survival rates from 10% to 20%. However, this should be compared to a 2% to 3% five-year survival rate for untreated disease. Even in a palliative setting, symptomatic improvement is often prompt and long-standing. Up to 89% of patients with curative intent have significant palliation from pain. Moreover, some of these patients achieve remarkable psychological improvement and an enhanced quality of life.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

Since it was first reported in 1948, pelvic exenteration has been used in the treatment of advanced pelvic malignancy. The original procedure has been modified in an attempt to preserve urinary or fecal continence. Rodriguez-Bigas and Petrelli (565) conducted a literature review on selected series of total pelvic exenterations and modified pelvic exenterations in order to assess and discuss the different types of pelvic exenterations and the indications, contraindications, morbidity, mortality, and results of these procedures. According to the series reviewed, morbidity after pelvic exenteration ranges between 32% and 84%, postoperative mortality ranges from 0% to 14%, and five-year survival varies from 23% to 68%. These numbers indicate that total pelvic exenteration and its modifications are a complex group of surgical procedures with significant early and late postoperative morbidity and mortality. While the authors think that these findings indicate that pelvic exenteration should only be undertaken by experienced surgeons at specialized centers, they also caution that, above all, their findings indicate that the potential curability of a patient with adjacent organ involvement should not be compromised by doing less than an en bloc resection. Kakuda et al. (566) reviewed 22 patients who underwent pelvic exenteration for recurrent rectal carcinoma. Seventeen underwent potentially curative resection, five were for palliation only. There was one operative death. Fifteen suffered at least one complication; nine suffered multiple complications. Ten patients required re-admission to the hospital. The overall disease-free interval was 11 months. Potentially curative and palliative resections resulted in median survivals of 20.4 and 8.4 months, respectively. While patients may derive oncologic and palliative benefits from exenteration, the price in terms of operative morbidity remains high. Although resection of locally recurrent rectal carcinoma has been associated with improved survival, symptomatic relief has been incompletely characterized. Miner et al. (567) reviewed 105 consecutive patients requiring repeat operation for locally recurrent rectal carcinoma. They were observed for a minimum of two years or until death. An operation was performed with palliative intent in 23% of patients. Before repeat operation, 79% of the palliative intent patients had symptoms: 21% bleeding, 42% obstruction, and 21% pain. After repeat operation with palliative intent, improvement was noted in 40% with bleeding, 70% with obstruction, and 20% with pain. Additional or recurrent symptoms were noted in 87% during follow-up. Seventy-seven percent of patients had an operation with nonpalliative intent. Before repeat operations, 57% of nonpalliative patients had symptoms, with 32% experiencing bleeding, 11% obstruction, and 19% pain. After repeat operation with nonpalliative intent, initial improvement was noted in 88% with bleeding, 78% with obstruction, and 40% with pain. During follow-up, symptoms arose in 37% of the initially asymptomatic patients and additional or recurrent symptoms were seen in 63% of those previously symptomatic. They noted, the recurrence or development of new symptoms makes a complete asymptomatic clinical course uncommon. Shoup et al. (568) reported survival after operation and intraoperative radiotherapy for recurrent rectal carcinoma. Of 634 patients undergoing resection for recurrent

287

rectal carcinoma, 111 received intraoperative radiotherapy with curative intent and 100 were available for follow-up. With a median follow-up of 23.2 months, 60% recurred: 33% locally and 45% distantly and 22% at both sites. Of all variables analyzed, only complete resection with microscopically negative margins and the absence of vascular invasion in the recurrent specimen predicted improved disease-free and disease-specific survival. Median diseasefree survival and median disease-specific survival were 31.2 and 66.1 months, respectively, for complete resection compared with 7.9 and 22.8 months for resection for microscopic or grossly positive margins. Median disease-free survival and median disease-specific survival were 6.4 and 16.1 months, respectively, in the presence of vascular invasion in the recurrent specimen compared with 23.3 and 57.3 months in the absence of vascular invasion. Complete resection and the absence of vascular invasion were the only predictors of improved local control. Kuehne et al. (569) evaluated the use of fractionated perioperative high-dose-rate brachytherapy in association with wide surgical excision (debulking). All patients had abdominal exploration, aggressive debulking of the carcinoma, and placement of afterloading brachytherapy catheters. Patients underwent simulation on postoperative day 3 and received 1200 to 2500 cGy of fractionated high-dose-rate brachytherapy between postoperative days 3 and 5. All patients had involvement of the lateral pelvic sidewall and/or the sacrum. Follow-up ranged from 18 to 93 months and was available in 27 patients of which 37% were alive at the time of the report. Nine patients were without evidence of disease. Eighteen percent died of noncarcinoma related causes without evidence of recurrent disease. Five complications potentially related to treatment (three abscesses, two fistulas) occurred in five patients. Cheng et al. (570) conducted a retrospective chart review of 142 patients with recurrent rectal carcinoma, 27 of whom had unilateral or bilateral hydronephrosis. Distant metastatic disease was present in 55% of these patients. Twelve patients (45%) with hydronephrosis and local recurrent disease on evaluation were analyzed. Six of the 12 patients underwent exploratory laparotomy with none found to have resectable disease. Their mean survival after diagnosis of recurrent disease was 14 months. Based on their results, the presence of hydronephrosis (unilateral or bilateral) in recurrent rectal carcinoma portends a survival equivalent to the presence of distant metastases. Therefore, they do not believe potentially curative resection has a role for patients with locally recurrent rectal carcinoma in the presence of hydronephrosis. Kelly and Nugent (571) described the use of intrarectal formalin instillation to control intractable rectal bleeding from advanced pelvic malignancy. The technique involves insertion of a foley catheter to the top of the rectum and inflating the balloon to 30 cc. Paraffin gauze and gel are applied to the anal skin following which 40 ML of 4% formalin is infused into the rectum for two minutes and then removed. The rectum is irrigated with normal saline for five minutes. They found formalin instillation into the rectum is an option when one is faced with rectal hemorrhage from inoperable pelvic malignancies. To provide local control and palliation of pain, a multimodality approach that includes external beam radiation

288

& PART I: COLORECTAL DISORDERS

therapy, surgical resection, and intraoperative electron irradiation has been used at the Mayo Clinic for patients with locally advanced anal or recurrent rectal carcinomas involving the sacrum (572). Sixteen consecutive patients underwent surgical exploration, sacrectomy, and intraoperative electron radiation therapy. Proximal extent of resection was S2-3 in four patients, S3-4 in five, and S4-5 in five. Two patients underwent resection of the anterior table of the sacrum. Margins were clear in 11, close in three, and microscopically involved in two patients. Operative times ranged from 6 to 17 hours (median, 12.5 hours), and blood loss ranged from 300 to 12,600 ML (median, 3350 ML). No operative deaths resulted. Major postoperative complications occurred in eight patients (50%): posterior wound infections and dehiscence, urinary leak, and ileal fistula. Five patients (31%) developed no complications, and three patients (19%) developed minor complications. The survival rate was 68% at one year and 48% at two years. Rodriguez-Bigas et al. (573) found that both unilateral and bilateral hydro nephrosis are contraindications for potentially curative resection for recurrent rectal carcinoma. Wanebo et al. (555) also identified lateral pelvic node and marrow involvement as contraindications for resection in patients with advanced or recurrent disease.

Management of Sacral and Perineal Defects The management of sacral and perineal defects following abdominoperineal resection and radiation can prove very difficult. Loessin et al. (574) presented their experience with treating persistent sacral and perineal defects secondary to radiation and abdominoperineal resection with or without sacrectomy. Fifteen consecutive patients were treated with an inferiorly based transpelvic rectus abdominis muscle or musculocutaneous flap. The technique as described by the authors is performed with the patient in the lithotomy position using two teams. The abdomen is opened in the midline, and the pelvis is cleared of all adhesions and bowel. The urologists place ureteral stents preoperatively in all patients to help avoid ureteral injury. After the pelvis has been cleared, very aggressive debridement of the perineal wound is performed so that fresh bleeding tissue and healthy bone are encountered. Leaving behind a nidus for infection or dead bone will result in failure. At this point, the rectus abdominis muscle or musculocutaneous flap is elevated. The flap is raised as an island flap connected only to the inferior epigastric pedicle (Fig. 31A). The entire muscle and skin unit is then passed into the pelvis via a 5 cm incision placed in the posterior sheath/peritoneum, starting at the point of the inferior epigastric vessels and running superiorly (Fig. 31B). The skin muscle unit is now carefully checked so that no twisting or kinking of the vascular leash is encountered. To ‘‘fit’’ the muscle and/or skin into the perineal wound, blunt dissection with the fingers must be performed ‘‘gingerly’’ so that a four-finger-breadth passage is created in the perineal wound (Fig. 31C). It is during this maneuver that bleeding or ureter or bladder injury may occur. Once this is accomplished, the pelvic cavity is copiously irrigated, bleeding observed and controlled, and the urinary tract checked for injury. At this point, the skin, muscle unit may be pulled and pushed into the defect. The perineal skin (Fig. 31D) and the abdomen can be closed.

Postoperatively, the patient is maintained on an air flotation mattress for five days. Patients are allowed to ambulate on day 7. Drains in the pelvis and abdomen are removed on day 10. Fourteen of the 15 patients healed, and seven patients had no complications. The remaining eight patients required one or more operative debridements and/or prolonged wound care to accomplish a healed wound. The authors concluded that difficult postirradiated perineal and sacral wounds would be healed with persistent surgical attention to adequate debridement, control of infections, and a well-vascularized muscle flap. The most satisfying aspects for patients are elimination of a foul-smelling discharge, discontinuation of multiple daily dressing changes, and reduction in the degree of chronic pain. Others have found this technique useful (575). Yeh et al. (576) used a variety of myocutaneous flaps to accomplish the same goal. Jain et al. (577) reported on the use of the transpelvic placement of the rectus abdominis muscle in a series of patients, 87% of whom received radiation therapy (muscle only was used in four patients, muscle with skin grafting in eight, and a musculocutaneous flap in three). Healing was complete in 12 of 15 patients at discharge. Yeh et al. (576) reported on nine patients who underwent 10 myocutaneous flap reconstructions for large perineal or pelvic defects following surgical extirpation of recurrent rectal carcinoma at Fox Chase Cancer Center. All nine patients had been previously treated with radiation therapy. Resection was determined by the extent of recurrence and included perineal resection, pelvic exenteration, cystectomy, sacrectomy, or coccygectomy. Extent of disease necessitated intraoperative radiation therapy in one case and placement of brachy-therapy catheters in four. Bilateral gracilis flaps were used in four, unilateral in three, gluteus maxi-mus in two, and combined gluteal and gracilis flaps in one patient. Six perineal and four combined perineal and vaginal defects were reconstructed. The length of hospitalization averaged 17.5 days. Acute surgical flap-related complications included three cases of minor wound infection or separation. Chronic flap complications consisted of one fistula from a ureterocolostomy, five persistent sinuses that ultimately healed, and one incidence of delayed minimal wound separation. One late abscess necessitated reoperation for drainage. Perioperative mortality was zero. There was a single case of delayed flap loss at two years in the setting of malnutrition and neglect. The median actuarial survival rate following reconstruction is 32 months. A single patient had local failure, two had distant disease, and two were disease free. Patients indicated that to a large extent they were able to pursue normal activities. Pain was nonexistent or minimal in all patients surveyed. Patients with pain, bleeding, or infection from locally recurrent rectal carcinoma may survive many months. Chessin et al. (578) cited a major source of morbidity after abdominoperineal resection after external beam pelvic radiation to be perineal wound complications, seen in up to 66% of cases. They determined the effect of rectus abdominus myocutaneous flap reconstruction on perineal wound morbidity in 19 patients with anorectal carcinoma treated with external beam pelvic radiation followed by abdominoperineal resection and rectus abdominus myocutaneous flap reconstruction of the perineum. A control group of

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

289

FIGURE 31 & Myocutaneous flap reconstruction. (A) Elevation of rectus abdominis myocutaneous flap. (B) Passage of flap through posterior sheath and peritoneum into the pelvis. (C) Securing myocutaneous flap into position. (D) Completed perineal closure.

59 patients with anorectal carcinoma treated with similar radiation doses that subsequently underwent an abdominoperineal resection without a rectus abdominus myocutaneous flap during the same time period was used. Perineal wound complications occurred in 15.8% of the flap patients and 44% of the control patients. The incidence of other complications was not different between groups (42.1% vs. 42.4%). They believe because other complications are not increased, flap closure of the perineal wound should be strongly considered in this patient population. The advantages of musculocutaneous flap reconstruction of the irradiated pelvis and perineal wound include reduction of dead space, interposition of wellvascularized, nonirradiated tissue, and replacement of

resected skin. The disadvantages of these flaps include the added time for the surgical procedure, the added costs, and the potential morbidity (such as infections, flap loss, seroma, ventral and perineal hernia, and abdominal bulge) of such procedures. These disadvantages are usually offset by lower rates of perineal wound complications (579).

Local Ablation The use of laser treatment for palliation in patients with unresectable recurrent carcinoma of the rectum has been described in some length in Chapter 23. Electrocoagulation has also been employed under these circumstances to avoid a colostomy (580). Of 15 patients so treated, nine were alive at 8 to 16 months without colostomy. Salvati and Rubin

290

& PART I: COLORECTAL DISORDERS

(581) reported that 16 of 19 patients so treated were spared a colostomy. Good palliation of obstruction, bleeding, and pain have been reported with the use of the urologic resectoscope, a procedure that required no anesthetic or long operative time. In one series a colostomy was avoided in 14 of 15 patients (582). The most common application of cryotherapy is treatment of recurrent, unresectable carcinomas of the rectum. The goal is the avoidance of a stoma. In a report of 219 patients so treated, Fritsch et al. (583) noted that local eradication of recurrent disease was achieved in 30% and symptoms were relieved in a further 24%. However, complications were not inconsequential, with bleeding in 14%, stenoses in 26%, and other complications, including peritonitis, in another 8%. Furthermore, the malodorous discharge was distressing. Heberer et al. (168) also reported avoidance of a stoma in 80% of 268 patients treated with cryotherapy.

Nonoperative Treatment Radiotherapy Radiotherapy is probably the most frequently adopted modality for the treatment of recurrent disease. It is reasonably effective in the control of pain and may shrink the mass and diminish drainage from ulcerated lesions. The ability to deliver radiation depends on whether the patient received adjuvant radiotherapy following the operation. In a series of 85 patients who developed pelvic recurrence after abdominoperineal resection, Villalon and Green (584) reported a 92% response to pain and an 80% response to mass (shrinkage). In a series of 143 patients, Pacini et al. (585) reported control of symptoms in 80% of patients. Dobrowsky and Schmid (586) obtained a 90% response rate. The crude survival time is 19.8 months following palliative radiotherapy (525,586). Chemotherapy Systemic chemotherapy has not demonstrated dramatic or long-term effects. Some palliation has been reported with the use of intraarterial 5-FU and mitomycin (587). The combination of hyperthermia and chemotherapy has been reported to offer better pain control (588). 5-FU and folinic acid (Leucovorin) is a frequently used combination at the present time (589). Hyperthermia and Chemoradiotherapy Nishimura et al. (590) reported on the addition of hyperthermia to radiation therapy for primarily unresectable as well as recurrent colorectal carcinoma. The authors found a response rate of 54% compared to 36% for radiotherapy alone. In a nonrandomized trial, Ichikawa et al. (591) compared the results of postoperative combined radiation plus 5-FU suppositories plus hyperthermia (35 patients) to no pretreatment (41 patients). In the pretreatment group, local recurrence was lower (10.4% vs. 27.1%), and survival was improved (81.8% vs. 67.6%). The authors believe that this is a promising modality for advanced carcinoma of the lower rectum. Gonzalez et al. (592) reported on 72 patients with either unresectable or recurrent colorectal carcinoma who were treated with combined radiotherapy (50 Gy) and hyperthermia (administered within 30 minutes of radiation intending to reach temperatures of 41 C over 30 minutes).

Toxicity consisted of local pain (33%) and general discomfort (17%). Hyperthermia was prematurely stopped in 17% of patients. Palliation was achieved in 75% of patients with a mean duration of 12 months. Objective (CT scan) remission was obtained in 15%. Median survival was four months, and 17% of the patients were alive at three years after treatment. Korenga et al. (593) retrospectively compared 23 patients treated with preoperative hyperthermia, chemotherapy, and radiotherapy with 48 patients treated by operation alone and operation and chemotherapy. The incidence of local recurrence was zero in the first group, but 15% and 10%, respectively, in the second group. The authors concluded that preoperative treatment decreased the frequency of local recurrence and the likelihood of spread during operation. Pain Control The principles of pain control and general comfort of the patient are described in Chapter 23. Narcotics should be prescribed liberally. For patients with intractable pain, neurosurgical procedures such as a chordotomy or rhizotomy are effective for pain control.

& RESULTS OF REOPERATION Controversy exists regarding the value of reoperation for recurrent carcinoma, and reported cure rates vary considerably. Pilipshen (525) reviewed a number of reports in the literature in which recurrences were re-resected by extended low anterior resection, abdominoperineal resection, or pelvic exenteration, even with partial sacrectomy. The overall ‘‘cure’’ was approximately 25%, but local repeat recurrence ranged from 25% to 50%. Schiessel et al. (522) reported on 109 patients treated by operation with a radical procedure in 53 cases. The operative mortality was 4.6%, with a 50% survival rate at 17 months. Gagliardi et al. (594) reported on 55 patients who had resection of locally recurrent rectal carcinoma. The five-year survival rate was 18% with a median survival of 24 months. At a median follow-up of 28 months, 53% of patients developed a second local recurrence and 24% metastases only. Operative mortality was 5% and complications occurred in 22%. The authors’ review of the literature revealed a probability of survival at one year to range from 30% to 98% and after three years, 0% to 62%. Williams et al. (595), in their review of the English literature, have suggested that the operative mortality rate should be less than 10% and can be less than 5% in primary cases of total pelvic exenteration. In their review of the literature regarding exenterative surgery for recurrent disease, Sagar and Pemberton (520) found an operative mortality rate ranging from 0% to 8.5%, a morbidity rate of 10% to 44%, survival rate of 18% to 49% at five years, and a local recurrence rate of 18% to 69%.

OTHER MALIGNANT LESIONS OF THE RECTUM

& CARCINOID Only those features specifically relevant to rectal as opposed to colonic carcinoids or carcinoids in general are

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

discussed here. Derivation and histochemical characteristics have been discussed in the section on colonic carcinoids (Chapter 23). When the embryologically derived divisions of foregut, midgut, and hindgut are considered, the carcinoid neoplasm arising in each area has a characteristic chemicopathologic manifestation (596). Hindgut carcinoids, which include rectal carcinoids, do not produce serotonin and they are not considered to give rise to the carcinoid syndrome (596).

& CLINICAL PRESENTATION Rectal carcinoids comprise only 0.1% of all rectal neoplasms (597). They occur with no sex predominance and are most commonly found in the sixth and seventh decades of life. Carcinoids will be discovered in 0.05% of all patients who undergo sigmoidoscopy (598). In sigmoidoscopic examination of 21,522 asymptomatic adults, Matsui et al. (599) found 16 carcinoids (prevalence, 0.07%). In Chapter 23 it was stated that the appendix is the most common location for gastrointestinal carcinoids but in a review of 175 patients with gastrointestinal carcinoids from the Ochsner Clinic in New Orleans, Jetmore et al. (600) found that the rectum represents the most common primary site. The authors found that the distribution of carcinoids was rectum, 55%; ileum, 12%; appendix, 12%; colon, 6%; stomach, 6%; jejunum, 2%; pancreas, 2%; and other, 5%. The typical finding during rectal examination is a solitary mobile submucosal nodule with an intact overlying mucosa. Its color varies from yellow-gray to tan-pink. Its size is usually from 1 to 1.5 cm. An erythematous change or a depression may be seen (599). In only 2% to 4% of cases do multiple nodules occur. Symptoms are encountered in 50% of patients and include bleeding, change in bowel habits, pain, pruritus, or weight loss (601). None of the 86 patients with rectal carcinoids removed by Fitzgerald et al. (601) developed a carcinoid syndrome, a finding noted by others (597). Malignancy has been reported in 39% of cases, and in these the significant findings related to a size greater than 2 cm in diameter with invasion through the muscularis propria. The malignant variety is frequently ulcerated, with bleeding being part of the clinical presentation. The incidence of metastatic disease is less in hindgut carcinoids (rectum, 18%) than in either midgut (jejunoileal, 34%; colon, 60%) or foregut (stomach, 23%; bronchus, 21%) (602). An association with long-standing chronic inflammatory bowel disease such as ulcerative colitis has been suggested (603). A secondary malignancy was found in eight of 43 patients with rectal carcinoids reported by Sauven et al. (603). & INVESTIGATION As with carcinoma of the rectum, intrarectal ultrasonography has been useful in the evaluation of carcinoids. Yoshikane et al. (604) found an overall accuracy of 75% in determining the depth of invasion as well as lymph node metastases. Kobayashi et al. (605) evaluated the diagnostic value of endoscopic ultrasonography in 66 rectal carcinoids to determine whether a lesion is indicated for endoscopic polypectomy. They concluded rectal carcinoids that satisfy the following three conditions are indicated for local resection, including endoscopic polypectomy: a maximum

291

diameter of less than or equal to 10 mm, no invasion of the muscularis propria, and no depression or ulceration in the lesion. Endoscopic ultrasonography also is useful for estimating the depth of invasion of rectal carcinoids.

& PATHOLOGY Macroscopically, carcinoids may have a yellow appearance, but in most cases this is not a striking feature. Microscopically, they consist of rosettes, ribbons, or masses of uniform small, round, or polygonal cells with prominent nuclei and acidophilic cytoplasmic granules. Immunohistochemically, carcinoids show focal or diffuse positivities for chromagranin A and/or neuron-specific enolase and often to pancreatic polypeptide (599). No absolute histologic difference exists between benign and malignant rectal carcinoids. Size and histologic evidence of invasion into the muscularis propria are helpful in distinguishing benign from malignant lesions. Indeed, the only consistent microscopic feature of malignancy is invasion of the muscularis propria (597). In a review of 24 rectal carcinoids, Shirouzu et al. (606) found that all lesions less than 2 cm in diameter had neither muscle-layer invasion nor lymph node metastases except for atypical carcinoids. In patients with metastatic carcinoid Gerstle et al. (607) found 77% of patients had invasion through the muscularis propria. Of 86 patients with carcinoids less than 2 cm reported by Fitzgerald et al. (601) only one with a 1.5 cm ulcerated lesion developed liver metastases. In the review by Spread et al. (608), 64% of patients already had metastatic disease at time of presentation and another 22% were Dukes’ C. & TREATMENT The pathologic data of these neoplasms indicate that those greater than 2 cm in diameter should, in general, be treated by abdominoperineal resection or low anterior resection, if feasible (609,610). Carcinoids less than 2 cm in diameter with an atypical histologic appearance also indicate the need for abdominoperineal resection because of the risk of lymphatic involvement (606). For small lesions, treatment consists of total excisional biopsy, leaving a margin of normal tissue. This treatment permits assessment of the depth of penetration. If there is invasion of the muscularis propria, a resection appropriate for carcinoma should be performed, because 47% of these lesions have been shown to have metastases (609). & RESULTS Nauneim et al. (609) reported on the results of treatment of 595 patients with rectal carcinoids. The five-year survival rate for patients with no metastases was 92%; for patients with regional lymph node involvement, 44%; and for patients with distant metastases, 7%. Tsioulias et al. (611) found that the nuclear DNA pattern was a factor of significant prognostic value. In a study of 22 patients with rectal carcinoids, three patients with synchronous or metachronous metastases had an aneuploid DNA pattern, whereas 19 patients without metastases showed a diploid DNA pattern. No other single clinical or pathologic feature could predict more accurately the malignant potential and subsequent

292

& PART I: COLORECTAL DISORDERS

course of the rectal carcinoid. On the other hand, Fitzgerald et al. (601) found DNA ploidy of no useful prognostic value. In a review of 35 patients with rectal carcinoids, only four lesions were malignant and all patients died despite radical operative treatment (597). Because radical treatment was largely unsuccessful, the authors of this review advocated a flexible operative approach. Sauven et al. (603) also questioned whether aggressive operation is warranted. In their management of 43 patients with anorectal carcinoids, no recurrence developed in the 18 patients who underwent local excision. Lesions that were more advanced than T2 or greater than 2 cm in diameter were always fatal. All 13 patients with involved lymph nodes died of metastatic disease with a median survival of 10 months. The authors concluded that if local excision permits complete resection, radical extirpation would provide little benefit. Matsui et al. (599) treated 16 small polypoid carcinoids by local excision and at an average 79 months follow-up, all were alive and disease free. Fitzgerald et al. (601) found that radical resection of rectal carcinoids with ulceration or greater than or equal to 2 cm is associated with a poor prognosis. However, survival may be long term, even in the presence of metastatic disease. Others have also reported that patients with carcinoids greater than 2 cm or invading beyond the muscularis propria die of metastatic disease despite radical resection (597,603). In a review of 85 patients with rectal carcinoids less than 2 cm, Jetmore et al. (600) found that treatment by local excision or fulguration resulted in no recurrences with an average five-year follow-up. Ten patients with metastasizing carcinoids fared poorly despite radical operation, with only one surviving five years.

& LYMPHOMA & CLINICAL PRESENTATION These lesions are less common than the rectal carcinoid. The disease is more common in women (2:1) and is usually seen beyond the age of 50. The incidence is reported to be 0.19% to 1.3% of all rectal neoplasms (602). The rectum is the site of 21.7% of lymphomas of the large intestine (612). In homosexual men with AIDS, non-Hodgkin’s lymphoma is one of the two most common malignancies but is nevertheless a rare neoplasm (613). The patient usually seeks treatment because of rectal bleeding or an alteration in bowel habits. However, a significant proportion may present with nothing more than generalized malaise and vague abdominal pain. The lesion is detected on rectal examination and through sigmoidoscopy. The macroscopic appearance bears no relationship to the histologic structure. The growth patterns seen in the rectum vary. The growth may be bulky and protuberant with ulceration, it may appear as annular or plaquelike thickenings of the bowel wall, or it may manifest as multiple malignant lymphomatous polyposis. The histologic varieties were described under the colonic lymphomas (Chapter 23). The aggressive nature of the AIDS-related lymphoma usually results in disseminated disease at the time of diagnosis. The extent of the disease should be assessed with CT scans, chest X-ray films, and a bone marrow biopsy.

& TREATMENT AND RESULTS The treatment of local or even regional disease is resection whenever possible (614). Devine et al. (615) reported on 61 patients with lymphoma involving the rectum; 49 had extensive lymphoma with secondary involvement of the rectum and 12 had lymphoma confined to the rectum. Those with widespread lymphoma were treated with chemotherapy, radiotherapy, or both. Five of the 12 patients with localized disease underwent resection. The overall five-year survival rate was 20%; the rate decreased to 15% with widespread disease and jumped to 50% when disease was confined to the rectum. Patients who underwent excision did better than those treated nonoperatively. Jinnai et al. (616) noted that the best prognosis occurred in patients with intraluminal lymphomas that were less than or equal to 5 cm and without lymph node metastases. For patients with widespread disease, those in whom the rectum is secondarily involved, or patients with AIDS, operation would seem inappropriate. These patients may benefit from chemotherapy and radiotherapy. Radiotherapy has a limited place in the primary treatment of rectal lymphoma and is not a substitute for operation. Radiotherapy is useful in the palliation of inoperable cases and as an adjunct to operation when resection is considered incomplete, either in relation to the primary lesion or the associated lymph node metastases. Adjuvant chemotherapy plays an important role in the treatment of these malignancies. It is important to realize that long-term survivors have been reported following incomplete removal of gastrointestinal malignant lymphomas, including patients with lymph node involvement. With advances in cytotoxic drug therapy and, to a lesser extent, in radiotherapy, resection has a prominent place even when the surgeon is doubtful as to the completeness of the resection and thus the ultimate success. Lymphosarcoma has been associated with chronic ulcerative colitis (617). Caution is emphasized when reviewing rectal biopsies for staging of colitis, because in some of the cases confusion has arisen between chronic inflammatory infiltrate and lymphosarcoma. In such cases radical excision of the colon is clearly required. Perry et al. (618) reported on 22 patients with primary lymphoma of the rectum. Twelve of the 22 died, three-quarters within two years of diagnosis with a follow-up of 10 months to 10 years. In a review of the Mayo Clinic experience, Devine et al. (615) found a five-year survival rate of 50% for lymphomas confined to the rectum, but only 15% when disease was widespread. Postoperative radiotherapy has been reported to enhance survival (615,619).

& SARCOMA As with the colon, a large variety of very uncommon neoplasms of mesenchymal origin may be found in the rectum. These have more recently been described as mural-based stromal neoplasms (620). The benign lesions tend to be submucosal, range in size from 1 to 7 cm (mean, 2.1 cm), lack nuclear atypia, and display a mitotic activity not exceeding one mitosis per 50 high-power microscopic fields. In contrast lesions exhibiting a malignant behavior are located in the muscularis propria, range in size from

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

1.6 to 11 cm (mean, 4.5 cm), manifest necrosis in 67% of patients, demonstrate moderate atypia, and display a mitotic count ranging from 5 to 58 mitoses per 50 highpower microscopic fields. The most common of these neoplasms is the leiomyosarcoma. Leiomyosarcoma of the rectum occurs at all ages and in both sexes, but it is more common beyond the sixth decade of life and represents approximately 0.07% to 0.1% of all malignancies (621,622). In a review of 81,000 cases of malignancy entered into the Armed Forces Central Medical Registry, Feldtman et al. (623) found seven cases of leiomyosarcoma of the rectum. This neoplasm accounted for 0.06% of all gastrointestinal malignancies and 0.3% of all rectal malignancies. Bleeding, constipation, and rectal pain are the most common symptoms. In addition, the patient may complain of a pressure sensation, diarrhea, dysuria, or difficult defecation (621). The majority of lesions are palpable on digital examination. They vary in size from 1 to 20 cm, are predominantly endorectal lesions, and in the more malignant forms have extensive perirectal invasion (621,624). Most lesions appear as submucosal masses and approximately 44% are ulcerated. The consistency may be firm, soft, or hard (621). Histologically, interlacing bands of distinct, spindleshaped cells are seen. The criterion of malignancy is simple in the high-grade forms. It is based upon the number of mitoses per high-power field (625). It is on this definition that the decision for a more definitive operation is based. Leiomyosarcoma of the rectum grows slowly and metastasizes late. Spread is by direct extension into contiguous structures and by bloodstream to liver, lungs, brain, and bone. Lymphatic metastases are rare. Blood-borne metastases to liver and lung are the most common cause of death (622). Attempts have been made to modify the treatment according to the histologic classification of the neoplasm. Because of the difficulty in determining malignancy, supposedly benign lesions treated by local excision have a high recurrence rate. When local recurrence develops and in lesions diagnosed primarily as malignant, the treatment of choice is wide local excision rather than lymphatic clearance, but abdominoperineal resection may be necessary. In a review of the literature, Labow and Hoexter (624) calculated a 20% five-year survival rate. Survival is generally poor, with a five-year survival rate between 20% and 40% and local recurrence rates as high as 20% to 86% (626). Randleman et al. (627) reported on 22 patients with leiomyosarcoma. Treatment consisted of wide local excision in 10 patients and radical excision in 11. One lesion was unresectable. None of the patients had lymph node involvement. The 5-year and 10-year survival rates were 62% and 40%, respectively. Adjuvant therapy had no effect on survival. The authors concluded that lesions less than 2.5 cm in size can be treated by wide local excision, whereas larger lesions require radical resection. In a review of the literature involving 135 cases, Khalifa et al. (621) found the same 5- and 10-year survival figures whether the patient was treated by radical or local excision. Survival at 5 and 10 years was 38.9% and 9.7%, respectively. The patient survival rate was related to the degree of differentiation of the neoplasm;

293

poorly differentiated lesions had a poor prognosis. However, local excision was followed by 67.5% recurrence. They concluded that abdominoperineal resection is still the procedure of choice. Local excision might be considered for small lesions of low-grade malignancy in poor-risk patients. Radiotherapy is of limited benefit. Chemotherapy [doxorubicin (formerly adriamycin)] may prove effective in one-third of cases (621). Yeh et al. (628) presented the prognosis after resection of 40 patients with rectal leiomyosarcoma. The mean age was 58.7 years. Anal bleeding and perianal pain were the two most common symptoms at initial diagnosis. Twentynine patients received a radical surgical resection, such as abdominoperineal resection and low anterior resection; the other 11 patients received a wide local excision, such as transrectal excision or Kraske’s operation. Sixteen lesions were classified as high-grade leiomyosarcoma, and 23 as low-grade. With a median follow-up of 35 months, 48% developed recurrence or metastasis postoperatively. The overall and disease-free (one year, three years, and five years) survival rates were 97%, 90%, and 75% and 90%, 59%, and 46%, respectively. Patients less than 50 years (n ¼ 9) and high-grade leiomyosarcoma, showed poor prognosis in the disease-free survival curve. There was a strong trend toward higher local recurrence rates for the wide local excision group than for the radical resection (55% vs. 24%). There was no difference in the incidence of distant metastasis between the two groups with different operation types. The metastasis rates of the wide local excision and radical excision groups were 27% and 38%, respectively.

& GASTROINTESTINAL STROMAL TUMOR (GIST) Changchien et al. (629) reported the outcome of curative resection of rectal GIST. Diagnostic immunohistochemical staining with CD34, CD117, S-100, desmin, and muscle-specific actin was performed in 46 consecutive patients with previously diagnosed rectal leiomyosarcoma who underwent curative resection. CD44, Bcl-2, P53, and Ki-67 staining were performed on neoplasms rediagnosed as GIST for the prognostic evaluation. There were 91.3% of patients with rectal GIST (18 females and 24 males; mean age 58.4 years). Twenty-nine patients underwent radical surgical resections, such as abdominoperineal resection or low anterior resection whereas the other 13 patients underwent wide local excision, such as transrectal excision or Kraske’s operation. Sixteen neoplasms were classified as high-grade GIST and 26 as low-grade. No neoplasms had a positive P53 stain. Recurrence or metastasis developed in 64.3% of patients postoperatively (median follow-up 52 months). The one-year, two-year, and five-year disease-free survival rates were 90.2%, 76.7%, and 43.9%, respectively. Of these patients with recurrence, subsequent resections in 12 patients with local recurrence, transarterial embolization of the neoplasm or STI-571 chemotherapies in three patients with distant metastasis were performed. The one-year, two-year and five-year overall survival rates were 97.4%, 94.3%, and 83.7%, respectively. The group with wide local excision had a higher local recurrence rate than that of the radical resection group (77% vs. 31%) despite smaller

294

& PART I: COLORECTAL DISORDERS

size (4.5 vs. 7.2 cm). There was no difference in the incidence of distant metastasis between the two groups. Younger age (less than 50 years), higher histologic grade, positive Bcl-2 status, and larger neoplasms (greater than 5 cm) were factors associated with significantly poor prognosis for rectal GIST. Radical resection was superior to wide local excision in the prevention of local recurrence, but not that of distant metastases. Lo et al. (630) described a case of GIST of the rectum who declined abdominoperineal resection. Neoadjuvant treatment with imatinib decreased the size of the lesion permitting sphincter-sparing transanal excision. The patient had no evidence of disease for 24 months postoperatively until she recurred with lung metastases. Microdissection genotyping of the recurrent lesion revealed a deletion in exon 11.

& SECONDARY CARCINOMA It is unusual for the rectum to be invaded secondarily by carcinoma. The most common malignancies to do so are large bowel carcinomas, usually of the sigmoid colon, which may involve pelvic structures, including the rectum. It is rare for prostatic carcinomas to invade Denonvilliers’ fascia and present as a rectal neoplasm. It is more common for prostatic carcinoma to grow in a circumferential fashion around the rectum, giving the appearance of an annular carcinoma without a break in mucosal continuity. This malignancy has the appearance of a bulky, infiltrating rectal carcinoma. The diagnosis can only be made following a biopsy (631). The pouch of Douglas is a common site for metastatic implantation following transperitoneal spread from carcinoma of any of the intra-abdominal organs. The most common sites of origin are the stomach and ovary. Any metastatic neoplasm can present in this site; the breast is a typical site for such a malignancy (632). Carcinoma of the cervix rarely involves the rectum (633). Metastatic renal cell carcinoma involving the rectum has been reported (634).

& MISCELLANEOUS NEOPLASMS Extraordinarily rare neoplasms such as granular cell neoplasms of the rectum have been reported (635,636). Primary teratoma of the rectum is exceedingly rare. Takao et al. (637) reported the 48th case in the world literature and provided the following information on the subject. It is thought rectal teratoma may arise from an aberrant germ cell in the embryonic digestive tract. Most cases are women with a wide range age range 8 to 80 years (mean 42.5 years). Initial symptoms may include prolapse of the lesion or hair, or bloody stool. On examination, it is often located on the anterior wall of the rectum. The diagnosis is easy if hair is found on the surface. Endorectal ultrasound demonstrates a heterogeneous echo pattern different from that of submucosal neoplasms. The treatment of choice is endoscopic excision as most teratomas are pedunculated. All primary rectal teratomas are pathologically benign. Primary rectal malignant melanoma is an exceptionally rare neoplasm associated with an extremely poor prognosis despite aggressive treatment. Hazzan et al. (638) presented two patients with bulky lesions of the lower rectum above the squamocolumnar junction. Both patients

underwent abdominoperineal resection and postoperatively were treated with autologous melanoma cell vaccine. One patient was considered disease free for months after operation; the second one developed supraclavicular lymph nodes and right lung metastasis after seven months.

& REFERENCES 1. Miles WE. Pathology of spread of cancer of rectum and its bearing upon surgery of cancerous rectum. Surg Gynecol Obstet 1931; 52:350–359. 2. Block IR, Enquist IF. Lymphatic studies pertaining to local spread of carcinoma of the rectum in female. Surg Gynecol Obstet 1961; 112:41–46. 3. Bucci L, Saifi R, Meraviglia F, et al. Rectal lymphoscintigraphy. Dis Colon Rectum 1984; 27:370–375. 4. Miscusi G, Masori L, Dell’Anna A, et al. Normal lymphatic drainage of the rectum and anal canal revealed by lymphoscintigraphy. Coloproctology 1987; 9:171–174. 5. Hermanek P. Evolution and pathology of rectal cancer. World J Surg 1982; 6:502–509. 6. Quer EA, Dahlin DC, Mayo CW. Retrograde intramural spread of carcinoma of the rectum and rectosigmoid. Surg Gynecol Obstet 1953; 96:24–30. 7. Grinnell RS. Distal intramural spread of carcinoma of the rectum and rectosigmoid. Surg Gynecol Obstet 1954; 99:421–430. 8. Hughes TG, Jenevein EP, Poulos E. Intramural spread of colon carcinoma. Am J Surg 1983; 146:697–699. 9. Madsen PM, Christiansen J. Distal intramural spread of rectal carcinoma. Dis Colon Rectum 1986; 29:279–282. 10. Pollet WG, Nicholls RJ. The relationship between the extent of distal clearance and survival and local recurrence rates after curative anterior resection for carcinoma of the rectum. Ann Surg 1984; 198:159–163. 11. Williams NS. The rationale of preservation of the anal sphincter in patients with low rectal cancer. Br J Surg 1984; 71:575–581. 12. Elliot MS, Todd IP, Nicholls RJ. Radical restorative surgery for poorly differentiated carcinoma of the mid-rectum. Br J Surg 1982; 69:564–568. 13. Grinnell RS. Lymphatic block with atypical and retrograde lymphatic metastasis and spread in carcinoma of the colon and rectum. Ann Surg 1966; 163:272–280. 14. Griffiths JD, McKinna JA, Rowbotham HD, et al. Carcinoma of the colon and rectum: circulating malignant cells and five-year survival. Cancer 1973; 31:226–236. 15. Dukes CE, Bussey HJR. The spread of rectal cancer and its effect on prognosis. Br J Cancer 1958; 12:309–320. 16. Beart RW, Steele GD Jr, Menck HR, et al. Management and survival of patients with adenocarcinoma of the colon and rectum: a national survey of the commission on cancer. J Am Coll Surg 1995; 181:225–236. 17. Mesko TW, Rodriguez-Bigas MA, Petrelli NJ. Inguinal lymph node metastases from adenocarcinoma of the rectum. Am J Surg 1994; 168:285–287. 18. Nicholls RJ, Mason AY, Morson BC, Dixon AK, Fry IK. The clinical staging of rectal cancer. Br J Surg 1982; 69:404–409. 19. Bailey HR, Huval WV, Max E, et al. Local excision of carcinoma of the rectum for cure. Surgery 1992; 111:555–561. 20. Taylor RH, Hay JH, Larsson SN. Transanal Local Excision of Selected Low Rectal Cancers. Am J Surg 1998; 175:360–363. 21. Isabel-Martinez L, Chapman AH, Hall RI. The value of barium enema in the investigation of patients with rectal carcinoma. Clin Radiol 1988; 39:531–533. 22. Beets-Tan RG, Beets GL. Rectal cancer: how accurate can imaging predict the T stage and the circumferential resection margin? Int J Colorectal Dis 2003; 18:385–391. 23. Cance WG, Cohen AM, Enker WE, et al. Predictive value of a negative computed tomographic scan in 100 patients with rectal carcinoma. Dis Colon Rectum 1991; 34:748–751. 24. Lupo L, Angelelli G, Pennarale O, et al. Improved accuracy of computed tomography in local staging of rectal cancer using water enema. Int J Colorectal Dis 1996; 11:60–64. 25. Tada M, Endo M. Ultrasonographic examination for lateral lymphatic spread and local recurrence of rectal cancer. Preoperative detection and evaluation. Dis Colon Rectum 1995; 38:1047–1052. 26. Tempero M, Brand R, Holderman K, et al. New imaging techniques in colorectal cancer. Semin Oncol 1995; 22:448–471. 27. Marohn MR. Endorectal ultrasound. Postgraduate course syllabus. SAGES 1997:126–153. 28. Garcia-Aguilar J, Pollack J, Lee SH, et al. Accuracy of endorectal ultrasonography in preoperative staging of rectal tumors. Dis Colon Rectum 2002; 45:10–15. 29. Kauer WK, Prantl L, Dittler HJ, Siewert JR. The value of endosonographic rectal carcinoma staging in routine diagnostics: a 10-year analysis. Surg Endosc 2004; 18:1075–1078. 30. Savides JJ, Master SS. Endoractal ultra sound in rectal cancer. Gastrointest Endosc 2002; 56(suppl):512–518. 31. Wong WD, Orrom WJ, Jensen LL. Preoperative staging of rectal cancer with endorectal ultrasonography. Perspect Colon Rectal Surg 1990; 3:315–334.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

32. Glaser F, Schlag P, Herfarth C. Endorectal sonography for the assessment of invasion of rectal tumors and lymph node involvement. Br J Surg 1990; 77:883–887. 33. Sentovich SM, Blatchford GJ, Falk PM, et al. Transrectal ultrasound of rectal tumors. Am J Surg 1993; 166:638–642. 34. Hunerbein M, Below C, Schlag PM. Three-dimensional endorectal ultrasonography for staging of obstructing rectal cancer. Dis Colon Rectum 1996; 39:636–642. 35. Herzog U, von Flue M, Tondelli P, et al. How accurate is endorectal ultrasound in the preoperative staging of rectal cancer? Dis Colon Rectum 1993; 36: 127–134. 36. Williamson PR, Hellinger MD, Larach SW, et al. Endorectal ultrasound of T3 and T4 rectal cancers after preoperative chemo-radiation. Dis Colon Rectum 1996; 39:45–49. 37. Bernini A, Deen KI, Madoff RD, et al. Preoperative adjuvant radiation with chemotherapy for rectal cancer: its impact on stage of disease and the role of endorectal ultrasound. Ann Surg Oncol 1996; 3:131–135. 38. Fleshman JW, Myerson RJ, Fry RD, et al. Accuracy of transrectal ultrasound in predicting pathologic stage of rectal cancer before and after preoperative radiation therapy. Dis Colon Rectum 1992; 35:823–829. 39. Martling A, Holm T, Bremmer S, Lindholm J, Cedermark B, Blomqvist L. Prognostic value of preoperative magnetic resonance imaging of the pelvis in rectal cancer. Br J Surg 2003; 90:1422–1428. 40. Bissett IP, Fernando CC, Hough DM, et al. Identification of the fascia propria by magnetic resonance imaging and its relevance to preoperative assessment of rectal cancer. Dis Colon Rectum 2001; 44:259–265. 41. Leite JS, Castro-Sousa F, Tralhao G, et al. Preoperative staging of rectal carcinoma: water enema CT vs. transrectal ultrasound. Br J Surg 1995; 82(suppl l):2–3. 42. Romano G, Esercizio L, Santangelo M, et al. Impact of computed tomography vs. intrarectal ultrasound on the diagnosis, resectability, and prognosis of locally recurrent rectal cancer. Dis Colon Rectum 1993; 36:261–265. 43. Meyenberger C, Boni RAH, Bertschinger P, et al. Endoscopic ultrasound and endorectal magnetic resonance imaging: a prospective, comparative study for preoperative staging and follow-up of rectal cancer. Endoscopy 1995; 27:469–479. 44. Hunerhein M, Pegios W, Rau B, Vagi TJ, Felix R, Schlag PM. Prospective comparison of endorectal ultrasound, three-dimensional endorectal ultrasound, and endorectal MRI in the preoperative evaluation of rectal tumors. Preliminary results. Surg Endosc 2000; 14:1005–1009. 45. Katsura Y, Yamada K, Ishizawa T, et al. Endorectal ultrasonography for the assessment of wall invasion and lymph node metastasis in rectal cancer. Dis Colon Rectum 1992; 35:362–368. 46. McNicholas MMJ, Joyce WP, Dolan J, et al. Magnetic resonance imaging of rectal carcinoma: a prospective study. Br J Surg 1994; 81:911–914. 47. Kim NK, Kim MJ, Park JK, Park SJ, Min JS. Preoperative staging of rectal cancer with MRI: accuracy and clinical usefulness. Ann Surg Oncol 2000; 7:732–737. 48. Brown G, Radcliflfe AG, Newcombe RG, DallimoreNS, Bourne MW, Williams GT. Preoperative assessment of prognostic factors in rectal cancer using highresolution magnetic resonance imaging. Br J Surg 2003; 90:355–364. 49. Branagan G, Chave H, Fuller C, McGee S, Finnis D. Can magnetic resonance imaging predict circumferential margins and TNM stage in rectal cancer? Dis Colon Rectum 2004; 47:1317–1322. 50. Thaler W, Watzka S, Martin F, et al. Preoperative staging of rectal cancer by endoluminal ultrasound vs. magnetic resonance imaging. Preliminary results of a prospective comparative study. Dis Colon Rectum 1994; 37:1189–1193. 51. Kasunoki M, Yanagi H, Kamikonya N, et al. Preoperative detection of local extension of carcinoma of the rectum using magnetic resonance imaging. J Am Coll Surg 1994; 179:653–656. 52. Billingham RP. Conservative treatment of rectal cancer. Extending the indications. Cancer 1992; 70:1355–1363. 53. Stergiou N, Haji-Kermani N, Schneider C, et al. Staging of colonic neoplasms by colonoscopic microprobe ultrasonography. Int J Colorectal Dis 2003; 18: 445–449. 54. Heriot AG, Hicks RJ, Drummond EG, et al. Does positron emission tomography change management in primary rectal cancer? A prospective assessment. Dis Colon Rectum 2004; 47:451–458. 55. Durdey P, Williams NS. The effect of malignant and inflammatory fixation of rectal carcinoma as prognosis after rectal excision. Br J Surg 1984; 71:787–790. 56. Kyzer S, Gordon PH. Hidden colostomy. Surg Gynecol Obstet 1993; 177: 181–182. 57. Tytherleigh MG, McC Mortensen NJ. Options for sphincter preservation in surgery for low rectal cancer. Br J Surg 2003; 90:922–933. 58. Williams NS, Dixon ME, Johnson D. Reappraisal of the 5 cm rule of distal excision for carcinoma of the rectum: a study of distal intramural spread and of patients’ survival. Br J Surg 1983; 70:150–154. 59. Wilson SM, Beahrs OH. The curative treatment of carcinoma of the sigmoid, rectosigmoid, and rectum. Ann Surg 1976; 183:556–565. 60. Purves H, Pietrobon R, Hervey S, Guller U, Miller W, Ludwig K. Relationship between surgeon caseload and sphincter preservation in patients with rectal cancer. Dis Colon Rectum 2005; 48:195–202.

295

61. Mettlin C, Mittelman A, Natarajan N, et al. Trends in the United States for the management of adenocarcinoma of the rectum. Surg Gynecol Obstet 1981; 153:701–706. 62. Jarvinen HJ, Ovaska J, Mecklin JP. Improvements in the treatment and prognosis of colorectal carcinoma. Br J Surg 1988; 75:25–27. 63. Nicholls RJ, Ritchie JR, Wadsworth J, et al. Total excision or restorative resection for carcinoma of the middle third of the rectum. Br J Surg 1979; 66:625–627. 64. Slanetz CA, Herter FP, Grinnell RS. Anterior resection versus abdominoperineal resection for cancer of the rectum and rectosigmoid. Am J Surg 1972; 123:110–117. 65. Stearns MW Jr. The choice among anterior resection, the pull-through and abdominoperineal resection of the rectum. Cancer 1974; 34:969–971. 66. Williams NS, Durdey P, Johnston D. The outcome following sphincter saving resection and abdominoperineal resection for low rectal cancer. Br J Surg 1985; 72:595–598. 67. Williams NS, Johnston D. Survival and recurrence after sphincter saving resection and abdominoperineal resection for recurrence of the middle third of the rectum. Br J Surg 1984; 71:278–282. 68. Wolmark N, Fisher B. An analysis of survival and treatment failure following abdominoperineal resection and sphincter-saving resection in Dukes’; B and C rectal carcinoma. Ann Surg 1986; 204:480–487. 69. Di Betta E, D’Hoore A, Filez L, Penninckx F. Sphincter saving rectum resection is the standard procedure for low rectal cancer. Int J Colorectal Dis 2003; 18:463–469. 70. Bergqvist D, Bohe M, Ekelund G. Compartment syndrome after prolonged surgery with leg supports. Int J Colorectal Dis 1990; 5:1–5. 71. Harrison JL, Hooks VH, Pearl RK, et al. Muscle fragment welding for control of massive presacral bleeding during rectal mobilization: a review of eight cases. Dis Colon Rectum 2003; 49:1115–1117. 72. Xu J, Lin J. Control of presacral hemorrhage with electrocautery through a muscle fragment pressed on the bleeding vein. J Am Coll Surg 1994; 179:351–352. 73. Qinyao W, Weijun S, Youren Z, et al. New concepts in severe hemorrhage during proctectomy. Arch Surg 1985; 12:1013–1020. 74. Nivatvongs S, Fang DT. The use of thumb tacks to stop massive presacral hemorrhage. Dis Colon Rectum 1986; 29:589–590. 75. Metzger PP. Modified packing technique for control of presacral pelvic bleeding. Dis Colon Rectum 1988; 31:981–982. 76. Zama N, Fazio VW, Jagelman DG, et al. Efficacy of pelvic packing in maintaining hemostasis after rectal excision for cancer. Dis Colon Rectum 1988; 31:923–928. 77. Junginger T, Kneist W, Heintz A. Influence of identification and preservation of pelvic autonomic nerves in rectal cancer surgery on bladder dysfunction after total mesorectal excision. Dis Colon Rectum 2003; 46:621–628. 78. Heald RJ, Husband EM, Ryall ROH. The mesorectum in rectal cancer surgery: The clue to pelvic recurrence. Br J Surg 1982; 69:613–616. 79. MacFarlane JK, Ryall RDH, Heald RJ. Mesorectal excision for rectal cancer. Lancet 1993; 341:457–460. 80. Agaba EA. Does rectal washout during anterior resection prevent local tumor recurrence? Dis Colon Rectum 2004; 47:291–296. 81. Ranbarger KR, Johnston WD, Chang JC. Prognostic significance of surgical perforation of the rectum during abdominoperineal resection for rectal carcinoma. Am J Surg 1982; 143:186–188. 82. Gordon PH, Dalrymple S. The use of staples for reconstruction after colonic and rectal surgery. In: Ravitch MM, Steichen FM, eds. Principles and Practice of Surgical Stapling. Chicago: Year Book Medical Publishers, 1987:402–431. 83. Gordon PH, Vasilevsky CA. Experience with stapling in rectal surgery. Surg Clin North Am 1984; 64:555–566. 84. Minichan DP. Enlarging the bowel lumen for the EEA stapler. Dis Colon Rectum 1982; 25:61. 85. Tchervenkov CI, Gordon PH. Simple techniques of enlarging the diameter of the bowel lumen for the performance of end-to-end anastomoses using the EEA stapler. Dis Colon Rectum 1984; 27:630–631. 86. Knight CD, Griffin FD. Techniques of low rectal reconstruction. Curr Probl Surg 1983; 20:391–456. 87. Griffin FD, Knight CD, Whitaker JM. The double stapling technique for low anterior resection. Ann Surg 1990; 211:745–752. 88. Moran BJ, Blenkinsop J, Finnis D. Local recurrence after anterior resection for rectal cancer using a double stapling technique. Br J Surg 1992; 79: 836–838. 89. Laxamana A, Solomon MJ, Cohen Z, et al. Long-term results of anterior resection using the double-stapling technique. Dis Colon Rectum 1995; 38: 1246–1250. 90. Julian TB, Ravitch MM. Evaluation of the safety of end-to-end (EEA) stapling anastomosis across linear stapled closures. Surg Clin North Am 1984; 64: 567–577. 91. Goldberg SM, Gordon PH, Nivatvongs S. Essentials of Anorectal Surgery. Philadelphia: JB Lippincott 1980; Chpt. 16, p. 182. 92. Devereux DF, Eisenstat T, Zinkin L. The safe and effective use of postoperative radiation therapy in modified Astler-Coller stage C3 rectal cancer. Cancer 1989; 63:2393–2396. 93. Sugarbaker PH. Intrapelvic prosthesis to prevent injury of the small intestine with high dosage pelvic irradiation. Surg Gynecol Obstet 1983; 157:269–271.

296

& PART I: COLORECTAL DISORDERS

94. Mcdonald PJ, Heald RJ. A survey of postoperative function after rectal anastomosis with circular stapling devices. Br J Surg 1983; 70:727–729. 95. McNamara DA, Parc R. Methods and results of sphincter-preserving surgery for rectal cancer. Cancer Control 2003; 10:212–218. 96. Renner K, Rosen HR, Novi G, Hobleng N, Schiessel R. Quality of life after surgery for rectal cancer: do we still need a permanent colostomy? Dis Colon Rectum 1999; 42:1160–1167. 97. Grumann MM, Noack EM, Hoffman IA, et al. Comparison of quality of life in patients undergoing abdominoperineal extirpation or anterior resection for cancer. Ann Surg 2001; 233:149–156. 98. Dehni N, McNamara DA, Schlegel RD, et al. Clinical effects of preoperative radiation therapy on anorectal function after proctectomy and colonic J-pouch-anal anastomosis. Dis Colon Rectum 2002; 45:1635–1640. 99. Parks AG. Transanal technique in low rectal anastomosis. Proc R Soc Med 1972; 65:975–976. 100. Miller AS, Lewis WG, Williamson MER, et al. Factors that influence functional outcome after coloanal anastomosis for carcinoma of the rectum. Br J Surg 1995; 82:1327–1330. 101. Patey PB, Enker WE, Cohen AM, et al. Treatment of rectal cancer by low anterior resection with coloanal anastomosis. Ann Surg 1994; 219:365–373. 102. Braun J, Treutner KH, Winketan G, et al. Results of inter-sphincteric resection of the rectum with direct coloanal anastomosis for rectal carcinoma. Am J Surg 1992; 163:407–412. 103. Leo E, Belli F, Andreola S, et al. Total rectal resection and complete mesorectum excision followed by coloendoanal anastomosis as the optimal treatment for low rectal cancer: the experience of the National Cancer Institute of Milano. Ann Surg Oncol 2000; 7:125–132. 104. Saito N, Ono M, Sugito M, et al. Early results of intersphincteric resection for patients with very low rectal cancer: an active approach to avoid a permanent colostomy. Dis Colon Rectum 2004; 47:459–466. 105. Gamagami R, Istvan G, Cabarrot P, Liagre A, Chiotasso P, Lazorthes F. Fecal continence following partial resection of the anal canal in distal rectal cancer: long-term results after coloanal anastomoses. Surgery 2000; 127:291–295. 106. Nathanson DR, Espat NJ, Nash GM, et al. Evaluation of preoperative and postoperative radiotherapy on long-term functional results of straight coloanal anastomosis. Dis Colon Rectum 2003; 46:888–894. 107. Olagne E, Baulieux J, de la Roche E, et al. Functional results of delayed coloanal anastomosis after preoperative radiotherapy for lower third rectal cancer. J Am Coll Surg 2000; 191:643–649. 108. Rullier E, Laurent C, Bretagnol F, Rullier A, Vendrely V, Zerbib F. Sphinctersavhig resection for all rectal carcinomas: the end of the 2-cm distal rule. Ann Surg 2005; 241:465–469. 109. Lazorthes F, Fages P, Chiotasso P, et al. Resection of the rectum with construction of a colonic reservoir and colo-anal anastomosis for carcinoma of the rectum. Br J Surg 1986; 73:136–138. 110. Lazorthes F, Chiotasso P, Gamagami RA, et al. Late clinical outcome in a randomised prospective comparison of colonic J pouch and straight coloanal anastomosis. Br J Surg 1997; 84:1449–1451. 111. Nicholls RJ, Lubowski DZ, Donaldson DR. Comparison of colonic reservoir and straight colo-anal reconstruction after rectal excision. Br J Surg 1988; 75:318–323. 112. Dehni N, Tiret E, Singland JD, et al. Long-term functional outcome after low anterior resection: comparison of low colorectal anastomosis and colonic J-pouch-anal anastomosis. Dis Colon Rectum 1998; 41:817–823. 113. Hallbook O, Pahlman L, Krog M, et al. Randomized comparison of straight and colonic J pouch anastomosis after low anterior resection. Ann Surg 1996; 224:58–65. 114. Harris GJ, Lavery IC, Fazio VW. Function of a colonic J pouch continues to improve with time. Br J Surg 2001; 88:1623–1627. 115. Nicholls J, Hall C. Treatment of cancer of the lower rectum. Br J Surg 1996; 83:15–18. 116. Hallbo¨o¨k O, Johansson K, Sjodahl R. Laser Doppler blood-flow measurement in rectal resection for carcinoma—Comparison between the straight and colonic J pouch reconstruction. Br J Surg 1996; 83:389–392. 117. Mortensen NJM, Ramirez JM, Tekeuchi N, et al. Colonic J pouch-anal anastomosis after rectal excision for carcinoma: Functional outcome Br J Surg 1995; 82:611–613. 118. Lewis WG, Holdsworth PJ, Stephenson BM, et al. Role of the rectum in the physiological and clinical results of coloanal and colorectal anastomosis after anterior resection for rectal carcinoma. Br J Surg 1992; 79:1082–1086. 119. Stebbing JF, Mortensen NJ McC. Carcinoma in a colon J pouch reservoir after low anterior resection for villous adenoma. Br J Surg 1995; 82:172. 120. Hida J, Yasutomi M, Fujimoto K, et al. Functional outcome after low anterior resection with low anastomosis for rectal cancer using the colonic J-pouch. Prospective randomized study for determination of optimum pouch size. Dis Colon Rectum 1996; 39:986–991. 121. Dehni N, Parc R. Colonic J-pouch-anal anastomosis for rectal cancer. Dis Colon Rectum 2003; 46:667–675. 122. Laurent A, Parc Y, McNamara D, Parc R, Tiret E. Colonic J-pouch-anal anastomosis for rectal cancer: a prospective, randomized study comparing handsewn vs. stapled anastomosis. Dis Colon Rectum 2005; 48:729–734.

123. Machado M, Nygren J, Goldman S, Ljungqvist O. Similar outcome after colonic pouch and side-to-end anastomosis in low anterior resection for rectal cancer: a prospective randomized trial. Ann Surg 2003; 238:214–220. 124. Machado M, Nygren J, Goldman S, Ljungqvist O. Functional and physiologic assessment of the colonic reservoir or side-to-end anastomosis after low anterior resection for rectal cancer: a two-year follow-up. Dis Colon Rectum 2005; 48:29–36. 125. Dehni N, McNamara DA, Schlegel RD, Guiguet M, Tiret E, Parc R. Clinical effects of preoperative radiation therapy on anorectal function after proctectomy and colonic J-pouch-anal anastomosis. Dis Colon Rectum 2002; 45: 1635–1640. 126. Sailer M, Fuchs KH, Fein M, Thiede A. Randomized clinical trial comparing quality of life after straight and pouch coloanal reconstruction. Br J Surg 2002; 89:1108–1117. 127. Heah SM, Seow-Choen F, Eu KW, Ho YH, Tang CL. Prospective, randomized trial comparing sigmoid vs. descending colonic J-pouch after total rectal excision. Dis Colon Rectum 2002; 45:322–328. 128. Ho YH, Seow-Choen F, Tan M. Colonic J-pouch function at six months versus straight coloanal anastomosis at two years: randomized controlled trial. World J Surg 2001; 25:876–881. 129. Wang JY, You YT, Chen HH, Chiang JM, Yeh CY, Tang R. Stapled colonic J-pouch-anal anastomosis without a diverting colostomy for rectal carcinoma. Dis Colon Rectum 1997; 40:30–34. 130. Hamel CT, Metzger J, Curti G, Degen L, Harder F, von Flue MO. Ileocecal reservoir reconstruction after total mesorectal excision: functional results of the long-term follow-up. Int J Colorectal Dis 2004; 19:574–579. 131. da Silva GM, Kaiser R, Borjesson L, et al. The effect of diverticular disease on the colonic J pouch. Colorectal Dis 2004; 6:171–175. 132. Hida J, Yoshifuji T, Tokoro T, et al. Long-term functional outcome of low anterior resection with colonic J-pouch reconstruction for rectal cancer in the elderly. Dis Colon Rectum 2004; 47:1448–1454. 133. Gervaz P, Rotholtz N, Wexner SD, et al. Colonic J-pouch function in rectal cancer patients: impact of adjuvant chemoradiotherapy. Dis Colon Rectum 2001; 44:1667–1675. 134. Harris GJ, Lavery IC, Fazio VW. Reasons for failure to construct the colonic J-poiich. What can be done to improve the size of the neorectal reservoir should it occur? Dis Colon Rectum 2002; 45:1304–1308. 135. Furst A, Suttner S, Agha A, Beham A, Jauch KW. Colonic J-pouch vs. coloplasty following resection of distal rectal cancer: early results of a prospective, randomized, pilot study. Dis Colon Rectum 2003; 46:1161–1166. 136. Mantyh CR, Hull TL, Fazio VW. Coloplasty in low colorectal anastomosis: manometric and functional comparis with straight and colonic J-pouch anastomosis. Dis Colon Rectum 2001; 44:37–42. 137. Z’graggen K, Maurer CA, Birrer S, Giachino D, Kern B, Buchler MW. A new surgical concept for rectal replacement after low anterior resection: the transverse coloplasty pouch. Ann Surg 2001; 234:780–785. 138. Remzi FH, Fazio VW, Gorgun E, et al. Quality of life, functional outcome, and complications of coloplasty pouch after low anterior resection. Dis Colon Rectum 2005; 48:735–743. 139. Localio SA, Eng K, Coopa GF. Abdominosacral resection for midrectal cancer: a 15 year experience. Ann Surg 1983; 198:320–324. 140. Hardy TG, Pace WC, Maney JW, et al. A bioffagmentable ring for sutureless bowel anastomosis. An experimental study. Dis Colon Rectum 1985; 28: 484–490. 141. Corman ML, Prager ED, Hardy TG, et al. Comparison of the Valtrac biofragmentable anastomosis ring with conventional suture and stapled anastomosis in colon surgery. Results of a prospective randomized clinical trial. Dis Colon Rectum 1989; 32:183–187. 142. Moriya Y, Hojo K, Sawada T. En bloc excision of lower ureter and internal iliac vessels for locally advanced upper rectal and rectosigmoid cancers. Use of ileal segment for ureteral repair. Dis Colon Rectum 1988; 31:872–878. 143. Geerdes BP, Zoetmulder FAN, Baeten CGMI. Double dynamic graciloplasty and coloperineal pull-through after abdominoperineal resection. Eur J Cancer 1995; 31A:1248–1252. 144. Martling A, Singnomklao T, Holm T, Rutqvist LE, Cedermark B. Prognostic significance of both surgical and pathological assessment of curative resection for rectal cancer. Br J Surg 2004; 91:1040–1045. 145. Pa¨hlman L, Glimelius B, Enblad P. Clinical characteristics and their relation to surgical curability in adenocarcinoma of the rectum and rectosigmoid. A population-based study on 279 consecutive patients. Acta Chir Scand 1985; 151:685–693. 146. Davis NE, Evans EB, Cohen JR, et al. Colorectal cancer: a large unselected Australian series. Aust N Z J Surg 1987; 57:153–159. 147. Enblad P, Adami HO, Bergstrom R, et al. Improved survival of patients with cancers of the colon and rectum. J Natl Cancer Inst 1988; 80:586–591. 148. Michelassi F, Block GE, Vannucci L, et al. A 5 to 21 year follow-up and analysis of 250 patients with rectal adenocarcinoma. Ann Surg 1988; 208: 379–386. 149. Amato A, Pescatori M, Butti A. Local recurrence following abdominoperineal excision and anterior resection for rectal carcinoma. Dis Colon Rectum 1991; 34:317–322.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

150. Tagliacozzo S, Accordino M. Pelvic recurrence after surgical treatment of rectal and sigmoid cancer. A prospective clinical trial in 274 patients. Int J Colorectal Dis 1992; 7:135–140. 151. Fandrich F, Schroder DW, Saliveros E. Long-term survival after curative resection for carcinoma of the rectum. J Am Coll Surg 1994; 178:271–275. 152. Adam IJ, Mohamdee MD, Martin IG, et al. Role of circumferential margin involvement in the local recurrence of rectal cancer. Lancet 1994; 344:707–711. 153. Clemmesen T, Sprechler M. Recording of patients with colorectal cancer on a database: Results and advantages. Eur J Surg 1994; 160:175–178. 154. Enker WE, Thaler HT, Craner ML, et al. Total mesorectal excision in the operative treatment of carcinoma of the rectum. J Am Coll Surg 1995; 181:335–346. 155. Hermanek P, Wiebelt H, Staimmer D, et al. Prognostic factors of rectum carcinoma—Experience of the German multicentre study SGCPC. Tumori 1995; 81(suppl):60–64. 156. Singh S, Morgan BF, Broughton M, et al. A 10-year prospective audit of outcome of surgical treatment for colorectal carcinoma. Br J Surg 1995; 82: 1486–1490. 157. Zaheer S, Pemberton JH, FaroukR, Dozois RR, Wolff BG, Ilstrup D. Surgical treatment of adenocarcinoma of the rectum. Ann Surg 1998; 227:800–811. 158. Killingback M, Barron P, Dent OF. Local recurrence after curative resection of cancer of the rectum without total mesorectal excision. Dis Colon Rectum 2001; 44:473–483. 159. Staib L, Link KH, Blatz A, Beger HG. Surgery of colorectal cancer: surgical morbidity and five- and ten-year results in 2400 patients—monoinstitutional experience. World J Surg 2002; 26:59–66. 160. Croxford M, Salerno G, Walson M, Sexton R, Heald RJ, Moran BJ. Does the height of the tumour influence the local recurrence rate in low rectal cancer treatment by total mesorectal resection? Br J Surg 2001; 91(suppl 1):1. 161. Wibe A, Syse A, Andersen E, TretH S, Myrvold HE, Soreide O. Norwegian Rectal Cancer Group. Oncological outcomes after total mesorectal excision for cure for cancer of the lower rectum: anterior vs. abdominoperineal resection. Dis Colon Rectum 2004; 47:48–58. 162. Law WL, Ho JW, Chan R, Auj G, Chu KW. Outcome of anterior resection for stage II rectal cancer without radiation: the role of adjuvant chemotherapy. Dis Colon Rectum 2005; 48:218–226. 163. Gordon PH, Shrier J, Obrand DI, Unpublished 2006. 164. Fedorov VD, Odarjuk TS. Sphincter saving operation for cancer of the rectum. Coloproctology 1983; 6:336–338. 165. Zhou XG, Yu BM, Shen YX. Surgical treatment and late results in 1226 cases of colorectal cancer. Dis Colon Rectum 1983; 26:250–256. 166. Malmberg M, Graffner H, Ling L, et al. Recurrence and survival after anterior resection of the rectum using the end to end anastomotic stapler. Surg Gynecol Obstet 1986; 163:231–234. 167. Heberer C, Denecke H, Demmel N, et al. Local procedures in the management of rectal cancer. World J Surg 1987; 11:499–503. 168. Belli L, Beati CA, Frangi M, et al. Outcome of patients with rectal cancer treated by stapled anterior resection. Br J Surg 1988; 75:422–424. 169. Jatzko G, Lisborg P, Wette V. Improving survival rates for patients with colorectal cancer. Br J Surg 1992; 79:588–591. 170. Isenberg J, Keller HW, Pichlmaier H. Middle and lower third rectum carcinoma: Sphincter saving or abdominoperineal resection? Eur J Surg Oncol 1995; 21:265–268. 171. Law WL, Chu KW. Abdominoperineal resection is associated with poor oncological outcome. Br J Sur 2004; 91:1493–499. 172. Elliot MS, Steven DM, Terblanche J. Abdominoperineal resection of the rectum for carcinoma at Groote Schuur Hospital, Cape Town, 1971–1982. S Afr Med J 1984; 65:411–413. 173. Huguier M, Depoux F, Houry S, et al. Adenocarcinoma of the rectum treated by abdominoperineal excision: Multivariate analysis of prognostic factors. Int J Colorectal Dis 1990; 5:144–147. 174. Dehni N, McFadden N, McNamara DA, Guiguet M, Tiret E, Pargj R. Oncologic results following abdominoperineal resection for adenocarcinoma of the low rectum. Dis Colon Rectum 2003; 46:867–874. 175. Rosen L, Veidenheimer MC, Coller JA, et al. Mortality, morbidity, and patterns of recurrence after abdominoperineal resection for cancer of the rectum. Dis Colon Rectum 1982; 25:202–208. 176. Hojo K. Anastomotic recurrence after sphincter saving resection for rectal cancer. Length of distal clearance of the bowel. Dis Colon Rectum 1986; 29: 11–14. 177. Sugihara K, Moriya Y, Akasu T, et al. Pelvic autonomic nerve preservation for patients with rectal carcinoma. Oncologic and functional outcome. Cancer 1996; 78:1871–1880. 178. Christiansen J. Place of abdominoperineal resection in rectal cancer. J R Soc Med 1988; 81:143–145. 179. Colombo PL, Foglieni CLS, Morone C. Analysis of recurrence following curative low anterior resection and stapled anastomosis for carcinoma of the middle third and lower rectum. Dis Colon Rectum 1987; 30:457–464. 180. Gillen P, Peel ALG. Comparison of the mortality, morbidity and incidence of local recurrence in patients with rectal cancer treated by either stapled anterior resection or abdominoperineal resection. Br J Surg 1986; 73:339–341. 181. Pescatori M, Mattana C, Maria G, et al. Outcome of colorectal cancer. Br J Surg 1987; 74:370–372.

297

182. Yeatman TJ, Bland KI. Sphincter saving procedures for distal carcinoma of the rectum. Ann Surg 1989; 209:1–18. 183. Kyzer S, Gordon PH. Experience with the use of the circular stapler in rectal surgery. Dis Colon Rectum 1992; 35:696–706. 184. Cutait DE, Cutait R, Da Silva JH, et al. Stapled anastomosis in colorectal surgery. Dis Colon Rectum 1981; 24:155–160. 185. Killingback M. Intrapelvic restorative resection for carcinoma of the large bowel, Hunterian Lecture. Royal College of Surgeons of England, April 1, 1981. 186. Dorricott NJ, Braddley RM, Keighley MRB, et al. Complications of rectal anastomoses with end-to-end anastomosis (EEA) stapling instrument. Ann R Coll Surg Engl 1982; 64:171–174. 187. Friis J, Hjortrup A, Nielson OV. Sphincter-saving resection of the rectum using the EEA auto staples. Acta Chir Scand 1982; 148:379–381. 188. Goligher JC. The use of circular staplers for the construction of colorectal anastomoses after anterior resection. In: Heberer G, Denecke H, eds. Colorectal Surgery. Berlin: Springer-Verlag, 1982:107–113. 189. Hamelmann H, Thiede A, Jostarndt L, et al. Stapler anastomoses in the low rectal third. In: Heberer G, Denecke H, eds. Colorectal Surgery. Berlin: Springer-Verlag, 1982:115–119. 190. Helm W, Rowe PH. Rectal anastomosis with the EEA stapling instrument. Ann R Coll Surg Engl 1982; 64:356–357. 191. Leff EI, Hoexter B, Labow SB, et al. The EEA stapler in low colorectal anastomoses: initial experience. Dis Colon Rectum1982; 162(235):704–707. 192. Polglase AL, Cunningham GE, Hughes ESR, et al. Initial clinical experience with the EEA stapler. Aust N Z J Surg 1982; 52:71–75. 193. Vezeridis M, Evans TJ, Mittelman A, et al. EEA stapler in low anterior anastomosis. Dis Colon Rectum 1982; 35:364–367. 194. Anderberg B, Enblad P, Sjodahl R, et al. The EEA stapling device in anterior resection for carcinoma of the rectum. Br J Surg 1983; 149:99–103. 195. Fegiz G, Angelini L, Bezzi M. Rectal cancer: Restorative surgery with the EEA stapling device. Int Surg 1983; 68:13–18. 196. Isbister WH, Beasley SW, Dowle CS. The EEA stapler—A Wellington experience. Coloproctology 1983; 6:323–326. 197. Kennedy HL, Rothenberger DA, Goldberg SM, et al. Colo-colostomy and coloproctostomy utilizing the circular intraluminal stapling devices. Dis Colon Rectum 1983; 26:145–148. 198. Resnick SD, Burstein AE, Viner YL. Use of the stapler in anterior resection for cancer of the rectosigmoid. Isr J Med Sci 1983; 19:128–133. 199. Fazio VW. Advances in the surgery of rectal carcinoma utilizing the circular stapler. Spratt JS, ed. In: Neoplasms of the Colon, Rectum and Anus. Philadelphia: WB Saunders, 1984:268–288. 200. Hedberg SE, Helmy AH. Experience with gastrointestinal stapling at the Massachusetts General Hospital. Surg Clin North Am 1984; 64:511–528. 201. Steichen FM, Ravitch MM. Stapling in Surgery. Chicago: Year Book Medical Publishers, 1984:271. 202. Fazio VW, Jagelman DG, Lavery IC, McGonale BA. Evaluation of the proximate-ILS circular stapler. Ann Surg 1985; 201:108–114. 203. McGinn FP, Gartell PC, Clifford PC, et al. Staples or sutures for low colorectal anastomoses: a prospective randomized trial. Br J Surg 1985; 73:603–605. 204. Antonsen HK, Kronborg O. Early complications after low anterior resection for rectal cancer using the EEA stapling device: a prospective trial. Dis Colon Rectum 1987; 30:579–583. 205. Zannini G, Renda A, Lepore R, et al. Mechanical anterior resection for carcinoma of the midrectun: long-term results. Int Surg 1987; 72:18–19. 206. Dehong Y, Yue T, Qinglan C, et al. Anterior resection with EEA stapler and manual sutures: comparison of two patient groups. In: Ravitch MM, Steichen FM, Welter R, eds. Current Practice of Surgical Stapling. Philadelphia: Lea & Febiger, 1991:101–103. 207. Steegmuller KW, Brown S. Experience with 140 stapled colorectal anastomoses. Ravitch MM, Steichen FM, Welter R, eds. Current Practice of Surgical Stapling. Philadelphia: Lea & Febiger, 1991:269–271. 208. Karanjia ND, Corder AP, Beam P, et al. Leakage from stapled low anastomosis after total mesorectal excision for carcinoma of the rectum. Br J Surg 1994; 81:1224–1226. 209. Fingerhut A, Hay JM, Elhadad A, et al. Supraperitoneal colorectal anastomosis: hand-sewn versu circular staples—a controlled clinical trial. Surgery 1995; 118:479–485. 210. Detry RJ, Kartheuser A, Delriviere L, et al. Use of the circular staples in 1000 consecutive colorectal anastomoses: experience of one surgical team. Surgery 1995; 117:140–145. 211. Goligher JC, Lee PWR, Lintott DJ. Experience with the Russian model 249 suture gun for anastomosis of the rectum. Surg Gynecol Obstet 1979; 148:517–524. 212. Beart RW Jr, Kelly KA. Randomized prospective evaluation of the EEA stapler for colorectal anastomosis. Am J Surg 1981; 141:143–147. 213. Adloff M, Arnaud JP, Beehary S. Stapled vs sutured colorectal anastomosis. Arch Surg 1980; 115:1436–1438. 214. Scher KR, Scott-Conner C, Jones CW, et al. A comparison of stapled and sutured anastomoses in colonic operations. Surg Gynecol Obstet 1982; 155:489–493. 215. Brown AA, Gasson JE, Brown RA. Experience with the EEA stapler in carcinoma of the lower rectum. S Afr Med J 1981; 21:258–261.

298

& PART I: COLORECTAL DISORDERS

216. Goligher JC. The use of stapling devices for the construction of low rectal anastomoses. Ann Chir Gynaecol 1980; 69:125–131. 217. Heald RJ. Towards fewer colostomies—The impact of circular stapling devices on the surgery of rectal cancer in the district hospital. Br J Surg 1980; 67: 198–200. 218. Odou MW, O’Connell TX. Changes in the treatment of rectal carcinoma and effects on local recurrence. Arch Surg 1986; 121:1114–1116. 219. Hurst PR, Prout WG, Kelly JM, et al. Local recurrence after low anterior resection using the staple gun. Br J Surg 1982; 69:275–276. 220. Heald RJ, Husband EM, Ryall RDH. The mesorectum in rectal cancer surgery: the clue to pelvic recurrence? Br J Surg 1982; 69:13–16. 221. Rosen CB, Beart RW, Duane M, et al. Local recurrence of rectal carcinoma after hand sewn and stapled anastomoses. Dis Colon Rectum 1985; 28:305–309. 222. Heald RJ, Leicester RJ. The low stapled anastomoses. Br J Surg 1981; 68:333–337. 223. Luke M, Kirkegaard P, Lendorf A, et al. Pelvic recurrence rate after abdominoperineal resection and low anterior resection for rectal cancer before and after introduction of the stapling technique. World J Surg 1983; 7:616–619. 224. Kennedy HL, Langevin JM, Goldberg SM, et al. Recurrence following stapled coloproctostomy for carcinomas of the mid portion of the rectum. Surg Gynecol Obstet 1985; 160:513–516. 225. Leff EI, Shaver JO, Hoexter B, et al. Anastomotic recurrences after low anterior resection: stapled vs hand sewn. Dis Colon Rectum 1985; 28:164–167. 226. Wolmark N, Gordon PH, Fisher B, et al. A comparison of stapled and hand sewn anastomoses in patients undergoing resection for Dukes’ B and C colorectal cancer: an analysis of disease free survival and survival from NSABP randomized clinical trials. Dis Colon Rectum 1986; 29:344–350. 227. Carlsson V, Lasson A, Ekelund G. Recurrence rates after curative surgery for rectal carcinoma, with special reference to their accuracy. Dis Colon Rectum 1987; 30:431–434. 228. Neville R, Fielding LP, Amendola C. Local tumor recurrence after curative resection for rectal cancer: a ten year hospital review. Dis Colon Rectum 1987; 30:12–17. 229. Akyol AM, McGregor JR, Golloway DJ, et al. Recurrence of colorectal cancer after sutured and stapled large bowel anastomoses. Br J Surg 1991; 78: 1297–1300. 230. Willett CG, Lewandrowski K, Donnelly S, et al. Are there patients with stage I rectal carcinoma at risk for failure after abdominoperineal resection? Cancer 1992; 69:1651–1655. 231. Arbman G, NilsSon E, HallbookO, Sjodahl R. Local recurrence following total mesorectal excision for rectal cancer. Br J Surg 1996; 83:375–379. 232. Heald RJ, Moran BJ, Ryall RD, Sexton R, MacFarlane JK. Rectal cancer: the Basingstoke experience of total mesorectal excision, 1978–1997. Arch Surg 1998; 133:894–899. 233. Hall NR, Finan PJ, Al-Jaberi T, et al. Circumferential margin involvement after mesorectal excision of rectal cancer with curative intent: predictors of survival but not local recurrence. Dis Colon Rectum 1998; 41:979–983. 234. Dahlberg M, Glimelius B, Pahlman L. Changing strategy for rectal cancer is associated with improved outcome. Br J Surg 1999; 86:379–384. 235. Martling AL, Holm T, Rutqvist LE, Moran BJ, Heald RJ, Cedemark B. Effect of a surgical training programme on outcome of rectal cancer in the County of Stockholm. Stockholm Colorectal Cancer Study Group, Basingstoke Bowel Cancer Research Project. Lancet 2000; 356:93–96. 236. Kapiteijn, Marijnen CA, Nagtegaal ID, et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001; 345:638–646. 237. Nesbakken A, Nygaard K, Westerheim O, Mala T, Lunde OC. Local recurrence after mesorectal excision for rectal cancer. Eur J Oncol 2002; 28:126–134. 238. Bulow S, Christensen IJ, Harling H, Kronborg O, Fenger C, Nielsen HJ; Danish TME Study Group; RANX05 Colorectal Cancer Study Group. Recurrence and survival after mesorectal excision for rectal cancer. Br J Surg 2003; 90: 974–980. 239. Cecil TD, Sexton R, Moran BJ, Heald RJ. Total mesorectal excision results in low local recurrence rates in lymph node-positive rectal cancer. Dis Colon Rectum 2004; 47:1145–1149. 240. Kockerling F, Reymond MA, Altendorf-Hofmann A, Dworak O, Hohenberger W. Influence of surgery on metachronous distant metastases and survival in rectal cancer. J Clin Oncol 1998; l6:324–329. 241. McCall JL, Cox MR, Wattchow DA. Analysis of local recurrence rates after surgery alone for rectal cancer. Int J Colorectal Dis 1995; 10:126–132. 242. Nagtegaal ID, van de Velde CJ, van der Worp E, Kapiteijn E, Quirke P, van Krieken JH. Cooperative Clinical Investigators of the Dutch Colorectal Cancer Group. Macroscopic evaluation of rectal cancer resection specimen: clinical significance of the pathologist in quality control. J Clin Oncol 2002; 4(20):1729–1734. 243. Wibe A, Eriksen MT, Syse A, Myrvold HE, Soreide O. Norwegian Rectal Cancer Group. Total mesorectal excision for rectal cancer-what can be achieved by a national audit? Colorectal Dis 2003; 5:471–477. 244. Ike H, Shimada H, Yamaguchi S, Ichikawa Y, Fujii S, Ohki S. Outcome of total pelvic exenteration for primary rectal cancer. Dis Colon Rectum 2003:474–480. 245. Manfredi S, Benhamiche AM, Meny B, Cheynel N, Rat P, Faivre J. Populationbased study of factors influencing occurrence and prognosis of local recurrence after surgery for rectal cancer. Br J Surg 2001; 88:1221–1227.

246. Bonadeo FA, Vaccaro CA, Benati ML, Quintana GM, Garione XE, Telenta MT. Rectal cancer: local recurrence after surgery without radiotherapy. Dis Colon Rectum 2001; 44:374–379. 247. Vironen JH, Halme L, Sainio P, et al. New approaches in the management of rectal carcinoma result in reduced local recurrence rate and improved survival. Eur J Surg 2002; 168:158–164. 248. Parks AG. Per anal anastomosis. World J Surg 1982; 6:531–538. 249. Sweeney JL, Ritchie JK, Hawley PR. Resection and sutured per anal anastomosis for carcinoma of the rectum. Dis Colon Rectum 1989; 32:103–106. 250. Enker WE, Stearns MW Jr, Janov AJ. Per anal coloanal anastomosis following low anterior resection for rectal carcinoma. Dis Colon Rectum 1985; 28: 576–581. 251. Cohen AM, Enker WE, Minsky BD. Proctectomy and coloanal reconstruction for rectal cancer. Dis Colon Rectum 1990; 33:40–43. 252. Drake DB, Pemberton JH, Beart RW, et al. Coloanal anastomosis in the management of benign and malignant rectal disease. Ann Surg 1987; 206:600–605. 253. Hautefeuille P, Valleus P, Perniceni T, et al. Functional and oncologic results after coloanal anastomoses for low rectal carcinoma. Ann Surg 1988; 207:61–64. 254. Vernava AM III, Robbins PL, Brabbee GW. Restorative resection: Coloanal anastomosis for benign and malignant disease. Dis Colon Rectum 1989; 32:690–693. 255. Wunderlich M, Hansuch JK, Schiessel R. Results of coloanal anastomosis. A prospective study. Int J Colorectal Dis 1986; 1:157–161. 256. Kohler A, Athanasiadis S, Ommer A, Psarakis E. Long-term results of low anterior resection with intersphincteric anastomosis in carcinoma of the lower one-third of the rectum: analysis of 31 patients. Dis Colon Rectum 2000; 43:843–850. 257. Ho P, Law WL, Chan SC, Lam CK, Chu KW. Functional outcome following low anterior resection with total mesorectal excision in the elderly. Int J Colorectal Dis 2003; 18:230–233. 258. Phillips PS, Farquharson SM, Sexton R, Heald RJ, Moran BJ. Rectal cancer in the elderly: patients’ perception of bowel control after restorative surgery. Dis Colon Rectum 2004; 47:287–290. 259. Rengan R, Paty JP, Wong WD, et al. Distal cT2N0 Rectal Cancer: Is There an Alternative to Abdominoperineal Resection? J Clin Oncol 2005; 23:4905–4912. 260. Engel J, Kerr J, Schlesinger-Raab A, Eckel R, Sauer H, Holzel D. Quality of life in rectal cancer patients: a four-year prospective study. Ann Surg 2003; 238:203–213. 261. Schmidt CE, Bestmann B, Kuchler T, Longo WE, Kremer B. Prospective evaluation of quality of life of patients receiving either abdominoperineal resection or sphincter-preserving procedure for rectal cancer. Ann Surg Oncol 2005; 12:117–123. 262. Porter GA, O’Keefe GE, Yakimets WW. Inadvertent perforation of the rectum during abdominoperineal resection. Am J Surg 1996; 172:324–327. 263. Phang PT, MacFarlane JK, Taylor RH, et al. Effect of emergent presentation on outcome from rectal cancer management. Am J Surg 2003; 185:450–454. 264. Vironen JH, Sainio P, Husa AI, Kellokumpu IH. Complications and survival after surgery for rectal cancer in patients younger than and aged 75 years or older. Dis Colon Rectum 2004; 47:1225–1231. 265. Eriksen MT, Wibe A, Syse A, Haffner J, Wiig JN. Norwegian Rectal Cancer Group Norwegian Gastrointestinal Cancer Group. Inadvertent perforation during rectal cancer resection in Norway. Br J Surg 2004; 91:210–216. 266. Leong AF. Total mesorectal excision (TME)—twenty years on. Ann Acad Med Singapore 2003; 32:159–162. 267. Kressner U, Graf W, Mahteme H, PahJman L, Glimelius B. Septic complications and prognosis after surgery for rectal cancer. Dis Colon Rectum 2002; 45:316–321. 268. Platell CF, Thompson PJ, Makin GB. Sexual health in women following pelvic surgery for rectal cancer. Br J Surg 2004; 91:465–468. 269. Schmidt CE, Bestmann B, Kuchler T, Longo WE, Kremer B. Ten-year historic cohort of quality of life and sexuality in patients with rectal cancer. Dis Colon Rectum 2005; 480:483–492. 270. Sharma A, Hartley J, Monson JR. Local excision of rectal tumors. Surg Oncol 2003; 12:51–61. 271. Killingback MJ. Indications for local excision of rectal cancer. Br J Surg 1985; 72(suppl):S54–S56. 272. Whiteway J, Nicholls RJ, Morson BC. The role of surgical local excision in the treatment of rectal cancer. Br J Surg 1985; 72:694–697. 273. Morson BC. Histological criteria for local excision. Br J Surg 1985; 72(suppl):S53–S54. 274. Minsky BD, Rich T, Recht A, et al. Selection criteria for local excision with or without adjuvant radiation therapy for rectal cancer. Cancer 1989; 63:1421. 275. Huddy SPJ, Hasband EM, Cook MG, et al. Lymph node metastases in early rectal cancer. Br J Surg 1993; 80:1457–1458. 276. Blumberg D, Paty PB, Guillem JG, et al. All patients with small intramural rectal cancers are at risk for lymph node metastasis. Dis Colon Rectum 1999; 42:881–885. 277. Nascimbeni R, Burgart LJ, Nivatvongs S, Larson DR. Risk of lymph node metastasis hi Tl carcinoma of the colon and rectum. Dis Colon Rectum 2002; 45:200–206. 278. Sakuragi M, Togashi K, Konishi F, et al. Predictive factors for lymph node metastasis in Tl stage colorectal carcinomas. Dis Colon Rectum 2003; 46: 1626–1632.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

279. Okabe S, Shia J, Nash G, et al. Lymph node metastasis in Tl adenocarcinoma of the colon and rectum. J Gastrointest Surg 2004; 8:1032–1039. 280. Lavery IC, Jones IT, Weakley FL, et al. Definitive management of rectal cancer by contact (endocavitary) irradiation. Dis Colon Rectum 1987; 30:835–838. 281. Gopaul D, Belliveau P, Vuong T, et al. Outcome of local excision of rectal carcinoma. Dis Colon Rectum 2004; 47:1780–1788. 282. Nascimbeni R, Nivatvongs S, Larson PR, Burgart LJ. Long-term survival after local excision for Tl carcinoma of the rectum. Dis Colon Rectum 2004; 47: 1773–1779. 283. Sengupta S, Tjandra JJ. Local excision of rectal cancer: what is the evidence? Dis Colon Rectum 2001; 44:1345–1361. 284. Biggers OR, Beart RW Jr, Ilstrup DM. Local excision of rectal cancer. Dis Colon Rectum 1986; 29:374–377. 285. Horn A, Halvorsen JF, Morild I. Transanal extirpation for early rectal cancer. Dis Colon Rectum 1989; 32:769–772. 286. Killingback MJ. Local excision of carcinoma of the rectum. World J Surg 1992; 16:437–446. 287. Lock MR, Ritchie JK, Hawley PR. Reappraisal of radical local excision for carcinoma of the rectum. Br J Surg 1993; 80:928–929. 288. Willett CG, Compton CC, Shellito PC, et al. Selection factors for local excision or abdominoperineal resection of early stage rectal cancer. Cancer 1994; 73:2716–2720. 289. Green JD, Riether RD, Rosen L, et al. Transanal diskectomy for cure of rectal adenocarcinoma: is clinical judgment accurate? [abstract]. Dis Colon Rectum 1994; 37:5. 290. Faivre J, Chaume JC, Pigot F, et al. Transanal electro resection of small rectal cancer: a sole treatment. Dis Colon Rectum 1996; 39: 270–278. 291. Obrand DI, Gordon PH. Results of local excision for rectal carcinoma. Can J Surg 1996; 39:463–468. 292. Chakravarti A, Compton CC, Shellito PC, et al. Long-term follow-up of patients with rectal cancer managed by local excision with and without adjuvant irradiation. Ann Surg 1999; 230:49–54. 293. Steele GD Jr, Herndon JE, Bleday R, et al. Sphincter-sparing treatment for distal rectal adenocarcinoma. Ann Surg Oncol 1999; 6:433–441. 294. Varma MG, Rogers SJ, Schrock TR, Welton ML. Local excision of rectal carcinoma. Arch Surg 1999; 134:863–867. 295. Mellgren A, Sirivongs P, Rothenberger DA, Madoff RD, Garcia-Aguilar J. Is local excision adequate therapy for early rectal cancer? Dis Colon Rectum 2000; 43:1064–1071. 296. Paty PB, Nash GM, Baron P, et al. Long-term results of local excision for rectal cancer. Ann Surg 2002; 236:522–529. 297. Chambers WM, Khan U, Gagliano A, Smith RD, Sheffield J, Nicholls RJ. Tumour morphology as a predictor of outcome after local excision of rectal cancer. Br J Surg 2004; 91:457–459. 298. Madbouly KM, Remzi FH, Erkek BA, et al. Recurrence after transanal excision of Tl rectal cancer: should we be concerned? Dis Colon Rectum 2005; 48: 711–719. 299. Graham RA, Garnsey L, Jessup JM. Local excision of rectal carcinoma. Am J Surg 1990; 160:306–312. 300. Muto T, Oya M. Recent advances in diagnosis and treatment of colorectal T1 carcinoma. Dis Colon Rectum 2003; 46(suppl):S89–S93. 301. Baxter NN, Stewart AR, Nelson H. Local Excision for rectal cancer. J Clin Oncol 2004; 22:145–273. 302. Minsky BD. Clinical experience with local excision and postoperative radiotherapy for rectal-cancer. Dis Colon Rectum 1993; 36:405–409. 303. Hanloser D, Wolff BG, Larson DW, Ping J, Nivatvongs S. Immediate radical resection after local excision of rectal cancer: an oncologic compromise? Dis Colon Rectum 2005; 48:429–437. 304. Cuthbertson AM, Simpson R. Curative local excision of rectal adenocarcinoma. Aust N Z J Surg 1986; 56:229–231. 305. Friel CM, Cromwell JW, Marra C, Madoff RD, Rothen berger DA, GarciaAguilar J. Salvage radical surgery after failed local excision for early rectal cancer. Dis Colon Rectum 2002; 45:875–879. 306. Weiser MR, Landmann RG, Wong WD, et al. Surgical salvage of recurrent rectal cancer after transanal excision. Dis Colon Rectum 2005; 48:1169–1175. 307. Baron PL, Enker WE, Zakowski MF, et al. Immediate vs. salvage resection after local treatment for early rectal cancer. Dis Colon Rectum 1995; 38: 177–181. 308. Rouanet P, Saint Aubert B, Fabre JM, et al. Conservative treatment for low rectal carcinoma by local excision with or without radiotherapy. Br J Surg 1993; 80:1452–1456. 309. Bennett CJ Jr, Sombeck MD, Mendanhall WM, et al. Conservative alternatives in the management of early carcinoma of the rectum. South Med J 1993; 86:409–413. 310. Wagman R, Minsky BD, Cohen AM, Safe L, Paty PB, Guillem JG. Conservative management of rectal cancer with local excision and postoperative adjuvant therapy. Int J Radiat Oncol Biol Phys 1999; 44:841–846. 311. Lamont JP, McCarty TM, Digan RD, Jacobson R, Tulanon P, Lichliter WE. Should locally excised T1 rectal cancer receive adjuvant chemoradiation? Am J Surg 2000; 80:402–405. 312. Coco C, Magistrelli P, Netri G, et al. Combined modality therapy in low risk (T2 NO) rectal cancer. Rays 1995; 20:156–164.

299

313. Graham RA, Atkins MB, Karp DD, et al. Local excision of rectal carcinoma. Early results with combined chemoradiation therapy using 5-fluorouraciI and leucovorin. Dis Colon Rectum 1994; 37:308–312. 314. Russell AH, Harris J, Rosenberg PJ, et al. Anal sphincter conservation for patients with adenocarcinoma of the distal rectum: long-term results of radiation therapy oncology group protocol 89–02. Int J Radiat Oncol Biol Phys 2000; 46:313–322. 315. Schell SR, Zlotecki RA, Mendenhali WM, Marsh RW, Vauthey JN, Copeland EM III. Transanal excision of locally advanced rectal cancers downstaged using neoadjuvant chemoradiotherapy. J Am Coll Surg 2002; 194:584–590. 316. Ruo L, Guillem JG, Minsky BD, Quan SH, Paty PB, Cohen AM. Preoperative radiation with or without chemotherapy and full-thickness transanal excision for selected T2 and T3 distal rectal cancers. Int J Colorectal Dis 2002; 17:54–58. 317. Papillon J. Endocavitary irradiation of early rectal cancer for cure: a series of 123 cases. Proc R Soc Med 1973; 66:1179–1181. 318. Papillon J. Present status of radiation therapy in the conservative management of rectal cancer. Radiother Oncol 1990; 17:275–283. 319. Sischy B, Hinson EJ, Wilkinson DR. Definitive radiation therapy for selected cancer of the rectum. Br J Surg 1988; 75:901–903. 320. Hull TL, Lavery IC, Saxton JP. Endocavitary irradiation: an option in select patients with rectal cancer. Dis Colon Rectum 1994; 37:1266–1270. 321. Horiot JC, Gerard JP, Maingen P. Conservative and curative management of rectal adenocarcinomas by local radiotherapy alone. Eur J Cancer 1995; 31A:1340–1342. 322. Reed NP, Cataldo PA, Garb JL, et al. The influence of local tumor ulceration on the effectiveness of endocautery radiation for patients with early rectal carcinoma. Cancer 1995; 76:967–971. 323. Schild SE, Martenson JA, Gunderson LL. Endocavitary radiotherapy of rectal cancer. Int J Radiat Oncol Biol Phys 1996; 34:677–682. 324. Mahajan A, Shenouda G, Gordon PH, et al. Long source-skin distance rectal irradiation technique: a review of results. Radiother Oncol 1996; 40:63–67. 325. Winslow ER, Kodner IJ, Mutch MG, Birnbaum EB, Fleshman JW, Dietz DW. Outcome of salvage abdominoperincal resection after failed endocavitary radiation in patients with rectal cancer. Dis Colon Rectum 2004; 47:2039–2046. 326. Lambrianidas HL, Ghilchick MG. Cryosurgery in the treatment of rectal carcinoma. Postgrad Med J 1983; 59:244–245. 327. Mathus-Vliegan EMH. Laser ablation of early colorectal malignancy. Endoscopy 1993; 25:462–468. 328. Cummings BJ. Radiation therapy and rectal carcinoma. The Princess Margaret Hospital experience. Br J Surg 1985; 72(suppl):S64–S66. 329. Sischy B. The role of radiation therapy in the management of carcinoma of the rectum. Contemp Surg 1987; 30:13–26. 330. Beart RW Jr. Rectal and anal cancers. In: Steele G Jr, Osteen RT, eds. Colorectal Cancer. New York: Marcel Dekker, 1986:170. 331. McDermott FT, Hughes ESR, Pihl E, et al. Local recurrence after potentially curative resection for rectal cancer in a series of 1008 patients. Br J Surg 1985; 72:34–37. 332. Heimann TM, Szporn A, Bolnick K, et al. Local recurrence following surgical treatment of rectal cancer. Comparison of anterior and abdominoperineal resection. Dis Colon Rectum 1986; 29:862–864. 333. Kirwan WO, Drumm J, Hogan JM, et al. Determining safe margin of resection in low anterior resection for rectal cancer. Br J Surg 1988; 75:720. 334. Shirouzu K, Isomoto H, Kakegawa T. Distal spread of rectal cancer and optimal distal margin of resection for sphincter-preserving surgery. Cancer 1995; 76:388–392. 335. Vernava AM, Moran M, Rothenberger DA, et al. A prospective evaluation of distal margins in carcinoma of the rectum. Surg Gynecol Obstet 1992; 175:333–336. 336. Kwok SPY, Lau WY, Leung KL, et al. Prospective analysis of the distal margin of clearance in anterior resection for rectal carcinoma. Br J Surg 1996; 83:969–972. 337. Enker WE, Laffer UT, Block GE. Enhanced survival of patients with colon and rectal cancer is based upon wide anatomic resection. Ann Surg 1979; 190: 350–360. 338. Tonak J, Gall FP, Hermanek P, et al. Incidence of local recurrence after curative operations for cancer of the rectum. Aust N Z J Surg 1982; 52:23–27. 339. Gibbs P, Chao MW, Tjandra JJ. Optimizing the outcome for patients with rectal cancer. Dis Colon Rectum 2003; 46:389–402. 340. Moore HG, Riedel E, Minsky BD, et al. Adequacy of 1-cm distal margin after restorative rectal cancer resection with sharp mesorectal excision and preoperative combined-modality therapy. Ann Surg Oncol 2003; 10:80–85. 341. Nakagoe T, Yamaguchi E, Tanaka K, et al. Distal intramural spread is an independent prognostic factor for distant metastasis and poor outcome in patients with rectal cancer: a multivariate analysis. Ann Surg Oncol 2003; 10:163–170. 342. Quirke P, Durdey P, Dixon MF, et al. Local recurrence of rectal adenocarcinoma due to inadequate surgical resection: Histologic study of lateral tumour spread and surgical excision. Lancet 1986; 2:996–999. 343. Wibe A, Rendedal PR, Svensson E, et al. Prognostic significance of the circumferential resection margin following total mesorectal excision for rectal cancer. Br J Surg 2002; 89:327–334. 344. Gordon PH. Is total mesorectal excision really important? J Surg Oncol 2000; 74:177–180.

300

& PART I: COLORECTAL DISORDERS

345. Abel AL. The modern treatment of cancer of the rectum. Milwaukee Proc 1931:296–300. 346. Heald RJ, Moran G, Brown G, Daniels IR. Optimal total mesorectal excision for rectal cancer is by dissection in fromt of Denonvilliers fascia. Br J Surg 2004; 91:121–123. 347. Hermanek P, Hermanek P, Hohenberger W, Klimpfinger M, Kockerling F, Papadopoulos T. The pathological assessment of mesorectal excision: implications for further treatment and quality management. Int J Colorectal Dis 2003; 18:335–341. 348. Sterk P, Kasperk R, Opitz T, Schubert F, Klein P. Vascular organization in the mesorectum: angiography of rectal resection specimens. Int J Colorectal Dis 2000; 15:225–228. 349. Cawthorn SJ, Parums DV, Gibbs NM, et al. Extent of mesorectal spread and involvement of lateral resection margin as prognostic factors after surgery for rectal cancer. Lancet 1990; 35:1055–1059. 350. Scott N, Jackson P, Al-Jaberi T, et al. Total mesorectal excision and local recurrence: a study of tumour spread in the mesorectum distal to rectal cancer. Br J Surg 1995; 82:1031–1033. 351. Choi JS, Kim SJ. Kim YI, et al. Nodal metastasis in the distal mesorecturn: need for total mesorectal excision of rectal cancer. Yonsei Med J 1996; 37:243–250. 352. Reynolds JV, Joyce WP, Dolan J, et al. Pathological evidence in support of total mesorectal excision in the management of rectal cancer. Br J Surg 1996; 83: 112–115. 353. Prabhudesai A, Arif S, Finlayson CJ, Kumar D. Impact of microscopic extranodal tumor deposits on the outcome of patients with rectal cancer. Dis Colon Rectum 2003; 46:1531–1537. 354. McCall JL, Cox MR, Wattchaw DA. Analysis of local recurrence rates after surgery alone for rectal cancer. Int J Colorectal Dis 1995; 10:126–132. 355. Leo E, Belli F, Andreola S, et al. Total rectal resection, mesorectal excision, and coloendoanal anastomosis: a therapeutic option for the treatment of low rectal cancer. Ann Surg Oncol 1996; 3:336–343. 356. Heald RJ. Rectal cancer: The surgical options. Eur J Cancer 1995; 31A:1189–1192. 357. Bokey EL, Ojerskog B, Chapuis PH, et al. Local recurrence after curative resection of the rectum for cancer withour adjuvant therapy: role of total anatomical resection. Br J Surg 1999; 86:1164–1170. 358. Krook JE, Moertel CG, Gunderson LL, et al. Effective surgical adjuvant therapy for high risk rectal carcinoma. N Engl J Med 1991; 324:709–715. 359. Fisher B, Wolmark N, Rockette H, et al. Postoperative adjuvant chemotherapy or radiation therapy for rectal cancer: results from NSABP protocol R-01. J Natl Cancer Inst 1988; 80:21–29. 360. Wibe A. On behalf of the Norwegian Recta Cancer Group and the Norwegian Cancer Registry presented at the ASCRS Annual Meeting. Washington. DC. May 1–6, 1999. 361. Ridgway PF, Darzi AW. The role of total mesorectal excision in the managment of rectal cancer. Cancer Control 2003; 10:205–211. 362. Colquhoun P, Wexner SD, Cohen A. Adjuvant therapy is valuable in the treatment of rectal cancer despite total mesorectal excision. J Surg Qncol 2003; 83:133–139. 363. Nesbakken A, Nygaard K, Lunde OC. Outcome and late functional results after anastomotic leakage following mesorectal excision for rectal cancer. Br J Surg 2001; 88:400–404. 364. Nesbakken A, Nygaard K, Bull-Njaa T, Carlsen E, Eri LM. Bladder and sexual dysfunction after mesorectal excision for rectal cancer. Br J Surg 2000; 87: 206–210. 365. Kim NK, AahnTW, Park JK, et al. Assessment of sexual and voiding function after total mesorectal excision with pelvic autonomic nerve preservation in males with rectal cancer. Dis Colon Rectum 2002; 45:1178–1185. 366. Kneist W, Heintz A, Junginger T. Major urinary dysfunction after mesorectal excision for rectal carcinoma. Br J Surg 2005; 92:230–234. 367. Peeters KC, Tollenaar RA, Marijnen CA, et al. Risk factors for anastomotic failure after total mesorectal excision of rectal cancer. Br J Surg 2005; 92:211–216. 368. von Flu¨e MO, Degen LP, Beglinger C, et al. Ileocecal reservoir reconstruction with physiologic function after total mesorectal cancer excision. Ann Surg 1996; 224:204–212. 369. Lopez-Kostner F, Lavery IC, Hool GP, et al. Total mesorectal excision is not necessary for cancer of the upper rectum. Surgery 1998; 124:612–618. 370. Hojo K, Sawada T, Moriya Y. An analysis of survival and voiding, sexual function after wide ileopelvic lymphadenectomy in patients with carcinoma of the rectum compared with conventional lymphadenectomy. Dis Colon Rectum 1989; 32:128–133. 371. Hojo K, Vernava AM III, Sugihara K, et al. Preservation of urine voiding and sexual function after rectal cancer surgery. Dis Colon Rectum 1991; 34: 532–539. 372. Glass RE, Ritchie JK, Thompson HR, et al. The results of surgical treatment of cancer of the rectum by radical resection and extended abdominosacral lymphadenectomy. Br J Surg 1985; 72:599–601. 373. Moreira LF, Hizuta A, Iwagaki H, et al. Lateral lymph node dissection for rectal carcinoma below the peritoneal reflection. Br J Surg 1994; 81:293–296. 374. Moriya Y, Sugihara K, Akasu T, et al. Nerve-sparing surgery with lateral node dissection for advanced lower rectal cancer. Eur J Cancer 1995; 31A: 1229–1232.

375. Morita T, Murata A, Koyama M, Totsuka E, Sasaki M. Current status of autonomic nerve-preserving surgery for mid and lower rectal cancers: Japanese experience with lateral node dissection. Dis Colon Rectum 2003; 46(suppl): S78–S87. 376. Hocht S, Mann B, Germer CT, et al. Pelvic sidewall involvement in recurrent rectal cancer. Int J Colorectal Dis 2004; 19:108–113. 377. Fujita S, Yamamoto S, Akasu T, Moriya Y. Lateral pelvic lymph node dissection for advanced lower rectal cancer. Br J Surg 2003; 90:1580–1585. 378. Nagawa H, Muto T, Sunouchi K, et al. Randomized, controlled trial of lateral node dissection vs. nerve-preserving resection in patients with rectal cancer after preoperative radiotherapy. Dis Colon Rectum 2001; 44:1274–1280. 379. Koda K, Saito N, Oda K, Takiguchi N, Sarashina H, Miyazaki M. Evaluation of lateral lymph node dissection with preoperative chemo-radiotherapy for the treatment of advanced middle to lower rectal cancers. Int J Colorectal Dis 2004:188–194. 380. Hida J, Yasutomi M, Fujimoto K, Maruyama T, Okuno K, Shindo K. Does lateral lymph node dissection improve survival in rectal carcinoma? Examination of node metastases by the clearing method. J Am Coll Surg 1997; 184:475–480. 381. Ruo L, Paty PB, Minsky BD, Wong WD, Cohen AM, Guillem JG. Results after rectal cancer resection with in-continuity partial vaginectomy and total mesorectal excision. Ann Surg Oncol 2003; 10:664–668. 382. Sokmen S, Terzi C, Unek T, Alanyali H, Fuzun M. Multivisceral resections for primary advanced rectal cancer. Int J Colorectal Dis 1999; 14:282–285. 383. Orkin BA, Dozois RR, Beart RW, et al. Extended resection for locally advanced primary adenocarcinoma of the rectum. Dis Colon Rectum 1989; 32:286–292. 384. Fedorov VD, Odaryuk TS, Shelygin YA. Results of radical surgery for advanced rectal cancer. Dis Colon Rectum 1989; 32:567–571. 385. Sugarbaker PH. Partial sacrectomy for en bloc excision of rectal cancer with posterior fixation. Dis Colon Rectum 1982; 25:208–211. 386. Stelzer M. Summary statement 2003 SSAT-AGA-ASGE Workshop on Palliative Therapy of Rectal Cancer Gastrointestinal Surg 2004; 8:253–258. 387. Moran MR, Rothenberger DA, Lahr CJ, et al. Palliation for rectal cancer. Resection? Anastomosis? Arch Surg 1987; 122:640–643,. 388. Longo WE, Ballantyne GH, Bilchik HJ, et al. Advanced rectal cancer: what is the best palliation? Dis Colon Rectum 1988; 31:842–847. 389. Nash GM, Salte LB, Kemeny NE, et al. Radical resection of rectal cancer primary tumor provides effective local therapv m patients with stage IV disease. Ann Surg Oncol 2002; 9:954–960. 390. Johnson WR, McDermott FT, Pihl E, et al. Palliative operative management in rectal carcinoma. Dis Colon Rectum 1981; 24:606–609. 391. Gordon PH, Fry RD, Segall MM, et al. Advanced carcinoma of the rectum. Perspect Colon Rectal Surg 1994; 7:235–244. 392. Heah SM, Eu KW, Ho VH, Leong AF, Seow-Choen F. Hartmann’s procedure vs. abdominoperineal resection for palliation of advanced low rectal cancer. Dis Colon Rectum 1997; 40:1313–1317. 393. Mahteme H, Pa¨hlman L, Gimelius B, et al. Prognosis after surgery in patients with incurable rectal cancer: a population-based study. Br J Surg 1996; 83: 1116–1120. 394. Nakfoor BM, Willett CG, Shellito PC, Kaufman DS, Daly WJ. The impact of 5-fluorouracil and intraoperative electron beam rradiation therapy on the outcome of patients with locally advanced primary rectal and rectosigmoid cancer. Ann Surg 1998; 228:194–200. 395. ReMine SG, Dozois RR. Hartmann’s procedure: Its use with complicated carcinomas of sigmoid colon and rectum. Arch Surg 1981; 116:630–633. 396. Doci R, Audisio RA, Bozzetti F, et al. Actual role of Hartmann’s resection in elective surgical treatment for carcinoma of rectum and sigmoid colon. Surg Gynecol Obstet 1986; 163:49–53. 397. Tottrup A, Frost L. Pelvic sepsis after extended Hartmann’s procedure. Dis Colon Rectum 2005; 48:251–255. 398. TurnbuIl RB, Weakley FL Jr. Atlas of Intestinal Stomas. St. Louis: CVMosby, 1967:161–165. 399. Grinnell RS, Hiatt RB. Ligation of the inferior mesenteric artery at the aorta in resections for carcinoma of the sigmoid and rectum. Surg Gynecol Obstet 1952; 94:526–534. 400. Grinnell RS. Results of ligation of inferior mesenteric artery at the aorta in resections of carcinoma of the descending and sigmoid colon and rectum. Surg Gynecol Obstet 1965; 120:1031–1036. 401. Busuttil RW, Foglia RP, Longmire WP Jr. Treatment of carcinoma of the sigmoid colon and upper rectum. Arch Surg 1977; 112:920–923. 402. Surtees P, Ritchie JK, Phillips RKS. High versus low ligation of the inferior mesenteric artery in rectal cancer. Br J Surg 1990; 77:618–621. 403. Bland KI, Polk HC. Therapeutic measures applied for the curative and palliative control of colorectal carcinoma. Surg Annu 1983; 15:123–161. 404. Corder AP, Karanjia ND, Williams JD, et al. Flush aortic tie versus selective preservation of the ascending left colic artery in low anterior resection for rectal carcinoma. Br J Surg 1992; 79:650–652. 405. Hida J, Yasutomi M, Maruyama T, et al. Indication for using high ligation of the inferior mesenteric artery in rectal cancer surgery. Examination of nodal metastases by the clearing method. Dis Colon Rectum 1998; 41: 984–987.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

406. Slanetz CA Jr, Grimson R. Effect of high and intermediate ligation on survival and recurrence rates following curative resection of colorectal cancer. Dis Colon Rectum 1997; 40:1205–1218. 407. Rothenberger DA, Wong WD. Rectal cancer—Adequacy of surgical management. Surg Annu 1985; 17:309–336. 408. McLean CM, Duncan W. Rectal cancer: a review of randomized trials of adjuvant radiotherapy. Clin Oncol 1995; 7:349–358. 409. Kerman HD, Roberson SH, Bloom TS, et al. Rectal carcinoma. Long-term experience with moderately high doses preoperative radiation and low anterior resection. Cancer 1992; 69:2813–2819. 410. Minsky BD. Oxaliplatin-based combined-modality therapy for rectal cancer. Semin Oncol 2003; 30(4 suppl 15):26–33. 411. Gerard A, Buyse M, Nordlinger B, et al. Preoperative radiotherapy as adjuvant treatment of rectal cancer. Ann Surg 1988; 308:606–612. 412. Medical Research Council. The evaluation of low-dose preoperative x-ray therapy in the management of operable rectal cancer; Results of a randomly controlled trial. Br J Surg 1984; 71:21–25. 413. Cummings BJ, Rider WD, Harwood AR, et al. Radical external beam radiation therapy for adenocarcinoma of the rectum. Dis Colon Rectum 1983; 26:30–36. 414. Fortier GA, Constable WC. Preoperative radiation therapy for rectal cancer. Arch Surg 1986; 121:1380–1385. 415. James RD, Schofield PF. Resection of ‘‘inoperable’’ rectal cancer following radiotherapy. Br J Surg 1985; 72:279–281. 416. Papillon J. The future of external beam irradiation as initial treatment of rectal cancer. Br J Surg 1987; 74:449–454. 417. Mendenhall WM, Bland KI, Copeland EM, et al. Does preoperative radiation therapy enhance the probability of local control and survival in high-risk distal rectal cancer. Ann Surg 1992; 215:696–706. 418. Friedmann P, Garb JL, Pare WC, et al. Survival following moderate dose preoperative radiation therapy for carcinoma of the rectum. Cancer 1985; 55: 967–973. 419. Cummings BJ. A critical review of adjuvant preoperative radiation therapy for adenocarcinoma of the rectum. Br J Surg 1986; 73:332–338. 420. Duncan W. Adjuvant radiotherapy and rectal cancer: the MRC trials. Br J Surg 1985; 72(suppl):S59–S62. 421. Stockholm Rectal Cancer Study Group. Preoperative short-term radiation therapy in operable rectal carcinoma. A prospective randomized trial. Cancer 1990; 66:49–55. 422. Cedermark B, Johansson H, Rutqvist L, et al. The Stockholm I Trial of preoperative short term radiotherapy in operable rectal carcinoma. Cancer 1995; 75:2269–2275. 423. Pa¨hlman L, Glimelius B. Pre- or postoperative radiotherapy in rectal and rectosigmoid carcinoma. Report from a randomized multicenter trial. Ann Surg 1990; 211:187–195. 424. Pa¨hlman L, Glimelius B. The value of adjuvant radio(chemo)-therapy for rectal cancer. Eur J Cancer 1995; 31A:1347–1350. 425. Pa¨hlman L. Initial report from a Swedish multicentre study examining the role of preoperative irradiation in the treatment of patients with resectable rectal carcinoma. Br J Surg 1993; 80:1333–1336. 426. Swedish Rectal Cancer Trial. Improved survival with preoperative radiotherapy in resectable rectal cancer. N Engl J Med 1997; 336:980–987. 427. Cedermark B, SRCS Group. Randomized study on preoperative radiation therapy in rectal carcinoma. Ann Surg Oncol 1996; 3:423–430. 428. Holm T, Rutqvist LE, Johansson H, et al. Abdominoperineal resection and anterior resection in the treatment of rectal cancer: results in relation to adjuvant preoperative radiotherapy. Br J Surg 1995; 182:1213–1216. 429. Holm T, Rutqvist LE, Johansson H, et al. Postoperative mortality in rectal cancer treated with or without preoperative radiotherapy: causes and risk factors. Br J Surg 1996; 83:964–968. 430. Holm T, Singnomklao T, Rutqvist LE, et al. Adjuvant preoperative radiotherapy in patients with rectal carcinoma. Adverse effects during long-term follow up of two randomized trials. Cancer 1996; 78:968–976. 431. Cohen AM, Minsky BD, Schilsky RL. Cancer of the rectum. In: deVita VT Jr, Hellman S, Rosenberg SA, eds. Cancer Principles and Practice of Oncology. Philadelphia: Lippincott-Raven, 1997:1197–1251. 432. Dahlberg M, Glimelius B, Graf W, Pahlman L. Preoperative irradiation affects functional results after surgery for rectal cancer: results from a randomized study. Dis Colon Rectum 1998; 41:543–549. 433. Goldberg P, Nicholls RJ, Porter N, et al. Long-term results of a randomised trial of short-course low-dose adjuvant preoperative radiotherapy for rectal cancer: reduction in local treatment failure. Eur J Cancer 1994; 30A:1602–1606. 434. Bannon JB, Marks GJ, Modiuddin M, et al. Radical and local excisional methods of sphincter-sparing surgery after high-dose radiation for cancer of the distal 3 cm of the rectum. Ann Surg Oncol 1995; 2:221–227. 435. Minsky BD, Cohen AM, Enker WE, et al. Sphincter preservation with preoperative radiation therapy and coloanal anastomosis. Int J Radiat Oncol Biol Phys 1995; 31:553–559. 436. Myerson RS, Michalski JM, King ML, et al. Adjuvant radiation therapy for rectal carcinoma: predictors of outcome. Int J Radiat Oncol Biol Phys 1995; 32: 41–50. 437. Izar F, Fourtanier G, Pradere B, et al. Preoperation radiotherapy as adjuvant treatment in rectal cancer. World J Surg 1992; 16:106–111.

301

438. Pacelli F, Di Giorgio A, Papa V, et al. Preoperative Radiotherapy Combined With Intraoperative Radiotherapy Improve Results of Total Mesorectal Excision in Patients with T3 Rectal Cancer. Dis Colon Rectum 2004; 47: 170–179. 439. Colorectal Cancer Collaborative Group. Adjuvant radiotherapy for rectal cancer: asystematic overview of 8507 patients from 22 randomised trials. Lancet 2001; 358:1291–1304. 440. van de Velde CJH. Preoperative radiotherapy and TME-surgery for rectal cancer. Detailed analyses in relation to quality control in a randomized trial. Proc Am Soc Clin Oncol 2002; 21:127a (abstract 506). 441. Marijnen CA, Kapiteijn E, van de Velde CJ, et al. Acute side effects and complications after short-term preoperative radiotherapy combined with total mesorectal excision in primary rectal cancer: report of a multicenter randomized trial. J Clin Oncol 2002; 20:817–825. 442. Hartley A, Giridharan S, Gray L, et al. Retrospective study of acute toxicity following short-course preoperative radiotherapy. Br J Surg 2002; 89:889–895. 443. Camma C, Giunta M, Fiorica F et al. Preoperative radiotherapy for resectable rectal cancer: a metaanalysis. JAMA 2000; 284:1008–1015. 444. Graf W, Dahlberg M, Osman MM, et al. Short-term preoperative radiotherapy results in down-staging of rectal cancer: a study of 1316 patients. Radiother Oncol 1997; 43:133–137. 445. Marijnen CA, Nagtegaal ID, Klein Kranenbarg E, et al. No downstaging after short-term preoperative radiotherapy in rectal cancer patients. J Clin Oncol 2001; 19:1976–1984. 446. Francois Y, Nemoz CJ, Baulieux J, et al. Influence of the interval between preoperative radiation therapy and surgery on downstaging and on the rate of sphincter-sparing surgery for rectal cancer: The Lyon R90–01 randomized trial. J Clin Oncol 1999; 17:2396–2402. 447. Guillem JG, Chessin DB, Shia J, et al. Clinical examination following preoperative chemoradiation for rectal cancer is not a reliable surrogate end point. J Clin Oncol 2005; 23:3475–3479. 448. Stein DE, Mahmoud NN, Anne PR, et al. Longer time interval between completion of neoadjuvant chemoradiation and surgical resection does not improve downstaging of rectal carcinoma. Dis Colon Rectum 2003; 46: 448–453. 449. Marijnen CA, van de Velde CJ, Putter H, et al. Impact of short-term preoperative radiotherapy on health-related quality of life and sexual functioning in primary rectal cancer: report of a multicenter randomized trial. J Clin Oncol 2005; 23:1847–1858. 450. Mella O, Dahl O, Horn A, et al. Radiotherapy and resection for apparently inoperable rectal adenocarcinoma. Dis Colon Rectum 1984; 27:663–668. 451. Friedmann P, Garb JL, McCabe DP, et al. Intestinal anastomosis after preoperative radiotherapy for carcinoma of the rectum. Surg Gynecol Obstet 1987; 164:257–260. 452. Roberson SH, Heron HC, Kerman HD, et al. Is anterior resection of the rectosigmoid safe after preoperative radiation? Dis Colon Rectum 1985; 28: 254–259. 453. Higgins GA, Humphrey EW, Dweight RW. Preoperative radiation and surgery for cancer of the rectum: Veterans’ Administration Surgical Oncology Group Trial II. Cancer 1986; 58:352–359. 454. Reis Neto JA, Guilici FA, Reis Neto JA Jr. A comparison of non-operative vs preoperative radiotherapy in rectal carcinoma. A 10-year randomized trial. Dis Colon Rectum 1989; 32:702–710. 455. Dahl O, Horn A, Morild I, et al. Low dose preoperative radiation postpones recurrences in operable rectal cancer. Results of a randomized multicenter trial in Western Norway. Cancer 1990; 66:2286–2294. 456. Marsh P, James R, Schofield P. Adjuvant preoperative radiotherapy for locally advanced rectal carcinoma. Dis Colon Rectum 1994; 37:1205–1214. 457. Gunderson LL, Sosin H. Areas of failure found at reoperation (second or symptomatic look) following ‘‘curative surgery’’ for adenocarcinoma of the rectum. Cancer 1974; 34:1278–1292. 458. Hoskins RB, Gunderson IX, Dosoretz DE, et al. Adjuvant postoperative radiation therapy in carcinoma of the rectum and rectosigmoid. Cancer 1985; 55:61–71. 459. GITSG. Prolongation of the disease-free interval in surgically treated rectal carcinoma. N Engl J Med 1985; 312:1465–1472. 460. Martijn H, de Neve W, Lybeert ML, et al. Adjuvant postoperative radiotherapy for adenocarcinoma of the rectum and rectosigmoid. A retrospective analysis of locoregional control, survival, and prognostic factors on 178 patients. Am J Clin Oncol 1995; 18:277–281. 461. De Neve W, Martijn H, Lybeert MM, et al. Incompletely resected rectum, rectosigmoid or sigmoid carcinoma: results of postoperative radiotherapy and prognostic factors. Int J Radiat Oncol Biol Phys 1991; 21:1297–1302. 462. Mak AC, Rich TA, Schultheiss JE, et al. Late complications of postoperative radiation therapy for cancer of the rectum and rectosigmoid. Int J Radiat Oncol Biol Phys 1994; 28:597–603. 463. Tang R, Wang J-Y, Chen JS, et al. Postoperative adjuvant radiotherapy in Astler-Coller Stage B2 and C rectal cancer. Dis Colon Rectum 1992; 35: 1057–1065. 464. Balsev I, Pedersen M, Teglbjaerg PS, et al. Postoperative radiotherapy in Dukes B and C carcinoma of the rectum and rectosigmoid. a randomized multicentre study. Cancer 1986; 58:22–28

302

& PART I: COLORECTAL DISORDERS

465. Treurniet-Donker AD, van Putten WL, Wereldsma JC, et al. Postoperative radiation therapy for rectal cancer: An interim analysis of a prospective randomized multicentre trial in The Netherlands. Cancer 1991; 67:2042–2048. 466. Gates G. On the behalf of the MRC Rectal Cancer Working Party. Results of the MRC trial of postoperative radiotherapy for operable rectal cancer. UKCCR Meeting on Colorectal Cancer Oxford, U.K.: March 20, 1995. 467. Sause WT, Pajak JF, Noyes RD, et al. Evaluation of preoperative radiation therapy in operable colorectal cancer. Ann Surg 1994; 220:668–675. 468. Bayer I, Turani H, Lurie H, et al. The Sandwich approach: irradiation-surgeryirradiation in rectal cancer. Four years experience. Dis Colon Rectum 1988; 28:222–224. 469. Botti C, Cosimelli M, Impiombata FA, et al. Improved local control and survival with the ‘‘sandwich’’ technique of pelvic radiotherapy for resectable rectal cancer. A retrospective multivariate analysis. Dis Colon Rectum 1994; 37(suppl 2):S6–S15. 470. Tepper JE, Cohen AM, Wood WC, et al. Intraoperative electron beam radiotherapy in treatment of unresectable rectal cancer. Arch Surg 1986; 121: 421–423. 471. Sischy B. Intraoperative electron beam radiation therapy with particular reference to treatment of rectal carcinoma—primary and recurrent. Dis Colon Rectum 1986; 29:714–718. 472. Valentini V, de Santis M, Morganti AG, et al. Intraoperative radiation therapy (IORT) in rectal cancer: Methodology and indications. Rays 1995; 20:73–89. 473. Gunderson LL, Nelson H, Martenson JA, et al. Intraoperative electron and external beam irradiation with or without 5-fluorouracil and maximum surgical resection for previously unirradiated, locally recurrent colorectal cancer. Dis Colon Rectum 1996; 39:1379–1395. 474. Sadahiro S, Suzuki T, Ishikawa K, et al. Intraoperative radiation therapy for curatively resected rectal cancer. Dis Colon Rectum 2001; 44:1689–1695. 475. Fuchs JR, Bleday R. Massive iliac artery bleeding in patients receiving intraoperative radiation therapy for advanced rectal cancer: report of two cases. Dis Colon Rectum 2004; 47:383–386. 476. Wolmark N, Rockette H, Fisher B, et al. The benefit of leukovorin-modulated fluorouracil as postoperative adjuvant therapy for primary colon cancer: results from National Surgical Adjuvant Breast and Bowel Project Protocol C-03. J Clin Oncol 1993; 11:1879–1887. 477. O’Connell MJ, Martensen JA, Wieand HS, et al. Improving adjuvant therapy for rectal cancer by combining protracted infusion fluorouracil with radiation therapy after curative surgery. N Engl J Med 1994; 331:502–507. 478. Windle R, Bell PRF, Shaw D. Five year results of a randomized trial of adjuvant 5-fluorouracil and levamisole in colorectal cancer. Br J Surg 1987; 74:569–572. 479. Takahashi T, Mizusawa H, Kato T, et al. Preoperative irradiation and 5fluorouracil suppository for carcinoma of the rectum. Am J Surg 1988; 156:58–62. 480. Sischy B, Graney MJ, Hinson J, et al. Preoperative radiation therapy with sensitizers in the management of carcinoma of the rectum. Dis Colon Rectum 1985; 28:56–57. 481. Wolmark N, Wieand HS, Hyams DM, et al. Randomized trial of postoperative adjuvant chemotherapy with or without radiotherapy for carcinoma of the rectum: National Surgical Adjuvant Breast and Bowel Project Protocol R-02. J Natl Cancer Inst 2000; 92:388–396. 482. Haller DG. Defining the optimal therapy for rectal cancer. JNCI 2000; 92: 361–362. 483. Theodoropoulos G, Wise WE, Padmanabhan A, et al. T-level downstaging and complete pathologic response after preoperative chemoradiation for advanced rectal cancer result in decreased recurrence and improved disease-free survival. Dis Colon Rectum 2002; 45:895–903. 484. Burmeister BH, Schache D, Burmeister EA, et al. Synchronous postoperative adjuvant chemoradiation therapy for locally advanced carcinoma of the rectum. Int J Colorectal Dis 2004; 19:55–59. 485. Tepper JE, O’Connell M, Niedzwiecki D, et al. Adjuvant therapy in rectal cancer:analysis of stage, sex and local control-final report of intergroup 0114. J Clin Oncol 2002; 20:1744–1750. 486. Boulis-Wassis, Ferard A, Loygue J, et al. Final results of a randomized trial on the treatment of rectal cancer with preoperative radiotherapy alone or in combination with 5-fluorouracil followed by radical surgery; Trial of the European Organization for Research and Treatment of Cancer. Gastrointestinal Tract Cancer Cooperative Group. Cancer 1984; 53:1811–1818. 487. Meade PG, Blatchford GJ, Thorson AG, et al. Preoperative chemoradiation downstages locally advanced ultrasound-staged rectal cancer. Am J Surg 1995; 170:609–613. 488. Chari RS, Tyler DS, Anscher MS, et al. Preoperative radiation and chemotherapy in the treatment of adenocarcinoma of the rectum. Ann Surg 1995; 221:778–787. 489. Stryker SJ, Kiel KD, Rodemaker A, et al. Preoperative ‘‘chemoradiation’’ for stages II and III rectal carcinoma. Arch Surg 1996; 231:514–519. 490. Kollmorgan CF, Meagher AP, Wolff BG, et al. The long-term effect of adjuvant postoperative chemoradiotherapy for rectal carcinoma on bowel function. Ann Surg 1994; 220:676–682. 491. Picciocchi A, Coco C, Magistrelli P, et al. Combined modality therapy of resectable high risk rectal cancer. Rays 1995; 20:182–189.

492. Enker WE. The elusive goal of preoperative staging in rectal cancer. Ann Surg Oncol 2004; 11:245–246. 493. Garcia-Aguilar J, Hernandez de Anda E, Sirivongs P, Lee SH, Madoff RD, Rothenberger DA. A pathologic complete response to preoperative chemoradiation is associated with lower local recurrence and improved survival in rectal cancer patients treated by mesorectal excision. Dis Colon Rectum 2003; 46:298–304. 494. Ruo L, Tickoo S, Klimstra DS, et al. Long-term prognostic significance of extent of rectal cancer response to preoperative radiation and chemotherapy. Ann Surg 2002; 236:75–81. 495. Stipa F, Zernecke A, Moore HG, et al. Residual mesorectal lymph node involvement following neoadjuvant combined-modality therapy: rationale for radical resection? Ann Surg Oncol 2004; 11:l87–191. 496. Rullier E, Goffre B, Bonnel C, Zerbib F, Caudry M, Saric J. Preoperative radiochemotherapy and sphincter-saving resection for T3 carcinomas of the lower third of the rectum. Ann Surg 2001; 234:633–640. 497. Read TE, Ogunbiyi OA, Fleshman JW, et al. Neoadjuvant external beam radiation and proctectomy for adenocarcinoma of the rectum. Dis Colon Rectum 2001; 44:1778–1790. 498. Volter V, Stupp R, Matter M, et al. Preoperative hyperfractionated accelerated radiotherapy (HART) and concomitant CPT-11 in rectal carcinoma. A phase I/II study. Proc Am Soc Clin Oncol 2001; 20:126b (abstr 2254). 499. Chau I, Chan S, Cunningham D. Overview of preoperative and postoperative therapy for colorectal cancer: the European and United States perspectives. Clin Colorectal Cancer 2003; 3:19–33. 500. Luna-Perez P, Rodriguez-Ramirez S, Hernandez-Pacheco F, Gutierrez De La Barrera M, Fernandez R, Labastida S. Anal sphincter preservation in locally advanced low rectal adenocarcinoma after preoperative chemoradiation therapy and coloanal anastomosis. J Surg Oncol 2003; 82:3–9. 501. Bedrosian I, Rodriguez-Bigas MA, Feig B, et al. Predicting the node-negative mesorectum after preoperative chemoradiation for locally advanced rectal carcinoma. J Gastrointest Surg 2004; 8:56–63. 502. Enker WE. The elusive goal of preoperative staging in rectal cancer. Ann Surg Oncol 2004; 11:245–246. 503. Onaitis MW, Noone RB, Fields R, et al. Complete response to neoadjuvant chemoradiation for rectal cancer does not influence survival. Ann Surg Oncol 2001; 8:801–806. 504. Hofheinz RD, von Gerstenberg-Helldorf B, Wenz F, et al. Phase I trial of capecitabine and weekly irinotecan in combination with radiotherapy for neoadjuvant therapy of rectal cancer. J Clin Oncol 2005; 23:1350–1357. 505. Chessin DB, Enker W, Cohen AM, et al. Complications after preoperative combined modality therapy and radical resection of locally advanced rectal cancer: a 14-year experience from a specialty service. J Am Coll Surg 2005; 200:876–882. 506. Habr-Gama A, Perez RQ, Nadalin W, et al. Long-term results of preoperative chemoradiation for distal rectal cancer correlation between final stage and survival. J Gastrointest Surg 2005; 9:90–99. 507. Moore HG, Gittleman AE, Minsky BD, et al. Rate of pathologic complete response with increased interval between preoperative combined modality therapy and rectal cancer resection. Dis Colon Rectum 2004; 47:279–286. 508. Sauer R. Adjuvant versus neoadjuvant combined modality treatment for locally advanced rectal cancer: first results of the German rectal cancer study (CAO/ARO/AIO-91). Int J Radiat Oncol Biol Phys 2003; 57(suppl 2):S124– S125. 509. Sauer R, Becker H, Hohenberger W, et al. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Eng J Med 2004; 351:1731–1740. 510. Ravasco R Monteiro-Grillo T. Vidal PM, Camilo ME. Dietary counseling improves patient outcomes: a prospective, randomized, controlled trial in colorectal cancer patients undergoing radiotherapy. J Clin Oncol 2005; 23: 1431–1438. 511. Gastrointestinal Tumor Study Group. Radiation therapy with fluorouracil with or without semustine for the treatment of patients with surgical adjuvant adenocarcinoma of the rectum. J Clin Oncol 1992; 10:549–557. 512. Tviet KM, Guldvog I, Hagen S, et al. Randomized controlled trial of postoperative radiotherapy and short-term time-scheduled 5-fluorouracil against surgery alone in the treatment of Dukes B and C rectal cancer. Norwegian Adjuvant Rectal Cancer Project Group. Br J Surg 1997; 84:1130–1135. 513. Holmes EC, Eilber FR, Morton DL. Immunotherapy of malignancy in humans. Current status. JAMA 1975; 232:1052–1055. 514. MacDonald JS. The immunobiology of colorectal cancer. Semin Oncol 1976; 2:421–431. 515. Martin DS. The necessity for combined modalities in cancer therapy. Hosp Pract 1973:129–136. 516. Reed RC, Gutterman JW, Mavligit GM, et al. Phase I trial of intravenous (IV) and subcutaneous (SC) Corynebacterium parvum (Cp). Am J Clin Oncol 1975; 16:228. 517. Mavligit GM, Gutterman JV, Burgess MA, et al. Adjuvant immunotherapy and chemo immunotherapy in colorectal cancer of the Dukes’ C classification: preliminary clinical results. Cancer 1975; 36(suppl):2421–2427. 518. Ooi B, Tjandra JJ, Green MD. Morbidities of adjuvant chemotherapy and radiotherapy for resectable rectal cancer. Dis Colon Rectum 1999; 42:403–418.

CHAPTER 4: MALIGNANT NEOPLASMS OF THE RECTUM &

519. Kent TS, Miller A, Bennett A, Weber T. Decision analysis for adjuvant therapy of rectal cancer. J Am Coll Surg 2003; 197:571. 520. Sagar PM, Pemberton JH. Surgical management of locally recurrent rectal cancer. Br J Surg 1996; 83:293–304. 521. Fick ThE, Baeten CGMI, von Meyenfeldt MF, et al. Recurrence and survival after abdominoperineal and low anterior resection for rectal cancer without adjuvant therapy. Eur J Surg Oncol 1990; 16:105–108. 522. Schiessel R, Wunderlich M, Herbst F. Local recurrence of colorectal cancer: effect of early detection and aggressive surgery. Br J Surg 1986; 73:342–344. 523. Castro-Sousa F, Leite JS, Alves FC, et al. Results of sphincter saving resection in rectal carcinoma: morbidity and local recurrence. Br J Surg 1995; 82(suppl 1):1. 524. Rullier E, Laurent C, Carles J, et al. Sphincter saving resection and local recurrence in the treatment of low rectal cancer stage B and C. Br J Surg 1995; 82(suppl 1):3–4. 525. Pilipshen S. Cancer of the rectum: local recurrence. In: Fazio VW, ed. Current Therapy in Colon and Rectal Surgery. Toronto: Brian C. Decker, 1990:137–149. 526. Heald RJ. Rectal cancer: anterior resection and local recurrence—a personal review. Perspect Colon Rectal Surg 1988; 1:1–26. 527. Pihl E, Hughes ESR, McDermott FT, et al. Recurrence of carcinoma of the colon and rectum at the anastomotic suture line. Surg Gynecol Obstet 1981; 153:495–496. 528. Feil W, Wunderlich M, Kovatz E, et al. Rectal cancer: factors influencing the development of local recurrence after radical anterior resection. Int J Colorectal Dis 1988; 3:195–200. 529. Twomey P, Burchell M, Strawn D, et al. Local control in rectal cancer. A clinical review and meta-analysis. Arch Surg 1989; 124:1174–1179. 530. Welch JP, Donaldson GA. Detection and treatment of recurrent cancer of the colon and rectum. Am J Surg 1978; 135:505–511. 531. Pilipshen SJ, Heilweil M, Quan SHQ, et al. Patterns of pelvic recurrence following definitive resection of rectal cancer. Cancer 1984; 53:1354–1362. 532. Luna-Perez P, Delgado S, Labastida S, et al. Patterns of recurrence following pelvic exenteration and external radiotherapy for locally advanced primary adenocarcinoma. Ann Surg Oncol 1996; 3:526–533. 533. Butch RJ, Wittenberg J, Mueller PR, et al. Presacral masses after abdominoperineal resection for colorectal carcinoma. The need for needle biopsy. AJR 1985; 144:309–312. 534. Ramirez JM, Mortensen NJ McC, Takeuchi N, et al. Endoluminal ultrasonography in the follow up of patients with rectal cancer. Br J Surg 1994; 81:692–694. 535. Lohnert MS, Doniec JM, Henne-Bruns D. Effective of endoluminal sonography in the identification of occult local rectal cancer recurrences. Dis Colon Rectum 2000; 43:483–491. 536. Cohen AM, Martin EW Jr, Lavery I. Radioimmunoguided surgery using iodine 125 B72.3 in patients with colorectal cancer. Arch Surg 1991; 126: 349–352. 537. Moore HG, Akhurst T, Larson SM, Minsky BD, Mazumdar M, Guillem JG. A case-controlled study of 18-fluorodeoxyglucose positron emission tomography in the detection of pelvic recurrence in previously irradiated rectal cancer patients. J Am Coll Surg 2003; 197:22–28. 538. Yamada K, Ishizawa T, Niwa K, Chumarn Y, Akiba T. Patterns of pelvic invasion are prognostic in the treatment of locally recurrent rectal cancer. Br J Surg 2001; 88:988–993. 539. Herefarth C, Schlag P, Hohenberger P. Surgical strategies in locoregional recurrences of gastrointestinal carcinoma. World J Surg 1987; 11:504–510. 540. Segall MM, Goldberg SM, Nivatvongs S, et al. Abdominoperineal resection for recurrent cancer following low anterior resection. Dis Colon Rectum 1981; 24:80–84. 541. Suzuki K, Dozois RR, Devine RM, et al. Curative reoperations for locally recurrent rectal cancer. Dis Colon Rectum 1996; 39:730–736. 542. Wiggers T, de Vries MR, Veeze-Kuypers B. Surgery for local recurrence of rectal carcinoma. Dis Colon Rectum 1996; 39:323–328. 543. Saito N, Koda K, Takiguchi N, et al. Surgery for local pelvic recurrence after resection of rectal cancer. Int J Colorectal Dis 1998; 13:32–38. 544. Huguier M, Houry S, Barrier A. Local recurrence of cancer of the rectum. Am J Surg 2001; 182:437–439. 545. Boyle KM, Sagar PM, Chalmers AG, Sebag-Montefiore D, Cairns A, Eardley I. Surgery for locally recurrent rectal cancer. Dis Colon Rectum 2005; 48:929–937. 546. Vermaas M, Ferenschild FT, Nuyttens JJ, et al. Preoperative radiotherapy improves outcome in recurrent rectal cancer. Dis Colon Rectum 2005; 48: 918–928. 547. Garcia-Aguilar J, Cromwell JW, Marra C, Lee SH, Madoff RD, Rothenberger DA. Treatment of locally recurrent rectal cancer. Dis Colon Rectum 2001; 44:1743–1748. 548. Hahnloser D, Nelson H, Giinderson LL, et al. Curative potential of multimodality therapy for locally recurrent rectal cancer. Ann Surg 2003; 237:502–508. 549. Lopez-Kostner F, Fazio VW, Vignali A, Rybicki LA, Lavery IC. Locally recurrent rectal cancer: predictors and success of salvage surgery. Dis Colon Rectum 2001; 44:173–178. 550. Mannaerts GH, Rutten HJ, Martijn H, Hanssens PE, Wiggers T. Comparison of intraoperative radiation therapy-containing multimodality treatment with historical treatment modalities for locally recurrent rectal cancer. Dis Colon Rectum 2001; 44:1749–1758.

303

551. Lopez MJ, Kraybill WG, Downey RS, et al. Exenterative Surgery for locally advanced rectosigmoid cancers. Is it worthwhile? Surgery 1987; 102:644–651. 552. Falk RE, Moffat FL, Makowka L, et al. Pelvic exenteration for advanced primary and recurrent adenocarcinoma. Can J Surg 1985; 28:539–541. 553. Temple WJ, Ketcham AS. Sacral resection for control of pelvic tumors. Am J Surg 1992; 163:370–374. 554. Pearlman NW, Steigmann GV, Donohue RE. Extended resection of fixed rectal cancer. Cancer 1989; 63:2438–2441. 555. Wanebo HJ, Koness RJ, Vezeridis MP, et al. Pelvic resection of recurrent rectal cancer. Ann Surg 1994; 220:586–597. 556. Temple WJ, Ketcham AS. Surgical palliation for recurrent rectal cancers ulcerating in the perineum. Cancer 1990; 65:1111–1114. 557. Tepper JE, O’Connell M, Hollis D, Niedzwiecki D, Cooke E, Mayer RJ. Intergroup Study 0114. Analysis of surgical salvage after failure of primary therapy in rectal results from Intergroup Study 0114. J Clin Oncol 2003; 21:3623–3628. 558. Jimenez RE, Shoup M, Cohen AM, Paty PB, Guillem J, Wong WD. Contemporary outcomes of total pelvic exenteration in the treatment of colorectal cancer. Dis Colon Rectum 2003; 46:1619–1625. 559. Ike H, Shimada H, Ohki S, Yamaguchi S, Ichikawa Y, Fujii S. Outcome of total pelvic exenteration for locally recurrent rectal cancer. Hepatogastroenterology 2003; 50:700–703. 560. Bakx R, van Lanschot JJ, Zoetmulder FA. Sacral resection in cancer surgery: surgical technique and experience in 26 procedures. J Am Coll Surg 2004; 198:846–851. 561. Mannaerts GH, Rutten HJ, Martijn H, Groen GJ, Hanssens PE, Wiggers T. Abdominosacral resection for primary irresectable and locally recurrent rectal cancer. Dis Colon Rectum 2001; 44:806–814. 562. Moriya Y, Akasu T, Fujita S, Yamamoto S. Total pelvic exenteration with distal sacretomy for fixed recurrent rectal cancer in the pelvis. Dis Colon Rectum 2004; 47:2047–2053. 563. Shirouzu K, Isomoto H, Kakegawa T. Total pelvic exenteration for locally advanced colorectal carcinoma. Br J Surg 1996; 83:32–35. 564. Yeung RS, Moffat FL, Falk RE. Pelvic exenteration for recurrent colorectal carcinoma: a review. Cancer Invest 1994; 12:176–188. 565. Rodriguez-Bigas MA, Petrelli NJ. Pelvic exenteration and its modifications. Am J Surg 1996; 171:293–301. 566. Kakuda JT, Lamont JP, Chu DZ, Paz IB. The role of pelvic exenteration in the management of recurrent rectal cancer. Am J Surg 2003; 186:660–664. 567. Miner TJ, Jaques DP, Paty PB, Guillem JG, Wong WD. Symptom control in patients with locally recurrent rectal cancer. Ann Surg Oncol 2003; 10:72–79. 568. Shoup M, Guiliem JG, Alektiar KM, et al. Predictors of survival in recurrent rectal cancer after resection and intraoperative radiotherapy. Dis Colon Rectum 2002; 45:585–592. 569. Kuehne J, Kleisli T, Biernacki P, et al. Use of high-dose-rate brachytherapy in the management of locally recurrent rectal cancer. Dis Colon Rectum 2003; 46:895–899. 570. Cheng C, Rodnguez-Bigas MA, Petrelli N. Is there a role for curative surgery for pelvic recurrence from rectal carcinoma in the presence of hydronephrosis? Am J Surg 2003; 182:274–277. 571. Kelly SR, Nugent KP. Formalin instillation for control of rectal hemorrhage in advanced pelvic malignancy: report of two cases. Dis Colon Rectum 2002; 45:121–122. 572. Magrini S, Nelson H, Gunderson LL, et al. Sacropelvic resection and intraoperative electron irradiation in the management of recurrent anorectal cancer. Dis Colon Rectum 1996; 39:1–9. 573. Rodriguez-Bigas MA, Herrera L, Petrelli NJ. Surgery for recurrent rectal adenocarcinoma in the presence of hydronephrosis. Am J Surg 1992; 164:18–21. 574. Loessin SJ, Moland NB, Devine RM, et al. Management of sacral and perineal defects following abdominoperineal resection and reduction with transpelvic muscle flaps. Dis Colon Rectum 1995; 38:940–945. 575. McAllister E, Wells K, Chaet M, et al. Perineal reconstruction after surgical extirpation of pelvic malignancies using the trans-pelvic transverse rectus abdominal myocutaneous flap. Ann Surg Oncol 1994; 1:164–168. 576. Yeh KA, Hoffman JP, Kusiak JE, et al. Reconstruction with myocutaneous flaps following resection of locally recurrent rectal cancer. Am Surg 1995; 61:581–589. 577. Jain AK, De Franzo AJ, Marks MW, et al. Reconstruction of pelvic exenterative wounds with transpelvic rectus abdominis flaps: a case series. Ann Plast Surg 1997; 38:115–123. 578. Chessin DB. Hartley J, Cohen AM, et al. Rectus flap reconstruction decreases perineal wound complications after pelvic chemoradiation and surgery: a cohort study. Ann Surg Oncol 2005; 12:104–110. 579. Butler CE, Rodriguez-Bjgas MA. Pelvic reconstruction after abdominoperineal resection: is it worthwhile? Ann Surg Oncol 2005; 12:91–94. 580. Christiansen J, Kirkegaard P. Treatment of recurrent rectal cancer by electroresection/coagulation after low anterior resection. Dis Colon Rectum 1983; 12:41–44. 581. Salvati EP, Rubin RJ. Electrocoagulation as primary therapy for rectal carcinoma. Am J Surg 1976; 132:583–586. 582. Kurz KR, Pitts WR, Speer D, et al. Palliation of carcinoma of the rectum and pararectum using the urologic resectoscope. Surg Gynecol Obstet 1988; 166:60–62.

304

& PART I: COLORECTAL DISORDERS

583. Fritsch A, Seidl W, Walzel C, et al. Palliative and adjunctive measures in rectal cancer. World J Surg 1982; 6:569–577. 584. Villalon AH, Green D. The use of radiotherapy for pelvic recurrence following abdominoperineal resection for carcinoma of the rectum. A ten year experience. Aust N Z J Surg 1981; 51:149–151. 585. Pacini P, Cionini L, Pirtoli L, et al. Symptomatic recurrences of carcinoma of the rectum and sigmoid. The influence of radiotherapy on the quality of life. Dis Colon Rectum 1986; 29:865–868. 586. Dobrowsky W, Schmid AP. Radiotherapy of presacral recurrence following radical surgery for rectal carcinoma. Dis Colon Rectum 1985; 28:917–919. 587. Patt YZ, Peters RE, Chuang VP, et al. Palliation of pelvic recurrence of colorectal cancer with intra-arterial 5-fluorouracil and mitomycin. Cancer 1985; 56:2175–2180. 588. Estes NG, Morphis JG, Hornback NB, et al. Intraarterial chemotherapy and hyperthermia for pain control in patients with recurrent rectal cancer. Am J Surg 1986; 152:597–601. 589. Machover D, Timus M, Schwartenberg L, et al. Treatment of advanced colorectal and gastric adenocarcinoma with 5-fluorouracil combined with high dose leucovorin. An update. In: Bruckner HW, Rustum YM, eds. Advances in Cancer Chemotherapy. The Current Status of 5-Fluorouracil, Leucovorin, Calcium Combination. New York: Park Row Publishers, 1984:55–64. 590. Nishimura Y, Hiraoka M, Akuta K, et al. Hyperthermia combined with radiation therapy for primarily unresectable and recurrent colorectal cancer. Int J Radiat Oncol Biol Phys 1992; 23:759–768. 591. Ichikawa D, Yamaguchi T, Yoshioka Y, et al. Prognostic evaluation of preoperative combined treatment for advanced cancer in the lower rectum with radiation, intraluminal hyperthermia, and 5-Fluorouracil suppository. Am J Surg 1996; 171:346–350. 592. Gonzales DG, van Dijk JDP, Blank LECM. Radiotherapy and hyperthermia. Eur J Cancer 1995; 31A:1351–1355. 593. Korenga D, Matsushima T, Adachi Y, et al. Preoperative hyperthermia combined with chemotherapy and radiotherapy for patients with rectal carcinoma may prevent early local pelvic recurrence. Int J Colorectal Dis 1992; 7:206–209. 594. Gagliardi G, Hawley PR, Hershman MJ, et al. Prognostic factors in surgery for local recurrence of rectal cancer. Br J Surg 1995; 82:1401–1405. 595. Williams LF, Huddleston CB, Sawyers JL, et al. Is total pelvic exenteration reasonable primary treatment for rectal carcinoma. Ann Surg 1988; 207:670–678. 596. Creutzfeldt W. Carcinoid tumors: Development of our knowledge. World J Surg 1996; 20:126–131. 597. Burke M, Shepherd N, Mann CV. Carcinoid tumors of the rectum and anus. Br J Surg 1987; 74:358–361. 598. Pronay G, Nagy G, Vjszaszy L, et al. Carcinoid tumors of the rectum. Ann Gastroenterol Hepatol 1982; 18:313–315. 599. Matsui K, Iwase T, Kitagawa M. Small, polypoid-appearing carcinoid tumors of the rectum: clinicopathologic study of 16 cases and effectiveness of endoscopic treatment. Am J Gastroenterol 1993; 88:1949–1953. 600. Jetmore AB, Ray JE, Gathright JB, et al. Rectal carcinoids: The most frequent carcinoid tumors. Dis Colon Rectum 1992; 35:717–725. 601. Fitzgerald SD, Meagher AP, Moniz-Pereira P, et al. Carcinoid tumor of the rectum. DNA ploidy is not a prognostic factor. Dis Colon Rectum 1996; 39:643–648. 602. Godwin JD II. Carcinoid tumors. An analysis of 2837 cases. Cancer 1975; 36:560–569. 603. Sauven P, Ridge JA, Quan SH, et al. Anorectal carcinoid tumors. Is aggressive surgery warranted? Ann Surg 1990; 21:67–71. 604. Yoshikane H, Tsukamoto Y, Niwa Y, et al. Carcinoid tumors of the gastrointestinal tract: evaluation with endoscopic ultrasonography. Gastrointest Endosc 1993; 39:375–383. 605. Kobayashi K, Katsumata T, Yoshizawa S, et al. Indications of endoscopic polypectomy for rectal carcinoid tumors and clinical usefulness of endoscopic ultrasonography. Dis Colon Rectum 2005; 48:285–291. 606. Shirouzu K, Isomoto H, Kakegawa T, et al. Treatment of rectal carcinoid tumors. Am J Surg 1990; 160:262–265. 607. Gerstle JT, Kaufman GL, Koltun WA. The incidence, management, and outcome of patients with gastrointestinal carcinoids and secondary primary malignancies. I Am Coll Surg 1995; 180:427–432. 608. Spread C, Berkel H, Jewell L, et al. Colonic carcinoid tumors. A populationbased study. Dis Colon Rectum 1994 37:482–491.

609. Nauneim KS, Zeitels ], Kaplan EL, et al. Rectal carcinoid tumors—Treatment and prognosis. Surgery 1983; 94:670–676. 610. Stinner B, Kisker O, Zielke A, et al. Surgical management for carcinoid tumors of small bowel, appendix, colon and rectum. World J Surg 1996; 20: 183–188. 611. Tsioulias G, Muto T, Kubota Y, et al. DNA ploidy pattern in rectal carcinoids. Dis Colon Rectum 1991; 34:31–36. 612. Bruneton JN, Thyss A, Bourry J, et al. Colonic and rectal lymphomas. A report of six cases and review of the literature. Fortschr Ro¨ntgenstr 1983; 138:283–287. 613. Lee MH, Waxman M, Gillooley JF. Primary malignant lymphoma of the anorectum in homosexual men. Dis Colon Rectum 1986; 29:413–416. 614. Onhi SK, Keane PF, Sackier JM, et al. Primary rectal lymphoma and malignant lymphomatous polyposis. Two cases illustrating current methods in diagnosis and management. Dis Colon Rectum 1989; 32:1071–1074. 615. Devine RM, Beart RW Jr, Wolff BG. Malignant lymphoma of the rectum. Dis Colon Rectum 1986; 29:821–824. 616. Jinnai D, Iwasa Z, Watanuko T. Malignant lymphoma of the large intestine— Operative results in Japan. Jpn J Surg 1983; 13:331–336. 617. Vieta JO, Delgado GE. Chronic ulcerative colitis complicated by—colonic lymphoma: report of a case. Dis Colon Rectum 1976; 19:56–62. 618. Perry PM, Cross RM, Morson BC. Primary malignant lymphoma of the rectum (22 cases). Proc R Soc Med 1972; 65:72. 619. Shepherd NA, Hall PA, Coates PJ, et al. Primary malignant lymphoma of the colon and rectum: a histopathological and immunohistochemical analysis of 45 cases with clinicopathologi-cal correlations. Histopathology 1988; 12:235–252. 620. Hague S, Dean PJ. Stromal neoplasms of the rectum and anal canal. Hum Pathol 1992; 23:762–767. 621. Khalifa AA, Bong WL, Rao VK, et al. Leiomyosarcoma of the rectum. Report of a case and review of the literature. Dis Colon Rectum 1986; 29:427–432. 622. Moore DO, Hilbun BM. Leiomyosarcoma of the rectum. A case report and review of the literature. Contemp Surg 1986; 29:132–137. 623. Feldtman RW, Oram-Smith JC, Teers RJ, et al. Leiomyosarcoma of the rectum: the military experience. Dis Colon Rectum 1981; 24:402–403. 624. Labow SB, Hoexter B. Leiomyosarcoma of the rectum: Radical vs conservative therapy and report of three cases. Dis Colon Rectum 1977; 20:603–605. 625. Akawri OE, Dozois RR, Weiland LH, et al. Leiomyosarcoma of the small and large bowel. Cancer 1978; 42:1375–1384. 626. Kessler KJ, Kerlakian GM, Welling RE. Perineal and perirectal sarcomas. Report of two cases. Dis Colon Rectum 1996; 39:468–472. 627. Randleman CD, Wolff BG, Dozois RR, et al. Leiomyosarcoma of the rectum and anus. Int J Colorectal Dis 1989; 4:91–96. 628. Yeh CY, Chen HH, Tang R, Tasi WS, Lin PY, Wang JY. Surgical outcome after curative resection of rectal leiomyosarcoma. Dis Colon Rectum 2000; 43: 1517–1521. 629. Changchien CR, Wu MC, Tasi WS, et al. Evaluation of prognosis for malignant rectal gastrointestinal stromal tumor by clinical parameters and immunohistochemical staining. Dis Colon Rectum 2004; 47:1922–1929. 630. Lo SS, Papachristou GL, Finkelstein SD, Conroy WP, Schraut WE, Ramanathan RK. Neoadjuvant imatinib in gastrointestinal stromal tumor of the rectum: report of a case. Dis Colon Rectum 2005; 48:1316–1319. 631. Lasser A. Adenocarcinoma of the prostate involving the rectum. Dis Colon Rectum 1978; 21:23–25. 632. Cervi G, Vettoretto N, Vinco A, et al. Rectal localization of metastatic lobular breast cancer: report of a case. Dis Colon Rectum 2001; 44:453–455. 633. Christodoulopoulos JB, Papaionnou AN, Drakopoulou EP, et al. Carcinoma of the cervix presenting with rectal symptomatology: Report of three cases. Dis Colon Rectum 1972; 15:373–376. 634. Rosito MA, Damin DC, Lazzaron AR, et al. Metastatic renal cell carcinoma involving the rectum. Int J Colorectal Dis 2002; 17:359–361. 635. Okano A, TakaKuwa H, Nishi A. Granular cell tumor of the rectum Gastrointest. Endosc 2001; 54:624. 636. Nakachi A, Miyazato H, Oshiro T, Shimoji H, Shiraishi M, Muto Y. Granular cell tumor of the rectum: a case report and review of the literature. J Gastroenterol 2000; 35:631–634. 637. Takao Y, Shimamoto C, Hazama K, et al. Primary rectal teratoma: EUS features and review of the literature. Gastrointest Endosc 2000; 51:353–355. 638. Hazzan D, Reissmann P, Halak M, Resnick MB, Lotem M, Shiloni E. Primary rectal malignant melanoma: report of two cases. Tech Coloproctol 2001; 5:51–54.

II: Anorectal Disorders

5

Perianal and Anal Canal Neoplasms Santhat Nivatvongs

& Introduction, 305

& INTRODUCTION

& Anatomic Landmarks, 305

Perianal and anal canal malignancies are uncommon. Although the literature is replete with reports on this subject, the use of terminology and classification has not been uniform; consequently, interpreting the results of treatment is difficult because the malignant neoplasms at different locations with different behaviors are grouped together.

& Incidence, 306 & Etiology and Pathogenesis, 306 & Staging, 307 & Screening for Anal Carcinoma Precursors, 307

Who Should Be Screened?, 307 Technique of Screening Cytology Test, 307 What to Do When the Biopsies Are Positive, 307 & Human Papilloma Virus Type 16 Vaccine, 307

& ANATOMIC LANDMARKS The anal area, although small, is rather complex due to differences in histologic features, characteristics, and lymphatic spread. Many reports of malignant neoplasms in this region use different terminologies to define the location of the malignancy. To overcome this confusion, the World Health Organization (WHO) and the American Joint Committee on Cancer (AJCC) have developed a universally accepted descriptive terminology for the histologic typing of intestinal neoplasms of the anal region (1,2). According to their terminology, ‘‘The anal canal is defined as the terminal part of the large intestine, beginning at the upper surface of the anorectal ring and passing through the pelvic floor at the anus. The lower part extends from the dentate line and downwards to the anal verge’’ (1). This is essentially the ‘‘surgical anal canal.’’ The perianal skin (the anal margin) is defined by the appearance of skin appendages (such as hairs). There exists no generally accepted definition of its outer limit (1). Some authors defined the lateral or distal extent of the perianal skin as 5 to 6 cm from the anal verge (3,4). This definition is in contrast to many series in the literature that use the dentate line as the dividing line describing the anal canal as the area above the dentate line, and the anal margin as the area below the dentate line (5–9). Numerous other reports never define the landmarks. The area above the dentate line up to the anorectal ring (the first 6–10 mm referred to as the transitional zone) has primarily cephalad lymphatic drainage via the superior rectal lymphatics to the inferior mesenteric nodes. It also has lesser drainage laterally along both the middle rectal vessels and inferior rectal vessels through the ischioanal fossa to the internal iliac nodes. Lymphatic drainage from the anal canal below the dentate line drains to the inguinal nodes. However, secondary drainage can follow the inferior rectal lymphatics to the ischioanal nodes and internal iliac nodes, and along the superior rectal nodes (see Fig. 20B in Chapter 1). Lymphatic drainage of the perianal skin is entirely to the inguinal nodes. The new edition of the WHO classification, which is similar to the American Joint Committee on Cancer (AJCC),

& Perianal Neoplasms (Anal Margin), 309

Anal Intraepithelial Neoplasia (AIN) of Perianal Skin (Bowen’s Disease), 309 Clinical Features, 309 Diagnosis, 309 Treatment, 309

Squamous Cell Carcinoma, 311 General Considerations, 311 Clinical Features, 311 Treatment, 312

Perianal Paget’s Disease, 312 General Considerations, 312 Clinical Features, 312 Diagnosis, 313 Treatment, 313 A Case of Circumferential Perianal Paget’s Disease, 313 Prognosis, 313

Basal Cell Carcinoma, 314 General Considerations, 314 Clinical Features, 315 Treatment and Prognosis, 315

Verrucous Carcinoma, 315 & Neoplasms of the Anal Canal, 315

Anal Canal Intraepithelial Neoplasia, 315 Squamous Cell Carcinoma, 316 Clinical Manifestations, 316 Diagnosis and Workup, 316 Character of Squamous Cell Carcinoma of Anal Canal, 316 Treatment, 317

Adenocarcinoma, 321 Rectal Type, 321 Anal Glands, 321 Anorectal Fistula, 322

Small Cell Carcinoma, 322 Undifferentiated Carcinoma, 322 Melanoma, 323 General Considerations, 323 Clinical Features, 323 Diagnosis, 323 Mode of Metastasis, 323 Treatment, 323 & References, 324

305

306

& PART II: ANORECTAL DISORDERS

has started to emerge but probably will take years before it is widely used and becomes standardized. Until then, a meaningful comparison of the anal carcinoma presented in different reports in the literature is difficult. WHO also recommends that the generic term ‘‘squamous carcinoma’’ be used for all subtypes of anal squamous cell (1).

& INCIDENCE In 2001, the new cases of anal carcinomas (anus, anal canal, and anorectum) in the United States was estimated at 3500 cases (2000 female, 1500 male). It accounted for 2.6% of carcinomas of the large bowel (10). In 2005, the new cases of anal carcinomas was estimated at 3990 cases (2240 female, 1750 male). It accounted for 2.7% of carcinomas of the large bowel (11). If it is accepted that the anal canal extends from the anorectal ring to the anal verge, as defined by WHO, 85% of anal carcinomas arise in the anal canal (12). The mean age of the patient at presentation varies from 58 to 67 years, and the age range is wide: 64 years of age and older, 58%; 45 to 64 years, 37%; and 25 to 44 years, 5%. Anal canal carcinomas show a marked female predominance, with the female-to-male ratio proximately 5:1. However, in areas with a large proportion of male patients at high-risk, the female-to-male ratio may approach 1:1. In contrast, perianal carcinomas are more common in men, with a male-tofemale ratio approximately 4:1 (12). In the United States, the incidence of squamous carcinoma of the anal canal and perianal skin in homosexual men has been estimated to be 11 to 34 times higher than the general male population. Human immunodeficiency virus (HIV)–infected homosexual men appear to be at particular risk. Other factors strongly associated with anal squamous carcinoma include the number of sexual partners, receptive anal intercourse, coexistence of sexually transmitted diseases, history of cervical, vulvar, or vaginal carcinoma, and use of immunosuppression after solidorgan transplantation (1,13).

& ETIOLOGY AND PATHOGENESIS There is strong evidence that human papilloma virus (HPV) infection causes anal carcinoma in a manner that closely parallels the role of HPV infection in the genesis of cervical carcinoma (14,15). Evidence supporting this observation includes the fact that many patients have simultaneous anal and genital viral infections and share common demographic characteristics, including an increased number of sexual partners. Furthermore, both anal and cervical carcinomas are associated with the specific ‘‘high-risk’’ HPV genotypes 16 and 18 (16–18). Over 60 different HPV genotypes have been identified, approximately 20 of which are known to infect the anogenital region. HPV types 6 and 11 are generally associated with benign lesions such as warts and a lowgrade anal intraepithelial neoplasia (AIN) that rarely progress to carcinoma. In contrast, HPV types 16, 18, 31,

33, 34, and 35 are most commonly associated with highgrade dysplastic AIN, carcinoma in situ, and carcinoma of anus and cervix. HPV-6 and -11 are maintained as extrachromosomal episomes, whereas HPV-16 and HPV-18 are integrated into host DNA, thus explaining the different propensity to initiate the development of carcinoma (16,19). The study by Palmer et al. (17), examining patients with invasive squamous-cell carcinoma of the anus, demonstrated that the majority of lesions in these patients contained HPV 16 and -18 DNA, which is confined to the nuclei of carcinoma cells and is predominantly integrated into the host cell DNA. Their study showed that none of the 56 control samples examined and none of the four nonsquamous cell primary anal malignancies contained any detectable HPV DNA. These observations add considerable weight to the concept of a specific association between HPV-16 and -18 and the development of anal squamous cell carcinoma. The observation that six of seven carcinomas of the upper anal canal contained HPV-16 or -18 DNA while only 8 of 18 carcinomas of the lower anal canal contained HPV-16 or -18 DNA is of interest because the epithelium of the transitional zone of the anal canal has both embryonic and histologic similarities with the transitional zone of the cervix. This study has also demonstrated a significant relationship between the absence of keratin and the presence of HPV DNA. The authors noted that all six carcinomas containing HPV-16 or -18 arising in the anal canal were nonkeratinizing. In contrast, only one of eight heavily keratinized lesions arising in the anal canal below the dentate line contained HPV-16 or -18 DNA. It is possible that these observations indicate a predilection of HPV-16 and -18 for the environment of the less stable epithelium of the upper anal canal rather than the modified skin of the lower anal canal. None of the cases of anal adenocarcinoma contains HPV-16 or -18 DNA (18,20). Immunocompromised patients such as those with renal transplantation, cardiac allograft recipients, and patients with carcinoma after chemotherapy have increased risk of HPV infection and increased progression to anal squamous cell carcinoma (21,22). They occur at a younger age, are multifocal, persistent, recurrent, and progress rapidly. Approximately 50% of patients positive for HIV have detectable HPV DNA (12). Penn (23) noted 65 anogenital (anal canal, perianal skin, or external genitalia) carcinomas in 2150 renal transplant recipients occurring an average of seven years after transplantation. Two-thirds of the patients were women and one-third men. These patients were much younger than those with similar malignancies in the general population, with the average age of 37 years for women and 45 years for men. Generally, the carcinomas are varieties of squamous cell carcinomas. Thirty two percent of the neoplasms are in situ lesions. Such carcinomas are biologically aggressive despite being histologically low grade. A study by Gervaz et al. (24) on molecular biology of squamous cell carcinoma of the anus between HIV-positive and HIV-negative patients revealed that allelic imbalance on chromosomes 17p, 18q, and 5q differ markedly. The data also demonstrated that DCC and p53 mutations were not required for anal squamous cell carcinoma progression in HIV-positive patients. These data suggest that immunosuppression may promote anal squamous cell carcinoma progression through an alternate

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

pathway and that persistence of HPV infection within the anal canal may play a central role in this process. In a study on anal squamous cell carcinoma, 47% of patients have a positive history for genital warts. In patients without a history of warts, the carcinoma is associated with a history of gonorrhea, herpes simplex type II virus, and Chlamydia trachomatis. Smoking is also a substantial risk factor (25). Anal intercourse itself does not carry an increased risk of anal squamous carcinomas. Rather, anal sex at young age carries an increased risk. However, most men and women with anal squamous carcinoma do not engage in anal sex. Thus, if HPV is truly a causative agent in anal squamous carcinoma, other modes of transmission to the anal area should be considered (18). Current evidence suggests that the etiology of anal carcinoma is a multifactorial interaction between environmental factors, HPV infection, immune status, and suppressive genes (12).

& STAGING The prognosis for survival in anal carcinoma deteriorates as the primary carcinoma enlarges. It worsens when the carcinoma metastasizes to the regional lymph nodes and to extrapelvic sites (26). Unlike carcinoma of the colon and rectum, the Dukes staging system is irrelevant because part of the lymphatic drainage is in the inguinal region and is outside the extent of the resection. The tumor, node, metastases (TNM) classification system has been proposed. It is important to note that the most recent edition of WHO standards, and the unified AJCC, introduced major changes in the staging of the primary carcinoma. T category is now determined by the largest diameter of the primary carcinoma measured in centimeters. Formerly, it was necessary to estimate clinically the circumferential extent of the anal carcinoma and whether the external sphincter was invaded. The current TNM system is explained in Box 1 (1). The best means of staging anal carcinoma remains a careful examination and, if necessary, under general anesthesia, supplemented by endorectal ultrasonography, computed tomography (CT), or magnetic resonance imaging (MRI) scanning. These procedures enable good biopsies to be taken and an appropriate decision to be made on the best mode of treatment. If the patient receives radiotherapy (RT) or chemoradiotherapy (CHT), further staging should be carried out eight weeks later to assess the results of treatment (27).

& SCREENING FOR ANAL CARCINOMA PRECURSORS For many years, the approach used to prevent cervical carcinoma has relied on the identification and treatment of cervical intraepithelial neoplasia before its progression to invasive carcinoma. Typically, women are screened at regular intervals using cervical cytology. Women with abnormal results are referred for colposcopy, to permit visualization of the lesion, and biopsy, to precisely ascertain the level of the disease. Based on the biologic similarity between anal and cervical carcinomas and their respective

307

precursors, similar methods might be used to identify the potential anal carcinoma precursors (28).

& WHO SHOULD BE SCREENED? Screening should be considered in the high risk group: (i) a history of homosexual activity in men and/or a history of receptive anal intercourse; (ii) all HIV-positive women, regardless of whether or not they have engaged in anal intercourse; (iii) all women with high-grade cervical or vulvar lesions or carcinoma (29). In homosexual and bisexual men, Goldie et al. (30) found that screening every two or three years for anal squamous intraepithelial lesions with anal cytology would provide life-expectancy benefits comparable to other accepted preventive health measures, and would be cost effective. & TECHNIQUE OF SCREENING CYTOLOGY TEST Because it is easy to perform and is relatively inexpensive, cytology remains the screening test of choice. A Cytette brush (Fig. 9 in Chapter 11) is moistened in saline solution or tap water , is rotated on the perianal skin for 10–20 revolutions with firm pressure abrading cells from the area. In the anal canal, very few AIN lesions occur solely in the transitional zone. The cytological preparation should be taken from the anal verge and lower anal canal only, to avoid fecal contamination in the smears. The brush is then smeared across a glass slide and the smear is fixed for standard fixative for Papanicoloau staining. When properly performed, the sensitivity and specificity is over 95% on the presence or absence of abnormal cells (31). Abnormal anal cytologic results should prompt anoscopic assessment, preferably with magnification and after application of 3% or 5% acetic acid. The identification of one or more well-demarcated lesions that appear white after application of 3% or 5% acetic acid and demonstrate features of vascular punctuation, leukoplakia, papillations, or other topographic irregularities should be biopsied, if (31,32) there is no contraindication, such as a bleeding disorder (29,30). & WHAT TO DO WHEN THE BIOPSIES ARE POSITIVE Only patients with high-grade AIN should be treated. This approach is based primarily on experience accumulated in natural history studies of cervical intraepithelial neoplasia that suggest that the majority of low-grade lesions regress spontaneously. Currently, it is most often treated with electrocautery or excisional biopsy. Low-grade AIN should be followed by a repeat screening cytology in three to six months (33). Screening HIV-positive homosexual and bisexual men for anal intraepithelial lesions and squamous cell carcinoma with anal Pap tests offers quality-adjusted life-expectancy benefits at a cost comparable with other accepted clinical preventive interventions (30).

& HUMAN PAPILLOMA VIRUS TYPE 16 VACCINE HPV-16 is most commonly linked with cancer, because it is present in 50% of cervical carcinomas and high-grade cervical intraepithelial neoplasias and in 25% of low-grade cervical intraepithelial neoplasias. A vaccine that prevents

308

& PART II: ANORECTAL DISORDERS

BOX 1 & TNM Cancer Staging System Anal Canal Primary Carcinoma (T) Tis Carcinoma in situ T0 No evidence of primary carcinoma T1 Carcinoma 2 cm or less in greatest dimension T2 Carcinoma more than 2 cm but not more than 5 cm in greatest dimension T3 Carcinoma more than 5 cm in greatest dimension T4 Carcinoma of any size invading adjacent organ(s) (e.g., vagina, urethra, bladder); involvement of the sphincter muscle(s) alone is not classified as T4 TX Primary carcinoma cannot be assessed Regional Lymph Node(s) (N) N0 No regional lymph node metastasis N1 Metastasis in perirectal lymph node(s) N2 Metastasis in unilateral internal iliac and/or inguinal lymph node(s) N3 Metastasis in perirectal and inguinal lymph nodes and/or bilateral internal iliac and/or inguinal lymph nodes NX Regional lymph nodes cannot be assessed Distant Metastasis (M) M0 No distant metastasis M1 Distant metastasis MX Presence of distant metastasis cannot be assessed Stage Grouping Stage 0 Tis Stage I T1 Stage II T2 T3 Stage IIIA T1-3 T4 Stage IIIB T4 Any T Any T Stage IV Any T

N0 N0 N0 N0 N1 N0 N1 N2 N3 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

Perianal Skin Primary Carcinoma (T) T0 No evidence of primary carcinoma Tis Carcinoma in situ T1 Carcinoma 2 cm or less in greatest dimension T2 Carcinoma more than 2 cm but not more than 5 cm in greatest dimension T3 Carcinoma more than 5 cm in greatest dimension T4 Carcinoma invades deep extradermal structure (e.g., cartilage, skeletal muscle, or bone) TX Primary carcinoma cannot be assessed Regional Lymph Node(s) (N) N0 No lymph node metastasis N1 Regional lymph node metastasis NX Regional lymph nodes cannot be assessed Distant Metastasis (M) M0 No distant metastasis M1 Distant metastasis MX Presence of distant metastasis cannot be assessed Stage Grouping Stage 0 Tis Stage I T1 Stage II T2–4 Stage III T4 Any T Any T

N0 N0 N0 N0 N1 Any N

M0 M0 M0 M0 M0 M1

Histopathologic Grade (G) G1 Well differentiated G2 Moderately differentiated G3 Poorly differentiated G4 Undifferentiated GX Grade cannot be assessed

Histopathologic Grade (G) G1 Well differentiated G2 Moderately differentiated G3 Poorly differentiated G4 Undifferentiated GX Grade cannot be assessed Source: Modified from Ref. 2.

persistent HPV-16 infection could substantially reduce the incidence of cervical carcinoma (34). In a double-blind study conducted by Koutsky et al. (34), 2392 young women were randomly assigned to receive three doses of placebo or HPV-16 virus-like particle vaccine given at day 0, month 2, and month 6. Genital samples to test for HPV-16 DNA were obtained at enrollment, one month after the third vaccination, and every six months thereafter. Women were referred for colposcopy according to a protocol. Biopsy tissue was evaluated for cervical intraepithelial neoplasia and analyzed for HPV-16 DNA. The primary end point was persistent HPV-16 infection.

The primary analysis was limited to women who were negative for HPV-16 DNA and HPV-16 antibodies at enrollment and HPV-16 DNA at month 7. The women were followed for a median of 17.4 months after completing the vaccination regimen. The incidence of persistent HPV-16 infection was 3.8 per 100 women-years in the placebo group and 0 per 100 (p < 0.001) in the vaccine group. All nine cases of HPV-16-related cervical intraepithelial neoplasia occurred among the placebo recipients. The authors concluded that immunizing HPV-16–negative women may reduce their risk of cervical carcinoma but caution that a larger study is required to prove that clinical

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

309

disease is prevented by vaccination. Its efficacy should also be similar for people with anal HPV-16 and AIN. If it becomes available, a vaccine to control HPV-16 infection will require universal immunization as opposed to the targeting of ‘‘high-risk’’ persons. Because these are sexually acquired pathogens, immunizing persons before they become sexually experienced will afford the greatest benefit (35).

& PERIANAL NEOPLASMS (ANAL MARGIN) & ANAL INTRAEPITHELIAL NEOPLASIA (AIN) OF PERIANAL SKIN (BOWEN’S DISEASE) High-grade AIN of perianal skin is synonymous with the old term perianal Bowen’s disease because of their indistinguishable histologic and immunohistochemical features (32). Bowen (36) described an intraepidermal squamous cell carcinoma (carcinoma in situ) in 1912 as a chronic atypical epithelial proliferation. The term anal canal intraepithelial neoplasia was proposed by Fenger and Nielsen in 1986 (37). Fenger, in 1990 (38) used the term perianal skin intraepithelial neoplasia in lieu of perianal Bowen’s disease. Most authors at the present time regard only high-grade AIN as Bowen’s disease, which is mostly caused by HPV-16 and -18, in contrast to lowgrade AIN which is mostly caused by HPV-6 and -11 (32,39,40). Scholefield et al. (41) found that high-grade AIN had a relatively low potential for malignant transformation in immunocompetent patients. The natural history of perianal AIN is unknown (42). Most data came from the study of HPV infection. Results from cross-sectional analyses before the era of highly active antiretroviral therapy (HAART) show that nearly all HIV-positive men as well as a substantial proportion of HIV-negative homosexual men harbor the infection (29). Whether perianal Bowen’s disease has higher incidence of malignancy in other organs, particularly internal organs, than the normal population is not clear. Although the studies by Arbesman and Ransohoff (43) and Chute et al. (44) found no increased subsequent risk of internal malignancy, these studies were on Bowen’s disease of the skin in general and not perianal Bowen’s disease. A survey by Marfing et al. (45) among members of the American Society of Colon and Rectal Surgery in 1987 yielded 106 cases of perianal Bowen’s disease. There were two cases of carcinoma of the colon and one case of carcinoma of the anus. Morganthaler et al. (46) reviewed 167 cases (age 34–56 yrs) of perianal Bowen’s disease in the literature; there were 31 (19%) associated carcinomas (not limited to internal organs) diagnosed concomitant or after treatment for the perianal Bowen’s disease, with a following of one to five years. Morganthaler et al. (46), in the series of 25 patients treated for perianal Bowen’s disease, found that five patients (20%) developed malignancy (one sigmoid carcinoma, four carcinoma of the vulva) during the average follow-up of four to seven years. It appears that perianal Bowen’s disease has significant associated malignancies, in spite of a relatively young age. Clinical Features Grossly, high-grade perianal AIN or Bowen’s disease appear as discrete, erythematous, occasionally pigmented,

FIGURE 1 & Bowen’s disease. Note atypical epithelial cells involving full thickness of the epidermis.

noninfiltrating, scaly, or crusted plaques, which sometimes have a moist surface. Foci of ulceration indicate that an invasive carcinoma has developed. Patients may complain of itching, which often is intense, burning, or spotted bleeding, but only a biopsy will confirm the diagnosis. In a series by Marchesa et al. (47), 25.5% of the cases of perianal Bowen’s disease were incidental findings, diagnosed after pathologic evaluation of tissue removed from different perianal diseases. The histologic picture is that of in situ squamous cell carcinoma that may have characteristic bowenoid cells, which are multinucleated giant cells with some vacuolization, giving a ‘‘halo’’ effect (Fig. 1).

Diagnosis Perianal high-grade AIN can involve perianal skin, anal verge, anoderm, and vulvar. A circumferential involvement of the disease is common (48). Although the gross lesions can be biopsied, their boundary and extent cannot be ascertained. This problem can be overcome by painting the anal canal, perianal region, and in women, including the vulva, with 3% or 5% acetic acid. The abnormal skin and mucosa will be demarcated by whitening of the tissue that should be biopsied. Chang and Welton (32) performed the examination in the operating room under sedation and perianal block. They examined the painted tissue through the operating microscope attached to a real-time video display. Changes in the vascular pattern suggestive of high-grade dysplasia are identified. The same tissue is then painted with Lugol’s solution (10% iodine). High-grade AIN does not take up the iodine solution because of the lack of glycogen in the dysplastic cells and they appear yellow or tan, whereas normal tissue or low-grade AIN appears dark brown or black. Lesions suspicious for high-grade AIN are biopsied. Treatment Treatment of Bowen’s disease or high-grade AIN has changed dramatically. Although the natural history of highgrade AIN is still not known, recent information has played an important role in the current management. High-grade AIN is preinvasive and requires further management. At the time of surgical treatment, 2–28% of patients already have an invasive squamous cell carcinoma (40,47,48). It is

310

& PART II: ANORECTAL DISORDERS

well known that the extent of the disease is usually beyond the gross demarcation of the lesion and can only be detected microscopically. Preoperative mappings were used in the past but were plagued with residual disease (39,47,48). Currently, the biopsies are performed with the aid of staining the perianal skin, perineum, and anal canal with 3% or 5% acetic acid (31,32). For low-grade AIN, treatment is not necessary if it is asymptomatic, but the patients should be followed periodically. There are several options on the treatment of highgrade AIN. Application of Imiquimod Imiquimod (Aldara, 3M Pharmaceuticals, St. Paul, Minnesota, U.S.A.) is an immune response modifier with potent antiviral and antitumor activity in animal models. It was approved by the FDA in 1997 for topical treatment of external genital and perianal warts in adults (49). The topical application has also been used for other skin conditions such as basal cell carcinoma, vulva intraepithelial neoplasms, invasive squamous cell carcinoma of skin, herpes simplex virus, and others, with excellent results (44–46). Pehoushek and Smith (50) reported treating an HIVpositive patient with squamous cell carcinoma in situ of perianal skin and the anal canal, with a combination of 5% Imiquimod cream three times a week combined with 5% fluorouracil daily with complete remission of the disease, and had no recurrence after a three-month follow-up. Imiquimod should be used as a preliminary treatment for perianal high-grade AIN (51,52). Imiquimod is formulated as 5% cream packaged as one box of 12 single-used sachets, each containing 250 mg of cream to cover an area of 20 cm2. A thin layer should be rubbed in until the cream is no longer visible. The application site should not be occluded. The recommended dosage is three times a week on nonconsecutive days (e.g., Monday, Wednesday, and Friday) at night for up to 16 weeks (49,53). Most patients tolerate the treatment well. Systemic reactions are rare but may cause fatigue, fever, influenzalike symptoms, headache, diarrhea, nausea, and myalgia. Local reactions are uncommon for applications of three times a week. It may cause itching, burning, and pain at the site of the application (49). In order to minimize local reactions and yet not compromising the therapeutic effects, Chen and Shumack (54) recommended application of Imiquimod three times weekly for three weeks, followed by a rest period of four weeks, to be repeated as necessary. The rest period allows any local skin reactions to subside. Local application of Imiquimod to a variety of skin diseases is safe and effective but it is new for perianal high-grade AIN. It remains to be seen whether it will become the treatment of choice; the prospect is promising. Topical 5-Fluorouracil Topical 5% 5-fluorouracil (5-FU) therapy has been found to be a safe and effective method to treat anal Bowen’s disease. Graham et al. (55) conducted a prospective study in 11 patients over a 6-year period. For one-half circumferential disease or greater, patients underwent topical 5% 5-FU therapy for 16 weeks. For smaller involvement, wide surgical excision was performed. All patients underwent

anal mapping biopsy one year after the completion of therapy. Of 11 patients, 8 (5 male) received 16 weeks of topical 5% 5-FU therapy. Three patients (3 female) underwent surgical excision for localized disease. All but one patient who was HIV-positive were free of Bowen’s disease one year after completion of therapy. One patient underwent total excision of a residual microinvasive squamous carcinoma after circumferential Bowen’s disease had resolved. One patient received 8 additional weeks of topical 5-FU therapy for incomplete resolution. All patients were followed yearly, with a mean follow-up of 39 months and the range of 12 to 74 months, and there have been no recurrences. There were no long-term side effects or morbidity from topical 5-FU. Cautery Ablation High-grade AIN is still benign and may take a long time to become malignant. Ablation, particularly in extensive disease, is attractive because it is less morbid than an extensive excision. One disadvantage of ablation is the lack of tissue diagnosis and an invasive carcinoma can be overlooked. Cautery ablation is the most convenient because it is available in every operating room compared to cryosurgery or laser vaporization. It should be performed with the aid of acetic acid painting to visualize the extent of the margins. The cauterization should not be too deep to cause a chronic unhealed wound. A circumferential involvement of the disease may require a staging ablation about three months apart to avoid anal stricture. Another drawback of cautery ablation is the findings that high-grade AIN has also the skin appendage involvement: 57% of hair follicles, 16% of sebaceous glands, and 25% of sweat glands (56). These structures can be missed by the cautery ablation. The most recent advancement in the treatment of high-grade AIN is the use of high-resolution anoscopy (32,57). In this way, change in the vascular pattern suggestive of high-grade dysplasia can be identified. The procedure is performed under 3% acetic acid and Lugol’s solution painting to delineate the site of the high-grade lesions. The suspicious lesions are biopsied and ablated with electrocautery. Using this technique, Chang and Welton (32) have treated over 400 patients. There were no recurrences among HIV-negative patients after 42 months but in the HIV-positive patient group, there was a projected 100% recurrence rate by the end of 60 months. The postoperative pain was significant in half of all patients. There was no sphincter dysfunction or anal stenosis. High-resolution anoscopy is new to colorectal surgeons and probably will become more popular in its use. Berry et al. (57) recommended, ‘‘ . . . . the technique is not particularly difficult to learn, but requires a level of clinical experience. It is important to perform high-resolution anoscopy on a large number of patients with the disease to learn the clinical pathologic correlations that signify high-grade squamous intraepithelial lesions, and potential areas of invasion. We recommend that interested providers first take an introductory colposcopy course. Providers should become familiar with the basics of colposcopy and be able to recognize and distinguish epithelial and vascular changes that are hallmarks of high-grade squamous intraepithelial lesions. To do high-resolution anoscopy well, individuals must be

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

thoroughly trained. It is unlikely that without some training in the basics of colposcopy, one can simply decide to use an operating microscope and recognize the changes that distinguish anal squamous intraepithelial lesions.’’ As good as it appears, high-resolution anoscopy is not available in most operating rooms and it is not cost effective to buy one to use in a practice that has low volume of this disease. Excision A survey of management of perianal Bowen’s disease among members of the American Society of Colon and Rectal Surgeons as recent as the year 2000 showed that 87% of the 663 respondents chose a wide local excision as the treatment of choice for a lesion larger than 3 cm (58). This approach has now been challenged because of high residual and recurrence rates, even when mapping biopsies have been used. Marchesa et al. (47) reported a recurrence rate of 34% in 41 patients who underwent local excision, with a median follow-up of 104 months. Samiento et al. (48), in a series of 19 patients, had a recurrence rate of 31% at five-year follow-up. Brown et al. (40) had histologic evidence of incomplete excision at the initial operation in 56% of 34 patients, and the recurrence rate of 40% with a median follow-up of 41 months. Morgenthaler et al. (46), in a series of studies, found that a clear margin was obtained in 23 of the 25 patients (92%), and the recurrence rate was 12% at three years’ follow-up; they gave credit for this relatively low residual and recurrence rate to aggressive mapping biopsies and wide margin of excision. Mapping with the application of 3% or 5% acetic acid followed by Lugol’s solution may help to determine the extent of involvement of the disease more accurately. Even with this, if the skin still harbors the HPV, a recurrence can occur later. For an extensive involvement of the perianal skin, a circumferential excision should include the anoderm up to the dentate line because of its frequent involvement of the disease especially in HIV-positive patients (51). In this situation, V–Y island subcutaneous skin flaps can be performed with good results and avoid the use of a splitthickness skin graft (59). Evidence now has shown that wide excision of highgrade AIN of perianal area does not preclude a high recurrence rate, even with negative margins. This is partly because it is difficult to accurately determine the margin of the disease, and because the remaining perianal skin may still harbor HPV, especially HPV-16 and -18. Wide excision of perianal skin and anal canal also has high complications, particularly anal stricture, ectropion, and fecal incontinence; some of these patients require a colostomy or a loop ileostomy (39,59). In a patient who is found to have a high-grade AIN in a routine hemorrhoidectomy specimen or other anorectal procedures, it is necessary to reexamine the area when the wound has had time to heal. Occasionally an invasive lesion is underreported and a persistent ulcerated area two months after the original procedure should lead to rebiopsy of the area. If the area has healed fully, a thorough inspection of the rest of the anogenital region should identify any remaining severely dysplastic lesions, and should be treated accordingly (60).

311

SUMMARY & The management of perianal high-grade AIN or perianal Bowen’s disease has changed from the standard, ‘‘surgical excision of high-grade AIN remains the treatment of choice’’ (56) to a more conservative, ‘‘close observation with regular biopsy of any suspicious areas to exclude invasive malignancy may be a better treatment option and is the policy currently advocated in this unit’’ (40). Unless there is an invasive carcinoma, it is reasonable to apply Imiquimod (51) or topical 5-FU (52) as an initial treatment. An alternative option may be cautery ablation. Surgical excision, especially an extensive one, should be reserved for patients with symptomatic disease such as untreatable itching, burning, or crusting of the skin (51). A long-term follow-up is essential because the skin may harbor the virus that perpetuates the disease. Subsequent development of malignancies in other organs should be kept in mind.

& SQUAMOUS CELL CARCINOMA General Considerations Squamous cell carcinomas of the perianal skin resemble those occurring in skin elsewhere in the body. Grossly, they typically have rolled, everted edges with central ulceration (Fig. 2). Any chronic unhealed ulcer should be considered a potential squamous cell carcinoma until proven otherwise by biopsy. Squamous cell carcinomas vary in size from as small as < 1 cm to large masses that completely surround and obstruct the anal orifice. The average age of the patient is between 62 and 70 years with a male-to-female ratio approximately equal (61,62). Clinical Features Despite their surface location, squamous cell carcinomas are usually diagnosed late; more than 50% of cases are detected more than 24 months after the onset of symptoms (63). The carcinoma is often discovered at a late stage

FIGURE 2 & Gross appearance of squamous cell carcinoma with rolled everted edges and central ulceration.

312

& PART II: ANORECTAL DISORDERS

FIGURE 3 & Histologic appearance of squamous cell carcinoma. Note the keratinization.

measuring 5 cm or larger in diameter (62). The usual presentations are a lump, bleeding, pain, discharge, and itching (61). On rare occasions, squamous cell carcinoma presents as a perianal abscess (64). Up to 28% of patients with perianal squamous cell carcinoma are misdiagnosed as having hemorrhoids, an anal fissure, an anal fistula, atopic eczema, an anorectal abscess, or a benign neoplasm (4). Histologically, these carcinomas are usually well differentiated, with well-developed patterns of keratinization (Fig. 3). Local invasion occurs, but the carcinoma is typically slow growing. Lymphatic spread from these carcinomas is directed mainly to the inguinal lymph nodes.

Treatment There are only a few articles in the literature devoted to perianal squamous cell carcinoma, and it is five times less common than squamous cell carcinoma of the anal canal (62). Only a handful of authors clearly define the perianal region according to the WHO and the AJCC standards. The treatment of perianal squamous cell carcinoma varies widely among different institutions. One common finding is clear. In advanced cases, local excision and abdominoperineal resection (APR) have high failure rates because of local recurrence, inguinal node metastasis, and distant metastasis (62,65,66). Wide local excision has remained the cornerstone of the treatment of perianal squamous cell carcinoma, but only in properly selected cases. For in situ or microinvasive carcinoma, local excision has a 100% cure rate (61,67). For superficial well-differentiated or moderately welldifferentiated squamous cell carcinoma up to 3 to 4 cm in diameter, Cummings (3) of Princess Margaret Hospital in Toronto has shifted away from radiation therapy toward local excision, supplemented with a skin graft when necessary, provided that the operation is not anticipated to interfere with anal sphincter function. Cummings reasons, ‘‘Although severe damage is infrequent following radiation, chronic irritation of the perianal skin and varying degrees of dysfunction of the anal region are common and can be troublesome’’ (3). For other less favorable lesions, he recommends chemoradiation. Most authors deliver 40 to 70 Gy (62,65,66,68). Chemoradiation has also been shown to be superior to radiation alone in nonrandomized studies (3). Residual or recurrent carcinoma after

radiation can be treated with local excision or an APR. Prophylactic radiation to the groin is also recommended, particularly for T2 and T3 lesions (62,66). In a series of 54 patients who received radiation with or without chemotherapy, Papillon and Chassard (62) achieved a cancer-specific five-year survival rate of 80%. The five- and ten-year cancer-specific survival rates of 86% and 77%, respectively, were reported by Touboul et al. (66) in a series of 17 patients. The size of the carcinoma determined the patient’s survival. In that study, 5- and 10-year survival rates for T1 lesions were 100% and 100%, respectively, compared to 60% and 40% for T2 lesions, respectively (4). With proper technique, serious complications from radiation therapy are uncommon. Radionecrosis and fecal incontinence have been claimed to occur only in a few patients (66).

& PERIANAL PAGET’S DISEASE General Considerations Perianal Paget’s disease is an intraepithelial neoplasm of the perianal skin. In 1874, Sir James Paget first described this disease in relation to the nipple of the female breast (69). George Thin, in 1881, was the first to describe the cytologic features of Paget’s cells, which appeared microscopically as large rounded cells with abundant palestaining cytoplasm and a large nucleus that is often displaced to the periphery of the cell (70). The first case of perianal Paget’s disease was reported by Darier and Couillaud in 1893 (71). Extramammary Paget’s disease may be found in the axilla and the anogenital region (labia majora, penis, scrotum, groin, pubic area, perineum, perianal region, thigh, and buttock). The histogenesis of perianal Paget’s disease is not fully understood, but ultrastructural and immunohistochemical studies have helped to clarify the debate. In contrast to Paget’s disease of the nipple, which is invariably associated with an underlying invasive or in situ ductal adenocarcinoma, perianal Paget’s disease starts out as a benign neoplasm. It may eventually become invasive and give rise to an adenocarcinoma. The immunohistochemical studies show that, in general, Paget’s cells stain positive for apocrine cells and in most cases stain negative for colorectal goblet cells (72). Unfortunately, many of the markers expressed by perianal Paget’s cells are also expressed by signet ring cell carcinoma of the anorectum (72). However, staining that is negative for all except a marker for anorectal goblet cells almost certainly indicates a spread of Paget’s cells from an anorectal mucinous adenocarcinoma. Most authors agree with the concept that Paget’s cells are of glandular and probably of apocrine origin (73–75). An alternative hypothesis implicating pluripotential intraepidermal cells cannot be excluded, but the evidence for such a histogenesis is lacking (73). Clinical Features Perianal Paget’s disease is an uncommon condition that is most commonly found in elderly people with an average age of 66 years. From 1963 to 1995, there have been 194 cases reported in the literature (76). The lesions appear as a slowly enlarging erythematous, eczematous, and often sharply demarcated perianal skin rash that may ooze or scale and

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

is usually accompanied by pruritus. It is normally located outside the anal canal but may extend up to the level of the dentate line. Because of its similarity to other perianal conditions such as idiopathic pruritus ani, hidradenitis suppurativa, condyloma acuminatum, Crohn’s disease, Bowen’s disease, and epidermoid carcinoma, the diagnosis of perianal Paget’s disease is often delayed because of clinical diagnostic error. In almost one-third of the cases in a series by Jensen et al. (77), the lesion involved the entire circumference of the anus.

Diagnosis The diagnosis must be confirmed by biopsy and by identification of the characteristic Paget’s cells through histologic examination (Fig. 4). Paget’s cells contain a mucoprotein (sialomucin) which stains with periodic acid-Schiff, and cytokeratin (CK) 7, CK20 immunohistochemical staining (78–80). True Paget’s disease of the perianal skin must not be confused with the downward intraepidermal spread of a signet-ring cell carcinoma of the rectum or with Bowen’s disease of the perianal skin. Immunohistochemistry for CK7 and CK20 can differentiate between the true perianal Paget’s disease and the downward spread of anorectal adenocarcinoma (pagetoid) (76–78). In general, H&E staining is sufficient to diagnose perianal Paget’s disease and distinguish it from the grossly similar appearance of perianal Bowen’s disease (79). A complete large bowel investigation with emphasis on thorough examination of the rectum and anal canal should be performed. The coexistence of visceral carcinomas is well known, with an incidence of about 50% (76). Treatment In the absence of invasive carcinoma, wide excision is the treatment of choice. A small lesion (< 25% of perianal area or anus) can be excised with the wound left open. Larger defects should be closed with a skin flap. Obtaining an adequate microscopically clear margin is important. Because perianal Paget’s disease may extend beyond the gross margin of the lesion, mapping the extent of involvement by obtaining multiple biopsies at least 1 cm from the edge of the lesion in all four quadrants, including the dentate line, the anal verge, and the perineum, is essential (76).

FIGURE 4 & Perianal Paget’s disease. Paget’s cells are found just above the basal layer.

313

A Case of Circumferential Perianal Paget’s Disease A 65-year-old woman with perianal Paget’s disease was prepared for a wide excision (Fig. 5A). The skin about 2 cm from the gross lesion was marked, and multiple small excisional biopsies showed no disease and thus became the lateral extent of the excision. The entire lesion was excised deep to the subcutaneous tissue. The proximal extent of the excision was at the dentate line (Fig. 5B). With the patient in prone position, a V sliding flap was created on each side of the wound. To accommodate the flap into the anal canal, a tongue of skin was excised at the base of each flap (shaded areas in Fig. 5C). The final flaps looked like an arrowhead (Fig. 5D). The flaps were slid into the anal canal and sutured to the mucosa of the anal canal all around, using interrupted 4–0 synthetic monofilament sutures (Fig. 5E). The skin was closed with interrupted subcuticular sutures of the same material in a Y fashion (Fig. 5F). No drain was placed. The same patient is shown 8 years after the operation (Fig. 5G). The patient should be prepared with the same bowel preparation and antibiotics administered before a colon resection. The results of a circumferential excision of the perianal skin and the anal canal, with bilateral V–Y island flaps are satisfactory. Of the 15 patients undergoing this technique (not all patients had perianal Paget’s disease) reported by Hassan et al. (59), none of them had flap loss or infection. Most complications were minor, including superficial wound separation, flap hematoma, and anal stricture. None of the patients had significant fecal incontinences at the time of follow-up, average 45 months. Perineal wound pain was a problem and five of the 15 patients required a diverting ileostomy or colostomy. Radiation therapy may be an option for perianal Paget’s disease. In an extensive review of literature by Brown et al. (81) they had difficulty to confirm or refute the benefit of this mode of treatment. This was because of the rarity of the disease; many were treated for recurrence after a surgical excision, or reserved for the medically unfit patients, lack of standardization, and techniques of radiation therapy. Nevertheless, in a limited cumulative series of nine patients of perianal Paget’s disease without invasive carcinoma treated with radiation or chemoradiation, two patients had recurrences at two years and six years. They believed that radiation therapy could be an alternative treatment for noninvasive Paget’s disease that has an extensive involvement to require an APR, and for patients who cannot tolerate a general anesthesia. A recurrence after local excision can also be successfully treated with radiation. Prognosis In a patient with noninvasive Paget’s disease, the lesion can be cured by wide local excision, although recurrence is high. The recurrence can be reexcised. Long-term followup is essential to detect local recurrence and development of invasive Paget’s disease or intercurrent invasive carcinoma of the rectum and anal canal (82–84). A patient with invasive perianal Paget’s disease has a poor prognosis despite APR, and in most cases, distant metastasis has already occurred at the time of diagnosis (56). Adjuvant chemoradiation does not seem to help (81).

314

& PART II: ANORECTAL DISORDERS

FIGURE 5 & (A) Circumferential perianal Paget’s disease. Note the sharply demarcated erythematous and eczematous skin rash. The dotted line marks the extent of the excision. (B) Excision deep to subcutaneous fat was made up to the dentate line. (C) V flaps were outlined. Shaded areas were tongues of skin to be excised to accommodate the flaps into the anal canal. (D) Sliding flaps were deep down to subcutaneous fat. Note the arrowhead shape of the flaps. (E) Note accommodation of the flaps into the anal canal. (F) The Y-shaped flaps at completion. (G) Ten months after operation. The patient experienced no fecal incontinence and there had been no recurrence after 8 years of follow-up.

& BASAL CELL CARCINOMA General Considerations Basal cell carcinomas of the perianal skin are rare. The Mayo Clinic has listed only 20 patients with this type of neoplasm in the 20 year period ending in 1996 (85). This type of carci-

noma occurs more frequently in men than in women, usually appearing in the sixth decade. The etiology is unknown but the most consistent factor was an association with basal cell carcinoma on other cutaneous sites (33% of patients) in the series reported by Paterson et al. (85).

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

Clinical Features Basal cell carcinomas are usually 1 to 2 cm in diameter and are localized to the perianal skin, although a large lesion may extend into the anal canal. Grossly, they are similar to cutaneous basal cell carcinomas found elsewhere in the body and are characterized by a central ulceration with irregular and raised edges. These carcinomas remain superficial and mobile and rarely metastasize. Histologically, they are similar to basal cell carcinomas of the skin elsewhere (Fig. 6). They are of long duration, have a low invasive potential, and must be distinguished from squamous cell carcinoma, which have an entirely different origin and behavior. In the study by Nielsen and Jensen (86), almost onethird of patients with basal cell carcinoma of the anal margin were misdiagnosed as having hemorrhoids, an anal fissure, or perianal eczema. The median delay in treatment caused by an erroneous diagnosis was eight months. Treatment and Prognosis Local excision with adequate margins is the treatment of choice for patients with basal cell carcinoma. Local recurrence after local excision is common and accounted for 29% of patients in a series of 27 patients reported on by Nielsen and Jensen (86); they recommend reexcision as treatment. Abdominoperineal resection and radiation therapy are reserved for large lesions. The five-year survival rate in the series by Nielsen and Jensen was 73%, but no patients died as the result of the basal cell carcinoma. In the series reported by Paterson et al. (85), no recurrence or death, occurred from perianal basal cell carcinoma after local excision; however, none of the patients had an invasive carcinoma. A novel approach is the use of topical Imiquimod 5% cream (Aldara1), an immune response modifier that induces cytokines related to cell-mediated immune response. In the multicenter phase II dose–response open-label trial (87) for small (<2 cm2) noninvasive basal cell carcinoma of the head, neck, trunk, or limbs, the cream was applied for six weeks, followed by an excision for histologic examination. The complete response was achieved in 100%, 88%, 73%, and 70% for the application of twice daily, daily, three times a week, and daily three times a week, respectively. All of the patients developed local skin reactions with erythema occurring most often and they were dose-related.

315

In the twice-daily group, all the local skin reactions were assessed as severe in two-thirds of the patients, some of whom had vesicles, ulcerations, and excoriations. However, only one patient of the total of 99 entered the study discontinued because of a medication-related skin reaction at the site of application. Whether this approach can be used for a large or invasive basal cell carcinoma is not known. It is possible to apply it to the basal cell carcinoma of the perianal skin.

& VERRUCOUS CARCINOMA Although there is no universal agreement, there is a growing consensus of opinion that the entity that has been termed a ‘‘giant condyloma acuminatum’’ or a ‘‘Buschke– Lo¨wenstein tumor’’ represents a verrucous carcinoma. Similar to anal warts, verrucous carcinoma is associated with HPV-6 and -11. These lesions typically present as large (8
8 cm), slow growing, painful, wart-like growths that are relatively soft and have a cauliflower-like appearance. The lesions may arise in the perianal skin, anal canal, or distal rectum and are frequently indistinguishable from condylomata acuminata. Although they are histologically benign, their behavior is clinically malignant. The clinical course of these lesions is one of relentless progression and expansion of the neoplasm by extensive erosion and pressure necrosis of surrounding tissues with invasion of the ischioanal fossa, perirectal tissues, and even the pelvic cavity. The invasive nature of these lesions may cause multiple sinuses or fistulous tracts, which may invade fascia, muscle, or the rectum and may cause inflammation, infection, and hemorrhage. The extent of involvement can be precisely determined by CT examination. Microscopically, the lesions bear a marked resemblance to condylomata acuminata with papillary proliferation, keratinization, acanthosis, parakeratosis, and vacuolization of superficial layers. Metastasis from these tumors has not been reported to date (88,89). The basic treatment is wide local excision. If the carcinoma involves the anal sphincters, an APR should be performed. At the present time, there have been no reports regarding the use of multimodality therapy in the treatment of verrucous carcinoma, but such consideration may obviate the need for radical extirpative surgery. Squamous cell carcinoma associated with condyloma acuminatum has been treated effectively with chemoradiation. The difference in the two entities may only be one of nomenclature (88,89). As an alternative treatment, particularly in unfit patients, Heinzerling et al. (90) reported a successful treatment with Imiquimod followed by CO2 laser ablation. In this case, 5% Imiquimod was applied to the lesion three times a week and from the second week on once daily. After six weeks of treatment, the persisting residual diseases were removed by CO2 laser under local anesthesia. To prevent recurrences, treatment with Imiquimod was continued for six weeks after the laser vaporization.

& NEOPLASMS OF THE ANAL CANAL FIGURE 6 & Basal cell carcinoma.

& ANAL CANAL INTRAEPITHELIAL NEOPLASIA AIN of the anal canal locates in the anal transitional zone as well as the anoderm below the dentate line. AIN of the anal

316

& PART II: ANORECTAL DISORDERS

canal has been found in minor surgical specimens such as hemorrhoidectomy, as well as in females with vulvar and cervical neoplasm. There is considerable evidence that AIN of the anal canal is the precursor of anal canal carcinoma, and possibly has a more aggressive behavior than perianal AIN (38). Similar to perianal AIN, the diagnosis should be confirmed by biopsy, using 3% or 5% acetic acid and Lugol’s solution staining as a guide (31,32). The treatment should be a cautery ablation or local excision. For an extensive involvement of the anoderm, it should be done in stages to avoid anal stricture. AIN of the anal canal found incidentally in hemorrhoidectomy requires no further treatment provided the rest of the anal canal has no high-grade AIN (91); a periodic follow-up with acetic acid staining is indicated. Using electrocautery ablation directed by high-resolution anoscopy, Chang et al. (92) reported 79% recurrence in HIVpositive patients, with a mean time recurrence of 12 months compared with zero recurrence in HIV-negative patients.

& SQUAMOUS CELL CARCINOMA This general category encompasses a number of different microscopic appearances, including large-cell keratinizing, large-cell nonkeratinizing (transitional), and basaloid (Fig. 7). The term ‘‘cloacogenic’’ carcinoma has been used especially for the basaloid and large-cell nonkeratinizing (transitional) forms of squamous carcinoma. Keratinizing squamous cell carcinomas are rare in the anal canal above the dentate line. Mucoepidermoid carcinoma is extremely rare, if it exists at all in this site. Previously reported examples probably represent squamous cell carcinoma with mucinous microcysts (Fig. 8). When carcinomas contain a mixture of cell types, they should be classified according to the appearance that predominates (1). Because of their similar response to treatment, it is reasonable to group them together as squamous cell carcinoma. To comply with the WHO and the AJCC terminology (1,2), we use the term squamous cell carcinoma instead of epidermoid carcinoma. Clinical Manifestations The presentation of squamous cell carcinomas generally follows a long history of minor perianal problems, such as bleeding, which occurs in approximately 50% of the

FIGURE 7 & Basaloid (cloacogenic) carcinoma. Note the nonkeratinization.

FIGURE 8 & Squamous cell carcinoma of anal canal with mucinous microcysts. This is not a mucoepidermoid carcinoma. Source: From Ref. 93.

patients (61,94,95). Other signs and symptoms include pruritus, discharge, pain, and an indurated anal mass. Discharge, incontinence, change in bowel habit, pelvic pain, and anovaginal fistula suggest advanced lesions with involvement of the anal sphincter. Almost one-third of the patients in the series study by Stearns and Quan (94) were initially incorrectly diagnosed as having benign or inflammatory disease.

Diagnosis and Workup The most important part of diagnosis is digital examination of the anal canal. The size, consistency, and fixation of the primary lesion and the presence or absence of pararectal lymph nodes can be determined. Proctoscopy should be done to confirm the digital findings, and the exact location of the neoplasm in relationship to the dentate line should be documented. Biopsy via proctoscope or transanal excision of the neoplasm must be done to determine the histologic type of the carcinoma, thus enabling an appropriate recommendation for treatment. A colonoscopic examination should be done to rule out more proximal associated lesions. Endorectal ultrasonography is useful in evaluating the depth of invasion and detection of lymph node metastasis. Both groins must be examined carefully to detect any enlargement of lymph nodes. Enlarged or suspicious lymph nodes in the groin area should be assessed by excision or biopsy because inguinal lymphadenopathy caused by reactive hyperplasia is common. Character of Squamous Cell Carcinoma of Anal Canal Even though squamous cell carcinoma of the anal canal is within the easy reach of a finger, the carcinoma is seldom diagnosed early. In a series from the Mayo Clinic, 88% of the carcinomas had already invaded beyond the mucosa before diagnosis (96). In approximately 50% of the cases, the carcinoma has penetrated the bowel wall or the perianal skin. Invasion of the vaginal septum is more common than involvement of the prostate or urethra. Extensive carcinoma may invade the muscular or bony walls of the pelvis (97). The anal canal has extensive lymphatic pathways. If the carcinoma is situated above the dentate line, metastases are found along the superior rectal vessels; if the carcinoma is at the dentate line, the lymphatic drainage is toward the

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

internal pudendal, hypogastric, and obturator nodes; and if the carcinoma is below the dentate line, the lymphatic drainage is via the inguinal nodes. A study of lymph node metastasis of carcinoma of the anal canal has been conducted by Wade et al. (98), who used a clearing technique. Their results showed that 44% of the lymph node metastases were found in lymph nodes measuring < 5 mm in diameter. Analysis of maps of lymph nodes allowed these authors to observe that most lymph nodes were located above the peritoneal reflection and were scant in perianal zones. Review articles by Cummings (97) revealed that inguinal metastases are found in 15% to 20% of patients at the time of diagnosis and become apparent later in an additional 10% to 25%. The risk of lymphatic metastases is correlated with the depth of the invasion, the size, and the histologic grade. Nodal metastases are present in 30% of the patients when the smooth muscle is infiltrated and in 58% when the infiltration is beyond the external sphincters. Boman et al. (96) found nodal metastases in only 3% of squamous cell carcinomas < 2 cm in diameter but in 25% to 35% of all the larger squamous cell carcinomas. The data from 1973 to 1998 in the Surveillance Epidemiology and End Results (SEER) cancer registry showed the prevalence by stage was localized in 53%, regional in 38%, and distant in 9% (13).

Treatment Local Excision This form of therapy should be reserved for early carcinomas or the well-differentiated type that have invaded only the submucosa. Review of literature reveals that the recurrence rate after local excision is 20% to 78% and the five-year survival rate is 45% to 85% (88,99). However, properly selected, local excision can be highly successful. Of 188 patients with squamous cell carcinoma of the anal canal treated at the Mayo Clinic, 13 patients with a superficially invasive carcinoma and 2 cm in size were selected for local excision. Although one required an APR for local recurrence, all were cured at a follow-up of five or more years (96). In the series from St. Mark’s Hospital, 8 of 145 patients with squamous cell carcinoma of the anal canal were treated by local excision. The cancer-specific five-year survival rate was 100% (8). In general, mobile lesions 2 cm are suitable for local excision. Abdominoperineal Resection In the past, APR with wide excision of perineal tissue formed the basis of treatment (3). Despite this aggressive approach, the results have been disappointing. The local recurrence rate is 27% to 50% and the five-year survival rate ranges from 24% to 62%, with a perioperative mortality rate of 2% to 6% (88). An APR is no longer the primary treatment for invasive squamous cell carcinoma of the anal canal. APR is reserved for those patients who cannot tolerate the chemoradiation and is the primary treatment for failed chemoradiation. To improve long-term survival and to save some patients from having a colostomy, use of APR as the primary treatment for squamous carcinoma has been challenged by the use of combined modality therapy.

317

Chemoradiation Regimen In 1974, Nigro et al. (100) initially attempted giving preoperative 5-FU with mitomycin C (MMC) and radiation in the management of squamous cell carcinoma of the anal canal to enhance the effectiveness of the operation. An APR was performed four to six weeks after the radiation therapy. Subsequently, they found that most of the surgical specimens contained no residual carcinoma. Nigro (9) then dropped the mandatory APR and instead excised the scar for histologic examination. Patients whose carcinoma grossly disappeared after the chemoradiation treatment required no further operation. Nigro’s subsequent regime of the chemoradiation is on the accompanying box. It should be noted that currently the dosage used for radiation varies among different practitioners. Review of the literature via MEDLINE search by Sato et al. (102) on the management of carcinoma of the anal canal showed that chemoradiotherapy with 5-FU and MMC was superior in local control, colostomy-free rate, progressive-free survival, and carcinomaspecific survival compared with radiation alone. Nigro’s Chemoradiation Therapy

External Irradiation & 3000 rads to the primary carcinoma and pelvic and inguinal nodes; start—day 1 (200 rads/day) (101)

&

Systemic Chemotherapy & 5-FU 1000 mg/m2/24 hr as a continuous infusion for four days; start—day 1 & Mitomycin C, 15 mg/m2 as an intravenous bolus; start— day 1 only & 5-FU, repeat four-day infusion; start—day 28

&

There have been questions regarding whether concomitant chemotherapy has the advantage over radiation alone, and whether MMC, which is toxic, is necessary. There were three well-conducted trials to test these questions. The Radiation Therapy Oncology Group and the Eastern Cooperative Oncology Group trial (103) studied the efficacy of MMC added to 5-FU and radiation. Of 291 assessable patients, 145 received 45 to 50.4 Gy pelvic radiation therapy plus 5-FU; 146 received radiation therapy, 5-FU, and MMC. Patients with residual disease on posttreatment biopsy were treated with a salvage regimen that consisted of additional pelvic radiation therapy (9 Gy), 5-FU, and Cisplatin. Posttreatment biopsies were positive in 15% of patients in the 5-FU arm versus 7.7% in MMC arm (p ¼ 0.135). At four years’ follow-up, colostomy rates were lower (9% vs. 22%, p ¼ 0.002), colostomy-free survival higher (71% vs. 59%, p ¼ 0.014) and disease-free survival higher (75% vs. 51%, p ¼ 0.0003) in MMC arm. However, a significant difference in overall survival was not observed at four years. Toxicity was greater in the MMC arm (23% vs. 7%, Grade 4 and 5 toxicity, p < 0.001). The authors concluded that despite greater toxicity, the use of MMC in a chemoradiation regime for squamous cell carcinoma of the anal canal was justified, particularly in patients with large primary carcinoma. Salvage chemoradiation should be attempted in patients with residual disease following definitive chemoradiation before resorting to radical surgery.

318

& PART II: ANORECTAL DISORDERS

The United Kingdom Coordinating Committee on Cancer Research (UKCCCR) trial (104) was designed to compare combined modality (5-FU þ MMC þ Radiation) with radiation therapy alone in patients with anal squamous cell carcinoma. In the multicenter trial, 585 patients were randomized to receive 45 Gy RT (290 patients) or radiotherapy (RT) combined with 5-FU þ MMC (295 patients). They assessed clinical response six weeks after initial treatment: good responders received a boost RT and the poor responders received a salvage surgery. The main end point was local failure rate (6 weeks after initial treatment); secondary end points were overall and causespecific survival. The results showed that after a median follow-up of 42 months (range 28–62 months) 59% radiation alone group had a local failure compared with 36% of combined modality group. This gave a 46% reduction in the risk of local failure in the patients receiving combined modality (relative risk 0.54, 95% CI 0.2–0.69, p < 0.0001). The risk of death from anal canal was also reduced in the combined modality therapy arm (0.71, 0.53–0.95, p ¼ 0.02). There was no overall survival advantage (0.86, 0.67–1.11, p ¼ 0.25). Early morbidity was significantly more frequent in the combined modality arm (p ¼ 0.03), but late morbidity occurred at similar rates. They concluded that the standard treatment for most patients with anal squamous cell carcinoma should be a combination of RT and infused 5-FU and MMC, with surgery reserved for those who failed this regimen. The European Organization for Research and Treatment of Cancer Radiotherapy and Gastrointestinal Cooperative Groups was a prospective randomized trial to validate the use of concomitant RT and chemotherapy in the treatment of squamous cell carcinoma of the anus (105). The study included 103 patients randomized between RT alone (n ¼ 51) and a combination of RT and chemotherapy (5-FU continuous infusion þ MMC, n ¼ 51). The patients had T3-T4 N0-N3 or T1-T2 N1-N3 squamous cell carcinoma of the anus. The radiation dosage was 45 Gy. The results showed that at five-year follow-up, the addition of chemotherapy to radiation therapy resulted in a significant increase in the complete remission rate from 54% for RT alone to 80% for chemoradiation therapy, and from 85% to 96%, respectively, if results are considered after surgical resections. Local recurrence improved by 18% while colostomy-free rate increased by 32% in the

chemoradiation group. Event-free survival, defined as free of local recurrence, no colostomy, and no severe side effects or death, showed significant improvement in favor of the chemoradiation group (p ¼ 0.03). The overall survival rates remained similar in both treatment arms. Summary of the three randomized trials is in Table 1. Currently, most investigators recommend combined modality treatment with continuous course radiation (45 Gy in 1.8 Gy fractions) plus two cycles of concurrent continuous infusion 5-FU (weeks 1 and 5) plus MMC (days 1 and 29). This regimen is considered by most to be the standard of care. For patients who have T3 to T4 disease (primary tumors, greater than 5 cm), it is reasonable to provide a radiation boost with an additional 5.4 to 9.0 Gy (106). Doses up to 66 to 70 Gy may be needed to treat patients with radiation alone (107). The average results of chemoradiation regimen include a complete response rate of 84% (81–87%), a local control rate of 73% (64–86%) and a five-year survival rate of 77% (66–92%). For patients who have T1 to T2 disease, the complete response rates are more than 90%. In patients who have T3 to T4 disease, approximately 50% will require a salvage APR. If they achieve a complete response following completion of combined modality therapy, only 25% will require a salvage APR (106). An example of chemoradiation treatment of squamous cell carcinoma of the anal canal is illustrated on the study by Myerson et al. (108). During 1975 to 1997, 106 patients with squamous cell carcinoma of the anal canal underwent radiation therapy. The dramatic response to low doses of radiation and concurrent chemotherapy led them to change in policy with most patients receiving chemoradiation, and surgery was reserved for salvage of persistent or recurrent disease. Since 1985, 81 patients have received definitive chemoradiation. Doses for definitive treatment without chemotherapy were 45 to 50 Gy to the carcinoma and the regional nodes, followed by a boost to a total dose of at least 65 Gy. Substantially lower radiation doses were used with concurrent chemotherapy. A dose of 30 Gy was used for T1/T2, followed by a boost of 10 to 20 Gy (mean 16.19 Gy) and 16 to 30 Gy (mean 23.67 Gy) for T3/T4 lesions. The results of 88 patients that excluded salvage surgery showed: T1-N0 (15 cases, 93  6%); T2-N0 (33 cases, 84  7%); T3-N0 (16 cases, 60  13%); T4-N0 (24 cases, 37  12%)(p ¼ 0.001) (Fig. 9).

TABLE 1 & Summary of Randomized Controlled Trials on Squamous Cell Carcinoma of the Anal Canal Reference (Yr) 103 (1996)

Study Group 5-FU þ radiation; n ¼ 145

105 (1997)

Results

Yes: disease-free survival; colostomy-free survival (favoring MMC) No: overall survival

51% vs. 73%, disease-free survival (favoring MMC); follow-up 4 yrs

Radiation alone n ¼ 285

Yes: local failure; disease-free survival (favoring chemoradiation)

46% reduction local failure; reduction death from cancer (favoring chemoradiation); follow-up 42 mos

Radiation þ 5-FU þ MMC; n ¼ 292

No: overall survival

Radiation alone n ¼ 52

Yes: local failure; cancer-free survival (favoring chemoradiation)

Radiation þ 5-FU þ MMC; n ¼ 51

No: overall survival

5-FU þ MMC þ radiation; n ¼ 146 104 (1996)

Statistically Significant

Abbreviations: 5-FU, 5-fluorouracil; MMC, mitomycin C.

80% vs. 54% complete remission; 18% improve in local recurrence; 32% increase in colostomy-free (favoring chemoradiation); follow-up 5 yrs

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

FIGURE 9 & Disease-free survival (Kaplan–Meier) vs. clinical stage. The contribution of surgery (planned or salvage) to the ultimate disease status is included in this figure. Source: From Ref. 109.

Of importance were the additional malignancies noted by Myerson et al. (108). There were a total of 33 additional malignancies in 26 patients. Nineteen of the lesions antedated the squamous carcinoma of the anal canal, and 14 postdated. Types of the malignancies are on Table 2. With exception of the two sarcomas within the radiation portals occurred 5 and 10 years after radiation without chemotherapy, the additional malignancies were not suggestive of the radiation-induced malignancy. This information underscores the importance of complete examination in general, in addition to a diligent long-term oncologic screening in patients with squamous cell carcinoma of the anal canal (108). Brachytherapy Brachytherapy is an ideal method to deliver conformal radiation for anal carcinoma while sparing the surrounding healthy structures, such as small intestine and bladder. The primary concern is anal necrosis, and anal necrosis rates reportedly vary from 2% to as high as 76% (110). Although there are no randomized data, the phase II Trials suggest that, even in experienced hands, brachytherapy is associated with higher complication rates than in external beam radiation (EBR) therapy (106). Kapp et al. (111) using brachytherapy to boost during a short split between the EBR  chemotherapy, gave satisfactory results. The rationale of this combination was to TABLE 2 & Additional Malignancies in Patients with Squamous Cell Carcinoma of the Anal Canal Type Gynecologic Head and neck Lung Colorectal Genitourinary Central nervous system Hematologic Sarcoma Breast TOTAL

No. of Lesions 9 6 5 3 3 2 2 2 1 33

319

improve tolerance of 50 Gy EBR in the pelvis, perineum, and groin. Intestinal or intraluminal brachytherapy was performed using a single high-intensity (192Ir). Dose calculation for needle implants was 6 Gy. EBR was delivered with one dose 1.8 or 2.0 Gy per day, five days a week. After 30 Gy, an interstitial or intraluminal 192Ir high dose-rate (HDR) boost was performed. Depending on the patient’s skin reaction, EBR was resumed within one to two weeks, with an additional 20 Gy. Patients who failed to achieve a complete response received additional brachytherapy. Follow-up ranged from 3 to 14 months (median 31 months). The median treatment radiation was 56 days. The five-year actuarial rates of locoregional control (LRC) and disease-specific survival (DSS) were 76% and 76%, respectively. The overall anal preservation was 77% and 97% in patients in whom LRC was achieved. Uncompromised anal function was recorded in 93% of these patients. Although acute toxicity occurred in 89% in radiochemotherapy group and in 64% of RT alone, predominantly the triad of pronounced skin reaction, diarrhea, and nausea, none of these patients required hospitalization. Late complications that required a temporary colostomy occurred in 7.6% of the patients, for pain relief because of ulceration of the lesions. The authors concluded that the integration of HDR brachytherapy boost in a split-course EBR regimen  chemotherapy resulted in excellent sphincter function without an increase of severe complications and with rates of locoregional control, disease-specific survival, and cancerfree survival compared favorably with those reported in the literature. Treatment in HIV-Positive Patients In general, HIV-positive patients have received lower doses of radiation and chemotherapy because of poorer tolerance to therapy and increased complication rate (106,112,113). In situ patients with CD-4 counts as low as 105 cells/ mL do well with local excision. A low CD-4 count at diagnosis without HAART predicts a poor prognosis because these patients appear to succumb to their HIV status and not the anal disease (112). Even in the era of HAART and with experienced clinicians, treatment of invasive anal squamous-cell carcinoma in HIV-infected patients remains a challenge but the techniques in management will undoubtedly improve in the near future. Currently, at the University of California, San Francisco (114), HIV-positive patients who had invasive anal squamous cell carcinoma with CD-4 counts greater than 200 cells/mm3 were treated with CMT in a similar way to patients that were not HIV infected. Patients with CD-4 counts less than 200 cells/mm3 were treated using more individualized approaches. In general, the dosage was kept tighter, and prophylactic radiation at nodal areas was not used. These patients had a higher rate of colostomies. Cisplatin was used instead of MMC in some cases. Blazy et al. (115) reported their experience of treating anal canal carcinoma in the era of new antiviral drugs, in patients with HIV-positive. Nine men on highly active antiretroviral therapies with good immune status before chemoradiotherapy received concomitant chemoradiotherapy consisting of 5-FU and Cisplatinum, and high-dose RT (60–70 Gy). Six carcinomas were Stage I, two were Stage II, and one was Stage III. CD-4þ cell counts were less than

320

& PART II: ANORECTAL DISORDERS

200/mL for four patients, between 200/mL and 500/mL for four, and more than 500/mL for one. All patients received the planned dose of radiation (60 Gy). The chemotherapy dose was reduced to 25% in six patients. The overall treatment time was 58 days. Grade III hematologic or skin toxicity occurred in four patients. No association was observed between high-grade toxicity and CD-4þ cell count. None of the patients developed opportunistic infections during follow-up. Eight patients were disease-free after a median follow-up of 33 months. Among them, four had no or minor anal function impairment at the last follow-up visit. One patient with T4-N2 disease relapsed locally one year after treatment and underwent salvage abdominoperineal excision. The authors concluded that high-dose chemoradiotherapy for anal canal carcinoma is feasible with low toxicity in HIV-positive patients, treated with highly active antiretroviral therapies. Toxicity of Chemoradiation Chemoradiation is not to be taken lightly. Reports of complications vary widely in the literature because they are dose and technique dependent. In general, severe complications are increased when the total dose is > 40 Gy. The common complications encountered are dermatitis and mucositis, diarrhea, fecal incontinence, fatigue, bone marrow depression, cystitis, small bowel obstruction, and major arterial stenosis. Death, although very rare, has been reported (116,117). Anorectal function was preserved in 88% of patients who underwent chemoradiation, as reported in the series of Cummings et al. (116); the dose was 45 to 55 Gy. Those with severe fecal incontinence required a colostomy. Less severe anorectal function, such as fecal urgency and occasional fecal incontinence, can be managed satisfactorily with antidiarrheal medications and the adjustment of diet. Complications of chemoradiation can be minimized by split-course radiation (116). Cisplatin has emerged as a potential replacement for mitomycin in the combination drug regimens. It is a radiation sensitizer and is less myelosuppressive than MMC. However, preliminary results showed the toxicity rate to be similar to MMC regimen (106,107). Pattern of Failure and Treatment The predominant sites of failure after chemoradiation are the pelvis, either the anal area or the regional lymph nodes. In a large series of 190 patients reported by Cummings et al. (116), 41% experienced recurrence at one or more sites. Of those recurrences, 62% were confined to the pelvis, 16% were outside the pelvis, and the rest occurred both inside and outside the pelvis. Those patients with residual or recurrent carcinoma confined to the pelvis or perianal area should undergo a salvage APR with or without a booster dose of radiation. The outcome is significantly related to the extent of the disease at the time of failure. The series from Memorial Sloan-Kettering Cancer Center showed that T stage did not appear to affect survival after APR (p ¼ 0.07) (118). On the other hand, Nguyen et al. (119) revealed size to be the only significant factor associated with the need for a stoma (p ¼ 0.01), and that node positivity was the only independent predictor of mortality (p ¼ 0.02). Inguinal node

metastasis at initial presentation, before the chemoradiation, predicted poor outcome after APR for treatment failure. Patients with disease fixed to the pelvic side wall on digital examination at the time of treatment failure fared poorly, with an eight-month median survival and no fiveyear survival. Among those with mobile lesions, the median survival is 40 months, with an overall five-year survival of 47% of those patients (118). Salvage APR following recurrence can have a longterm survival rate (120,121). A dismal result was described by the series by Zelnick et al. (122) where there was no fiveyear survival. Inguinal Lymph Node Because of the high morbidity and low yield in the prevention of death from cancer, prophylactic groin dissection is not recommended (94,95). The simultaneous appearance of inguinal metastasis is an ominous sign. In the series of Gerard et al. (123), of 270 patients with squamous cell carcinoma of the anal canal, synchronous inguinal metastasis occurred in 10%, and metachronous inguinal metastasis in 7.8%. The five-year overall survival in patients without inguinal lymph node involvement was 73% versus 54% in patients with synchronous lymph node metastasis. For patients with synchronous lymph node metastasis, the authors recommended a unilateral lymph node dissection immediately followed by a cycle of continuous infusion of 5-FU (day 1–4) and bolus of cisplatinum (day 2–5). Radiation to the involved groin was initiated after completion of the chemotherapy. The dose of the radiation was 45 to 50 Gy over five weeks. The results showed local control of the inguinal area in 86%; the five-year overall survival was 54%. For patients with a metastatic metachronous inguinal lymph node metastasis, the initial treatment was inguinal lymph node dissection. Irradiation started after the wound has healed, delivering 45 to 50 Gy over five weeks. No prophylactic irradiation of the contralateral inguinal area was performed. The local control of the inguinal area was observed in 68% of patients so treated and the five-year overall survival rate was 41% (123). Whether prophylactic groin radiation should be performed is controversial. Elective radiation of clinically normal inguinal nodes reduces the risk of late node failure and carries little morbidity. Only one of 38 such patients had a late recurrence in the inguinal area after undergoing combination chemotherapy and RT (124). In series in which the inguinal nodes were not treated electively, the late nodal recurrence rate was 15% to 25% (96,125,126). Ulmer et al. (127) evaluated the feasibility of the sentinel lymph node technique for groin metastasis of anal squamous cell carcinoma. The lesion in the anal canal was injected submucosally or subdermally with 1 mL Tc99m sulfur colloid in four sites, using a 27-gauge needle and an insulin syringe. Scintigraphy was recorded with a gamma camera. Seventeen hours later, patients with detectable radiocolloid enrichment in the groin underwent lymph node biopsy guided by a handheld gamma probe. Sentinel lymph nodes were detected in 13 of 17 patients (76.5%) metastases were found in the sentinel lymph nodes of 5 of 12 biopsied patients (42%); in two patients the metastases were detected by serial sectioning or immunohistochemical

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

staining after H&E results were negative for carcinoma. Compared with staging using ultrasonography and CT, assessment of the sentinel lymph node provides more reliable staging of inguinal lymph nodes because 44% of all lymph nodes metastasis in anal carcinoma are smaller than 5 mm in diameter (127). This technique may prove to be a valuable diagnostic workup for anal squamous cell carcinoma. Further studies are warranted. Drugs for Metastasis The most frequent sites of visceral metastasis include the liver, lung, bone, and subcutaneous tissues. The prognosis is poor, with a median survival of approximately nine months. Twenty percent of patients with recurrent carcinoma of the anal canal die from distant metastasis, but most carcinoma-related deaths are secondary to uncontrolled pelvic and perianal disease (128). Drugs used to treat metastasis include 5-FU, bleomycin, methyl-CCNU, vincristine, doxorubicin, and cisplatin. Combinations of agents such as bleomycin, vincristine, or methotrexate, plus leucovorin, had also been used. All combinations had resulted in only partial responses (129). A phase III randomized study is being completed in the United Kingdom for patients who have locally advanced squamous cell carcinoma of the anal canal using different drugs and radiation, including 5-FU/MMC/radiation; 5-FU/cisplatin/radiation (107). SUMMARY & Surgeons must be familiar with the anatomic

landmarks of the anal canal and the perianal skin as defined by the WHO and the AJCC, as well as with TNM staging. Currently, most reports and studies use these systems. Local excision is still the treatment of choice for carcinoma in situ or microscopic invasive carcinoma that has occurred in the anal canal or the perianal area. Unfortunately, only small numbers of the lesions are suitable for this option because most are too large or too advanced at the time of diagnosis. APR is no longer the primary treatment for invasive carcinoma of the anal canal and most perianal carcinomas. It not only has a high recurrence, but local recurrence after APR has a less favorable prognosis. It responds less favorably to chemoradiation than the primary carcinoma, as 52% of patients have persistent disease after the treatment (130). APR is reserved for local treatment failure of chemoradiation, for anorectal complications of the treatment, especially fecal incontinence, and for those patients who cannot tolerate chemoradiation. Chemoradiation, which was originally used as the primary treatment of anal canal squamous cell carcinoma, is now also applied to its counterpart in the perianal area.

321

cinoma of the large intestine. It is generally difficult or impossible to separate adenocarcinoma of the anal canal from adenocarcinoma of the lower rectum (1).

Anal Glands The ducts of the anal glands are lined by squamous epithelium close to their opening in the crypts, by transitional epithelium more deeply, and by mucin-secreting columnar epithelium in the depth of the gland. The histologic picture of these lesions, therefore, may be one of adenocarcinoma or mucoepidermoid carcinoma. It can be differentiated from other types of anal lesions by its haphazardly dispersed, small glands with scant mucin production invading the wall of the anorectal area without an intraluminal component. The glands are positive for CK7 (134). The most characteristic feature of anal duct carcinoma is its extramucosal adenocarcinoma without the involvement of the surface epithelium, except when the lesion has become advanced (Fig. 10). If there is a break in the surface epithelium, as is often seen clinically, greater perianal involvement or deeper infiltration may provide the only clue to the anal duct origin of these lesions. In a series of 21 patients reported by Jensen et al. (136), nine of the neoplasms were localized in the ischioanal space, seven were in the anal canal, and five were in a fistula-in-ano. Patients usually present with complaints of pain and an extra-anorectal lump, perianal induration, or a perianal abscess. Despite its distal location and its accessibility to digital examination, most anal gland and duct carcinomas are detected late. In the series by Jensen et al. (136), the median duration of symptoms before correct diagnosis was 18 months, the sensation of having a perianal mass, bleeding, pain, soiling, pruritus ani, change in bowel habits, prolapse, and weight loss were common symptoms. Like other malignancies of the anal region, most adenocarcinomas of the anus are diagnosed erroneously by physicians and surgeons, as well as by patients, with a resulting delay in the correct diagnosis. The average size of the carcinoma of the anal canal in the series by Jensen et al. (136) was 5 cm and in the perianal area was 10 cm. Sixty-two percent of the patients already had a regional or distant metastasis. Because of the late stage of the carcinoma 20 of 21 patients died within 18 months after treatment (136).

& ADENOCARCINOMA Primary adenocarcinomas of the anus are very rare, constituting 3% to 9% of all anal carcinomas (131–133). The WHO classifies these malignancies into the rectal type, the anal glands, and those within an anorectal fistula (1). Rectal Type This type is the most common adenocarcinoma found in the anal canal. It arises within the upper zone lined by colorectal-type mucosa. Its histology is that of an adenocar-

FIGURE 10 & Anal duct carcinoma. Note the intact epithelium. Source: From Ref. 135.

322

& PART II: ANORECTAL DISORDERS

TABLE 3 & Survival of Three Modes of Treatment of Primary Anal Canal Adenocarcinoma Mode of Treatment

FIGURE 11 & Adenocarcinoma in a fistula-in-ano. Note pools of mucin. Source: From Ref. 135.

The primary treatment is an APR, with wide excision of the perineal part (137). The role of radiation or chemoradiation is unknown.

Anorectal Fistula Well-differentiated mucinous adenocarcinomas (Fig. 11) occasionally develop within an anorectal fistula that may be developmental or acquired (1). Most often these carcinomas arise in patients with long-standing perianal disease, especially fistulas (138,139). Some authors believe that they originate in the anal glands and ducts (140,141). Treatment For a rectal type or a primary adenocarcinoma of anal canal, a wide local excision can be performed for small and welldifferentiated carcinomas that have not invaded the muscular wall of the anorectum. In a multicenter study, data collected for primary adenocarcinoma of the anal canal from patients reported by Belkacemi et al. (142) included: 18 T1 (18%), 34 T2 (42%), 22 T3 (27%), and 11 T4 (13%). There were three treatment categories: radiation and surgery, 45 patients; chemoradiation, 31 patients; and APR, 6 patients. The patients’ characteristics were evenly distributed among the three groups. The results showed that recurrence occurred in 37%, 36%, and 20%, respectively, at four-year follow-up (not statistically significant). Both the overall and the disease-free five-year survival were significantly better in the chemoradiation group (Table 3). Multivariate analysis revealed four independent prognostic factors for survival: T stage, N stage, histologic grade, and treatment modality. From this study, it is apparent that from Stage T2 onwards, the primary treatment should be chemoradiation and the APR should be reserved for salvage treatment. A small series from Memorial Sloan-Kettering Cancer Center (133) showed that 6 of 13 patients were free of disease after chemoradiation and APR, with a follow-up of 26 months. For adenocarcinomas of the anal gland or an anal fistula, the role of adjuvant therapy is not yet defined due to the uncommon disease. (143). The Nigro chemoradiation regime has been successfully used by Tarazi and Nelson (144). Of nine patients who used this protocol, six patients were free of disease on follow-up of two to four years. Papagikos et al. (145) recommended a preoperative chemoradiation followed by APR.

Survival

Five Year (%)

Ten Year (%) 23 p ¼ 0.02 39 21 18 p ¼ 0.038 20 22

RT/APR

Overall

29

RT/CHT APR RT/APR

Overall Overall Disease-free

58 21 25

RT/CHT APR

Disease-free Disease-free

54 22

Abbreviations: RT, radiotherapy; CHT, chemotherapy; APR, abdominoperineal resection.

& SMALL CELL CARCINOMA This very rare carcinoma may arise in the anorectal region. It is similar in histology, behavior, and histochemistry to small-cell (oat cell) carcinoma of the lung (Fig. 12). It is a neuroendocrine carcinoma or a Merkel cell carcinoma. The diagnosis can be confirmed by immunohistochemistry and electron microscopy. This type of lesion has been known to have early and extensive dissemination. The case presented by Paterson et al. (146) showed that a 1 cm Merkel cell carcinoma of the anal canal had already metastasized to the liver. Based on Merkel cell carcinoma of other organs, ‘‘it is extremely lymphophile; lymph node relapses occur in approximately 80% of the patients with recurrent disease’’ (147). Merkel cell carcinoma has been known to respond to RT (147). Whether this mode of treatment is useful for Merkel cell carcinoma of the anal canal is not known. & UNDIFFERENTIATED CARCINOMA Also very rare, this type of malignant lesion has no glandular structure or other features to indicate definite differentiation (Fig. 13). Undifferentiated carcinoma may be distinguished from poorly differentiated carcinoma, small-cell carcinoma, lymphoma, or leukemic deposits by the use of mucin stains or immunohistochemical methods (1). The treatment is the same as for adenocarcinomas. The prognosis can be expected to be poor.

FIGURE 12 & Small cell carcinoma of the large intestine. Source: From Ref. 93.

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

323

FIGURE 13 & Undifferentiated carcinoma of the large intestine. Source: From Ref. 93.

& MELANOMA General Considerations Malignant melanoma is the most depressing of all anorectal malignancies. It is a rare malignant neoplasm of the anorectum that constitutes 1% to 3% of all melanomas. The anal canal is the third most common site, exceeded only by the skin and eyes (148). The female-to-male ratio is approximately 2:1, and the average age at presentation is approximately 63 years (149). Evaluation of the NCI SEER data from 1973 to 1992 showed female-to-male ratio as 1.72:1. The mean age was 66  16 years. Mean age by gender was lower for males (57 years) than for females (71 years; p < 0.001). The incidence of anorectal melanoma in young males ages between 25 and 44 years tripled in the San Francisco area when compared with all other locations (14.4 vs. 4.8 per 10 million population; p ¼ 0.06). There was indirect evidence that implicated HIV infection as a risk factor (150). Malignant melanoma arises from epithelium of the anal canal, both above and below the dentate line (151). A few reports describe these lesions as arising from, and being situated in, the rectum (152,153). Electron microscopy shows that normal melanocytes are present in the rectal mucosa (152). Clinical Features Rectal bleeding, a mass in the anal canal, and anorectal pain are the three most common, and most consistent signs and symptoms of malignant melanoma (149,154). Only 25% of patients have lesions < 1 cm in diameter. The remainder have melanomas as large as 6 cm in diameter, with an average size of 4 cm (153). Often the mass protrudes through the anus. Weight loss is also a common finding. Diagnosis Melanomas are suspected when a pigmented polypoid lesion is noted (Fig. 14). Unless an ulceration with raised edges is present, this disease may be confused with a thrombosed hemorrhoid. The majority of melanomas, however, are only lightly pigmented or nonpigmented and are often misdiagnosed as being polyps or other neoplasms of the anal canal. Between 40% and 70% of melanomas of the anal canal are amelanotic and in only 25% of the pigmented

FIGURE 14 & Malignant melanoma of the anal canal. The arrows point at the dentate line.

lesions is there abundant melanin (151,155). If melanin is seen on microscopic examination, then the diagnosis is simple. In the amelanotic melanoma, sheets of anaplastic cells may be misinterpreted as undifferentiated squamous cell carcinoma. The most helpful diagnostic feature is the presence of malignant cells in clusters (155). Endorectal ultrasound is useful to determine the depth of invasion and possible adjacent lymph node metastasis (156,157).

Mode of Metastasis Anal canal melanomas have a marked tendency to spread submucosally along the rectum but rarely invade adjacent organs. Review of the literature by Cooper et al. (158) showed that 46 of 120 patients (38%) had metastasis at the time of diagnosis. Perirectal, perianal, and mesenteric lymph nodes were the most common sites of metastasis, followed by inguinal lymph nodes, liver, and lung. Widespread systemic metastases are early and rapid, most commonly to the liver, lung, and bone. Wolff (159) raised a question whether, using thickness of the lesion similar to melanoma of the skin can be useful to determine options of treatment and prognosis. Treatment There is no survival benefit of adjuvant therapy for melanomas of the anal canal partly because of associated distant metastases in most patients. For the majority of cases, there appears to be no clear-cut choice of surgical treatment between a wide local excision and an APR. Both treatments yield a five-year survival between 0% and 22% (149). In the larger series reported by Brady et al. (160) of 85 cases and Thibault et al. (149) of 50 cases, the five-year survival rates were 17% and 22%, respectively. Both options of treatment have their proponents (Table 4). Thibault et al. (149) reviewed the series from the Mayo Clinic attempting to find predictive factors of survival, including gender, size of the lesion, presence of melanin, depth of penetration, positive perirectal lymph nodes, wide local excision versus APR, and there was none. This is in contrast to the finding by Brady et al. (160) at Memorial Sloan-Kettering Cancer Center that all long-term survivals

324

& PART II: ANORECTAL DISORDERS

TABLE 4 & Selected Series of Anal Melanoma

Author/Reference Pessaux et al., 2004 (161) Bullard et al., 2003 (156) Thibault et al., 1997 (149) Ward et al., 1996 (162) Roumen, 1996 (163) Brady et al., 1995 (160) Ross et al., 1990 (164) Slingluff et al., 1990 (165) Goldman et al., 1990 (166)

No. of Patients 16 40 50 21 63 85 32 24 49

No. Curative WLE and Disease-Free at 5 Yr (%) 11 21 11 3 16 13 12 7 18

(29%) (16%) (18%) (0%) (13%) (8%) (8%) (0%) (6%)

No. Curative Abdominoperineal Resection and Disease-Free at 5 Yr (%) 4 9 26 4 18 43 14 12 15

(0%) (33%) (19%) (0%) (11%) (26%) (0%) (0%) (7%)

Recommended Treatment WLE WLE WLE WLE WLE APR WLE Multimodality APR

Abbreviation: WLE, wide local excision.

occurred in women. Indeed, in their study, women with operable disease had a five-year survival rate of 29%. Although the authors recommend an APR, only one of nine patients who underwent such an operation had positive mesenteric nodes. It appears that local control of the disease after the operation is not as much a problem as distant metastasis, which is the major cause of death (156,163). A reasonable approach is to perform local excision of the lesion, only if this can be accomplished with wide margin and full thickness without causing fecal incontinence. Otherwise, an APR should be performed. Pessaux et al. (161) reported 30 patients who underwent APR or wide local excision. Features that showed significantly better results in five-year survival included: negative inguinal node (p ¼ 0.031), duration of symptoms less than three months (p ¼ 0.046), Stage I versus Stage II (p ¼ 0.025), nonmelanomic (p ¼ 0.033). Tumor size, depth of invasion (in mm), age, gender, wide local excision versus APR, were not statistically different.

& REFERENCES 1. Fenger C, Frisch M, Marti MC, Parc R. Tumors of the anal canal. Hamilton SR, Aaltonen LA, eds. WHO classification of tumors. Pathology and genetics of tumors of the digestive system. Lyon, France: International Agency for Research on Cancer (IARC), 2002:146–155. 2. Green FL, Page DL, Fleming ID, et al. American Joint Committee on Cancer. 6th ed. New York: Springer-Verlag, 2002:139–144, 231–237. 3. Cummings BJ. [Editorial]. Oncology 1996; 10:1853–1854. 4. Jensen SL, Hagen K, Shokouh-Amiri MH, Nielsen OV. Does an erroneous diagnosis of squamous-cell carcinoma of the anal canal and anal margin at first physician visit influence prognosis?. Dis Colon Rectum 1987; 30:345–351. 5. Williams GR, Talbot IC. Anal carcinoma—a histological review. Histopathology 1994; 25:507–516. 6. Brown DK, Ogelsby AB, Scott DH, Dayton MT. Squamous cell carcinoma of the anus. A twenty-five year retrospective. Am Surg 1988; 54:337–342. 7. Greenall MJ, Quan SHQ, Urmacher C, DeCosse JJ. Treatment of epidermoid carcinoma of the anal canal. Surg Gynecol Obstet 1985; 161:509–517. 8. Pintor MP, Northover JMA, Nicholls RJ. Squamous cell carcinoma of the anus at one hospital from 1948 to 1984. Br J Surg 1989; 76:806–810. 9. Nigro ND. Multidisciplinary management of cancer of the anus. World J Surg 1987; 11:446–451. 10. Bal DG. Cancer statistics 2001: quo vadis or whither goest thou? CA Cancer J Clin 2001; 51:11–14. 11. Jemal A, Tiwari RC, Murray T, et al. Cancer statistics 2005. CA Cancer J Clin 2005; 5:10–30. 12. Deans GT, McAlee JJA, Spence RAJ. Malignant anal tumors. Br J Surg 1994; 81:501–508. 13. Maggard M, Beanes SR, Ko CY. Anal canal cancer: a population-based reappraisal. Dis Colon Rectum 2003; 46:1517–1524.

14. Frisch M. On the etiology of anal squamous carcinoma. Dan Med Bull 2002; 49:194–209. 15. Chang GJ, Sheldon A, Welton ML. Epidemiology and natural history of anal HPV infection and ASIL and cancer in the general population. Semin Colon Rect Surg 2004; 15:210–214. 16. Saclarides TJ, Klem D. Genetic alterations and virology of anal cancer. Semin Colon Rectal Surg 1995; 6:131–134. 17. Palmer JG, Scholefield JH, Coates PJ, et al. Anal cancer and human papillomaviruses. Dis Colon Rectum 1989; 32:1016–1022. 18. Shroyer KR, Kim JG, Manos MM, Greer CE, Pearlman NW, Franklin WA. Papillomavirus found in anorectal squamous carcinoma, not in colon adenocarcinoma. Arch Surg 1992; 127:741–744. 19. Bjorget B, Engeland A, Luostarinen T, et al. Human papilloma virus infection as a risk factor for anal and perianal skin cancer in a prospective study. Br J Cancer 2002; 87:61–64. 20. Fisch M, Glimelius B, Van Den Brule AJ, et al. Sexually transmitted infection as a cause of anal cancer. NEJM 1997; 337:1350–1358. 21. Welton ML. Etiology of human papilloma virus infections and the development of anal squamous intraepithelial lesions. Semin Colon Rectal Surg 2004; 15:193–195. 22. Mullerat J, Northover J. Human papilloma virus and anal neoplastic lesions in the immunocompromised (Transplant) patient. Semin Colon Rectal Surg 2004; 15:215–217. 23. Penn I. Cancers of the anogenital region in renal transplant recipients. Analysis of 65 cases. Cancer 1986; 58:611–616. 24. Gervaz P, Hahnloser D, Wolff BG, et al. Molecular biology of squamous cell carcinoma of the anus: a comparison of HIV positive and HIV-negative patients. J Gastrointest Surg 2004; 8:1024–1031. 25. Noffsinger A, Witte D, Fenoglio-Preiser CM. The relationship of human papillomaviruses to anorectal neoplasia. Cancer 1992; 70:1276–1287. 26. Cummings BJ. Anal canal carcinoma. Hermanek P, Gaspodarowicz MK, Henson DE, Hutter RVP, Sobin LH, eds. Prognostic Factors in Cancer. New York: Springer, 1995:80–87. 27. Carter PS. Anal cancer—current perspectives. Dig Dis 1993; 11:239–251. 28. Welton ML, Winkler B, Darragh TM. Anal-rectal cytology and anal cancer screening. Semin Colon Rectal Surg 2004; 15:196–200. 29. Polefsky JM. Anal squamous intraepithelial lesions in human immunodeficiency virus-positive men and women. Semin Oncol 2000; 27:471–479. 30. Goldie SJ, Kuntz KM, Weinstein MC, Freedberg KA, Palefasky JM. Cost-effectiveness of screening for anal squamous intraepithelial lesions and anal cancer in human immunodeficiency virus-negative homosexual and bisexual men. Am J Med 2000; 108:634–641. 31. Sholefield JH, Johnson J, Hitchcock A, et al. Guidelines for anal cytology—to make cytological diagnosis and follow up much more reliable. Cytopathology 1998; 9:15–22. 32. Chang GJ, Welton ML. Anal neoplasia. Seminars Colon Rectal Surg 2003; 14:111–118. 33. Goldstone SE, Winkler B, Wifford LJ, Alt E, Polefsky JM. High prevalence of anal squamous intraepithelial lesions and squamous-cell carcinoma in men who have sex with men as seen in a surgical practice. Dis Colon Rectum 2001; 44:690–698. 34. Koutsky LA, Ault KA, Wheeler CM, et al. A controlled trial of a human papillomavirus type 16 vaccine. NEJM 2002; 347:1645–1651. 35. Stanberry LR. A human papillomavirus type 16 vaccine. Editorial. NEJM 2003; 348:1404. 36. Bowen JT. Precancerous dermatoses: a study of two cases of chronic atypical epithelial proliferation. J Cutan Dis 1912; 30:241–255. 37. Feuger C, Nielsen VT. Intraepithelial neoplasia in the anal canal. The appearance and relation to genital neoplasia. ACTA Pathol Microbiol Immunol Scand 1986; 94:343–349.

CHAPTER 5: PERIANAL AND ANAL CANAL NEOPLASMS &

38. Feuger C. Intraepithelial neoplasia in the anal canal and perianal area. Curr Topic Pathol 1990; 81:91–102. 39. Halverson AL. Perianal Bowen’s disease then and now: evolution of the treatment for anal high-grade intraepithelial neoplasia. Semin Colon Rectal Surg 2003; 14:213–217. 40. Brown SR, Skinner P, Tidy J, Smith JH, Sharp F, Hosie KB. Outcome after surgical resection for high-grade anal intraepithelial neoplasia (Bowen’s disease). Br J Surg 1999; 86:1063–1066. 41. Scholefield JH, Castle MT, Watson NFS. Malignant transformation of highgrade anal intraepithelial neoplasia. Br J Surg 2005; 92:1133–1136. 42. Abbasakoor F, Boulos PB. Anal intraepithelial neoplasia. Br J Surg 2005; 92:277–290. 43. Arbesman H, Ransohoff DF. Is Bowen’s disease a predictor for the development of internal malignancy? JAMA 1987; 257:516–518. 44. Chute CG, Chuang TY, Bergstralh EJ, Su WPD. The subsequent risk of internal cancer with Bowen’s disease. JAMA 1991; 266:816–819. 45. Marfing TF, Abel ME, Gallagher DM. Perianal Bowen’s disease and associated malignancies. Results of survey. Dis Colon Rectum 1987; 30:782–785. 46. Morgenthaler JA, Dietz DW, Matthew GM, Birmbaum EH, Kodner IJ, Fleshman JW. Outcomes, risk of other malignancies, and need for formal mapping procedures in patients with perianal Bowen’s disease. Dis Colon Rectum 2004; 47:1655–1661. 47. Marchesa P, Fazio VW, Oliart S, Goldblum JR, Lavery IC. Perianal Bowen’s disease: a clinico-pathologic study of 47 patients. Dis Colon Rectum 1997; 40:1286–1293. 48. Samiento JM, Wolff BG, Burgart LJ, Frizelle FA, Ilstrup DM. Perianal Bowen’s disease. Associated tumors, human pappilloma virus, surgery, and other controversies. Dis Colon Rectum 1997; 40:912–918. 49. Gupta AK, Browne M, Bluhm R. Imiquimod: a review. J Cutan Med Surg 2002; 6:554–560. 50. Pehoushek J, Smith KJ. Imiquimod and 5% fluorouracil therapy for anal and perianal squamous cell carcinoma in situ in an HIV positive man. Arch Dermatol 2001; 137:14–16. 51. Gottesman L. [Editorial]. Dis Colon Rectum 2004; 47:1660–1661. 52. Chang LK, Gottesman L, Breen EL, Bledag R. Anal dysplasia: controversies in management. Semin Colon Rectal Surg 2004; 15:233–238. 53. Nouri K, O’Connell C, Rivas MP. Imiquimod for the treatment of Bowen’s disease and invasive squamous cell carcinoma. Case J Drugs Dermatol 2003; 2:669–673. 54. Chen K, Shumack S. Treatment of Bowen’s disease using a cycle regimen of imiquimod 5% cream. Clin Exp Dermatol 2003; 28(suppl):10–12. 55. Graham BD, Jetmore AB, Foote JE, Arnold LK. Topical 5-fluorouracil in the management of extensive anal Bowen’s disease: a preferred approach. Dis Colon Rectum 2005; 48:444–450. 56. Skinner PP, Ogunbiyio A, Scholefield JH, et al. Skin appendage involvement in anal intraepithelial neoplasia. Br J Surg 1997; 84:675–678. 57. Berry JM, Jay N, Polefsky JM, Welton ML. State-of-the-art of high-resolution anoscopy as a tool to manage patients at risk for anal cancer. Semin Colon Rectal Surg 2004; 15:218–226. 58. Cleary RK, Schaldenbrand JD, Fowler JJ, Schuler JM, Lampman RM. The treatment options for perianal Bowen’s disease: survey of American Society of Colon and Rectal Surgeons members. Am Surg 2000; 66:686–688. 59. Hassan I, Horgan AF, Nivatvongs S. V-Y island flags for repair of large perianal defects. Am J Surg 2001; 181:363–365. 60. Scholefield JH. Anal intraepithelial neoplasia. Br J Surg 1999; 86:1364–1364. 61. Beahrs OH, Wilson SM. Carcinoma of the anus. Ann Surg 1976; 184:422–428. 62. Papillon J, Chassard JL. Respective roles of radiotherapy and surgery in the management of epidermoid carcinoma of the anal margin. Dis Colon Rectum 1992; 35:422–429. 63. Moller C, Saksela E. Cancer of the anus and anal canal. Acta Chir Scand 1970; 136:340–348. 64. Nelson RL, Prasad L, Abcarian H. Anal carcinoma presenting as a perianal abscess or fistula. Arch Surg 1985; 120:632–635. 65. Fuchshuber PR, Rodriguez-Bigas M, Weber T, Peorelli NJ. Anal canal and perianal epidermoid cancers. Collective review. J Am Coll Surg 1997; 185:494–505. 66. Touboul E, Schlienger M, Buffat L, et al. Epidermoid carcinoma of the anal margin: 17 cases treated with curative-intent radiation therapy. Radiother Oncol 1995; 34:195–202. 67. Schraut WH, Wang C, Dawson PJ, Block GE. Depth of invasion, location, and size of cancer of the anus dictate operative treatment. Cancer 1983; 51: 1291–1296. 68. Cummings BJ, Keane TJ, Hawkins NV, O’Sullivan B. Treatment of perianal carcinoma by radiation (RT) or radiation plus chemotherapy (RTCT). Int J Radiat Oncol Biol Phys 1986; 12:170–173. 69. Paget J. On disease of the mammary areolar preceding cancer of the mammary gland. St. Bartholomew’s Hosp Report 1874; 10:87–89. 70. Tjandra J. Perianal Paget’s disease. Report of three cases. Dis Colon Rectum 1988; 31:462–466. 71. Darier J, Couillaud P. Sur un cas de maladie de Paget de la region kerineo-anal et scrotale. Ann de Dermatole et de Syph 1893; 4:25–31. 72. Armitage NC, Jass JR, Richman PI, Thomson JPS, Phillips RKS. Paget’s disease of the anus: a clinicopathological study. Br J Surg 1989; 76:60–63.

325

73. Morson BC, Dawson IMP, Day DW, Jass JR, Price AB, Williams GT. Morson and Dawson’s Gastrointestinal Pathology. London: Blackwell Scientific, 1990:673–675. 74. Rosai J. Ackerman’s Surgical Pathology. 8th ed. Louis: CV Mosby, 1996: 808–809. 75. Miller LR, McCunniff A, Randall ME. An immunohistochemical study of perianal Paget’s disease. Cancer 1992; 69:2166–2171. 76. Beck DE. Paget’s disease and Bowen’s disease of the anus. Semin Colon Rectal Surg 1995; 6:143–149. 77. Jensen SL, Sjolin KE, Shokouh Amiri MH, Hagen K, Harling H. Paget’s disease of the anal margin. Br J Surg 1988; 75:1089–1092. 78. Park JS, Kerner BA. Perianal Paget’s disease. Semin Colon Rectal Surg 2003; 14:218–221. 79. Ohnishi T, Watanabe S. The use of cytokeratins 7 and 20 in the diagnosis of primary and secondary extramammary Paget’s disease. Br J Dermatol 2000; 142:243–247. 80. Goldblum JR, Hart WR. Perianal Paget’s disease. A histologic and immunohistochemical study with and without associated rectal adenocarcinoma. Am J Surg Pathol 1998; 22:170–179. 81. Brown RSD, Lankester KJ, McCormack M, Power DA, Spittle MF. Radiotherapy for perianal Paget’s disease. Clin Oncol 2002; 14:272–284. 82. Samiento JM, Wolff BG, Burgart LJ, Frizelle FA, Ilstrup DM. Paget’s disease of the perianal region—An aggressive disease? Dis Colon Rectum 1997; 40: 1187–1194. 83. McCarter MD, Quan SHQ, Busam K, Paty PH, Wong D, Guillem JG. Longterm outcome of perianal Paget’s disease. Dis Colon Rectum 2003; 46:612–616. 84. Marchesa P, Fazio VM, OliartS, Goldblum JR, Lavery IC, Milsom JW. Longterm outcome of patients with perianal Paget’s disease. Am Surg Oncol 1997; 4:475–480. 85. Paterson CA, Young-Fadok TM, Dozois RR. Basal cell carcinoma of the perianal region. 20-year experience. Dis Colon Rectum 1999; 42:1200–1202. 86. Nielsen OV, Jensen SL. Basal cell carcinoma of the anus—a clinical study of 34 cases. BR J Surg 1981; 68:856–857. 87. Marks R, Gebauer K, Shumack S, et al. Imiquimod 5% cream in the treatment of superficial basal cell carcinoma: results of a multicenter 6-week doseresponse trial. J Am Acad Dermatol 2001; 44:807–813. 88. Gordon PH. Current status—perianal and anal canal neoplasms. Dis Colon Rectum 1990; 33:799–808. 89. Cintron J. Buschke-Loewenstein tumor of the perianal and anorectal region. Semin Colon Rectal Surg 1995; 6:135–139. 90. Heinzerling LM, Kempf W, Kamarashev J, Hafner J, Nestle FO. Treatment of verrucous carcinoma with Imiquimod and CO2 laser ablation. Dermatology 2003; 207:119–122. 91. Foust R, Dean PJ, Stoler MH, Monuddin SM. Intraepithelial neoplasia of the anal canal in hemorrhoidal tissue: a study of 19 cases. Hum Pathol 1991; 22:528–534. 92. Chang GJ, Berry JM, Jay N, Polefsky JM, Welton ML. Surgical treatment of high-grade anal squamous intraepithelial lesions. Dis Colon Rectum 2002; 45:453–458. 93. Jass JR, Sobin LH. . Histological Typing of Intestinal Tumors. 2nd Heidelberg: Springer-Verlag, 1989:90. 94. Stearns MW Jr, Quan SHQ. Epidermoid carcinoma of the anorectum. Surg Gynecol Obstet 1970; 131:953–957. 95. Welch JP, Malt RA. Appraisal of the treatment of carcinoma of the anus and anal canal. Surg Gynecol Obstet 1977; 145:837–844. 96. Boman BM, Moertel CG, O’Connell MJ, et al. Carcinoma of the anal canal: a clinical and pathologic study of 188 cases. Cancer 1984; 54:114–125. 97. Cummings BJ. Treatment of primary epidermoid carcinoma of the anal canal. Int J Colorectal Dis 1987; 2:107–112. 98. Wade DS, Herrera I, Castillo NB, Petrelli NJ. Metastases to the lymph nodes in epidermoid carcinoma of the anal canal studied by a clearing technique. Surg Gynecol Obstet 1989; 169:238–242. 99. Jensen SL, Hagen K, Harling H, Shokouh-Amiri MH, Nielsen OV. Long-term prognosis after radical treatment for squamous cell carcinoma of the anal canal and anal margin. Dis Colon Rectum 1988; 31:273–278.. 100. Nigro ND, Vaitkevicuis VK, Considine B. Combined therapy for cancer of the anal canal: a preliminary report. Dis Colon Rectum 1974; 17:354–356. 101. Nigro ND. An evaluation of combined therapy for squamous cell cancer of the anal canal. Dis Colon Rectum 1984; 27:763–766 102. Sato H, Koh PK, Bartolo DCC. Management of anal canal cancer. Dis Colon Rectum 2005; 48:1301–1315. 103. Flam M, John M, Pajak TF, et al. Role of mitomycin in combination with fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study. J Clin Oncol 1996; 14:2527–2539. 104. UKCCCR Anal Cancer Trial Working Party. Epidermoid anal cancer: results from the UKCCCR randomized trial of radiotherapy alone versus radiotherapy, 5-flourouracil, and mitomycin. UKCCCR anal cancer trial working party. UK Co-ordinating Committee on Cancer Research. Lancet 1996; 348: 1049–1054. 105. Bartelink H, Roelofsen F, Eschwege F, et al. Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally

326

106. 107. 108.

109. 110. 111.

112.

113.

114.

115.

116.

117. 118.

119. 120.

121.

122.

123.

124. 125. 126. 127.

128.

129. 130. 131. 132. 133.

& PART II: ANORECTAL DISORDERS

advanced anal cancer: results of a phase III randomized trial of the European organization for research and treatment of cancer radiotherapy and gastrointestinal cooperative groups. J Clin Oncol 1997; 15:2040–2049. Eng C, Abbruzzese J, Minsky BD. Chemotherapy and radiation of anal canal cancer: the first approach. Surg Onc Clin N Am 2004; 13:309–320. Stafford S, Martenson JA. Combined radiation and chemotherapy for carcinoma of the anal canal. Oncology 1998; 12:373–389. Myerson RJ, Kong F, Birnbaum EH, et al. Radiation therapy for epidermoid carcinoma of the anal canal, clinical and treatment factors associated with outcome. Radiother Oncol 2001; 61:15–22. Myerson RJ et al. Carcinoma of the anal canal. Am J Clin Oncol 1995; 18:32–39. Roed H, Engelholm SA, Svendsen LB, et al. Pulsed dose rate (PDR) brachy therapy of anal carcinoma. Radiother Oncol 1996; 41:131–134. Kapp KS, Geyer E, Gebhart FH, et al. Experience with split-course external beam irradiation Ir- chemotherapy and integrated in-192 high dose-rate brachytherapy in the treatment of primary carcinoma of the anal canal. Int J Radiat Oncol Bio Phys 2001; 49:997–1005. Place RJ, Gregory SG, Huber PJ, Simmang CL. Outcome analysis of HIV-positive patients with anal squamous cell carcinoma. Dis Colon Rectum 2001; 44:506–512. Kim JH, Sarani B, Orkin BA, et al. HIV-positive patients with anal carcinoma have poorer treatment tolerance and outcome than HIV-negative patients. Dis Colon Rectum 2001; 44:1496–1502. Berry JM, Polefsky JM, Welton ML. Anal cancer and its precursors in HIVpositive patients: perspectives and management. Surg Onc Clin N Am 2004; 13:355–373. Blazy A, Hennequine, Gornet JM, et al. Anal carcinomas in HIV-positive patients: high-dose chemoradiotherapy is feasible in the era of highly active antiretroviral therapy. Dis Colon Rectum 2005; 48:1176–1181. Cummings BJ, Keane TJ, O’Sullivan B, Wong ES, Cotton CN. Epidermoid anal cancer: treatment by radiation alone or by radiation and 5-fluorouracil with and without mitomycin C. Int J Radiat Oncol Biol Phys 1991; 21:1115–1125. Tanum G, Tveit KM, Karlsen KO. Chemo-radiotherapy of anal carcinoma: tumor response and acute toxicity. Oncology 1993; 50:14–17. Ellenhorn JD, Enker WE, Quan SHQ. Salvage abdominoperineal resection following combined chemotherapy and radiotherapy for epidermoid carcinoma of the anus. Ann Surg Oncol 1994; 1:105–110. Nguyen WD, Mitchell KM, Beck DE. Risk factors associated with requiring a stoma for the management of anal cancer. Dis Colon Rectum 2004; 47:843–846. Longo WE, Vernava AM, Wade TP, Coplin MA, Virgo KS, Johnson FE. Recurrent squamous cell carcinoma of the anal canal. Predictors of initial treatment failure and results of salvage therapy. Ann Surg 1994; 220:40–49. Ghouti L, Houvenaeghel G, Moutardier V, et al. Salvage abdominoperineal resection after failure of conservative treatment in anal epidermoid cancer. Dis Colon Rectum 2005; 48:16–22. Zelnick RS, Hass PA, Ajlouni M, Szilagyi E, Fox TA Jr. Results of abdominoperineal resections for failures after combination chemotherapy and radiation therapy for anal cancers. Dis Colon Rectum 1992; 35:574–578. Gerard JP, Chapet O, Samiei F, et al. Management of inguinal lymph node metastasis in patients with carcinoma of the anal canal. Experience in a series of 270 patients treated in Lyon and review of the literature. Cancer 2001; 92: 77–84. Cummings BJ, Thomas GM, Keane TJ. Primary radiation therapy in the treatment of anal canal carcinoma. Dis Colon Rectum 1982; 25:778–782. Papillon J, Montbarbon JF. Epidermoid carcinoma of the anal canal. A series of 276 cases. Dis Colon Rectum 1987; 30:324–333. Stearns MW, Urmacher C, Sternborg SE, Woodruff J, Attiyeh FF. Cancer of the anal canal. Curr Probl Cancer 1980; 4:1–44. Ulmer C, Bembenek A, Gretschel S, et al. Refined staging by sentinel lymph node biopsy to individualize therapy in anal cancer. Ann Surg Onc 2004; 11(3 suppl):259s–262s. Gupta N, Longo WE, Vernara AM, Wade TP, Johnson FE. Treatment of recurrent epidermoid carcinoma of the anal canal. Semin Colon Rectal Surg 1995; 6:160–165. Gordon PH. Squamous cell carcinoma of the anal canal. Surg Clin North Am 1988; 68:1391–1399. Tanum G, Tveit K, Karlsen KO, Hauer-Jensen M. Chemotherapy and radiation therapy for anal carcinoma. Cancer 1991; 67:2462–2466. Basik M, Rodriguez-Bigas MA, Penetrante R, Pitrelli NJ. Prognosis and recurrence patterns of anal adenocarcinoma. Am J Surg 1995; 169:233–237. Tarazi R, Nelson R. Adenocarcinoma of the anus. Semin Colon Rectal Surg 1995; 6:169–173. Beal KP, Wong D, Guillem JG, et al. Primary adenocarcinoma of the anus treated with combined modality therapy. Dis Colon Rectum 2003; 46: 1320–1324.

134. Hobbs CM, Lowry MA, Owen D, Sobin LH. Anal gland carcinoma. Cancer 2001; 92:2045–2049. 135. Morson BC, Sobin LH. Histologic Typing of Intestinal Tumors. Geneva: World Health Organization, 1976. 136. Jensen SL, Shokouh-Amiri MH, Hagen K, Harling H, Nielsen OV. Adenocarcinoma of the anal ducts: a series of 21 cases. Dis Colon Rectum 1988; 31: 268–272. 137. Perkowski PE, Surrells DL, Evans JT, Nopajaronsri C, Johnson LW. Anal duct carcinoma: case report and review of the literature. Am Surg 2000; 66: 1149–1152. 138. Fenger C. Anal canal tumors and their precursors. Pathol Annu 1988; 23:45–66. 139. Schaffzin DM, Stahl TJ, Smith LE. Perianal mucinous adenocarcinoma: unusual case presentations and review of the literature. Arch Path Lab Med 2001; 125:1074–1077. 140. Jensen SL, Nielsen OV. Anal duct carcinoma [editorial]. Dis Colon Rectum 1989; 32:355–357. 141. Getz SB Jr, Ough YD, Patterson RB, Kovalcik PJ. Mucinous adenocarcinoma developing in chronic anal fistula: report of two cases and review of the literature. Dis Colon Rectum 1981; 24:562–566. 142. Belkacemi Y, Berger C, Poortmans P, et al. Management of primary anal canal adenocarcinoma: a large retrospective study from the rare cancer network. Int J Radiat Oncol Biol Phys 2003; 56:1274–1283. 143. Abel ME, Chiu YSY, Russell TR, Volpe PA. Adenocarcinoma of the anal glands. Results of a survey. Dis Colon Rectum 1993; 36:383–387. 144. Tarazi R, Nelson RL. Anal adenocarcinoma: a comprehensive review. Semin Colon Rectal Surg 1994; 10:235–240. 145. Papagikos M, Crane CH, Skibber J, et al. Chemoradiation for adenocarcinoma of the anus. Int J Radiat Oncol Biol Phys 2003; 55:669–678. 146. Paterson C, Musselman L, Chorneyko K, Reid S, Rawlinson J. Merkel cell (neuroendocrine) carcinoma of the anal canal. Dis Colon Rectum 2003; 46:676–678. 147. Coquard R. Merkel cell carcinoma of the anal canal: importance of radiotherapy [Editorial]. Dis Colon Rectum 2004; 47:256–257. 148. Mason JK, Helwig EB. Anorectal melanoma. Cancer 1966; 19:39–50. 149. Thibault C, Sagar P, Nivatvongs S, Wolff BG. Anorectal melanoma: an incurable disease? Dis Colon Rectum 1997; 40:661-–668. 150. Cagir B, Whiteford MH, Topham A, Rakinic J, Fry RD. Changing epidemiology of anorectal melanoma. Dis Colon Rectum 1999; 42:1203–1208. 151. Ward MWN, Romano G, Nicholls RJ. The surgical treatment of anorectal malignant melanoma. Br J Surg 1986; 73:68–69. 152. Werdin C, Limas C, Knodell RG. Primary malignant melanoma of the rectum. Evidence for origination from rectal mucosal melanocytes. Cancer 1988; 61:1364–1370. 153. Quan SHQ. Malignant melanoma of the anorectum. Semin Colon Rectal Surg 1995; 6:166–168. 154. Antoniuk PM, Tjandra JJ, Webb BW, Petras RE, Milsom JW, Pazio VW. Anorectal malignant melanoma has a poor prognosis. Int J Colorectal Dis 1993; 8:81–86. 155. Chiu YS, Unni KK, Beart RW Jr. Malignant melanoma of the anorectum. Dis Colon Rectum 1980; 23:122–124. 156. Bullard KM, Tuttle TM, Rothenberger DA, et al. Surgical therapy for anorectal melanoma. J Am Coll Surg 2003; 196:206–211. 157. Malik A, Hull TL, Milsom J. Long-term survivor of anorectal melanoma. Report of a case. Dis Colon Rectum 2002; 45:1412–1417. 158. Cooper PH, Mills SE, Allen MS Jr. Malignant melanoma of the anus. Report of 12 patients and analysis of 255 additional cases. Dis Colon Rectum 1982; 25:693–703. 159. Wolff BG. Anorectal melanoma [Editorial]. Dis Colon Rectum 2002; 45:1415. 160. Brady MS, Kavolius JP, Quan SHQ. Anorectal melanoma. A 64 year experience at Memorial Sloan-Kettering Cancer Center. Dis Colon Rectum 1995; 38: 146–151. 161. Pessaux P, Pocard M, Elias D, et al. Surgical management of primary anorectal melanoma. Br J Surg 2004; 91:1183–1187. 162. Ward MW, Romano G, Nichollo RJ. The surgical treatment of anorectal malignant melanoma. Br J Surg 1996; 73:68–69. 163. Roumen RMH. Anorectal melanoma in the Netherlands: a report of 63 patients. Eur J Surg Onc 1996; 22:598–601. 164. Ross M, Pezzi C, Pezzi T, Meurer D, Hickey R, Balch C. Patterns of failure in anorectal melanoma. A guide to surgical therapy. Arch Surg 1990; 125: 313–316. 165. Slingluff CL Jr, Vollmer RT, Seigler HF. Anorectal melanoma: clinical characteristics and results of surgical management in twenty-four patients. Surgery 1990; 107:1–9. 166. Goldman S, Glimelius B, Pahlman L. Anorectal malignant melanoma in Sweden. Report of 49 patients. Dis Colon Rectum 1990; 33:874–877.

6

Transanal Techniques Santhat Nivatvongs

& Introduction, 327

& INTRODUCTION

& Rectal Biopsy, 327

Procedures such as rectal biopsy, electrocoagulation, and snaring of rectal polyps are usually performed in the office or an outpatient clinic. Thus the office or clinic must be well equipped with essential instruments and spare parts, and the clinician should be proficient in the use of these instruments. Providing sedation or anesthesia to the patient is rarely, if ever, necessary. If a barium enema study is indicated, it should be done one day before or three to four weeks after the procedure to avoid the risk of perforation. However, if a biopsy is performed on a carcinoma, a barium enema can be safely performed anytime thereafter because the rectal wall itself is not damaged by the biopsy. Anticoagulants, including aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) should be stopped for one week or until the prothrombin time and partial thromboplastin time have reached normal or nearnormal values. In many healthy people the colon contains a variable amount of hydrogen and methane (1). The packaged sodium biphosphate enema (Fleet enema) that is frequently used is not adequate to eliminate these gases, and explosions from electrocoagulation or snaring, although rare, can occur (2,3). Explosions can be prevented by the frequent use of suctioning to evacuate the gases before the electric current is applied. Patients are usually sent home soon after the procedure. No dietary restrictions are necessary. After removal of larger sessile lesions, a liquid diet may be in order for 24 to 48 hours. The patients are forewarned to report profuse rectal bleeding, persistent abdominal or shoulder pain, or fever. When the patient’s status is in doubt, a flat plate, an upright view of the abdomen, and lateral decubitus films should be obtained.

Technique, 328 Biopsy of Rectal Carcinoma, 328 Biopsy of Rectal Mucosa, 328 & Electrocoagulation of Rectal Polyps, 328

Technique, 328 & Snare Polypectomy, 328

Pedunculated Polyp, 328 Sessile Polyp, 328 & Transanal Excision of Rectal Adenoma, 328

General Considerations, 329 Sessile Adenoma of Lower Rectum, 329 Sessile Adenoma of Middle Rectum, 330 Circumferential Villous or Tubulovillous Adenoma of Rectum, 330 & Transanal Excision for Carcinoma

of the Low Rectum, 332 Preparation, 332 Technique, 332 & Posterior Approach to the Rectum, 332

Posterior Proctotomy Approach (Kraske’s Approach), 332 Posterior Transsphincteric Approach (York-Mason’s Approach), 335 & Transanal Endoscopic Microsurgery, 335 & Electrocoagulation of Carcinoma of

the Rectum, 337 Selection of Patients, 338 Preparation, 338 Technique, 338 Postoperative Management, 338 Failure of Electrocoagulation, 338 Results, 338 & References, 338

& RECTAL BIOPSY Many local and systemic disorders can be diagnosed by rectal biopsy. Conditions in which rectal biopsy is essential or useful for diagnosis or management include malignant or benign neoplasms of the rectum, chronic ulcerative colitis, Crohn’s colitis, ischemic colitis, pseudomembranous colitis, radiation proctitis, collagenous colitis, amebiasis, schistosomiasis, bacillary dysentery, pneumatosis cystoides intestinalis, amyloidosis, solitary ulcer of the rectum, Hirschsprung’s disease, and colitis cystica profunda. Biopsy specimens are inevitably small, rarely > 8 mm. However, with an adequate biopsy and proper orientation of the specimen, an accurate interpretation can be achieved. 327

328

& PART II: ANORECTAL DISORDERS

& TECHNIQUE For a rectal biopsy, the 1.9  25 cm proctoscope is used, and the patient is placed in the prone jackknife or a left lateral position. A cup-shaped or an alligator biopsy forceps is suitable (see Fig. 8 in Chapter 3). & BIOPSY OF RECTAL CARCINOMA The best area for taking a biopsy is the edge of the carcinoma where there is no necrosis. Multiple areas should be sampled. Bleeding is usually minimal but may require electrocoagulation. & BIOPSY OF RECTAL MUCOSA The lower valve of Houston is frequently thought to be the ideal location for biopsy. Other surgeons believe that the best location for biopsy is the posterior part of the middle valve of Houston. At the middle valve, if bleeding occurs, the proctoscope can be pressed against the sacrum to constrict blood vessels, and electrocoagulation can be applied accurately in a bloodless field (4). Frosted glass, nylon mesh, or filter paper is applied lightly to the submucosal surface of the specimen. The mounted specimen is then dropped into the fixative.

& ELECTROCOAGULATION OF RECTAL POLYPS During a routine proctoscopy it is common to find small rectal polyps, ranging in size from 1 to 5 mm. The majority are hyperplastic polyps or lymphoid hyperplasia and have no malignant potential. Small adenomatous polyps 1 to 5 mm in size have an extremely low incidence of carcinoma. However, because of the minimal risks of the complications during a properly performed excisional biopsy, all such polyps within the reach of the proctoscope should be obtained for biopsy so that the histologic morphology is known. If several polyps are present, biopsy is performed for one or two and electrocoagulation is performed for the rest.

& TECHNIQUE For electrocoagulation of rectal polyps, the patient is placed in either a prone jackknife position or a Sims’ position. A 1.9  25 cm proctoscope is suitable for the procedure. A ball-tip or a suction-coagulation electrode, which is connected to the electrocoagulation unit set at an optimal power, is used (see Fig. 6 in Chapter 3). After the biopsy, the electrode is placed directly in contact with the polyp to coagulate it. The duration of each burning should not be more than a few seconds, but this step can be repeated until the entire polyp becomes white. A ‘‘pop’’ noise may be heard, which results from the explosion of the cells due to the pressure of the steam generated within them.

& SNARE POLYPECTOMY The patient is placed in either a prone jackknife or a Sims’ position. A 1.9  25 cm proctoscope is preferred for snare polypectomy. Different commercial snaring devices are

available (see Fig. 7 in Chapter 3). They are connected to the coagulator, which is set at the appropriate power, depending on the size of the polyp. Before transection of a polyp, the stalk or the base must be visualized fully to avoid drawing the bowel wall into the snare. If the polyp is at the limit of the proctoscopic visualization and the base is not adequately visualized, it may be safer to perform a colonoscopic polypectomy.

& PEDUNCULATED POLYP The snare wire is looped around the pedunculated polyp and is positioned onto the stalk a few millimeters from the bowel wall. Once the loop is in a satisfactory position, the snare is pushed cephalad so that the base of the wire loop touches the stalk (Fig. 1A). This fixes the snare wire in the proper position for closing the loop. It always must be ascertained that no mucosal fold has inadvertently been caught in the snare (Fig. 1B). Before coagulation is done, the polyp is manipulated toward the center of the rectal lumen to avert the polyp from resting on the rectal wall. This maneuver avoids the possibility of burning the mucosa of the opposite wall because the current is transmitted through the polyp (Fig. 1C). The coagulator is applied no longer than a few seconds. The snare is gently tightened while the electric current is applied. If the polyp is not cut, approximately 10 to 15 seconds are allowed for the tissue to cool; then coagulation and tightening of the snare are repeated until the stalk of the polyp is completely transected. The polyp can be retrieved easily by using suction or a polyp grasper or a biopsy forceps. The polyp site should be checked for bleeding or perforation. & SESSILE POLYP A small sessile polyp can be snared in the same way as a pedunculated polyp by including a small part of the underlying mucosa tented as a pseudopedicle. A sessile polyp > 2 cm should be removed piecemeal. The snare wire cuts through the substance of the polyp to divide it into multiple pieces 1 to 1.5 cm in size (Fig. 1D). The residual polyp at the base can be electrocoagulated. It may be necessary to remove a large sessile polyp in more than one session. The patient should return after four to six weeks for examination and removal of any residual polyp.

& TRANSANAL EXCISION OF RECTAL ADENOMA A villous or a tubulovillous adenoma has a characteristic velvety appearance and soft consistency. If the neoplasm is not indurated or ulcerated, the chance that it is benign is 90% (5,6). The entire lesion should be removed in one piece or at least piecemeal in its entirety in the submucosal plane for an adequate histopathologic examination. A preoperative biopsy is unreliable and has high false positive and false negative reports of malignancy (7). It also makes a subsequent transanal excision more difficult. Indurated or ulcerated areas are signs of invasive carcinoma, and it is useful to obtain a biopsy of these areas. An adenoma in the low rectum and in some cases in the midrectum is suitable for transanal excision. The techniques used vary according to the size and site of the lesion (8).

CHAPTER 6: TRANSANAL TECHNIQUES &

329

FIGURE 1 & Snare polypectomy. (A) Base of snare loop is pushed against stalk of the polyp. (B) Mucosal fold is entrapped in the snare. (C) Electrical current is transmitted through the polyp to opposite bowel wall. (D) Piecemeal excision of a large sessile polyp.

& GENERAL CONSIDERATIONS All patients who have rectal villous or tubulovillous adenomas should be thoroughly examined by total colonoscopy to rule out other lesions located more proximally. The rectum is prepared with two Phospho-soda enemas. An antibiotic is not indicated. Inducing general or regional anesthesia is preferred, but in selected patients local anesthesia suffices. The patient is placed in the prone jackknife position, but if the adenoma is located on the posterior rectum, a lithotomy position may be preferred. A Pratt anal speculum provides excellent exposure, but other retractors, such as the Fansler anal speculum, the Sawyer anal retractor, the Hill-Ferguson retractor, and the Gelpi retractor should be available and used as appropriate (Figs. 2 and 3).

FIGURE 2 & Fansler anal speculum (above) and Pratt bivalve anal speculum (below).

Postoperative care is simple. A clear liquid diet is given for one day to ensure that there are no signs of rectal perforation. The first postoperative check should be in three months. Any residual polyp can be electrocoagulated in the clinic or office. Thereafter, proctosigmoidoscopy or flexible sigmoidoscopy at six months and then annually is desirable.

& SESSILE ADENOMA OF LOWER RECTUM A sessile adenoma with the proximal margin up to 7 cm from the anal verge can be excised in one piece. A scissors

FIGURE 3 & Hill-Ferguson anal speculum (above) and Sawyer anal speculum (below).

330

& PART II: ANORECTAL DISORDERS

the completion of the procedure to make sure that the lumen is patent. The excised specimen is pinned flat on a piece of cardboard and then placed in a fixative solution (Fig. 6).

FIGURE 4 & Infiltration of a villous adenoma, using 1:200,000 epinephrine solution to raise the submucosa.

or an electrocautery blade is used to raise the mucosa and submucosa from the underlying muscular wall of the rectum. A dilute epinephrine solution (1:200,000) is infiltrated into the submucosa to minimize bleeding (Fig. 4). A Pratt anal speculum is inserted to expose the anorectum. An incision is made to encompass a margin of about 1 cm of normal mucosa and submucosa. The distal margin of the adenoma is grasped with an Allis forceps for traction. A Sawyer or a Hill-Ferguson retractor may work better at this point to allow prolapse of the rectal wall. The entire lesion is then excised (Fig. 5). Depending on its size and shape, the wound may be completely closed, marsupialized, or left open. Proctosigmoidoscopy is performed at

& SESSILE ADENOMA OF MIDDLE RECTUM Excising a lesion in the middle rectum 7 to 11 cm from the anal verge is difficult, if not impossible, to do per anum. The anal speculum cannot reach the adenoma for adequate exposure. However, if the technique as described by Faivre is used, it is possible to remove such a lesion (9,10). The anorectal submucosa and the anal canal are infiltrated with 0.25% bupivacaine (Marcaine) or 0.5% lidocaine (Xylocaine) containing 1:200,000 epinephrine to obtain complete relaxation of the anal canal and hemostasis. A Fansler or a Pratt anal speculum is used to expose the anorectum. An elliptical excision is made with a scissors or an electrocautery blade, starting at the dentate line or at the anal verge similar to the technique for hemorrhoidectomy. The mucosa and the submucosa are dissected from their underlying muscle. With this anoderm or submucosal pedicle used as traction, the dissection can be carried up to 8 to 10 cm from the anal verge without much difficulty. When the lower margin of the polyp is reached, the dissection should encompass it with a 1 cm rim of normal mucosa (Fig. 7). If the anorectal wall does not prolapse when the submucosal pedicle is pulled, a Sawyer or Hill-Ferguson retractor should be used instead. The wound is closed with running 3-0 synthetic braided or monofilament sutures. A large wound should be closed transversely. A diverting colostomy is not indicated. Pigot et al. (10) performed this technique in 207 consecutive patients (100 males), mean age 68 years (range, 24–90 years), for an apparently benign villous rectal adenoma. The mean distance of lower edge of the lesion from the anal margin was 5.6 cm (range, 0–13 cm) and was < 10 cm in 82%. Immediate postoperative course was uneventful for 96%. The mean size of the resected lesion was 5.4 cm (range, 1–17 cm). & CIRCUMFERENTIAL VILLOUS OR TUBULOVILLOUS ADENOMA OF RECTUM Good exposure is essential for the removal of a large polyp in the rectum. A Lone Star self-retaining retractor (Fig. 8)

FIGURE 5 & (A) Flat villous adenoma of the low rectum. (B) Dissection of the adenoma in the submucosal plane with 1 cm margin of normal mucosa. (C) Transverse approximation of the wound with continuous sutures.

CHAPTER 6: TRANSANAL TECHNIQUES &

FIGURE 6 & Specimen is pinned on cardboard.

helps the exposure tremendously (11). If that instrument is not available, two Gelpi retractors placed at right angles to each other at the anal verge can be used as an alternative. A circumferential villous or tubulovillous adenoma with the lower margin in the lower rectum can be removed even if the proximal margin extends to the midrectum (Fig. 9A). Although Parks and Stuart (12) removed it in three to four longitudinal strips, this type of lesion can be removed in one piece as a submucosal tube. A circumferential incision is made around the dentate line in the submucosal plane with an electrocautery blade. Diluted epinephrine may be injected to raise the submucosal plane. Using a Pratt anal speculum to stretch the anal canal makes the dissection in the submucosal plane easier, all around up to about 7 cm. The submucosal tube is grasped with an Allis clamp, and a submucosal

331

dissection is carried as far proximally as needed. The circumferential dissection of the submucosa apparently allows the rectum to prolapse during traction. It is relatively easy to dissect the submucosal tube from the dentate line to the anorectal ring. From the anorectal ring to the rectal ampulla (approximately 10 cm from the anal verge), the dissection is difficult, and the submucosal plane may be easily lost. Packing sponges into the submucosal tube facilitates traction on the tube during the dissection. A Pratt speculum, Sawyer or Hill-Ferguson retractor, and a small Deaver retractor should be used freely as appropriate. From the rectal ampulla to upper rectum, dissection is usually easier. After the entire adenoma has been removed, the proximal cut end of the mucosa and submucosa can be brought down, along with denuded muscle wall, to approximate circumferentially with the lower cut end at the dentate line (Fig. 9B) At the completion of the procedure, the anorectal wall will be imbricated in the longitudinal plane (Fig. 9C). Incontinence of gas and liquid stool may last for a few weeks. Circumferential submucosal dissection of the anorectum can be carried from the dentate line up to 10 cm, and occasionally up to 12 to 15 cm, from the anal verge.

FIGURE 7 & (A) Sessile adenoma in the middle rectum. Elliptical excision starts at the dentate line. (B) With traction to prolapse the anorectal mucosa, the specimen is completely excised. (C) Closure of wound with running absorbable sutures. Larger wound should be closed transversely.

332

& PART II: ANORECTAL DISORDERS

FIGURE 8 & An excellent exposure of the anus using a Lone Star selfretaining retractor. Patient is in prone position.

& TRANSANAL EXCISION FOR CARCINOMA OF THE LOW RECTUM Certain carcinomas of the low rectum have not yet spread to the regional lymph nodes at the time of diagnosis. Thus they are amenable to local excision provided the extirpation can be done with adequate margin in all directions. From a technical point of view, suitable lesions are those that are < 3 cm in greatest dimension, invade only the submucosa and have a favorable pathologic grade (grade 1 or 2). Although preoperative intrarectal ultrasonography is useful to determine the depth of the invasion and the lymph node metastasis, the ultimate decision whether additional treatment is indicated relies on the complete histopathologic examination. Approximately 5% of carcinomas of the rectum are suitable for local excision with the intent to cure (13). With proper selection of specific patients in conjunction with meticulous and precise surgical technique, the carcinoma five-year survival rate can be expected to be 74% to 90% (14–17). This approach should be used with caution because it has been recently reported that local recurrence rate is high, between 7% and 29% (14–17) (see Chapters 22 and 24).

an electrocautery blade (coagulation current). The Allis clamp is used for traction and exposure and should be moved around as appropriate. This technique is simpler and more precise than using multiple sutures around the lesion for traction. The tissue is incised all around in the perianorectal plane (deeper then the internal sphincter and rectal wall) until the entire carcinoma is removed. It is important that the excision is full thickness, exposing the fat (Fig. 10C). The wound is closed transversely with running 3-0 or interrupted monofilament or braided synthetic absorbable material (Fig. 10D). No packing or drain is placed. The electrocautery allows an accurate excision with minimal bleeding. The specimen is carefully placed on a towel and marked for locations, right, left, caudad, and cephalad aspects. The pathologist then checks the margins to determine that they are adequate. If adequate margins cannot be achieved, and if the carcinoma has adverse risk factors a radical resection should be considered in goodrisk patients. Postoperative care is generally simple. The anorectal pain is usually minimal, although mild fever can occur. The Foley catheter is removed on the first postoperative day. Clear liquids are started on the first postoperative day, provided there are no signs of peritoneal irritation, followed by a regular diet the next day and discharge from the hospital. A follow-up proctoscopy or flexible sigmoidoscopy should be performed at three and six months. The patient then begins a routine follow-up schedule appropriate for high-risk patients.

& POSTERIOR APPROACH TO THE RECTUM

& PREPARATION The patient should undergo mechanical bowel preparation similar to a colon resection, including prophylactic antibiotics. The prone jackknife position is for lesions in the anterior and lateral locations. For lesions in the posterior rectum, the lithotomy position with candy cane stirrups is preferred.

A large villous or sessile adenoma of the high rectum, with the lower margin more than 10 or 11 cm from the anal verge, is impossible to excise via a transanal approach unless the rectum can be prolapsed through the anus. This higher lesion, however, can be reached by opening the rectum posteriorly. The bowel is prepared as for a colon resection. Because of the potential serious complications, the posterior approach to the rectum should be reserved for cases in which other alternatives are not suitable. In good-risk patients, a low anterior resection can be performed. Other alternatives include piecemeal snaring and electrocoagulation of the adenoma via a proctoscope or a colonoscope and using laser ablation (18). The posterior approach is most useful for local excision of neoplasms in the middle rectum.

& TECHNIQUE A Foley catheter is placed in the bladder. The anorectum is irrigated with saline solution. Exposure of the anorectum is achieved by a Pratt anal speculum or a Parks retractor. Transanal excision of carcinoma is suitable for an early lesion with an upper margin within 7 cm from the anal verge. A line of excision is marked around the carcinoma, with at least a 1cm grossly normal margin, using electrocautery (Fig. 10A). The crucial part of the technique is to grasp the internal sphincter muscle at the inferior margin with an Allis clamp (Fig. 10B). An incision is made using

& POSTERIOR PROCTOTOMY APPROACH (KRASKE’S APPROACH) In 1874 Theader Kocher (19) of Bern recommended excising the coccyx to facilitate access to the rectum when excision of a high rectal neoplasm was contemplated. Kraske, however, found that removal of the coccyx alone is not adequate to expose the upper rectum. In 1885 he reported his posterior approach technique. Kraske found that he could obtain an excellent exposure of the upper rectum by cutting the lower margin of the gluteus maximus muscle and the sacrospinous and sacrotuberous ligaments. The

CHAPTER 6: TRANSANAL TECHNIQUES &

333

FIGURE 9 & (A) Circumferential villous adenoma of the rectum. (B) Starting at the dentate line, entire adenoma is excised circumferentially in the submucosal plane with electrocautery. Denuded muscle wall is plicated with interrupted absorbable sutures. (C) At completion the anorectal wall is pleated.

FIGURE 10 & (A) The line of excision is outlined with electrocautery. (B) Incision is made in full thickness. Allis clamp is used for traction and exposure. (C) A full-thickness disk of the lesion is removed, exposing the perirectal fat. (D) The defect is closed transversely with running 3-0 monofilament or braided synthetic absorbable material.

334

& PART II: ANORECTAL DISORDERS

exposure was even better when the lowermost part of the left wing of the sacrum was excised (20). The patient is placed in a prone jackknife position. A transverse or a midline incision is made from the sacrococcygeal joint to a point just proximal to the anus (Fig. 11A). The anococcygeal ligament is separated along the midline (Fig. 11B). If necessary, the coccyx can be excised to gain

exposure. The incision is deepened to expose the levator ani muscle, which is incised in the midline to expose the posterior rectal wall (Fig. 11B and C). The rectum is opened through the posterior wall, and the polyp is excised submucosally (Fig. 11D and E). The defect is approximated, if possible, using 3-0 sutures (Fig. 11F). The posterior proctotomy wound is closed vertically or transversely with a

FIGURE 11 & Kraske’s approach. (A) Patient is placed in prone position. A midline or a transverse skin incision is used. (B) Incision through skin, with division of anococcygeal ligament. (C) Levator ani muscle is divided in the midline plane. Coccyx can be excised as indicated. (D) Posterior rectal wall is incised, exposing adenoma in the anterior rectal wall. (E) Adenoma is excised in the submucosal plane. (F) Closure of anterior rectal mucosa. (G) Closure of posterior rectal wall. (H) Closure of levator ani muscle. (I) Approximation of anococcygeal ligament.

CHAPTER 6: TRANSANAL TECHNIQUES &

one-layer synthetic absorbable 3-0 suture, depending on which method is easier to accomplish without compromising the lumen (Fig. 11G). A two-layer closure may be performed if desired. Hemostasis is achieved, and the entire wound is irrigated with normal saline solution. A closed suction drain is placed. The muscles are closed in layers using 3-0 sutures (Fig. 11H and I). The skin is closed with a subcuticular 3-0 monofilament synthetic absorbable sutures. A classic Kraske’s approach, excising the coccyx and part of the lower sacrum is seldom practiced anymore. However, when needed, one should not hesitate to excise the coccyx. There have been sporadic reports on Kraske’s approach for rectal neoplasm. Arnaud et al. (21) used the modified Kraske’s approach in 35 patients with various anorectal diseases, including 24 patients with villous adenomas of the rectum situated between 8 and 15 cm from the anal verge. Significant complications included four patients who developed wound infection, three patients who developed a fistula requiring a sigmoid colostomy, and two patients who developed a perianal hernia that did not require surgical repair. There was no postoperative mortality. Christiansen (22) operated on 17 patients with various neoplasms of the rectum located between 6 and 15 cm from the anal verge. No operative deaths resulted. Two patients developed wound infections, and two had fecal fistulas that spontaneously closed. Bergamaschi and Arnaud (23) successfully resected the midrectum with a circular stapled anastomosis for circumferential adenomas of the midrectum. The coccyx and the fifth and fourth sacrum were osteotomized to fully mobilize the rectum.

& POSTERIOR TRANSSPHINCTERIC APPROACH (YORK-MASON’S APPROACH) The posterior transsphincteric approach of the rectum was first introduced by Bevan (24) in 1917 but lacked popular success. In 1970 York-Mason (25) reintroduced this technique and subsequently reported satisfactory results with 89 patients in 1977. He strongly emphasized accurately suturing the divided sphincters, with restoration of the anatomic layers. This enthusiasm was also shared by Allgower et al. (26). The patient is placed in the prone jackknife position. A left parasacral incision is made from middle sacrum, passing obliquely downward to the anal verge in the midline (Fig. 12A). The exposure can be widened by incising the lower part of the gluteus muscle. The incision is deepened to expose the external sphincteric complex and the puborectalis and levator ani muscles. They are divided along the lines of incision, and the divided edges are tagged step by step (Fig. 12B). It is important to mark each layer and component of the muscles accurately with stay sutures to allow correct identification when reapproximating them. The nerve supply to these muscles lies lateral to the incision and is therefore safe from injury. Next the internal sphincter is divided, with proximal extension to the thinner muscle wall of the rectum. The submucosa and mucosa are incised to expose the interior of the rectum and the anal canal (Fig. 12C). The lesion is excised (Fig. 12D). The bowel is closed in layers. The mucosa and submucosa are closed with running or interrupted 3-0 synthetic absorbable sutures; the internal

335

sphincter is closed with interrupted or running 3-0 braided or monofilament synthetic sutures (Fig. 12E). The external sphincter and the puborectalis and levator ani muscles are closed with the same kind of sutures. A suction drain is placed in the subcutaneous space, and the skin is closed with subcuticular 4-0 absorbable sutures. York-Mason’s approach is used more widely than Kraske’s approach because of its superior exposure. Despite cutting the sphincteric muscle during the procedure, anal incontinence has not been a problem. York-Mason excised or resected carcinomas of the rectum via the transsphincteric approach in 50 patients and removed villous adenomas in another 39 patients. The results were satisfactory, but the outcome of the operation was not given (25). Heij et al. (28) and Bergman and Solhaug (29) used this technique on 6 and 21 patients, respectively. There was no mortality. No patients developed a fecal fistula, sepsis, or anal incontinence. Thompson and Tucker (30) operated on 26 patients; one patient developed temporary anal incontinence. Fecal fistulas occurred in seven patients. Five closed spontaneously and two patients required a temporary colostomy. Allgower et al. (26) reported no deaths in 79 patients, but complications were not mentioned. The middle and upper rectum can also be mobilized and resected with the York-Mason approach. Either a handsewn or a circular stapled anastomosis can be performed (31). The posterior approach is seldom required. A low rectal lesion can be excised with a transanal approach. The upper rectal lesion is too high for this approach. The midrectal lesion at the 7 to 10 cm level may be appropriate. I prefer to snare the benign lesion, even a large sessile type, via a colonoscope. For a malignant lesion, a low anterior resection is more appropriate unless there are contraindications.

& TRANSANAL ENDOSCOPIC MICROSURGERY Transanal endoscopic microsurgery (TEM) is a relatively new operative technique to remove sessile lesions of the rectum, developed by Buess et al. (32) of Germany. It has been applied clinically since 1983. The system uses a special rectoscope, 4 cm in diameter and 10 or 20 cm long. The scope is connected to the operative table via a supporting arms system, which can be adjusted to any desired position. The scope has a closed system of sealed caps with individual ports for forceps, suction, scissors, needle holder, and electrocautery. It is connected to a stereoscopic angulated optical system for visualization and can be hooked up to a television screen (Fig. 13). During the procedure, a continuous pressure-controlled insufflation of carbon dioxide keeps the rectum open for exposure. The patient is placed in a prone lateral or lithotomy position, according to the location of the polyp. The procedure is used for lesions in the middle and high rectum and occasionally in the low sigmoid colon up to 20 cm from the anal verge. The excision is done by electrocautery. Lesions located in the extraperitoneal part of the rectum are excised by full thickness. Only adenomas of the anterior wall located higher than 12 cm are excised in the submucosal plane. Invasive carcinomas of this location cannot be adequately treated because a full-thickness excision loses

336

& PART II: ANORECTAL DISORDERS

FIGURE 12 & York-Mason’s transsphincteric approach. (A) With patient in prone position, a parasacral skin incision is used. (B) Incision is made through subcutaneous tissue, external sphincter muscles, and levator ani muscle. Each pair of muscles is marked on each side with sutures for accurate approximation. Posterior wall of rectum is exposed. Part of the gluteus muscle is incised to gain exposure. (C) Posterior wall of rectum is incised to expose the lumen. (D) Adenoma is excised in the submucosal plane. (E) Posterior wall of rectum is closed with running absorbable sutures, followed by closure of the wound in layers. Source: From Ref. 27.

gas into the peritoneal cavity. The defect is closed with 3-0 synthetic monofilament suture, which is cut to 6 cm in length. A silver clip is used instead of tying a knot (Fig. 14). TEM is suitable for a T1 lesion of the rectum. Buess et al. (33) reported their experience in 326 patients with only one death from a pulmonary embolism. There was no intraoperative complication. Seventy-four patients had an invasive adenocarcinoma. The recurrence rate in the adenoma group was 4%. Twenty-nine of the 74 patients had T1 lesions that did not receive further treatment, and only one patient developed a recurrence. Major complications included breakdown of suture lines and rectovaginal fistulas. Complications occurred in 9% of patients who had invasive carcinoma. The TEM technique has gained acceptance worldwide, including the United States (34,35). Smith et al. (35) reported the initial registry in the United States from 1990 to 1994. There were 153 patients: 82 with an adenoma, 54

with a carcinoma, and 17 with other entities. Approximately 50% of the lesions were located in the lower half of the rectum (below 8 cm). The sizes of the lesions ranged from 1 to 15 cm. The recurrence rate for adenomas was 11%. According to the TNM staging system, 30 of the malignant lesions were classified as T1, 15 were T2, and 6 were T3 carcinomas. There were three T1 recurrences (10%), six T2 recurrences (40%), and four T3 recurrences (66%). The intraoperative complications were surprisingly low. Eight procedures were converted to celiotomy because of the lack of proximal exposure. There was one perforation and one massive hemorrhage. Early postoperative complications were minor. A review of selected series in the literature by Saclarides (36) on T1 carcinoma of the rectum treated by TEM showed recurrent rates of 0–12.5%, 5-year disease free survival of 96–100%, and 5-year survival of 96-100%. Middleton et al. (37) conducted a systematic review of comparative studies and the case of TEM from 1980 to 2002.

CHAPTER 6: TRANSANAL TECHNIQUES &

FIGURE 13 & Operative system for transanal endoscopic microsurgery.

Three comparative studies (including one randomized, controlled trail) and 55 case series were included. At the first area of study was a safety and efficacy of adenomas. In the randomized, controlled trial, no difference could be detected in the rate of early complications between the TEM (10.3%) and the direct local excision (17%) (relative risk, 0.61; TEM resulted in less local recurrence (6/98; 6%) than direct local excision (20/90; 22%) (relative risk, 0.28). The 6% rate of local recurrence for TEM in this trial is consistent with the rates found in case series of TEM (median, 5%). The second area of study was a safety and efficacy of carcinomas. In the randomized, controlled trial, no difference could be detected in the rate of complications between TEM and direct local excision (relative risk for overall early complication rates, 0.56). No difference in survival for local recurrence rate between TEM and anterior resection could be detected in either the randomized, controlled trial (hazard ratio, 1.02

337

for survival) or nonrandomized, comparative study. There were 2 of 25 (8%) TEM recurrences in the randomized, controlled trial, but no figures were given for recurrence after anterior resection. In the case series, the median local recurrence rate for TEM was 8.4%, ranging from 0% to 50%. The authors concluded that the evidence regarding TEM is very limited, being largely based on a single relatively small randomized, controlled trial. Despite the large size of the operating scope (4 cm diameter), only three patients had mild incontinence (35). A prospective study to evaluate the anorectal function was conducted by Banerjee et al. (38). The evaluation included the clinical, manometric, and proctographic results of 36 patients presenting for TEM. Anorectal manometry showed no difference in maximal squeeze pressure before and 12 months after operation, but resting pressures were lower after operation (mean, 86.1 vs. 67.2; p < 0.5). The rectoanal inhibitory reflex was lost in a significant number of patients (34 of 36 before operation, 27 of 36 after operation, p < 0.05). The results of proctography showed no significant difference. Clinically, continence to solid stool or flatus was not affected after the operation. Completeness of emptying was not affected by TEM. This is in agreement with the prospective study by Cataldo et al. (39). Surgeons who have attempted to learn TEM agree that this is the most difficult technique that they have tried to master (35,40). The problem lies with an inability to use the usual muscles of the arms and shoulders that are used in either open or laparoscopic surgery. In TEM the surgeon’s fingers and wrists are the primary movers of the instruments, so a new coordination must be developed (40,41). Any one wishing to incorporate TEM into their practice must devote significant time and energy to overcome a steep learning curve (40). TEM is a choice in the armamentarium that can obviate the need to perform a more radical operation, such as low anterior resection, and the approaches of Kraske or York-Mason. According to Saclarides (36), currently, there were about 510 TEM systems in use worldwide. As expected, it was most commonly used in the United Kingdom and Europe where approximately 110 units were present in Germany and 300 were in use in England and remaining areas of the continent. In Japan and Southeast Asia, approximately 58 units were in use. In the United States, 45 systems were functional; however, usage has been slow to catch on for a variety of reasons.

& ELECTROCOAGULATION OF CARCINOMA OF THE RECTUM

FIGURE 14 & Technique of TEM excision. (A) Extent of excision is marked by electrocautery. (B) Submucosal dissection of the polyp using electrocautery and forceps. A suction tube is placed close at hand. (C) Defect is closed with continuous sutures. Note that a silver clip is used as a knot.

Since the report by Strauss et al. (42) in 1935 on the excellent results with electrocoagulation of carcinoma of the rectum, there have been sporadic reports of using this approach. In 1957, Madden and Kandalaft (43) reported on a small series of patients with carcinoma of the rectum who were treated by electrocoagulation as the primary and preferred method. Other investigators also reported favorable results in selected cases (44–46). Use of this method of treatment was stimulated by the less than satisfactory long-term survival rate and the high morbidity

338

& PART II: ANORECTAL DISORDERS

and mortality rates, especially in the elderly, with abdominoperineal resection. With modern advances in medicine, the morbidity and mortality rates for major operations have dropped dramatically. Indeed, the mortality rate for elective surgery for carcinoma of the rectum is reported as low as < 1% (47). Electrocoagulation should be reserved for palliative treatment of rectal carcinoma or for poor surgical risk patients. However, there are proponents who use electrocoagulation as the a primary treatment for care in highly selected patients (46).

& SELECTION OF PATIENTS Chinn and Eisenstat (46) have outlined the criteria for electrocoagulation of distal rectal carcinomas, with the intention for cure: & & & & &

Well- or moderately well-differentiated adenocarcinoma Located 7.5 cm from the anal verge (below peritoneal reflection) 4 cm in diameter Absence of pelvic/pararectal adenopathy Stage T1 or T2 on endorectal ultrasonography

The obvious advantages of electrocoagulation for carcinoma of the lower part of the rectum are the 0% mortality rate and the avoidance of a permanent colostomy (48). The disadvantages are that electrocoagulation may fail to eradicate the primary malignancy, and that electrocoagulation fails to cure patients who have lymph node metastasis.

Preparation Electrocoagulation of rectal carcinoma is not an office procedure. The bowel is prepared in the same manner as for a colonic resection. All patients should be hospitalized after the procedure. Spinal or general anesthesia is required, and the operative time may range from one to two hours. Technique The prone jackknife position is used for carcinomas located anteriorly and laterally. The lithotomy position is used for posterior ones. A Pratt speculum is used to visualize lowlying lesions, whereas an operative proctoscope is used for higher ones. Exposure is the key to success. The entire lesion must be seen. In the low-lying region, saline injection into the rectal wall adjacent to the carcinoma is helpful to protect thermal injury. A needlepoint electrode is used in preference to the ball-tip type because of its greater heat production and deeper penetration. A spatula tip, which is available in almost every operating room, is also effective. The coagulator is set at an appropriate power. The cautery tip is embedded in the carcinoma, with the depth dependent on the extent of penetration into the wall. The coagulator is applied repeatedly in an orderly manner until the entire carcinoma becomes flat and soft. Suction is used frequently to evacuate the smoke and the possible explosive gases. A curette or a biopsy forceps can be used to remove the coagulum. Carcinomas located in the posterior or lateral quadrant of the lower and middle portions of the rectum can be coagulated vigorously because they are extraperitoneal, whereas lesions located in the anterior wall of the middle rectum must be treated more cautiously because of their intraperitoneal location. In women, carcinomas of

the anterior wall may require coagulation in multiple stages to avoid the development of a rectovaginal fistula.

Postoperative Management Patients usually have no significant pain following operation. Fever from the reaction to the coagulation may last for several days and should be treated only symptomatically. Only clear liquids are allowed for the first day until it is certain that no perforation has occurred. When the patient has advanced to a regular diet, a bulk-producing agent is prescribed. The patient should be reexamined in the clinic two months after the coagulation. A biopsy should be obtained of any suspicious area. If the area is malignant, repeat coagulation is performed. It may be necessary to repeat the coagulation several times for complete eradication of the malignancy. Failure of Electrocoagulation The site of the electrocoagulation must be reexamined by proctoscopy every two to three months during the first year and every three to six months during the second year. If coagulation fails to control the lesion and if digital examination or endorectal ultrasonography strongly suggests lymph node metastasis, an abdominoperineal resection or a low anterior resection is indicated. In the series by Chinn and Eisenstat (46), 31 of 114 patients required conversion to an abdominoperineal resection because of recurrences. The five-year survival rate in that group of patients was 29%, with almost 90% having lesions originally > 4 cm, and 21 of 31 patients demonstrating nodal metastasis. Results Madden and Kandalaft (48) reported on a series of 204 patients. Electrocoagulation was used as the primary curative method, with a follow-up of 5 to 27 years. The lesions included those in 12 patients with squamous cell carcinoma of the anal canal. The carcinomas were 2 to 10 cm from the anal verge. Fifty-nine percent of the patients required two to four electrocoagulations. No deaths resulted, and the overall complication rate was 24%. Among the frequent complications were bleeding 16%, half of which stopped spontaneously; rectal stricture, 2%; rectovaginal fistula, 2%; and perforation, 1%. The overall five-year survival rate was 64%, with 72% for lesions < 3 cm in diameter and 62% for lesions > 3 cm. Of the 81 patients treated with an intention for cure in the study by Chinn and Eisenstat (46), the five-year survival rate was 65%. The complications of electrocoagulation included bleeding, 7%; stricture, 6%; urinary retention, 2.6%; electrical burns, 2.6%; perianal abscess, 0.9%; and perforation; 0.9%. The overall mortality rate was 2.7%.

& REFERENCES 1. Levitt MD. Volume and composition of human intestinal gas determined by means of an intestinal washout technic. N Engl J Med 1971; 284:1394–1398. 2. Bond JH, Levy M, Levitt MD. Explosion of hydrogen gas in the colon during proctosigmoidoscopy. Gastrointest Endosc 1976; 23:41. 3. Carter HG. Explosion in the colon during electrodessication of polyps. Am J Surg 1952; 84:514.

CHAPTER 6: TRANSANAL TECHNIQUES &

4. Nivatvongs S. Technique of rectal biopsy. Dis Colon Rectum 1981; 24:132. 5. Galandiuk S, Fazio VW, Jagelman DG, et al. Villous and tubulovillous adenomas of the colon and rectum. A retrospective review, 1964–1985. Am J Surg 1987; 153:41–47. 6. Nivatvongs S, Nicholson JD, Rothenberger DA, et al. Villous adenomas of the rectum: The accuracy of clinical assessment. Surgery 1980; 87:549–551. 7. Taylor EW, Thompson H, Oates GD, et al. Limitations of biopsy in preoperative assessment of villous papilloma. Dis Colon Rectum 1981; 24:259–262. 8. Sakamoto G, MacKeigan JM, Senagore AJ. Transanal excision of large, rectal villous adenomas. Dis Colon Rectum 1991; 34:880–885. 9. Faivre J. Transanal electro-resection of rectal tumor by means of tractable mucocutaneous anal flap. Coloproctology 1980; 2:77–80. 10. Pigot F, Bouchard D, Mortaji M, et al. Local excision of large villous adenomas. Long-term results. Dis Colon Rectum 2003; 46:1345–1350. 11. Roberts PL. Mucosectomy for treatment of rectal neoplasia. Semin Colon Rectal Surg 1996; 7:210–214. 12. Parks AG, Stuart AE. The management of villous tumors of the large bowel. Br J Surg 1973; 9:688–695. 13. Nivatvongs S, Wolf BG. Technique of per anal excision for carcinoma of the low rectum. World J Surg 1992; 163:447–450. 14. Madbouly KM, Remji FH, Erkek BA, et al. Recurrence after transanal excision of T1 rectal cancer: Should we be concerned: Dis Colon Rectum 2005; 48:711–721. 15. Nascimberi R, Nivatrongs S, Larson DR, Burgart LJ. Long-term survival after local excision for T1 carcinoma of the rectum. Dis Colon Rectum 2004; 47:1773–1779. 16. Paty PS, Nash GM, Barm P, et al. Long-term results of local excision for rectal cancer. Ann Surg 2002; 236:522–530. 17. Mellgren A, Sirivongs P, Rothenberger DA, Modoff RD, Garcia-Aguilar J. Is local excision adequate therapy for early sectal cancer? Dis Colon Rectum 2000; 43:1064–1074. 18. Brunetaud JM, Maunoury V, Cochelard D, Boniface B, Cartot A, Paris JC. Endoscopic laser treatment for rectosigmoid villous adenoma: Factors affecting the results. Gastroenterology 1989; 97:272–277. 19. Kocher TD. Exstirpatic Recti Nach Vorheriger Excision des Steissbeins. Cent Chir 1874; 10:145–147. 20. Perry EG, Hinrichs B. A new translation of Professor Dr. P. Kraske’s Zur Exstirpation Hochsitzender Mastdarmkrebse. Aust N Z J Surg 1989; 59:421–424. 21. Arnaud A, Fretes IR, Joly A, Sarles JC. Posterior approach to the rectum for treatment of selected benign lesions. Int J Colorectal Dis 1991; 6:100–102. 22. Christiansen J. Excision of mid-rectal lesion by the Kraske sacral approach. Br J Surg 1980; 67:651–652. 23. Bergamaschi R, Arnaud A. Management of large encircling mid-rectal adenomas in frail patients. Int J Colorectal Dis 1995; 10:53–54. 24. Bevan AD. Carcinoma of the rectum—Treatment by local excision. Surg Clin North Am 1917; 1:1233–1240. 25. Mason AY. Transsphincteric approach to rectal lesion. Surg Annu l977; 9:171–194. 26. Allgower M, Durig M, Hochstetter AV, Huber A. The parasacral sphincter-splitting approach to the rectum. World J Surg 1982; 6:539–548.

339

27. York-Mason A. In: Malt R, ed. Surgical techniques illustrated. Boston: Little, Brown, 1977, pp 75–77. 28. Heij HA, Tan KG, Houten HV. The transsphincteric approach to rectal villous adenomas. Neth J Surg 1982; 34:4–7. 29. Bergman L, Solhaug JH. Posterior transsphincteric resect small tumors of the lower rectum. Acta Chir Scand 1986; 152:313–316. 30. Thompson BW, Tucker WE. Transsphincteric approach to lesions of the rectum. South Med J 1987; 80:41–43. 31. Kusunoki M, Yanogi H, Shoji Y, Sakanoue Y, Yamamura T, Utsunomiya J. Stapled anastomosis for trans-sphincteric resection of the rectum. Surg Gynecol Obstet 1991; 173:325–326. 32. Buess G, Kipfmuller K, Ibald R, et al. Clinical results of transanal endoscopic microsurgery. Surg Endosc 1988; 2:245–250. 33. Buess G, Mentzes B, Manncke K, Starlinger M, Becker HD. Technique and results of transanal endoscopic microsurgery in early rectal cancer. Am J Surg 1992; 163:63–70. 34. Mayer J, Mortensen NMcC. Transanal endoscopic microsurgery: A forgotten minimally invasive operation. Br J Surg 1995; 821:435–437. 35. Smith LE, Ko ST, Saclarides T, Caushaj P, Orkin BA, Khanduja KS. Transanal endoscopic microsurgery. Initial registry. Dis Colon Rectum 1996; 39:579–584. 36. Saclarides TJ. Transanal endoscopic microsurgery. Semin Colon Rectal Surg 2005; 16:20–25. 37. Middleton PF, Sutherland LM, Madder GJ. Transanal endoscopic microsurgery: A systemic review. Dis Colon Rectum 2005; 48:270–284. 38. Banerjee AK, Jehle BC, Kreis ME, et al. Prospective study of the proctographic and functional consequences of transanal endoscopic microsurgery. Br J Surg 1996; 83:211–213. 39. Cataldo PA, O’Brien S, Oster T. Transanal endoscopic microsurgery: A prospective evaluation of functional results. Dis Colon Rectum 2005; 48:1366–1371. 40. Buess GF, Raestrup H. Transanal endoscopic microsurgery. Surg Oncol Clin N Am 2001; 10:709–731. 41. Smith LE. Transanal endoscopic microsurgery for selected rectal tumors. Semin Colon Rectal Surg 1996; 7:221–226. 42. Strauss AA, Strauss SF, Crawford RA, Strauss HA. Surgical diathermy of carcinoma of the rectum: Its clinical end results. JAMA 1935; 104:1480–1484. 43. Madden JL, Kandalaft S. Electrocoagulation. A primary and preferred method of treatment for cancer of the rectum. Ann Surg 1967; 166:413–419. 44. Crile G Jr., Turnbull RB. The role of electrocoagulation treatment of carcinoma of the rectum. Surg Gynecol Obstet 1972; 135:391–396. 45. Hughes EP Jr., Veidenheimer MC, Corman ML, Collet JA. Electrocoagulation of rectal cancer. Dis Colon Rectum 1982; 25:215–218. 46. Chinn B, Eisenstat TE. Electrocoagulation of selected adenocarcinoma of the distal rectum. Semin Colon Rectal Surg 1996; 7:233–237. 47. Pollard CW, Nivatvongs S, Rojanasakul A, Ilstrup DM. Carcinoma of the rectum: Profiles of intraoperative and early postoperative complications. Dis Colon Rectum 1994; 37:866–874. 48. Madden JL, Kandalaft SI. Electrocoagulation as a primary curative method in the treatment of carcinoma of the rectum. Surg Gynecol Obstet 1983; 157:164–179.

III: Minimally Invasive Surgery

7

Laparoscopic Colon and Rectal Surgery Lee E. Smith and Philip H. Gordon

& Background and Rationale, 342

Total Abdominal Colectomy and Ileorectal Anastomosis, 362

Historical Review, 342 Advantages, 342 Disadvantages, 343 Patient Information, 343

& Robotics, 363 & Postoperative Care, 364

& Indications, 343

& Results, 364

Carcinoma, 344 Benign Neoplasms, 344

Conversion Rates, 364 Postoperative Ileus, 365 Laparoscopic Procedures in the Elderly, 366 Learning Curve, 366 Morbidity and Mortality, 368

& Equipment and Instrumentation, 344

Laparoscope, 344 Video Camera, 344 Video Monitors, 344 Recording Media, 345 Insufflators, 345 Power Instruments, 346 Insufflation Needles, 346 Cannulas and Trocars, 346 Hand-Assisted Ports, 346 Operating Instruments, 347

Combined Reports, 368 Carcinoma, 371 Stomas, 379

Hand-Assisted Techniques, 380 Obesity, 381 Quality of Life, 381 Cost Issues, 382 & Laparoscopic Complications and Their Prevention, 383

Retractors, 347 Graspers, 347 Clip Appliers, 347 Staplers, 348 Needles and Sutures, 348

Trocar Complications, 383 Visceral Injury, 384 Vascular Injury, 384 Urinary Tract Injury, 384 Abdominal Wall (Port-Site) Recurrence, 384 Bleeding, 385 Anastomotic Dehiscence, 385 Missed Lesions, 386 Electrosurgical Injury, 386 Ultrasonic Energy Injury, 386 Wound, 386 Miscellaneous, 387 Nonspecific Complications, 387

& Operative Procedure, 348

Preoperative Preparation, 348 Preoperative Evaluation, 348 Positioning and Readying the Patient, 348 Operating Room Setup and Conduct of Operation, 348 Pneumoperitoneum, 349 Gasless Laparoscopy, 350 Trocar–Cannula Insertion, 350 Additional Ports, 351

& References, 387

Hand-Assisted Port, 351

Specimen Localization, 351 Completion of Laparoscopy, 353 & Specific Colorectal Procedures, 353

Diagnostic Laparoscopy, 353 Right-Sided Colectomy, 353 Left-Sided Colectomy, 355 Abdominoperineal Resection, 359 Stoma Creation, 360 Stoma Closure, 361

341

342

& PART III: MINIMALLY INVASIVE SURGERY

& BACKGROUND AND RATIONALE & HISTORICAL REVIEW Laparoscopic approaches to operation vaulted into our surgical armamentarium much faster than other technologic innovations and without the clinical scrutiny of other therapeutic modalities. For example, laparoscopic cholecystectomy catapulted from almost an experimental procedure to the gold standard in a few short years, without rigid studies of its advantages and disadvantages. From that procedure stemmed a whole catalog of operations and now virtually every abdominal organ is deemed fair game for a laparoscopic approach. One of the most exciting technologies to be introduced into the armamentarium of colon and rectal surgeons is that of minimal access surgery. Although introduced over a decade ago in 1991, the acceptance of laparoscopic surgery in the field of colon and rectal surgery has been slower when compared with other areas of surgery. Several technical factors have played a role in this development (1). Some of the issues have been the uniquely complex technical nature of the colon resection, the necessity of doing a bowel anastomosis in most circumstances, the necessity to divide large blood vessels, and the need for extraction of a bulky specimen. Many of the early learning curve-related issues have been overcome and with the availability of new instrumentation, the safety and technical feasibility of this approach has now been confirmed. No doubt, with increasing expertise and better instrumentation, the laparoscopic approach will become more attractive for an even broader range of conditions. The first series of 20 patients who underwent laparoscopic-assisted colectomy was published in 1991 by Jacobs et al. (2). The authors provided a detailed description of their technique. Other investigators were then stimulated to follow their lead. In the same year, others reported on laparoscopic colon operations (3–8). Subsequently, the literature has been replete with reports of series of varying sizes, innovative techniques for virtually every colorectal operation, and the results, complications, and consequences associated with the new technology. Colectomy is termed an advanced laparoscopic technique because the colon is a large, mobile organ, spanning each abdominal quadrant, a specimen must be removed intact, and an anastomosis is often required; therefore, the visualization and dissection of the colon must move from abdominal quadrant to quadrant. The ability to see the magnified, slightly distorted anatomy is the initial major hurdle to overcome. & ADVANTAGES Advocates of laparoscopic colectomy cite numerous potential advantages. Among those are less pain postoperatively, faster patient recovery with less postoperative disability, shorter hospitalization, less intraoperative trauma, the technical capability to perform an operation comparable to an open technique, earlier return to work, better cosmetic results, a reduction in postoperative adhesions, and a better quality of life for the patient. Proponents believe that these considerations are sufficient reason to support adoption of the new technology. However, attempts to quantify these parameters have not been unequivocally demonstrated.

One of the benefits attributed to laparoscopic procedures is early postoperative feeding, with some patients permitted fluids as early as the evening of the day of operation, with diet advanced as tolerated. However, a number of investigators have suggested that early feeding might not be a unique benefit of laparoscopic procedures. In a randomized prospective trial, Ortiz et al. (9) found that patients who underwent a standard midline incision tolerated feedings equally well. A frequently cited advantage of laparoscopic colectomy is a shorter postoperative ileus. Hotokezaka et al. (10) studied this issue by placing recording electrodes in the distal antrum, the proximal side of the colonic anastomosis, and the rectosigmoid for postoperative myoelectric recordings. In this study, the potential benefit of using a laparoscopic approach to colon resection was not clearly confirmed because it only offered modest increases in the rapidity of recovery of gastrointestinal function. Early feeding and early discharge from the hospital may be encouraged to satisfy the surgeon, but it may be that this philosophy is equally applicable to open procedures. The same arguments apply to the length of time until the patient returns to work. In this regard, motivation may be the critical factor since self-employed individuals tend to return to work much more quickly than individuals, compensated by ‘‘sick leave.’’ Using serum levels of interleukin (IL)-6, IL-10, C-reactive protein, and granulocyte elastase as indicators of surgical stress, Hildebrandt et al. (11) showed significantly lower levels of these markers after laparoscopic colon resection which was most evident for IL-6 and granulocyte elastase. By using these parameters they found a significant reduction in surgical trauma after laparoscopic surgery compared with the open procedure. Duepree et al. (12) defined the incidence of accessrelated complications in a cohort of 716 consecutive patients undergoing either laparoscopic-assisted bowel resection (n ¼ 211) or open bowel operation (n ¼ 505). The mean follow-up in the laparoscopic-assisted and open groups were 2.7 years and 2.4 years, respectively. The incidence of wound hernia was significantly higher in open cases (12.9%) compared with laparoscopic-assisted ones (2.4%). The incidence of surgical repair of ventral hernia was also significantly higher in the open group (5.5%) compared with the laparoscopic-assisted group (1.9%). Postoperative small bowel obstruction requiring hospitalization with conservative management occurred significantly less frequently in laparoscopic-assisted patients (1.9%) compared with the open resection (6.1%). The need for surgical release of small bowel obstruction was similar between the open and laparoscopic-assisted groups (1.6% vs. 1.4%). The overall reoperation rate for these two complications was two times higher in the open group than in the laparoscopic-assisted group (7.7% vs. 3.8%). The data demonstrate that laparoscopic access for bowel operation significantly reduces the incidence of ventral hernia and small bowel obstruction rates compared with laparotomy. It is uncertain whether cost should be considered under the heading of advantage or disadvantage. Advocates of laparoscopic colectomy believe that the shorter hospital stay and better quality of life clearly qualify as advantages, but equipment is clearly more expensive, and an exact evaluation is often difficult because of the

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

sometimes-perverted calculations that may involve actual cost or charges to patients. Nevertheless, the decreased length of hospital stay is usually stated to more-than-compensate for the expensive technology and results in overall decreased health care costs. On the other hand, if the case has to be converted to an open procedure, the costs of both laparoscopic and open operation are incurred. Recently the hand-assisted technique has been employed with some added advantages beyond those of standard laparoscopy. The tactile sense permits locating pathology easier, evaluation of the abdomen for metastases, better retraction, the ability for isolation and control of the vessels, and ease of blunt dissection.

& DISADVANTAGES Laparoscopic colectomy is technically demanding and may be more difficult. The learning curve may be steep and prolonged. An analysis of total operative time as an indication of learning shows that approximately 11 to 15 completed laparoscopic-assisted colectomies are needed to comfortably learn the procedure (13). Another report suggested that the learning curve is longer than appreciated by many surgeons, requiring as many as 35 to 50 procedures to decrease operative time to baseline (14). However, complications can be kept at an acceptably low level while on the curve if a cautious approach is taken. In a series of 60 laparoscopic colectomies, Fowler and White (3) reported a morbidity rate of 11.6% (small bowel obstruction, anastomotic staple line bleeds) and a mortality rate of 1.6% (pulmonary embolus). Length of stay, complications, and operative time all decreased with experience, suggesting a steep learning curve. For example, because of small bowel obstruction, these investigators recommended closure of mesenteric defects. Agachan et al. (15) also studied the impact of surgical experience on complications. Over successive time periods, the total complication rate decreased from 29% during the first period to 11% by the second period and to 7% during the third period. Thus, the learning curve appeared to have required more than 50 cases for the surgeon to become proficient. A major disadvantage of the laparoscopic approach is the loss of tactile sensation. Operative times are often very prolonged. As opposed to most other laparoscopic procedures, laparoscopic colorectal surgery requires dissection in more than one quadrant and there is need for intraoperative repositioning of instruments, ports, and personnel. The colonic mesentery includes numerous large vessels and vascular control requires considerable time and quite often much more cost than do other procedures. An anastomosis is required for any colorectal operation and retrieval of the resected segment of colon with its attached mesentery requires an incision with the exception of an abdominoperineal resection. Furthermore, it involves specialized equipment. The hand-assisted technique has become popular with some surgeons by restoring some tactile ability and facilitating the operation (16–20). However, a 6- to 7-cm incision is necessary to admit the hand. & PATIENT INFORMATION The surgeon should discuss with the patient the reasonable expectations for the pathologic condition for which the

343

patient is undergoing the operation. The patient should be informed that intraoperative circumstances may mandate conversion of the technique to an open operation. Because of the new nature of this technology, some patients will decline the option of a laparoscopic approach. Age is not a contraindication as older patients fare as well as younger ones (21).

& INDICATIONS The reputed advantages of laparoscopic operations have been discussed but despite the cited advantages, clear-cut indications for such an approach are far from accepted. Some protagonists forge ahead with evangelical fervor, whereas some ultra-conservative surgeons only grudgingly concede its existence. The following discussion attempts to present a balanced portrayal of current evidence. What is certain is that the indications for operation in any given condition should not be modified simply because of the availability of a minimally invasive technique. Virtually the entire spectrum of colorectal disorders has been approached laparoscopically. But does the capability to perform a procedure laparoscopically mean that all procedures should be conducted in this fashion? Almost certainly not. Cogent concerns from an oncologic view have been raised against adopting this technique for malignancy. Is the lengthy time required for a restorative proctocolectomy justified? These, among other questions, have made the subject controversial, and only time and critical evaluation of currently performed procedures will permit the establishment of reasonable indications for the use of the laparoscopic approach. Procedures such as sigmoid resection, rectopexy, abdominoperineal resection, Hartmann’s reversal, or laparoscopic-assisted right hemicolectomy have been reported to show potential advantages. Complex procedures such as restorative proctocolectomy and completely intracorporeal right hemicolectomy require advanced technical expertise, and an advantage over open procedures has been questioned (22). For all procedures, the sometimes rather steep ‘‘learning curve’’ must be passed before the benefits of comparable operative times, morbidity, and shorter hospital stay can be achieved. Certainly, initially it would seem prudent to limit laparoscopic colectomy to elective situations. Disease processes for which the laparoscopic approach has been applied include colonic polyps not amenable to colonoscopic excision, carcinoma, diverticulitis, Crohn’s disease, rectal procidentia, volvulus, and diversion ileostomy or colostomy. Absolute contraindications include carcinoma infiltrating into adjacent structures, a large phlegmenous mass and perforation. Factors that could contraindicate laparoscopic approach include urgent intervention, mid-rectal carcinomas, locally advanced carcinomas, previous intestinal surgery, and carcinomas greater than 10 cm in size. Severe uncorrectable coagulopathy is also a contraindication. Contraindications may be relative or absolute and depend on the experience of the surgeon. Relative contraindications include obstruction, marked obesity, multiple previous laparotomies, extensive intra-abdominal adhesions, and carcinomatosis.

344

& PART III: MINIMALLY INVASIVE SURGERY

& CARCINOMA In recent years, studies have suggested that laparoscopic resection is equivalent to open resection when the extent of colon margins and the number of lymph nodes retrieved from the specimen are compared. Of more importance is the finding of no clinically significant difference in recurrence rates or survival. Furthermore, it is unclear whether the difference in lymph node count is clinically significant because the number of lymph nodes harvested has not been shown to affect survival. A number of studies have reported that the lymphovascular clearance in laparoscopic-assisted colectomy is similar to an open operation. In a detailed histopathologic study of laparoscopic-assisted right hemicolectomy specimens, Moore et al. (23) found no clinically significant difference between the groups in terms of margins of clearance or the number of lymph nodes harvested. The hand-assisted technique reestablishes tactile sense in operations for malignancies. The exploration to find the carcinoma and search for metastases is markedly improved. The surgeon’s actions parallel open technique, but the patient’s outcome is comparable to a laparoscopic procedure. Peritoneal dissemination of malignant cells must be considered to have an eventual fatal outcome. The early studies called attention to port-site recurrences. This has not been evident in recent studies (24). Twenty-one patients with emergent complications for advanced colorectal carcinoma were reviewed by Gonzalez et al. (25). Ten patients with perforation, 7 with bleeding, and 4 with obstruction were given a proximal ostomy, and 18 had concomitant colectomy. The complication rate was 30% and mortality was zero. & BENIGN NEOPLASMS A number of benign neoplasms such as adenomas or lipomas have been removed either through a colotomy or resection.

& EQUIPMENT AND INSTRUMENTATION During the past five years, the engineering and design of the equipment used for laparoscopy have been changing rapidly. Purchase of equipment is based on acceptable quality versus budget. If the purchaser waits for the final best system, he or she will wait forever. The best way to select equipment is to use it in the operating room and to give all potential users the opportunity to evaluate it. One should avoid buying equipment based on the salesperson’s recommendations. Generally, the various pieces of equipment are stacked on a cart with wheels (Fig. 1A) for ease of use and movement between operating rooms (26). Soon many hospitals will invest in dedicated laparoscopy rooms (Fig. 1B), which will contain overhead booms to hold all the equipment and permit easier placement of equipment for different procedures. The following section describes these pieces of equipment.

& LAPAROSCOPE Laparoscopes must deliver a high-quality image to be safe and effective (i.e., perfect vision is necessary for dissection and for identifying bleeding points). Not much has changed from the Hopkins rod-lens introduced in the early

1950s. Laparoscopes in diameters of 5, 7, and 10 mm are available. Viewing angles may be 0
, 12
, 30
, 45
, 70
, 90
, and 120
. Most surgeons use the 0
and 30
views. A flexible-tipped laparoscope is available. For diagnosis, a new class of minilaparoscopes that measure as small as 1.9 mm in diameter and pass through a 2 mm trocar are now available. Laparoscopes with larger rods deliver more light. Brighter light is a trade-off for depth of field and visual acuity. If the laparoscope is inserted cold, the tip fogs up, so it should be warmed before operation. If it becomes fogged, the lens can be cleaned with a sponge saturated with an alcohol-based antifog solution. Currently, the most used light source is an automatic xenon high-intensity machine; 200, 300, or 400 W lamps may be used. The life of these lamps is 400 to 500 hours. Spare lamps must be available for an unexpected light failure. The machine dispenses automatic light control, but it may be regulated manually. Smaller, less expensive xenon lamps have been designed for purposes requiring less light, as in the case of a choledochoscope, or to provide light behind the mesentery to help define the blood vessels. Light is transmitted via either fiber or fluid cables. The fiber cable is a bundle of glass fibers bound in a metal jacket that makes them rigid. These may be steam sterilized. The bundles are 1.6, 2.5, 3.5, or 4.5 mm. The cable bundle should be slightly larger than the lens system. However, if the fiber bundle is too large, light intensity is reduced. Fiberoptic cables suppress blue, and the illumination provides a yellow tinge. If 15% of fibers are fractured, a new cable should be obtained. Fluid cables transmit more light, but are rigid and difficult to use. The fluid tends to turn yellow if gassed, so fluid cables must be soaked to sterilize them. If the fluid is yellowed or the quartz is cracked, it must be replaced.

& VIDEO CAMERA In advanced procedures, such as colon resection, the assistant and primary surgeon work from the same monitor image. The perfect camera is still being sought, but many of the goals are being approached. An ideal camera will be sturdy, small, lightweight, simply controlled, sterilizable by soaking if necessary, well insulated and will bring forth a sharp, high-resolution, true-colored picture. The life of a good camera is approximately two years or 400 cases. Multiple types of cameras have been designed that use single-chip, single-chip digitized, three-chip, three-chip digitized with interchangeable sized focus lenses, zoom lenses, beam splitters, and viewers. Expensive three-dimensional systems are exciting to view, but the first generation requires use of special image glasses, which are uncomfortable, or nauseating for the surgeon. The second-generation three-dimensional cameras with special viewing monitors are under development, thus avoiding the need for special glasses. In the final analysis, the users of this equipment must conduct trials in the operating room to see which characteristics best suit them with regard to providing good vision and thus better and safer operations. & VIDEO MONITORS The monitor must generate high-resolution images with connectors to S-VHS systems. The monitor’s horizontal

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

345

FIGURE 1 & (A) The laparoscopy cart may contain a monitor (A), a component monitor or picture in picture (B), camera box (C), light source (D), electrosurgical unit (E), carbon dioxide insufflator (F), carbon dioxide tank (G), and control panel (H) (Courtesy of Olympus Corp.). (B) New laparoscopy room with equipment mounted on a five-arm beam. Same labels as in (A) (Courtesy of Olympus Corp.). (C) Vessel sealing/ cutting unit, harmonic scalpel (Courtesy of Ethicon Endosurgery). (D) Vessel sealing/cutting unit with graspers. (Courtesy of Ligasure, Valley Laboratory, Boulder, Colorado, U.S.A.)

resolution should exceed that of the camera. Most single-chip cameras produce 400 to 600 lines, which is twice that of standard consumer monitors. The size of the monitor screen is based on how close the surgeon works from it. Screen sizes vary from 8 to 20 in., but for colon surgery, the surgeon works 6 to 7 ft. away using a 19 in. screen. Generally, a second monitor is placed to face assistants on the opposite side of the operating table.

& RECORDING MEDIA Recording is a part of most laparoscopic systems, but it can exhaust the budget. The best recordings today are obtained by expensive digitized images stored on magnetic or optical disks. However, inexpensive videotapes or films in analog form are most often used. The original VHS, U-Matic, Beta-cam recorders are being replaced by

S-VHS, which gives good quality with 400 horizontal lines at a reasonable price. The best quality video photography for making 35 mm slides for publications or lecturing is taken from 35 mm single-reflex cameras attached to the telescopic eyepiece. This method is superior, but is inconvenient. To match most surgeon’s preferences (i.e., fast and easy techniques) electronic photography using the video signal from a still video recorder with a 35 mm camera may be used to capture the field of a moving video image. There are video printers to match the budget. These vary from inexpensive lowresolution printers to high-resolution digital printers.

& INSUFFLATORS Insufflation machines should be able to generate a flow of at least 10 L/min. Carbon dioxide is the gas usually used

346

& PART III: MINIMALLY INVASIVE SURGERY

because it is not combustible and is readily absorbed. The machine should have automatic pressure and volume regulators to maintain a constant intraperitoneal pressure. An acoustic and visual alarm system should be a standard part of the machine, which is activated if the preselected settings are exceeded. Gauges keep the surgeon apprised of insufflation pressure, gas flow per minute, intra-abdominal pressure, and gas volume used.

& POWER INSTRUMENTS Cutting and dissecting may be achieved using water jet, electricity, laser, or harmonic vibration instruments (see Chapter 7 for the principles of electrosurgery and laser). Currently, the most commonly used method is monopolar electrocoagulation. No particular advantage has been shown for using lasers in laparoscopic surgery. Harmonic scalpels (Fig. 1C) and shears use highenergy ultrasonic vibrations to mechanically denature proteins in tissues, thus sealing vessels and causing cavitation. Cutting–coagulation requires vibration frequencies of 55,500/sec. Ultrasonic shears with a blunt vibrating blade and a narrow closure blade have been developed. These shears can coagulate most vessels encountered. As cells are selectively fragmented, leaving nerves, ducts, and larger blood vessels, the cell debris and liquid are suctioned away. The new generation of more efficient harmonic scalpels fit 5 mm ports, have hand control buttons, and seal up to 5 mm vessels. Further advantages include minimal vapor production and increased safety. Avoidance of electric monopolar currents through the body reduces the possibility of burns, but the harmonic scalpel becomes hot enough to burn adjacent bowel if touched. A vessel-sealing device (Ligasure; Valley Laboratory, Boulder, Colorado, U.S.A.) delivers an electrical energy through special graspers (Fig. 1D). A high coaptive pressure with a temperature < 100
C denatures collagen and elastin forming a permanent seal of the vessel walls. This seal has a high tensile strength. The device has a triggeroperated blade to cut and separate the sealed vessel ends. Waterjet (hydrodynamic) dissection separates and fragments parenchymatous tissues with low collagenous fiber content. Its best use is in solid organs. It is less useful for endoscopic surgery because of mist obscuring the visual field, a limited cutting depth, potential contamination of the accumulated fluid, adjacent tissue bloating, and because hemostasis is not achieved. & INSUFFLATION NEEDLES The Veress needle, invented in 1938, is the most popular needle for establishing pneumoperitoneum (27). These needles are available in 10, 12, and 15 cm lengths with a diameter of 1.8 mm. The outer needle is sharp for cutting through the abdominal wall, and there is a blunt-tipped, spring-loaded inner stylet, which pops out beyond the tip when the sharp needle enters the free peritoneal cavity, preventing puncture of internal organs. The needle usually has an indicator that can be seen when the bare needle point is exposed without protrusion of the blunt internal stylet. The needle has a stopcock for connection to a carbon dioxide source. In turn, the carbon dioxide is insufflated into the needle, which egresses from a hole in the tip of the blunt stylet.

& CANNULAS AND TROCARS Access to the abdominal cavity is provided by a large variety of trocars and cannulas. The sizes vary, but colon surgery is performed with a variety of 5, 10, 12, and 15 mm cannulas. The design of these important instruments hinges on changes in the valves and ‘‘safety shields.’’ The cost of these instruments depends on whether they are reusable or disposable. Reusable trocars are more expensive, but require much labor to clean and maintain them. Because they are reusable the long-term cost is less. Disposable instruments generally have new, uninjured valves and sharp tips. Trumpet, trap-door, flap, chip, ball, slit, and leaflet valves have been devised. The sharp trocars present a risk for injury to abdominal viscera; so ‘‘safety shields’’ were developed to minimize that risk. There are variations on the principal theme of immediate shielding of the sharp trocar tip when it penetrates to the free peritoneal cavity. The new trocar points are not sharp. Modular port systems have been designed so that parts of a port that wear can be interchanged. This concept saves money and is particularly beneficial for cleaning and sterilization. Cannulas are held in place by several ‘‘anchor concepts.’’ Knitted sleeves, threaded sleeves, balloons, and mechanical expanders maintain the cannula in place and prevent gas leaks during a procedure. Flexible cannulas permit the insertion of curved instruments. The flexibility is afforded by creating a trocar from a tightly coiled spring. These instruments usually, require an outer distender sheath to fix them in place and prevent gas leak. & HAND-ASSISTED PORTS Hand-assisted laparoscopic surgery (HALS) is a technique that involves the intra-abdominal placement of a hand or a forearm through a mini-laparotomy incision while pneumoperitoneum is maintained. In this way, the hand can be used in a manner similar to an open procedure to palpate organs, retract viscera, identify vessels, provide traction to permit dissection in the appropriate plane, and apply finger pressure to control bleeding points. This approach may be more economical than a totally laparoscopic approach, reducing both the number of laparoscopic ports and number of instruments required. Some advocates of the technique claim that it is easier to learn and perform than totally laparoscopic approaches and that it may increase patient safety (28). Using a new generation of hand-access devices that extend the options for hand-assisted techniques, these devices facilitate hand insertion, protect the wound, act as the retrieval site for the specimen and serve as the portal for construction of extracorporeal anastomoses. In addition, these new devices can serve as laparoscopic trocar sites. This permits selective use of hand-assisted and laparoscopic-assisted techniques at various times in the same operation (29). Hand-assisted laparoscopic (HAL) colectomy has evolved into a clinically useful surgical technique. It enables the easy insertion and handling of a large surgical towel inside the peritoneal cavity (30). The towel successfully retracts the small intestine and enables the surgeon to concentrate the use of his or her hand on the targeted structures.

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

With the introduction of HALS, a series of hand ports have been devised. Six have been Food and Drug Administration approved and are sold for use in HALS. These include Lapdisc (Ethicon Endosurgery Inc., Cincinnati, Ohio, U.S.A.), Handport (Smith & Nephew Endoscopy, Andover, Massachusetts, U.S.A.), Gelport (Applied Medical, Rancho Santa Margarita, California, U.S.A.), Omniport (Advanced Surgical Concepts, Dublin, Ireland), Pneumosleeve (Dexterity Inc., Roswell, Georgia, U.S.A.), and Intromit (Applied Medical). These products need be compared in regard to size, ease of hand insertion, comfortability to introduce instruments, and rapidity of specimen extraction. These devices each have their own advantages and disadvantages. Ideally, the hand port provides easy passage of the hand into the abdominal cavity, protects the wound, maintains a pneumoperitoneum with or without a hand in, and allows easy extraction of the specimen. Nakajima et al. (31) reviewed 33 HALS colorectal procedures including total colectomy (n ¼ 16) and low anterior resection (n ¼ 10). In their study, 9.1% of 33 HALS procedures were converted to open surgery and 13.3% of 30 HALS procedures required minimal enlargement of incisions to facilitate extracorporeal procedures. The average operative time was 263 minutes. There were 9.1% major complications and 21% minor wound infections noted postoperatively. The mean hospital stay was 7.9 days. They concluded Gelport HALS is safely and reliably applicable for various colorectal procedures. Chang et al. (32) compared the outcome of 66 patients undergoing HAL sigmoid resection with that of 85 undergoing laparoscopic sigmoid resection. There were slight differences in operative time favoring hand-assisted (189 minutes vs. 205 minutes) and the extraction incision was larger in the hand-assisted group (8.1 cm vs. 6.2 cm). There was no difference in time for return of bowel function (hand-assisted 2.5 days vs. laparoscopic 2.8 days) or length of hospital (hand-assisted 5.2 days vs. laparoscopic 5.0 days). Complications were similar in the two groups (hand-assisted 21% vs. laparoscopic 23%), but there were fewer conversions in the hand-assisted group (0% vs. 13%). They concluded HAL sigmoid resection yields the same outcomes as standard laparoscopic techniques but with fewer conversions. Hand-assistance is a helpful innovation that may expand the application of laparoscopic colectomy.

347

FIGURE 2 & (A) Laparoscopic retractors. (B) Laparoscopic staplers. (Courtesy of Ethicon Endosurgery, Cincinnati, Ohio, U.S.A.) (C) Reusable 60 mm linear cutter. (Courtesy of Power Medical Interventions, Langhorne, PA.)

& OPERATING INSTRUMENTS Retractors Colon surgery requires movement from one abdominal site to another. To achieve this mobility with vision, it is necessary to retract adjacent organs, especially the small bowel. The positioning of the table makes gravity direct organs dependently, while a pneumoperitoneum helps ‘‘push’’ and hold organs away. Yet, some retraction locally often becomes necessary. Retractors should be blunt and should have atraumatic tips and blunt edges (Fig. 2A). Unfortunately, for an instrument to fit through a cannula, it must be blade-like. The blades may be made to curve, curl, or fan out, but the edges are still potentially sharp and must be checked for position frequently during an operation. Inflatable retractors may help solve the problem of sharp edges.

Graspers For the purposes of colon surgery, grasping forceps to push, pull, and dissect are necessary. Although forceps are available with traumatic or atraumatic tip designs, the atraumatic jaws required to avoid colon injury are most often needed. In general, the shafts can be rotated 360
to direct the tip appropriately. The tips can be electrified for electrocoagulating bleeding points. The diameter of the graspers is 5 or 10 mm, and they fit through cannulas of like diameter. Clip Appliers Clips are used to occlude larger vessels. The clip appliers may be single load or multifire. Generally, the single load

348

& PART III: MINIMALLY INVASIVE SURGERY

is reusable and the multifire are disposable. In a colon resection, multiple clips are needed, so a multifire clip applier is appropriate. The clips may be metal, such as titanium, or be absorbable.

Staplers Laparoscopic linear staplers (Fig. 2B) apply rows of staples, and cut between them. The staple lines may be 3 to 6 cm long. The 6-cm stapler only fits through a 12-mm cannula, and the 3-cm stapler fits through a 10-mm cannula. These are used for transecting bowel or performing functional end-to-end anastomoses. Flexible distal stapler shafts allow deeper insertion into the pelvis for rectal stapling with transection. Power driven staplers have been created that perform actions comparable to hand-powered staplers (Fig. 2C). End-to-end circular staplers may be used in laparoscopic-assisted anastomoses. The anvil is pursestring sutured into the proximal bowel, which is drawn through a small incision, and the stapler is inserted via the anus. Some of these staplers are extra long and reach well up into the abdomen. Special staplers with tighter and closer rows and shorter staple lines have been designed for use on larger vessels; however, these are expensive and add significant costs if many are used. Needles and Sutures Standard gastrointestinal needles and sutures generally can be used. Ski-shaped needles have been developed for ease of use with certain needle holders. For running suture and extracorporeal knot tying, polydiaxanone or polypropylene is best. Automatic sewing devices are being introduced. Needle shuttling devices are available, but they present enough problems and the applications are not well defined; so their use has been minimal. Fortunately, suturing is seldom needed for colectomy.

& OPERATIVE PROCEDURE

endoluminal ultrasonography is used to help define the relationships of that pathology to the layers of the rectal wall and other pelvic organs.

& POSITIONING AND READYING THE PATIENT The table is equipped with devices to hold the patient in place, even when marked tilting of the table is necessary (Fig. 3). A ‘‘bean bag’’ torso holder and shoulder braces work well. The patient is placed in a modified supine position with the legs in stirrups. The perineum is placed just off the inferior end of the table to allow access if intraoperative colonoscopy is deemed necessary and to facilitate a stapled anastomosis. The patient’s legs are positioned so that the surgeon or the assistant can stand between the legs. The hips and knees are flexed less than usual to avoid impediment to the surgeon’s arm movement and to keep the patient’s thigh from obstructing the movement of instruments. The legs are wrapped in compression, stockings (pneumatic if available) to enhance blood flow. The stirrups may be of further help if there is a need to cleanse the colorectum or mark a neoplasm. Both arms are tucked and padded to the sides, permitting more mobility by the surgeons. A gastric tube is inserted and removed at the end of the case, and a urinary catheter is inserted. The rectum may be washed out if rectal pathology is the reason for the operation. The abdomen is washed, prepared, and draped just as for celiotomy. Larger drapes have been designed which cover the legs and body, have straps to tether all the lines and tubes, have a perineal pouch, and a large opening to accommodate well-separated cannulas. & OPERATING ROOM SETUP AND CONDUCT OF OPERATION The placement of the operating team around the table varies with the site of the pathology (Fig. 4). The specifics of personnel placement and port placement are described later for individual colorectal procedures. However, generally the surgeon stands diagonally opposite the pathology. The first assistant stands opposite the surgeon. The cameraman is next to the surgeon, but positions change, depending on

& PREOPERATIVE PREPARATION The mechanical and antibiotic preparation of the colon is the same as that for laparotomy as discussed in Chapter 4. Generally, this entails the use of laxatives, most commonly sodium phosphate, and enemas. Oral polyethylene glycol solutions are used less frequently. In some patients the laparoscopy procedure is converted to an open procedure, and the informed consent should reflect that. & PREOPERATIVE EVALUATION Because the internal organs cannot be palpated during standard laparoscopy, the viscera must be inspected by other means. Colonoscopy is used to determine the pathology that is evident from the mucosal surface. Tattoos around the pathology are very helpful. Computed tomography (CT) scans are used to evaluate the liver and may reveal other pathology or enlarged lymph nodes. A CT scan is more meaningful if oral and intravenous contrast dyes are used, in concert. If the rectum contains pathology,

FIGURE 3 & Ready position may include ‘‘bean bag’’ torson holder (A), shoulder braces (B), stirrups with thighs level and legs angled down (C), nasogastric or orgastric tube (D), urinary catheter (E), compression stockings (F), and the arms are padded and tucked at the patient’s side (G).

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

349

view is possible without the need to turn his or her head. The camera should face the monitor that the surgeon uses.

FIGURE 4 & Orientation of the operating team. The surgeon’s sight, instruments, body, and monitor are aligned across the point of operation. This changes if the point of operation moves to a different abdominal quadrant The team is the surgeon (A), assistant surgeon (B), cameraman (C), scrub nurse (D), and anesthesiologist (E). Note that the surgeon uses a two-hand technique.

where the surgeon moves (i.e., to the right side, left side, or between the patient’s legs). The cameraman has responsibility for several functions. Because the view is critical, the laparoscope must be white-balanced, focused, and defogged. If the tip is fogged, the tip can be warmed, the cold CO2 lines can be moved to another cannula, and additional anti-fog solution can be applied to the tip. The view is frequently changed by moving the laparoscope close to the field for magnification or drawn back to give a wideangle view. At the beginning of the procedure, the entry sites for trocars must be observed during insertion, and at the end to see whether there is bleeding from the port sites internally as cannulas are withdrawn. The sterile operating room technician must have access to both the instrument table and the operating field. This technician must assemble the selected laparoscopic instruments, plus have instruments available if the laparoscopic procedure is converted to an open procedure. The significant amount of equipment results in clutter on the operating field; thus creating a standard positioning and arrangement of the various tubes and lines aids in smoother start of the operation. The larger pieces of equipment such as the camera, insufflation machine, and photography equipment are stacked on a mobile cart (Fig. 1A). The monitors are placed on the side opposite from the surgeon and assistant so that a direct

& PNEUMOPERITONEUM Pneumoperitoneum is achieved with either the Veress needle or the open technique. The Veress needle is easy to introduce, but it is a ‘‘blind’’ technique. For safety, the open technique is used. It is certainly preferred in cases of reoperative surgery, bowel obstruction with distended loops, known adhesions, and pregnancy. The easiest site for placement of the Veress needle is at the inferior midline margin of the umbilicus, where the fascia is most accessible (Fig. 5). A small incision is made with a No. 11 blade. Sharp, pointed towel clips are used to grasp the skin and umbilical fascia so that both can be lifted and essentially fixed. Without fixation, the peritoneum is more likely to ‘‘tent up’’ off the needle tip and fail to penetrate. The stopcock is opened, and the Veress needle is grasped like a pen, using the other fingers as supports to prevent a rapid thrust into the abdomen. By gentle advancement at 90
to the abdominal wall, the needle traverses the fascia and peritoneum. As the needle penetrates free into the peritoneal cavity, the ‘‘pop’’ of the blunt stylet springing forward to protect the sharp, tip is felt, and heard. When the stopcock is opened, a hiss may be heard. Simple tests should be performed to verify correct placement of the Veress needle in the free peritoneal cavity. The aspiration test, using a syringe half filled with water, determines whether blood or gas enters the syringe, suggesting that either a blood vessel or the bowel has been penetrated. If neither blood nor gas returns, the fluid is easily injected into the peritoneal cavity. The drop test is performed by placing a drop of saline on the top of the needle with the stopcock closed. When the stopcock is opened, the negative pressure in the abdomen draws the fluid

FIGURE 5 & Veress needle insertion. The needle is held like a pen with the hand braced on the abdominal wall. The towel clip is used to provide counter traction.

350

& PART III: MINIMALLY INVASIVE SURGERY

down. If this is not observed, the towel clips may be lifted, creating a negative pressure, and the drop of fluid should be drawn in. If these tests confirm correct placement in the free peritoneal cavity, the insufflation tube is attached to the stopcock. However, initially, the settings should be at a low pressure of 5 mmHg and a low volume of 1 L/min of carbon dioxide. The pressure should read 0 mmHg. If the pressure is > 5 mmHg or rises rapidly, the needle tip is likely in the preperitoneal space. When the site of the tip is properly shown to be in place, the rate of carbon dioxide insufflation can be increased to 6 to 10 L/min until a pressure of 10 to 15 mmHg is reached. The needle is withdrawn, and the incision is widened for insertion of the trocar. However, the ‘‘blind’’ insertion of the Veress needle for insufflation of the peritoneal cavity can result in severe major vascular and visceral injuries. It is therefore our preference to use the open technique to minimize potentially avoidable complications. The open technique is performed through an incision immediately beneath or above the umbilicus (Fig. 6) (33). Using narrow retractors the fascia is identified. The dense umbilical raphe is grasped with a Kocher clamp and pulled cephalad. A scissors can be used to incise the fascia where the umbilical raphe joins it. Sutures are placed through the fascia on each side, which are tied onto the cannula to prevent accidental extraction. Penetration into the free peritoneum can be seen; thus the blunt Hasson trocar–cannula can be inserted under direct vision. The cannulas are generally sewn into place with a heavy fascial suture. This suture through the fascia may be used for closure later. Thereafter, the abdomen can be insufflated rapidly up to a pressure of 15 mmHg.

& GASLESS LAPAROSCOPY Some surgeons have devised methods to perform laparoscopy without the pneumoperitoneum (34). These methods

depend on a mechanical means of pulling or lifting the abdominal wall. Various rods, pins, and bars may be configured through the abdominal wall or through the peritoneal cavity and attached to wires connected to a pulley system, which draws up the abdominal wall. Then small incisions can be made or cannulas can be inserted through which the surgeon can work. Hydraulic lifters were invented to simplify gas-less pneumoperitoneum. In this technique, a small incision is made and blades are inserted, fanned out, and attached to the power lifter so that an anterior space is created.

& TROCAR–CANNULA INSERTION The patient is placed in Trendelenburg position with complete pneumoperitoneum. For insertion of the trocar, the intraperitoneal pressure may be increased to 20 mmHg, but after insertion, the pressure is decreased to the 12 to 15 mmHg level. The infraumbilical site of the Veress needle is usually selected for the camera when this technique is used. The incision is widened to a size slightly greater than the cannula. The cannula with the stopcock closed is grasped between the index and middle fingers like a syringe, with the heel of the hand positioned over the trocar base (Fig. 7A). The shield mechanism is checked to ensure that it is ‘‘armed.’’ The other hand is used at the entry point into the skin as a brace to prevent sudden thrusts. The clamps holding the fascia are lifted by the assistant to fix the entry point. The sharp point of the trocar is impaled into the fascia at a 90
angle, but is redirected to an angle 30
inferior toward the pelvis. A rotating and twisting motion is used to advance easily. As the trocar enters the free cavity, the safety shield springs over the tip, causing a pop that can be felt and heard. A prewarmed laparoscopic camera is introduced, and the organs are checked for injury. The newer trocars do not have blades; thus tissue is separated rather than cut (Fig. 7B).

FIGURE 6 & Open Hasson trocar insertion technique. (A) Skin incision of the infraumbilical site, but this could be at any site. (B) Umbilical fascia exposed. (C) Umbilical fascia incision. (D) After a peritoneal incision, the traction sutures in the fascia are held taut, and the Hasson trocar is inserted.

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

351

and posterior rectus fascias are incised. To approach the posterior fascia the rectus muscle must either be retracted or split. The peritoneum and fascia are bound by sutures at each end and the midpoints of the incision to permit introduction of the hand-port without inadvertent splitting of any tissue plane with the deep edge of the device. Careful hemostasis is needed to avert hidden hematomas or hidden hemorrhage into the peritoneal cavity. The gloves used for the internal hand should be of dark color to prevent glare.

FIGURE 7 & (A) The penetrating trocar insertion. The thenar eminence is placed over the base of the trocar using the index finger as a guide. The other hand is used as a brace. The camera must focus on the entry site to be indicated when the trocar point is safely within. (B) Penetrating trocar. The components are optically guided tip (A), bilateral tissue separators (B), stability threads (C), clear cannula (D), insufflation valve (E), stopcock (F), friction lock for the laparoscope (G), and obturator release button (H).

& ADDITIONAL PORTS For colon surgery, usually three additional 10 to 12 mm ports are needed but 5 mm ports are often adequate except when stapling instruments are required: two for the surgeon’s instruments and one for the assistant’s instruments. The use of large ports permits the interchange and reposition of the camera and instruments. Port placement must be planned so that the ports are directed at the pathology in a semicircle. Placing the ports at least 7 to 8 cm apart, a hand’s width, is preferred to prevent instruments from crossing (‘‘sword fighting’’) (Fig. 8B). Before insertion, the camera should be used to transilluminate the site to identify large underlying abdominal wall vessels. With the camera in place, the other trocars can be observed as they penetrate the abdominal wall. The strategic placement of ports permits their incorporation into a later incision for specimen retrieval or stoma creation, if needed. Hand-Assisted Port When the hand-assisted technique is chosen (Fig. 8A), the number of ports can be decreased to a laparoscope port and two instrument ports for the surgeon. This requires the surgeon with a hand in the abdomen to be thoroughly familiar with the anatomy. The hand port is usually placed through the midline or through the rectus toward the pathology. The surgeon chooses the site based upon the need to mobilize and retract the specimen with the hand, but also draw the end of the bowel into the wound to permit preparation for anastomosis. The incision length is just short of the surgeon’s glove size. In the midline, the skin, subcutaneous tissue, and fascia are incised. The hand-assisted device generally can be directly introduced. If the chosen site is off midline, the skin, subcutaneous fat, and anterior

& SPECIMEN LOCALIZATION Visual cues such as erythema, puckering of the serosa, or adhesions may identify the point of pathology. Because palpation is not possible, localization may depend on the use of the colonoscope. Preoperatively, dye, preferably India ink, may be injected on a margin of the pathology; the mark may be seen on the serosal surface. If the site is uncertain, the colonoscope can be passed during the operation, and the site can be marked where the colonoscope light is placed adjacent to the pathology. Monson et al. (35) recommend a preoperative barium enema to confirm the location of the colonoscopically diagnosed lesions because of the inability to palpate the colon directly during laparoscopic-assisted colectomy. The identification of small neoplasms or areas presenting lesions associated with previous endoscopic polypectomy is one of the major problems in laparoscopic colon resection. Munegato et al. (36) evaluated the effectiveness of tattooing with India ink, to identify the area of the colon that is the site of the lesion in order to be able to perform colonic resections with oncologically correct margins. Marking was performed during preoperative colonoscopy in 84 patients by injecting 1 mL of the India ink solution with a sclerotherapy needle into each of the four quadrants of the colon wall. This method invariably allowed easy identification of the site of the neoplasm at laparoscopic colon resection. Only one complication due to paucisymptomatic microperforation was discovered during the operation. Feingold et al. (37) reviewed 50 consecutive patients with colorectal neoplasms who underwent endoscopic tattooing prior to laparoscopic resection. No complications related to endoscopy or tattooing were incurred. Tattoos were visualized intraoperatively and accurately localized the neoplasm in 88% of patients. In the context of laparoscopic colorectal resection, preoperative endoscopic tattooing is a safe and reliable method of localizing colon lesions and may be preferable to other localizing techniques including intraoperative endoscopy. Zmora et al. (38) evaluated the use of intraoperative lower endoscopy in 233 patients who underwent laparoscopic segmental colectomy. Lower endoscopy was employed in 24% of them compared with 17% in the laparotomy-matched group. The diseased segment was successfully identified in all of the patients in whom the main indication for endoscopy was localization (65% of cases). Endoscopy was judged to have changed the operative management in 66% of the 57 cases in whom it was employed and especially in 88% of the 37 patients for whom the main indication had been localization.

352

& PART III: MINIMALLY INVASIVE SURGERY

FIGURE 8 & (A) The hand port is inserted into a short incision, measured in centimeters to match the glove size, with the intent for hand mobilization, specimen and tissue retraction, and anastomosis construction. The touch sensation is reestablished. (B) Port placements for a right colectomy: ports are separated by the width of the fingers of a hand. Ports are placed along an imaginary semicircular line (D) facing the pathology. The initial port for the camera is usually placed above or below the umbilicus hatched lines (B) represent incision sites for either hand port or specimen retrieval. (C) Mobilization of the right colon, lateral to medial approach. (D) Alternative mobilization of the right colon, medial to lateral approach. First, the peritoneum is incised at the base of the ileal mesentery. (E) Alternative mobilization of the right colon, medial to lateral. Dissection separating the retroperitoneum from the right colon mesentery, pushing the ileocolic and right colic vessels anterior, the ureter posterior, and the duodenum medially. (F) Alternative mobilization of the right colon, medial to lateral. The ileocolic and right colic vessels are isolated and divided prior to incising the right lateral line of Toldt. (G) Alternative mobilization of the right colon, medial to lateral. The vessels are divided.

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

& COMPLETION OF LAPAROSCOPY The camera is used to look in all quadrants for injury or hemorrhage. As cannulas are moved, the site is observed for bleeding, and the skin site is plugged to maintain pneumoperitoneum. All port sites 10 mm usually have the fascia closed to prevent hernias. The skin is closed with subcuticular sutures and adhesive strips or skin staples.

& SPECIFIC COLORECTAL PROCEDURES & DIAGNOSTIC LAPAROSCOPY Diagnostic laparoscopy is valuable when the etiology is unknown or suspected in a patient who is ill from a possible intra-abdominal disease (39). A simple peek may resolve the issue. Such an examination may be combined with colonoscopy, upper endoscopy, ultrasonography, washings, and/or biopsy (40). The patient is positioned supine in stirrups as previously described. To perform diagnostic laparoscopy, pneumoperitoneum is obtained by the previously described techniques. The size of the cannula to be inserted is predicated on the size of the camera to be used. If looking around the abdomen is the only activity, a very small camera inserted through a 2.5 to 3.0 mm needle may be used. Larger cannulas or secondary cannulas are needed if instruments are to be inserted. To insert a laparoscopic ultrasonographic probe, a 10 to 12 mm cannula is needed at the infraumbilical site. After the camera has been inserted, all four quadrants are carefully inspected. Exploration is the first step after the ports are placed. The finding of marked adhesions, abnormal anatomy, or carcinoma characteristics may prompt early conversion to hand-assisted or open laparotomy. If colonoscopy has not been completed, the colonoscope tip can be followed in its movements, and sometimes a grasping instrument may help negotiate loops and turns in the colon. The small bowel should be clamped to prevent reflux filling in the event that the ileocecal valve is incompetent. In cases in which a malignancy is suspected, effort must be made to see the entire liver surface. To see over the dome of the liver, a 30
laparoscope or flexible-tipped scope may be used. Intraoperative ultrasound probes have been designed to fit through 10 to 12 mm cannulas, which permits sharper images for the detection of hepatic metastases (41). If the liver has a suspicious site visible on direct inspection, previous or intraoperative ultrasonography, or CT scan, a biopsy may be needed. If the lesion is visible, a biopsy needle can be inserted percutaneously or through a small port aimed directly over the site to be biopsied. If the lesion is not visible, an ultrasonography-guided biopsy can be performed. If a specimen is to be obtained, suction, irrigation, and electrocautery equipment as well as clips and staplers should be immediately available in the operating room if hemorrhage control is necessary. To look directly into the pelvis and see the depths of the cul-de-sac, a rigid sigmoidoscope can be used to point out the rectum and its junction into the peritoneal cavity. The uterus can be pushed up and out of the posterior pelvis by a special uterine manipulator introduced transvaginally. The small bowel may be inspected directly by ‘‘running the bowel’’ techniques. Either the terminal ileum at the cecum or the jejunum at the ligament of Treitz is chosen

353

as a beginning point. Two grasping instruments at wellseparated cannula sites are used in hand-over-hand fashion along the small bowel. Pneumoperitoneum must be well maintained, and the table may need to be rotated and/or tilted to use gravity to keep the bowel separated and fixed in a quadrant.

& RIGHT-SIDED COLECTOMY The principles of right colectomy should be the same as in the open procedure; the difference is access (42). As in all laparoscopic colectomy procedures, the patient is placed in supine position with the legs in stirrups. A gastric tube and a urinary catheter are placed. The abdomen is prepared and draped in sterile fashion, just as if open colectomy were to be performed. The surgeon stands on the left side facing the right upper quadrant. The surgeon may move to a position between the patient’s legs. Likewise, the camera operator and assistant surgeon take other positions as needed. The scrub technician is positioned over the patient’s right leg. A monitor is placed facing the right shoulder position for the surgeon. Most of the team uses this monitor. The second monitor can be placed over the left shoulder. If one member of the team is positioned in the right upper quadrant site for retraction, that person uses the second monitor. As with all laparoscopic colon operations, a variety of trocar positionings have been recommended. These choices are often selected because of individual preference, the specific site of the pathology in the colon, previous operations, incisions that may steer the surgeon away from a particular area, or the body habitus of the patient (Fig. 8B). The first trocar site is placed at an infraumbilical or supraumbilical site with the patient in a head-down position. A 10 to 12 mm trocar is used to accommodate the laparoscope. Other trocars are placed in a semicircle to create a triangle with the pathology as the apex of the triangle. Ports may be placed in the right lower quadrant, midline at a suprapubic site, a left upper abdominal site in the midclavicular line, and high in the midline. These ports are placed so that the surgeon and first assistant can both use a two-handed technique. The 10 to 12 mm trocars may be used for all ports, but 5 mm ports are usually adequate for grasping and dissecting functions. At least one should be a 10 to 12 mm port for possible rapid clip application in the event of acute hemorrhage, or possibly the application of a vascular stapler. The hand-assisted technique may be chosen, which permits tactile sensation, provides the best of retractors, the human hand, and permits guiding and checking of instrument tip position. The site for the hand port is usually the upper midline or through the right, upper rectus. The hand-assisted device is inserted, and it also serves as a wound protector during specimen extraction. The patient is placed in steep head-down position and rolled to the left. This causes the small bowel to fall away from the cecum and terminal ileum. The patient is repositioned to the head-up position as the operative field moves up the right colon to the hepatic flexure and the transverse colon. Even in this instance the Trendelenburg position may be preferred to prevent the omentum from falling into the field of vision.

354

& PART III: MINIMALLY INVASIVE SURGERY

The dissection begins in the right lower quadrant with dissection to free up the terminal ileum and appendix (Fig. 8C). As this mobilization progresses superiorly along the white line of Toldt, the right colon mobilizes medially. As the cecum is lifted, the ureter and gonadal vessels should be avoided. The right ureter is often not visualized during mobilization. As the hepatic flexure is approached, the table is tilted to a head-up position, which helps keep the small bowel dependent toward the pelvis by gravity. The duodenum is evident posteriorly as the transverse colon is mobilized by taking down the gastrocolic ligament. Electrosurgical scissors are most often used for the dissection; graspers are used to reflect and hold the bowel. Some surgeons prefer to use the harmonic scalpel for dissection. When mobilization is complete, the colon is suspended by the mesentery where the three named vessels (i.e., the ileocolic, right colic, and right branch of the middle colic vessels) reside; these must be cut. These vessels are individually freed up and clipped, ligated, or stapled (Fig. 9). The ileocolic vessels are always present, but the right colic is often absent and the middle colic is quite variable. When the vessels are divided, windows are evident in the mesentery. The mesentery is divided from the selected point for transection on the ileum to the point on the transverse colon (Fig. 10). An alternative is the medial to lateral approach. In the beginning, the terminal ileal mesentry is lifted anteriorly so that an incision can be made from the caudal and dorsal attachments of the cecum up medially toward the duodenum (Fig. 8D). An avascular plane behind the ileocolic vessels can be bluntly separated revealing the ureter and gonadal vessels that are swept posteriorly (Fig. 8E). This

FIGURE 9 & Division of the vessels in the lateral to medial approach.

FIGURE 10 & Intracorporeal division of the ileum and transverse colon.

dissection proceeds anterior to Gerota’s fascia, and extends cephalad to the liver and medially to the second portion of the duodenum, which is pushed medially. By pulling the cecum laterally the ileocolic vessels are drawn taut and stand out so that they can be traced from their origins at the superior mesenteric vessels to the cecum (Fig. 8F). Thus, they can be divided 1.0 to 1.5 cm from their origins. If the right colic vessels exist, they can be isolated and divided near their origins (Fig. 8G). Care is always necessary to spare the superior mesenteric vessels. The hepatic branches of the middle colic artery are encountered and divided as dissection proceeds to the transverse colon. Likewise, the marginal vessels adjacent to the colon are divided. The omentum is hanging from the transverse colon and may be incised, separating it off close to the colon in an avascular plane. Now only the lateral attachments of the white line need be taken down. Bowel transections and the remainder of the operation proceeds as in the lateral to medial approach, either as pure laparoscopic or laparoscopic-assisted technique. Colectomy can be performed with a laparoscopicassisted technique (4,43–47). In fact, our preference is laparoscopic-assisted surgery, which means that when adequate mobilization has been accomplished to allow delivery of the bowel to the surface, a 6-cm incision is made in the mid abdominal wall for specimen removal. If the procedure is performed for carcinoma, the incision should be shielded by a wound protector to avoid contamination by malignant cells. Mesenteric vessel ligation can be achieved through this incision. Via this incision, the bowel can be transected, usually with staplers, and anastomosis accomplished, usually with a functional end-to-end anastomosis using staplers (48). This type of stapled anastomosis

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

should be performed as shown in Chapter 23. Subsequently, the mesenteric defect may be closed, but many surgeons do not make an effort to close this defect. Extracorporeal anastomosis and vascular ligation have the advantage of laparoscopic mobilization and avoid the difficulties associated with time-consuming intracorporeal mesenteric dissection, ligation, and anastomosis. The right hemicolectomy may be performed entirely intraperitoneally, except for specimen removal. The resected tissue specimen may be placed in a special specimen bag and removed later through an incision. Linear cutting endoscopic staplers can be used to divide the bowel (Fig. 10) and the disease-bearing portion of colon removed (Fig. 11). Then the two limbs of bowel can be positioned end to end for an anastomosis with endoscopic stapling (i.e., functional end-to-end stapling). The antimesenteric corner of each transected and stapled bowel end is excised (Fig. 12A) to permit insertion of each limb of the linear laparoscopic stapler (Fig. 12B). When the side-to-side anastomosis has been created (Fig. 12C), the remaining opening at the tip can be closed with the linear stapler (Fig. 12D). Bernstein et al. (49) reported that intracorporeal division of the mesentery and anastomosis confer no advantage over the laparoscopic-assisted procedures. Data were prospectively collected on 102 consecutive laparoscopic colon resections. There was no statistically significant difference in the length of hospital stay or the duration of postoperative ileus, regardless of whether intracorporeal or extracorporeal mesenteric division and anastomosis were undertaken. These data demonstrate that a completely laparoscopic procedure does not appear to offer any advantage when compared with a laparoscopicassisted one. Furthermore, intracorporeal anastomoses are considerably more demanding and time-consuming.

FIGURE 11 & Specimen freed for removal.

355

& LEFT-SIDED COLECTOMY The preoperative preparation, evaluation, positioning, setup, pneumoperitoneum, and trocar insertion technique have been described earlier. The following techniques are used for sigmoid resection, left hermicolectomy, and the abdominal portion of an abdominoperineal resection (2,50). For left-sided pathology, the surgeon stands on the right side of the patient. The surgeon’s hands and eyes must be aligned facing a monitor. The second monitor is positioned as needed for assistants (Fig. 13A). The patient is placed in a steep head-down position, and the table is rotated to the right, causing the small bowel to fall away from the left lower quadrant. If great difficulty is encountered in retracting the small bowel out of the pelvis during a left-sided resection, the cecum may be mobilized to facilitate the exclusion of the small bowel from the operative field (Dr. Jaap Bonjer, personal communication). Ports are placed in a semicircle around the umbilical port, which usually holds the camera (Fig. 13B). A site for an incision to remove the specimen or for assisted resection is planned; the incision generally incorporates one or two of the port sites. Depending on whether the sigmoid colon alone or the descending colon or rectum is also to be resected, the length of mobilization varies and port placement is modified. To resect the sigmoid colon, ports are placed below the umbilicus and in the right lower quadrant, right upper quadrant adjacent to the umbilicus, and left upper quadrant adjacent to the umbilicus. Frequently, a left lower quadrant trocar is added so that the assistant can aid in retraction and traction on the colon. For the hand-assisted technique, the hand port is made in the left lower quadrant or midline and is used in place of the left-sided ports. If the patient is thin, the ureters may be visible, especially on the right; thus an incision medial to the ureter can be made as a marker to exclude the ureter during further dissection. The left ureter is more difficult to find, but as mobilization proceeds, the gonadal structures and ureter on the left must be identified and separated from the specimen. If a dense, adherent phlegmon is expected, ureteral stents may be inserted prior to beginning. This aids the surgeon if hand-assisted techniques are employed or the procedure is converted to an open approach. The sigmoid is retracted medially, exposing the left lateral attachments of the descending colon and sigmoid colon. An electrified scissors or harmonic scalpel is used to incise the white line of Toldt (Fig. 13C). After the left lateral attachments are freed down to the rectum, the sigmoid is retracted laterally, exposing the vessels, taut under the traction. The mesentery is scored across their base near the aorta, and the incision is carried down along the rectum on the right. The vessels are skeletonized and ligated, clipped, or stapled with a vascular stapler (Fig. 13C insert). Either ligatures or clips or both may be used. An alternative approach is the medial to lateral mobilization. The sigmoid mesentery is pulled anteriorly and inferiorly to make the inferior mesenteric vessels and sigmoid vessels taut and thus stand out. On the right side of the mesorectum an incision is made beneath these vessels near the sacral promentory. By bluntly dissecting across the retroperitoneum the nerve plexuses can be swept posteriorly and the vessels can be pushed anteriorly. As

356

& PART III: MINIMALLY INVASIVE SURGERY

FIGURE 12 & Intracorporeal anastomosis. (A) Ileum and colon are aligned. (B) The ileum and colon are stapled and opened side to side. (C) The completed side-to-side anastomosis. (D) The open ends are stapled.

the dissection progresses toward the left side, the ureter and gonadal vessels are identified so that they can be swept lateral and posterior to the specimen (Fig. 14A). The vessels under stretch stand out so that the lateral mesenteric attachments can be taken down permitting the inferior mesenteric vessels to be encircled. The vessels can be divided using previously mentioned methods while carefully observing that the ureter is not picked up (Fig. 14B). Thereafter, the lateral white line can be incised as high as necessary to permit an anastomosis without tension. The operation then proceeds in the same fashion as lateral to medial resections. Patients requiring the takedown of the colonic splenic flexure to facilitate a left hemicolectomy have an initial incision in the mesentery medial to the inferior mesenteric vein, ligation of vessels in no-touch isolation fashion, subsequent medial-to-lateral extension of retroperitoneal dissection along Gerota’s fascia, opening of the lesser sac by transection of the gastrocolic ligament, dissection of the mesenteric root of the distal transverse colon, and the final separation of splenocolic ligament and the lateral attachments of the descending colon (51). If an anterior resection or a left colon resection is necessary, the splenic flexure is mobilized. The team must change position, and the monitors must be realigned. The

surgeon usually stands between the patient’s legs and uses the left-sided ports for the instruments. The assistant moves to the right upper quadrant to grasp the bowel. The camera person is in the right lower quadrant. The table is shifted to a head-up position to cause the small bowel to slide low and to the right. Traction on the splenic flexure exposes the attachments to the side wall, spleen, omentum, and stomach. Layer after layer is incised until the splenic flexure descends easily. Dissection close to the transverse colon along the omentum is relatively bloodless. Care must be taken to avoid pulling on the spleen and causing a capsular tear and hemorrhage. Reissman et al. (40) described the use of a colonoscope to retract the splenic or hepatic flexures to facilitate exposure, dissection, and mobilization of these flexures. Dissection into the pelvis is performed under direct vision, using electrocautery or harmonic scalpel. Some surgeons use vessel sealing devices serially down the lateral stalks for cutting and coagulation (Fig. 15). The nerves at the pelvis can be seen and spared. If the rectum must be mobilized, the lateral attachments of the rectum are incised to the pelvic floor. The rectum is pulled anterior so that the presacral space can be entered over the sacral prominence. The rectum is separated from the presacral fascia down to the coccyx (Fig. 15A). The lateral stalks are

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

357

FIGURE 13 & (A) Positioning for left colectomy or sigmoid colectomy. (A) Surgeon; (B) first assistant; (C) cameraman; (D) nurse; (E) anesthetist. (B) Port placement and specimen removal incision or hand port (B). The incision is made early in the procedure if it is to serve as a hand port or late in the procedure if it is for specimen removal. (D) is the imaginary semicircular line facing the pathology; (S) is the surgeon’s port; (C) is the camera port; (A) the assistant’s port. (C) Mobilization of the left colon and sigmoid, lateral to medial approach. Insert: Division of the inferior mesenteric vessels, lateral to medial approach.

cut, with special attention to coagulating the middle hemorrhoidal vessels. Anteriorly, the dissection is carried down the rectovaginal septum or below the seminal vesicles. The dissection is the same as that in the open procedure. When the two sites for transection are decided upon (i.e., the colon and rectum), the mesenteric fat is dissected off in preparation for an anastomosis. The

transection of the mesorectum is performed with harmonic scalpel, vascular stapler, or a vessel sealing device. In the case of malignancy, the total mesorectal excision (TME) is performed as an en bloc resection. This is extremely difficult in the narrow male pelvis and the fat laden pelvis. The bowel can be transected with a 30 to 60 mm linear or laparoscopic stapler.

358

& PART III: MINIMALLY INVASIVE SURGERY

FIGURE 14 & (A) Alternative mobilization of the left colon, medial to lateral. Early exposure of the left ureter beneath the inferior mesenteric and sigmoid vessels. (B) Alternative mobilization of the left colon, medial to lateral. Early division of the inferior mesenteric vessels, prior to lateral mobilization along the left line of Toldt.

An alternative approach is to use laparoscopicassisted colectomy techniques. In those cases, the planned incision for specimen removal is made, and the mobilized colon is drawn into the wound. Then the vessels can be transected, transection sites dissected clean of fat, and standard stapler transection accomplished (Fig. 15D). The hand-assisted technique is valuable for pathology that is adherent to the peritoneal side walls. Detecting points where the colon is adherent and where retraction is necessary are more readily and speedily decided (53). Anastomosis can be performed using a doublestapling technique. The anastomosis may be made at low levels if dissection can be carried out and a staple line applied. Laparoscopic staplers today can be directed squarely across the bowel with the aid of flexible stapler shafts. Often more than one application is necessary to staple completely across rectum or enlarged colon. The

laparoscopic-assisted colectomy permits the proximal colon to be drawn out of the incision where a pursestring suture is applied and tightened around the shaft of a standard end-toend stapler anvil (Fig. 15D). The colon is returned to the abdominal cavity and the pneumoperitoneum is reestablished. The stapler can be introduced via the anus and the stapler trocar opened to penetrate the distal linear staple line. The anvil and stapler trocar are connected (Fig. 15E). The stapler is closed, fired, opened, and withdrawn. The rings of excised tissue (‘‘doughnuts’’) are inspected for completeness, and the anastomosis is tested by cross-clamping above the anastomosis and introducing air under pressure via the sigmoidoscope (Fig. 15F). Bubbles will be seen if water is placed over the anastomosis and a leak is present. With respect to another approach, there is a report of performing a low extracorporeal anastomosis with the rectal stump everted (52). Generally, laparoscopic surgeons believe

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

359

FIGURE 15 & Anterior resection. (A) The rectum is mobilized. (B) The bowel is divided below the pathology with a margin. (C) The colon and splenic flexure are mobilized. (D) The specimen is pulled through a small incision, and the anvil of the circular stapler is inserted into the proximal colon. (E) The stapler is introduced through the anus and aligned with the anvil. (F) The double-stapled anastomosis is completed and tested.

that a left-sided colectomy is more technically demanding than a right hemicolectomy because with the latter procedure the bowel can be more easily delivered to the abdominal incision. For low anterior resections and for patients who had neoadjuvant chemoradiation a proximal, diverting ostomy, usually a loop ileostomy, is created.

& ABDOMINOPERINEAL RESECTION Equipment and personnel positioning is shown in Figure 16. In laparoscopic abdominoperineal resection of the rectum, an abdominal incision is completely avoided because the carcinoma is delivered through the perineal incision. However, hand-assisted laparoscopy through a low abdominal

360

& PART III: MINIMALLY INVASIVE SURGERY

incision can be useful for retraction, control of hemorrhage and decrease the operation time. Darzi et al. (54) believe that the view provided in the pelvis by laparoscopy is significantly better than at laparotomy and allows excellent anatomic definition and meticulous dissection. In an abdominoperineal resection, a colostomy is created at a properly selected site (i.e., away from bones, scars, creases, and umbilicus) (55–57). A small 3-cm-diameter disk of skin is removed, and the opening is carried down through the subcutaneous fat, fascia, muscle, and peritoneum. A Babcock clamp is inserted through the hole, and the tip of the colon is grasped and drawn through the wall. After wound closures, the colostomy is matured. Ideally, one of the trocar sites should be planned as the colostomy site. Then the trocar site can be widened to accept the colon with its partial mesentery. The perineal resection is conducted as in the open abdominoperineal resection. Often the perineal dissection is more difficult, because dissection from the abdominal side may not have been carried as far distally as it might have been with the open technique.

FIGURE 16 & Positioning for abdominoperineal resection: (A) surgical assistant, (B) scrub nurse, (C) surgeon, (D) cameraman, (E) anesthesiologist.

& STOMA CREATION The site for an ostomy is planned using the basic principles of bringing the stoma through the rectus muscle, but avoiding bones, scars, the umbilicus, and folds. An ileostomy is created through the right lower quadrant, a sigmoid colostomy through the left lower quadrant, and a transverse colostomy through the upper quadrants (58–61). The ports are placed at different sites for the various ostomies (Fig. 17). A 10-mm trocar is inserted through an umbilical site. After diagnostic laparoscopy has been completed, a 10-mm

FIGURE 17 & Trocar placements for ostomies. (A) Loop ileostomy or end ileostomy. (B) End-sigmoid colostomy or loop-sigmoid colostomy. (C) Loop transverse colostomy.

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

trocar is inserted through the previously selected stoma site. An additional one or two 5-mm trocars are introduced. The segment of bowel is selected, and an umbilical tape is passed through the mesentery, at the point at which the ostomy is to be opened (Fig. 18A). The tape around the loop of bowel is passed through the cannula at the preselected ostomy site. Some surgeons simply grasp the bowel with a Babcock clamp and hold it in position while the opening around the cannula is enlarged (Fig. 18B). The segment of bowel is drawn up to determine whether the loop will reach through the abdominal wall. If it does not reach, mesenteric attachments, adhesions, or vessels may need to be divided. To open the abdominal wall, the pneumoperitoneum is released. A 3-cm-diameter disk of skin around the cannula site is excised and the subcutaneous fat is retracted. A 2  2 cm gridiron incision is made through the fascia. With the camera observing from the abdominal side, the bowel is pulled through the wound. The bowel may be fixed to the fascia with interrupted sutures. The cannula sites are closed. The ostomy is then matured as for a loop ostomy. Mattingly et al. (62) evaluated the possibility of performing fecal diversion with the assistance of a colonoscope and without the additional morbidity of abdominal exploration or general anesthesia. There were 15 patients diverted using a colonoscope to identify a site of the sigmoid colon that could be approximated to the anterior abdominal wall as confirmed by transillumination of the abdominal wall. A small skin disk was then removed at this location and a loop colostomy was made. The colonoscope was also used as a guide to identify the proximal and distal limbs of the loop colostomy.

361

& STOMA CLOSURE When a stoma is necessary, the stoma and mucous fistula or the oversewn distal bowel may be placed at sites apart from each other after the diseased segment is resected. The divided, separated bowel may be brought back together at a later date if the patient’s health status permits. For the properly selected patient, laparoscopy may be the operation of choice (63–67). The extent of adhesions secondary to extensive surgery, inflammation, or contamination may make laparoscopy a poor choice for closure. When the camera is inserted, the anticipated level of difficulty may be assessed, and the decision to open or not can be made. For reestablishing intestinal continuity the patient is placed on the table in stirrups as described earlier. The surgeon often has better control positioned between the patient’s legs. The assistants are placed on the patient’s left with monitors at the patient’s upper right and left shoulders. The surgeon lines up the laparoscope, ileostomy, previously transected colon, and the monitor. The ileostomy is taken down via a circumferential incision, which is gradually carried down into the peritoneal cavity. Adhesions around the stoma are lysed. A 10-mm cannula is inserted through the opening, and a pursestring suture is applied and tied tightly to hold a pneumoperitoneum. When pneumoperitoneum has formed, the camera is inserted. The table is placed head down with rotation to the left. The table is rolled to the right or left, depending on which way the small bowel is most easily shifted. After visual exploration, the decision to proceed with a laparoscopic or an open procedure is made. If laparoscopy is possible, ports are placed in the suprapubic midline, the left lower quadrant, and the left upper quadrant. The ileum is dissected free so that it easily

FIGURE 18 & Laparoscopic ileostomy. (A) Tape holds the ileum beneath the cannula at the proposed site of the ileostomy. (B) Babcock clamp pulls the ileum into position beneath the cannula.

362

& PART III: MINIMALLY INVASIVE SURGERY

reaches to the colon for anastomosis. A small incision is made around the mucous fistula and carried down into the peritoneal cavity or opened over the closed proximal end of the colon. The two ends are pulled into the same wound, where a functional end-to-end anastomosis can be performed extracorporeally. For a sigmoid colostomy, closure is performed with the patient in the supine position in stirrups as described previously. The surgeon stands on the patient’s right side with an assistant on the left. The table is tilted head down and to the right. The surgeon, stoma, and monitor are lined up. The procedure starts with mobilization of the colostomy. Through this opening, a cannula can be introduced. Ports are placed in the right lower quadrant, right upper quadrant, left upper quadrant, and possibly suprapubic sites. Adhesions are freed up. Often the vagina, uterus, or bladder have become adherent across the pelvis. A sigmoidoscope can be used to identify the rectal or bowel stump (Fig. 19A). Inserting a sizer or other suitable instrument into the vagina may help identify the vaginal vault. A place on the distal bowel must be cleaned of adhesions and fat for an end-to-end stapled anastomosis. The largest possible anvil is pursestringed into the proximal colon (the old colostomy segment). The stapler is introduced through the anus and brought into the previously selected point for anastomosis. The trocar is opened through the wall of the cleaned distal segment, the anvil is snapped onto the stapler, and a standard stapling anastomosis is completed (Fig. 19B). The anastomosis is checked for leaks by insufflating air through a sigmoidoscope. The fascia of the colostomy site is closed, but the skin is left open. The port sites are closed both at the fascia and at the skin.

& TOTAL ABDOMINAL COLECTOMY AND ILEORECTAL ANASTOMOSIS Laparoscopic total abdominal colectomy with ileorectal anastomosis requires that many of the previously described resections be performed at one operation (68). Thus the steps to achieve this are reviewed. The patient is positioned in stirrups with the perineum accessible off the end of the operating table. The

position of the patient’s knees is low in the stirrups so that they will not interfere with arm motion of the surgeon standing between the patient’s legs. The team will frequently need to shift positions, and thus move the monitors and tilt and rotate the table, depending on the segment of bowel being mobilized or resected. After pneumoperitoneum, the patient is placed in a head-down position while the infraumbilical trocar is inserted. With the laparoscope in place, cannulas are placed in the right upper quadrant, left upper quadrant, left lower quadrant, right lower quadrant, and suprapubic sites according to the surgeon’s preference. With the handassisted approach, the two most caudad port sites are replaced with an incision of about the length of the surgeon’s glove size in the low midline or Pfannenstiel site (69). The ureters must be seen at some point, and it is preferable to locate them early and score the peritoneum along each side of the rectum so that the ureters are known to be out of the operative field. The rectum is mobilized by incising the peritoneum on each side of the rectum medial to the ureters and extending the incision downward on each side to connect anteriorly in the cul-de-sac. When the rectum has been retracted, the attachments are divided and the rectum is freed up off the sacral promontory. The incision is carried to the base of the inferior mesenteric vessels. The vessels are seen as they become taut when the rectum is pulled inferiorly. The peritoneum can be scored across the base of these vessels to a point on their right where the mesentery is very thin. This thin spot in the mesentery is opened and, through this window, the posterior aspect of the inferior mesenteric vessels is exposed. When the rectum has been freed up, it can be stapled. A rigid sigmoidoscope can be introduced to mark the 15-cm level where it is cleared of fat. The laparoscopic stapler can be fired once or more to transect it. Some surgeons select a site for specimen removal and open it early to use standard staplers for the rectal and ileal transection. The wound is closed to make the abdomen gas-tight again. Attention is turned to taking down the left-sided attachments of the rectum and sigmoid colon. The rectum and sigmoid colon are tethered by the inferior mesenteric

FIGURE 19 & Colostomy closure. (A) A rigid sigmoidoscope introduces light to aid in identification of the rectal stump, and it may lever the rectum up to aid in lysis of adhesions. (B) The stapler is inserted via the anus and the anastomosis is completed.

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

vessels. These vessels are divided together or separately by a vascular stapler, ligations, or clips. The left colon is mobilized medially and laterally as high as possible around the splenic flexure. The surgeon may move to a position between the patient’s legs and the assistants to a right-sided position. The table may need to be tilted head up and to the right to approach the splenic flexure. When the left colon has been mobilized, the table is rolled to the left to expose the cecum and right colon. The head of the table may need to be shifted head down to move the small bowel out of the way. Often the inferior margins of the ileum, appendix, and cecum are exposed earlier and freed up during exposure of the inferior mesenteric vessels. The surgeon stands between the patient’s legs, and the assistants move to the left side. The ileum is grasped and lifted to expose the ileocolic, attachments, which are taken down. The terminal ileum is cleaned of fat and transected with the laparoscopic linear stapler. With the right colon pulled medially, the white line of Toldt is incised to a point around the hepatic flexure. The table is shifted from a head-down to a head-up position as the dissection progresses. The ileocolic vessels are transected and the right mesentery is taken down. The right colic vessels, if present, are transected. As the colon is mobilized medially, the duodenum is seen and avoided. The ligaments attaching the liver to the colon are divided. The omentum can be removed from the transverse colon in a relatively bloodless plane immediately adjacent to the colon. As the transverse colon mesentery is mobilized, the lesser sac is entered. The middle colic vessels, which can be stapled, ligated, or clipped, will be evident. The remainder of the dissection entails removal of the attachments along the greater curvature of the stomach and the spleen. When the colon has been totally freed up, an incision is made in the lower abdomen through which the colon is removed. If a previous incision was made for stapling, it is reopened for colon removal. The terminal ileum is drawn out of the wound, opened, and a pursestring suture applied. The largest possible stapler anvil is inserted, and the pursestring suture is tied tight around the anvil shaft. The stapler is introduced through the anus, and the anastomosis is completed using the standard double-stapling technique.

& ROBOTICS Robotics has been applied to clinical surgery in the past few years, including colectomy. Weber et al. (70) presented the first two reported cases of telerobotic-assisted laparoscopic colectomies performed on March 6 and 8, 2001—a sigmoid colectomy for diverticulitis and a right colectomy for cecal diverticulitis. The Da Vinci telerobotic surgical system was used in both cases to mobilize the bowel. The mesenteric division, bowel transection, and anastomoses were accomplished with standard laparoscopic-assisted techniques. Both operations were completed with a three trocar technique. They found that the Da Vinci system adequately replaced the camera holder. The combination of the threedimensional imaging and the hand-like motions of the

363

telerobotic surgical instruments facilitated dissection. Operative time for the sigmoid colectomy was 340 minutes and for the right hemicolectomy 228 minutes. Merola et al. (71) compared the use of voice-controlled robotic camera holder (AESOP 3000; Computer Motion, Inc., Santa Barbara, California, U.S.A.) to a human camera holder in a series of laparoscopic colectomies. The use of the voice-controlled robotic camera did not alter the length of the operative procedure, the patient’s length of stay, or postoperative morbidity. However, surgeons often have a subjective sense that there is less smudging, fogging, and inadvertent movements of the laparoscope when it is controlled by a robotic system. Advocates suggest that advantages include a threedimensional view and easy instrument manipulation. The disadvantages are the current need for large diameter instruments and the availability of only three robotic arms. Major problems to be solved are the tactile feedback to the operator and the need for more arms. Ultimately, the value of robotics will need to be expressed in terms of safety, operating time, cost, and outcomes. The capital cost is significant with a purchase cost of over US $1,300,000 and maintenance costs of over US $100,000 annually (72,73). Munz et al. (74) operated on six patients with fullthickness rectal prolapse using the robotic-assisted suture rectopexy. All operations were completed successfully using the robotic system. There were no major complications and no deaths. Mean set-up time was 28 minutes, mean operation time was 127 minutes, and mean hospital stay was 6 days. At 3 to 6 months of follow-up there were no signs of recurrence. Delaney et al. (73) reported six robot-assisted laparoscopic operations (two right hemicolectomies, three sigmoid colectomies, and one Wells rectopexy) using standard laparoscopic procedures with robot-assisted laparoscopic colon mobilization and vascular ligation. There was no associated morbidity. Operative time was increased from a median of 108 minutes for standard laparoscopic colorectal surgery to 165 minutes for robot-assisted laparoscopic procedures. Blood loss, length of stay, and hospital costs were not significantly different between groups. Additional direct equipment costs for robot-assisted colectomy cases included robotic laparoscopic instruments and sterile drapes (approximately US $350.00 per case) without including acquisition and maintenance cost for the robot. They concluded robot-assisted laparoscopic colectomy is a feasible and safe procedure. Although three-dimensional vision and dexterity are facilitated, operative time is increased, and the overall additional expense of robotics is of concern. D’Annibale et al. (75) compared the traditional laparoscopic approach and robotic techniques with the Da Vinci robotic system (Intuitive Surgical) in the treatment of colorectal diseases in 106 patients (53 in each group). No differences were observed in the total time of operation (laparoscopic group 222 minutes vs. robotic group 240 minutes), the specimen length (laparoscopic group 29 cm vs. robotic group 27 cm), or the number of lymph nodes retrieved (laparoscopic 16 vs. robotic 17). It took significantly longer to prepare the operating room and patient in the robotic group (24 minutes in the robotic group vs. 18 minutes in the laparoscopic group). There were three

364

& PART III: MINIMALLY INVASIVE SURGERY

conversions to laparotomy in the laparoscopic group and in the robotic group two cases were converted to laparoscopy and three to hand-assisted laparoscopy. No significant differences were observed between the two groups in terms of recovery of bowel function and postoperative hospital stay. They found that because of its dexterity and threedimensional view, the Da Vinci system was particularly useful in specific stages of the procedure, e.g. takedown of the splenic flexure, dissection of a narrow pelvis, identification of nervous plexus, and hand-sewn anastomosis. Hanly et al. (76) used the robot to assist in 35 colon resections in cases that did not require multiquadrant dissection. They believe the robot may play a role in shortening the learning curve. Gutt et al. (77) reviewed all papers about telemanipulators used in visceral surgery. Most papers presented case series demonstrating the feasibility of robotic technology in performing a specific procedure. Comparative studies of robot-assisted surgery versus standard laparoscopic or open surgery were usually matched cohort studies. They generally showed an increased operating time for robot-assisted procedures but with similar rates of conversion, intraoperative and postoperative complications, and mortality in comparison with those of laparoscopic surgery. Consistent long-term follow-up data were missing and only one randomized clinical trial was conducted. Robot-assisted surgery appears safe and feasible for certain standard surgical procedures. However, they concluded that at its current level of development, it offers no clear, significant advantage over standard laparoscopic techniques.

& POSTOPERATIVE CARE The nasogastric tube is removed following completion of the procedure. Postoperative feeding has been initiated as early as the night of the operation, but the rush for this seems unfounded. Oral intake can often begin on the first or second postoperative day if there is no evidence of nausea or abdominal distention. Progression of diet is introduced as tolerated. Pain control is the same as that used after the open technique with a patient-controlled analgesia pump or systemic medication, but less medication is usually required.

& RESULTS Assessing and comparing the results of laparoscopic colorectal resections are Herculean tasks. Authors tend to lump procedures together, disparate disease processes together, and simple and complicated cases together. Therefore, a clear picture for a given procedure in a given disease process is almost never presented. Notwithstanding these limitations, a number of selected reports have been reviewed.

& CONVERSION RATES Information about conversion rates and the reasons for conversion is possibly not representative of the overall experience of surgeons performing laparoscopic procedures. Publications at this stage of evolution of the art are

presented by those most highly skilled in laparoscopic procedures. Although this criticism may be true of many publications, it is probably more applicable to laparoscopic operations. Notwithstanding this limitation, information is available. Reasons cited for conversion from a laparoscopic to an open procedure include poor exposure for whatever reason (often adhesions or obesity), uncontrollable bleeding, injury to other structures (viscera or vessels), absence of a lesion in the resected specimen, instrumentation difficulties, or lack of progress. The presence of an extensive carcinoma has also been cited, but the wisdom of using an initial laparoscopic approach under these circumstances might best be questioned. Conversion to an open technique should not be viewed as a failure but more appropriately as exercising good surgical judgment. Conversion rates between 0% and 42% (average 14%) have been reported in a review of 26 studies on laparoscope resection colon carcinoma (78). Conversion rates clearly decrease with experience. Patient selection and operative expertise profoundly influence conversion rates. Slim et al. (79) reviewed 16 of 65 laparoscopic-assisted colorectal operations that were converted to open procedures. When compared with 252 planned open operations, a higher postoperative morbidity rate (50% vs. 21%) and more anastomotic leaks (25% vs. 8%) were apparent in the group who underwent conversion to an open procedure. Operating time, postoperative ileus, and hospital stay were longer in those requiring a converted operation. The authors’ poor results suggest the need for careful preoperative patient selection and for rapid decision to convert difficult cases to open procedures. Pandya et al. (80) reviewed their indications for conversion to laparotomy during laparoscopic colectomy. From a registry of 200 patients who underwent laparoscopic colon surgery, 23.5% were converted. The 200 patients were divided into four cohorts of 50 consecutive patients to analyze changes with time. The conversion rate was statistically greater in the first quarter (36.0%) than in the subsequent quarters (16%). The rate of conversion to laparotomy for segmental resection of the ascending and descending colon (20.3%) has been equivalent and less than the conversion rate for other procedures (48.5%). The distribution of patients by operative indication has been fairly constant. The indication for operation has not influenced the need for conversion. The indications for conversion were technical problems in 15 patients (hypercarbia, unclear anatomy, and stapler misfire), laparoscopic complications in 9 patients (bleeding, cystotomy, and enterotomy), and problems that exceeded the limits of laparoscopic dissection in 23 patients (phlegmon, adhesions, obesity, and adjacent organ involvement by carcinoma). Although obesity accentuates the technical limitations of a laparoscopic dissection, it is an infrequent cause for conversion to laparotomy. Marusch et al. (81) reported on the significance of conversion for the results obtained with laparoscopic colorectal surgery and identified the risk factors that established the need for conversion. Within the framework of the Laparoscopic Colorectal Surgery Study Group, a total of 1658 patients were recruited to a multicenter study. The observed conversion rate was 5.2%. The patients requiring conversion were significantly heavier [body mass index (BMI) 26.5 vs. 24.9] than those undergoing pure

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

laparoscopy. Resections of the rectum were associated with a higher risk for conversion (20.9% vs. 13%). Intraoperative complications occurred significantly more frequently in the conversion group (27.9% vs. 3.8%). The duration of the operation was significantly increased after conversion in a considerable portion of the procedures performed. Postoperative morbidity (47.7% vs. 26.1%), mortality (3.5% vs. 1.5%), recovery time, and postoperative hospital stay were all negatively influenced by conversion. Institutions with experience of more than 100 laparoscopic colorectal procedures proved to have a significantly lower conversion rate than those with experience of fewer than 100 such interventions (4.3% vs. 6.9%). Although of itself, conversion is not considered to be a complication of laparoscopic surgery, it is true that the postoperative course after conversion is associated with appreciably poorer results in terms of morbidity, mortality, convalescence, blood transfusion requirement, and postoperative hospital stay. Conversion rates following laparoscopic colorectal surgery vary widely between studies, and the outcome of converted patients remains controversial. Gervaz et al. (82) conducted a comprehensive search of the English-language literature in which 28 studies of 3232 patients were considered for analysis. The overall conversion rate was 15.38%. Seventy-nine percent of the studies did not include a definition for conversion; in these studies the conversion rate was significantly lower than in the studies where a specific definition was considered (13.7% vs. 18.9%). Converted patients had a prolonged hospital stay (11.4 vs. 7.4 days) and operative time (209 minutes vs. 189 minutes) in comparison with laparoscopically completed patients. The factors associated with an increased rate for conversion were left colectomy (odds ratio ¼ 1.06), anterior resection of the rectum (odds ratio ¼ 1.09), diverticulitis (odds ratio ¼ 1.30), and carcinoma (odds ratio ¼ 2.94). They concluded the rate of laparoscopically completed colorectal resections is close to 85% in nonrandomized studies. Because converted patients have a distinct outcome, a clear definition of conversion is required to compare the results of randomized trials. Schlachta et al. (83) developed a simple model for clinical use in predicting the individual risk of conversion to open surgery in patients undergoing laparoscopic colorectal resections. From their experience of 367 laparoscopic colon resections, a scoring system was developed on the basis of the three factors found to be predictive of the risk of conversion to open surgery: diagnosis of malignancy (odds ratio 3.23; one point), surgeon experience with 50 cases or fewer (odds ratio 2.26; one point), and weight level (odds ratio 3.42; 60 to 90 kg, one point, 90 kg or more, two points). The predicted conversion rates for the cumulative scores of 0 to 4 points were 1.1%, 3.3%, 9.8%, 25.4%, and 49.7%, respectively. More recently, Tekkis et al. (84) developed a mathematical model for predicting the conversion rate for patients undergoing laparoscopic colorectal surgery using clinical data collected from 1253 patients undergoing laparoscopic surgery. Exclusion criteria for laparoscopic colectomy included a BMI > 50, lesion diameter exceeding 15 cm, and multiple prior major laparotomies (exclusive of appendectomy, hysterectomy, and cholecystectomy). The average conversion rate for the study population was 10%. The inde-

365

pendent predictors of conversion of a laparoscopic to open surgery were the BMI (odds ratio 2.1 per 10 American Society of Anesthesiology (ASA) units increase), ASA grade 3 or 4, 1 or 2 (odds ratio 3.2, 5.8), type of resection (low rectal, left colorectal, right colonic vs. small/other bowel procedures; odds ratio 8.82, 4.76, 2.98), presence of intraoperative abscess (odds ratio 3.6), or fistula (odds ratio 4.73), and surgeon seniority (junior vs. senior staff, odds ratio 1.56). Casillas et al. (85) conducted a study to determine whether conversion is associated with increased morbidity and higher hospital costs. In a series of 430 laparoscopic colectomies (12%) of cases were converted to open operation. There were no significant differences between the groups for age, male to female ratio, or American Society of Anesthesiology score. Conversions occurred before defining the major vascular pedicle/ureter (50%), in relation to intracorporeal vascular ligation (15%), or during bowel transection or presacral dissection (35%). Specific indications for conversion were technical (41%), followed by adhesions (33%), phlegmon or abscess (23%), bleeding (6%), and failure to identify the ureter (6%). Median hospital stay was 5 days for both groups. In-hospital complications (converted 11.6%; open 8%), 30-day readmission rate (converted 13% vs. 8%), and direct costs were similar between groups. There were no mortalities. They concluded, conversion of a laparoscopic colectomy does not result in inappropriately prolonged operative times, increased morbidity or length of stay, increased direct costs or unexpected readmissions compared with similarly complex laparotomies.

& POSTOPERATIVE ILEUS Basse et al. (86) assessed gastrointestinal transit in 32 patients randomized to laparoscopic or open colonic resection. They received 4 MGq of 111 indium diethylenetriamine pentaacetic acid, a tracer, at the end of operation. Images of the abdomen were obtained 24 and 48 hours postoperatively. Defecation occurred on median day 2 postoperatively in both groups. At 48 hours postoperatively, 53% of the tracer was excreted by patients in the laparoscopic group when compared with 26% in the open group. Postoperative ileus and gastrointestinal transit normalized within 48 hours after colonic resection in the patients who received multimodal rehabilitation (epidural analgesia, early oral nutrition and mobilization, and laxative use). No significant difference was observed between the patients who underwent the laparoscopic procedure than those who underwent the open procedure. Kasparek et al. (87) investigated whether colonic motility increases more rapidly following laparoscopicassisted sigmoid resection in 11 patients compared with open sigmoid colectomy in nine patients. During operation, a manometry catheter was inserted into the colon via the anus, and the tip was placed in the splenic flexure. Continuous manometric recordings were performed from the day of operation until postoperative day 3 with a fourchannel microtransducer manometry system combined with a portable data logger. The postoperative colonic motility was 101, 199, and 163 mmHg/min on days 1, 2, and 3 after laparoscopic-assisted sigmoid resection, respectively, which was increased compared with indexes of 53, 71, and 76 following open sigmoid colectomy. The

366

& PART III: MINIMALLY INVASIVE SURGERY

amplitude but not the frequency of contractions was higher following laparoscopic-assisted sigmoid resection compared with open sigmoid colectomy. Following laparoscopicassisted sigmoid colectomy, patients requested a similar amount of pain medication but resumed oral food more rapidly on postoperative days 2 and 3 and they were discharged from the hospital earlier. Colonic motility in particular and the patient’s condition in general seem to improve more rapidly following laparoscopic-assisted sigmoid colectomy compared with the open procedure. In a novel study Asao et al. (88) evaluated the effect of gum chewing as a convenient method to enhance postoperative recovery from ileus after laparoscopic colectomy. A total of 19 patients were randomly assigned to one of two groups: a gum-chewing group (n ¼ 10) or control group (n ¼ 9). The patients in the gum-chewing group chewed gum three times a day from the first postoperative morning until oral intake. The first passage of flatus was seen, on average, on postoperative day 2.1 in the gum-chewing group and day 3.2 in the control group. The first defecation was 2.7 days sooner in the gum-chewing group (postoperative day 3.1) than in the control group (5.8 days). All patients tolerated gum-chewing on the first operative morning. They concluded gum-chewing aids early recovery from postoperative ileus and is an inexpensive and physiologic method for stimulating bowel motility. Raue et al. (89) reported their experience with a multimodal rehabilitation program (‘‘fast-track’’) with epidural analgesia, early oral feeding, and enforced early mobilization following laparoscopic colorectal resection. Twenty-nine standard-care patients and 23 fast-track patients were evaluated. On the first postoperative day, pulmonary function was improved in fast-track patients. Oral feeding was achieved early and defecation occurred earlier in the fasttrack group. Visual analogue scale scores for pain were similar for the two groups, but fatigue was increased in the standard-care group on the first and second postoperative days. Morbidity was not different for the two groups. Fast-track patients were discharged on day 4 and standard-care patients on day 7. They concluded, multimodal rehabilitation can improve further on the excellent results of laparoscopic sigmoidectomy and decrease the postoperative hospital stay.

& LAPAROSCOPIC PROCEDURES IN THE ELDERLY Reissman et al. (90) assessed the outcomes of laparoscopic colorectal operations in patients younger than 60 years with those older than 60 years. There were no statistically significant differences relative to the incidence of complications (11% vs. 14%), conversion (8% vs. 11%), length of ileus (2.8 days vs. 4.2 days), or length of hospitalization (5.2 days vs. 6.5 days). Peters and Fleshman (91) evaluated 103 patients aged 65 years of age or older who underwent attempted minimally invasive colectomy (right colectomy, 53; left colectomy, sigmoid colectomy, or anterior resection, 36; abdominoperineal resection, 12; and total proctocolectomy with ileostomy, 2). Eighty-one procedures were successfully completed. Complications occurred in 23% of patients converted to laparotomy (including one death) and in 25% undergoing successful minimally invasive

colectomy (two deaths). The average length of postoperative stay was 5.3 days in the minimally invasive colectomy group and 8.1 days in patients converted to laparotomy. These results compare favorably with published results of traditional colectomy for elderly patients. Delgado et al. (92) found that morbidity rates were similar in patients less than 70 years of age but found a significantly lower incidence of complications after laparoscopic colectomy compared with open colectomy in patients over 70 years of age. There may be a selected benefit of laparoscopic left colectomy in the elderly and laparoscopic right hemicolectomy in the young (93). Law et al. (94) also found laparoscopic colorectal resections a safe operation for elderly patients and is especially associated with more favorable short-term outcomes in terms of early return of bowel function, earlier resumption of diet and shorter hospital stay as well as associated with less cardiopulmonary morbidity. Seshadri et al. (95) identified 62 octogenarians from a large prospective database comprising 507 consecutive laparoscopic colorectal resections. The mean age and weight were 85 years and 63 kg, respectively. Seven patients (11%) were converted to an open procedure. Four (6%) intraoperative complications occurred in four patients (one colon perforation, one small bowel perforation, one burned gallbladder serosa, and one missed lesion), necessitating two conversions. Twenty-four postoperative complications occurred in 19 patients (31%) [six ileus (10%), five wound infections (8%), five cardiac problems (8%), two urinary retentions (3%), two hemorrhages (3%), one abscess (2%), one pneumonia (2%), and two other (3%)]. Intraoperative complications did not increase postoperative morbidity. Three patients (5%) died within 30 days of operation. When the procedure was completed laparoscopically, the overall median postoperative hospital stay was 10 days; occurrence of a postoperative complication increased the median length of stay to 15 days. They concluded these results are superior to the published historical controls involving open colorectal resections in octogenarians. Overall mortality, lung, and urinary tract complications were decreased and there were no reoperations for small bowel obstruction. Vignali et al. (96) defined the benefits in terms of early outcome for laparoscopic colectomy in patients over 80 years old compared with open colectomy. They matched 61 patients undergoing laparoscopic colectomy for colorectal carcinoma with 61 open colectomy patients. The mean age was 82 years in the laparoscopy group and 83 years in the open group. The conversion rate was 6.1%. Operative time was 49 minutes longer in the laparoscopy group. The overall mortality rate was 2.4%. The morbidity rate was 21.5% in the laparoscopy group and 31.1% in the open group. Patients in the laparoscopy group had a faster recovery of bowel function and a significant reduction of the mean length of hospital stay (9.8 days vs. 12.9 days for the open group). Laparoscopy allowed a better preservation of postoperative independence status compared with that of the open group. They concluded laparoscopic colectomy for carcinoma in octogenarians is safe and beneficial.

& LEARNING CURVE The learning curve may be defined in different ways but is essentially that point when the operating time, conversion

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

rates, and complications have stabilized. The consensus number appears to be about 30 laparoscopic colon resections (78). Schlachta et al. (97) reviewed their data to define the learning curve for laparoscopic colorectal resections. A total of 461 consecutive resections were evenly distributed among three surgeons. Cases 1 to 30 were considered ‘‘early experience’’ whereas cases 31 and higher were combined as ‘‘late experience.’’ There were greater proportions of rectal resections performed (14% vs. 32%) in the late experience. Trends toward declining rates of intraoperative complications (9% vs. 7%) and conversion to open surgery (13.5% vs. 9.7%) were observed with experience. Median operative time (180 vs. 160 minutes), and overall length of postoperative hospital stay (6.5 vs. 5 days) declined significantly with experience. There was no difference in the rate of postoperative complications between the early and late experience (30% vs. 32%). They concluded the learning curve for performing colorectal resections was approximately 30 procedures in their study. Bennett et al. (98) reported that there is a learning curve for laparoscopic-assisted colectomy with respect of intraoperative and postoperative outcomes. As might be expected surgeons who performed higher volumes of laparoscopic-assisted colectomy have lower rates of intraoperative and postoperative complications. They analyzed data of 1194 patients operated on by 114 surgeons from a prospective registry sponsored by the American Society of Colon and Rectal Surgeons. In 75% of cases, operation was completed laparoscopically with no difference between high-volume surgeons (40 cases) and low-volume surgeons. Length of stay (average 6 days) did not vary according to surgeon volume. Postoperative complications occurred in 15% of cases with a significantly lower rate for high-volume surgeons (10% vs. 19%). Intraoperative complications occurred in 5% of cases, with a nonsignificant trend toward a lower rate of high-volume surgeons (3.7% vs. 6.3%). A multivariate regression analysis, adjusting for type of disease (carcinoma vs. inflammatory vs. polyps) and for level of difficulty of the procedure (high vs. low) showed that for high-volume surgeons, there is a lower probability of both intraoperative complications (adjusted odds ratio 0.56) and postoperative complications (adjusted odds ratio 0.48). Dincler et al. (99) defined a multidimensional learning curve for sigmoid resection performed by two surgeons with experience in laparoscopic surgery. Surgeon A performed 199 and surgeon B 139 sigmoid resections. The operation time decreased from 225 minutes to 169 minutes after approximately 90 operations for surgeon A and from 270 minutes to 223 minutes after 110 operations for surgeon B. Based on a decline in intraoperative complications and conversion rate, the steady state was reached after approximately 70 to 80 interventions for both surgeons. They concluded the assessment of a learning curve should not be limited to measurement of a decrease in operation time but also should include the conversion and complication rates. Concern for prevention of untoward sequelae associated with the ‘‘learning curve’’ prompted establishment of the Laparoscopic Bowel Surgery Registry by the American Society of Colon and Rectal Surgeons, The Society of American Gastrointestinal Endoscopic Surgeons, and

367

The American College of Surgeons Commission on Cancer to identify as early as possible the pattern of practice and acute complications of laparoscopic colectomy (100). Cases were voluntarily registered by community and academic surgeons. Of 1056 cases contributed by 118 surgeons, 763 patients were completed laparoscopically. Indications for operation were carcinoma (43%), inflammatory diseases (23%), polyps (19%), other diagnoses (13%), and unknown indication (2%). The site of pathology was the right colon in 36%, sigmoid colon in 28%, rectum in 15%, left colon in 8%, transverse colon in 2%, and small bowel in 3%. Resection was performed in 91%, bypass in 4%, and polypectomy in 1%. The operative procedure was not reported in 4%. Completion of the intended procedure was possible in 73%. Extracorporeal anastomoses (61%) and intracorporeal anastomoses (18%) were performed. Laparoscopic-initiated procedures were converted to open laparotomies in 24%. Reasons stated for conversion to open operation were unclear anatomy in 23%, bleeding in 4%, perforation in 2%, and other reasons in 42%. No precise reason for conversion was stated in 2%. Intraoperative laparoscopic complications developed in 5%. Intraoperative bleeding, perforation, or contamination occurred with overall rates of 1.2%, 1.1%, and 0.1%, respectively. The overall incidence of postoperative complications was 15% and conforms to other experiences in open operation. The incidence of postoperative complications directly related to the laparoscopic approach was 2.7%. Patients whose bowel operation was performed and completed laparoscopically averaged 5.6 days of hospitalization postoperatively, whereas those converted to open laparotomy went home an average of 8.4 days postoperatively. In an invited editorial regarding the registry, Abcarian noted that laparoscopic colectomy has been associated with an array of complications heretofore rare in ‘‘open’’ colectomy. These include division of the ureters, perforation of the aorta, vena cava, and iliac artery and vein, perforation of the small bowel and colon, gross contamination of the peritoneal cavity with feces and carcinoma, uncontrolled bleeding, and recurrence at the trocar or port sites. The rate of these complications in the literature is significant and cannot be reconciled with data presented at the registry. Abcarian continues to note that anecdotal reports of complications heard in the hallways of hospitals and meeting rooms during any surgical society meeting far exceed those reported in the registry, even if one takes into consideration the infamous ‘‘learning curve.’’ When cases are contributed to a registry on a voluntary basis, the registry is subject to a worst case bias because surgeons are inherently reluctant to report worst case scenarios. Another gross misrepresentation in the registry is the personal subjective opinion of surgeons performing laparoscopic colectomy and declaring that they did not violate the principles of oncologic surgery in 88% of cases. Abcarian stated that in the absence of objective data, such as proximal and distal margins, extent and number of lymph nodes removed, local, regional, anastomotic, or port-site recurrences, and most important, without prospective data comparing the 5-year survival rates of colon carcinoma patients operated on in conventional open technique vs. laparoscopic procedure, the data presented in the registry are unacceptable because of the lack of scientific merit. It must be remembered that data from a registry

368

& PART III: MINIMALLY INVASIVE SURGERY

are probably not representative of results in the general population of surgeons. Participants are surgeons experienced in laparoscopic procedures. Kuhry et al. (101) assessed the impact of hospital case volume on short-term outcome after laparoscopic operation for 536 colon carcinoma patients within the Colon Cancer Laparoscopic or Open Resection (COLOR) trial. Median operating time was 240, 210, and 188 minutes in centers with low, medium, and high case volumes, respectively. A significant difference in conversion rate was observed among low, medium, and high case volume hospitals (24% vs. 24% vs. 9%). A higher number of lymph nodes were harvested at high case volume hospitals. After operation, fewer complications and a shorter hospital stay were observed in patients treated at hospitals with high caseloads. They concluded laparoscopic operation for colon carcinoma at hospitals with high caseloads appears to be associated with improved short-term results.

& MORBIDITY AND MORTALITY Most publications consist of patients with grouped pathologic processes and procedures. However, several authors have addressed individual disease processes or procedures and hence will be presented accordingly. Attempts were made to quote the most recent work from an institution with multiple publications and when possible to use reports with substantial numbers of patients. Combined Reports Jacobs et al. (2) reported the first series of laparoscopicassisted colectomies. Of the 20 patients, 9 patients underwent a right hemicolectomy; 8 a sigmoid colectomy, and 1 each underwent a low anterior resection, Hartmann’s procedure, and abdominoperineal resection. Indications for operation were large villous adenomas or adenocarcinoma in 12 patients, diverticular disease in 5, sigmoid endometrioma in one, cecal volvulus in one, and inflammatory bowel disease (IBD) in one. Eighty percent of patients were able to tolerate a liquid diet on the first postoperative day, and 70% were discharged within 96 hours after operation eating a regular diet and having normal bowel movements. There were three operative complications. A patient with a 3-unit postoperative bleed was managed without operation, one patient developed marked edema of the rectosigmoid anastomosis requiring decompression with a rectal tube, and one individual with metastatic colon carcinoma was operated on for a mechanical small bowel obstruction 7 days after the initial laparoscopic operation. Wexner et al. (102) reported the results of their first 140 consecutive patients who underwent laparoscopic or laparoscopic-assisted colorectal operations. Indications for operation included IBD in 47 patients, colorectal carcinoma in 19, diverticular disease in 17, polyps in 16, familial polyposis in 7, colonic inertia in 7, feed incontinence in 11, sigmoidocele in 3, irradiation proctitis in 3, rectal prolapse in 2, intestinal lymphoma in 2, and miscellaneous conditions in 6. The procedures included 38 total abdominal colectomies (ileoanal reservoir in 28, ileorectal anastomosis in 8, and end ileostomy in 2); 70 segmental resections of the colon, small bowel, and rectum; 18 diverting stoma creations; 10 reversals of Hartmann’s procedure; and

four other procedures. In 15 cases, the laparoscopic procedure was converted to a laparotomy (11%). Thirty-one patients (22%) sustained 37 complications, which included enterotomies (seven), hemorrhage (10), intra-abdominal abscess (four), prolonged ileus (six), wound infection (four), intestinal obstruction (two), anastomotic leak (one), aspiration (one), cardiac arrhythmia (one), and upper intestinal bleeding (one). There was no morality. The overall complication rate in total abdominal colectomy cases was significantly higher (42%) when compared with all other procedures (segmental resection, 17%; others, 9%). The mean length of operating time was 4.0 hours (range, 2.5 to 6.5 hours) for total abdominal colectomy, 2.6 hours (range, 1.5 to 5.5 hours) for segmental colonic resections, and 1.7 hours (range, 0.7 to 4 hours) for all other procedures. The length of ileus was 3.5 days (range, 2 to 7 days) after total abdominal colectomy, 3 days (range, 2 to 7 days) after the segmental resections, and 2 days (range, 1 to 4 days) after the other procedures. The mean length of hospital stay was 6.8 days (range, 2 to 40 days)(8.4, 6.5, and 6.3 days for total abdominal colectomies, segmental resections, and other procedures, respectively). Ramos et al. (103) studied the role of laparoscopy in colorectal surgery and found that of 200 patients assessed, 94% were considered appropriate for laparoscopic surgery, 65% of which were successfully completed (right hemicolectomy, 24 of 30; sigmoid resection, 22 of 36; appendectomy, 9 of 10; anterior resection, 3 of 8; abdominoperineal resection, 3 of 5; and left hemicolectomy, 1 of 2). Complications attributed to laparoscopy were 6.3%. Overall morbidity was 28% in the open procedure group, 18% in the converted group, and 15% in the laparoscopic group. Huscher et al. (104) reported on 200 patients who received laparoscopic colorectal resections for benign (49) or malignant (127) lesions. Twenty-one of 200 patients were converted to open surgery (10.5%), 37 patients had a complete laparoscopic procedure (17.1%), 137 had an assisted resection (68.5%), and the remaining five patients had a facilitated resection. The mean operative time was 208 minutes (range, 90 to 480 minutes) for assisted resection and 275 minutes (range, 54 to 550 minutes) for complete laparoscopic resection. The mortality rate was 1.7% and the overall morbidity rate was 19.6% (the rate of major complications was 11.2%). All patients quickly became ambulatory and demonstrated a prompt resumption of gastrointestinal functions and less postoperative pain compared with converted cases. The average number of lymph nodes was 12.1 (range, 1 to 32 nodes). The mean hospital stay was 8.6 days (range, 5.0 to 14.5 days). The mean follow-up was 16 months (range, 6 to 24 months). The recurrence rate was 11.7%. Begos et al. (105) reported on their first 50 patients undergoing laparoscopic techniques. These patients were compared with 34 consecutive patients undergoing open resections during the same time period. Overall, 33 patients (66%) were completed laparoscopically. This rate increased to 87% after the first 20 patients. Patients undergoing laparoscopic procedures showed significant improvement compared with patients undergoing open and converted procedures in several areas. Operative blood loss was decreased. Patients ate sooner (3.7 days) and required less postoperative pain medication. Major complications were less common

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

after laparoscopic operations. The average length of stay was 8.3 days compared with 13.9 days and 14.5 days in the converted and open groups, respectively. There was no difference in the operative time between laparoscopic and open cases; the time for converted cases was significantly longer. There was no difference in lymph node counts in patients who underwent resection for carcinoma. Zucker et al. (46) reported on 65 patients who underwent attempted laparoscopic colon resection. Indications for surgical intervention included carcinoma in 39 patients, adenomatous polyps in 14, diverticulosis in 10, stricture in one, and foreign-body perforation in one. A laparoscopicassisted technique in which the specimen was removed and the anastomosis was completed outside the abdomen was used in all patients. A dilated umbilical opening was used for right-sided lesions and a left lower quadrant muscle-splitting incision was used for descending and sigmoid colon resections. Two patients required conversion to open laparotomy. There were no deaths and only four complications (one case each of pneumonia, urinary tract infection, prolonged ileus, and subfascial abscess). The mean postoperative stay was 4.4 days (range, 3.0 to 8.0 days), and the average interval for return to normal activity was 8 days. Tucker et al. (106) retrospectively analyzed the results of their first 114 laparoscopic-assisted bowel procedures. The procedures performed consisted of partial colectomy in 85 patients, total or subtotal abdominal colectomy in eight, total proctocolectomy with J-pouch ileal reservoir in 11, and diverting procedures in 10. Forty-nine procedures were performed for malignancy. The rate of conversion to laparotomy was 13.2%. Oral feedings were resumed in 2.4 days (range, 1.0 to 5.0 days), and bowel function returned in 3.8 days (range, 2.0 to 8.0 days). The average length of stay was 4.2 days for partial colectomy and 6 days for total, subtotal, and proctocolectomy. The mean return to normal activity for all groups was 16.7 days (10.8 days for partial colectomy). There were no deaths. Major morbidity (6%) consisted of abscess (three), anastomotic leak (two), and hemorrhage (one). Mean operative costs analyzed for the initial 37 patients were higher for laparoscopic colectomy whencomparedwithtraditionalcolectomy;however,themean total hospital costs were less for laparoscopic procedures. Lumley et al. (107) summarized the outcome for the first 240 patients who underwent a laparoscopic colorectal procedure. All laparoscopic data were collected prospectively, and in selected studies data were compared with open surgical controls. Nineteen patients required open conversion (7.9%). There was a significant decrease in wound infection rates in patients who underwent laparoscopic-assisted colectomy (3.6%) compared with historical controls (7.9%). There were five anastomotic leaks, five laparotomies performed for postoperative adhesive obstruction, and four perioperative deaths. A total of 103 patients had a procedure for colorectal carcinoma. Of the 79 potentially curative procedures, there have been five (6.3%) recurrences to date. Slim et al. (108) reported their initial 40 cases of laparoscopic-assisted colorectal resection that were prospectively evaluated. The operations were performed for colonic neoplasms of the right segment (four), sigmoid (11), or rectum (seven), diverticular disease (17), and chronic constipation (one). Of 22 neoplasms, 11 were malignant.

369

The operative procedures were four right hemicolectomies, 28 segmental left colectomies, five anterior resections, two abdominoperineal resections, and one total colectomy. Thirty-one patients (77.5%) underwent a successfully completed laparoscopic-assisted resection. The reasons for conversion in the majority of the cases (66.6%) were difficulties in dissection. In the entirely laparoscopic-assisted procedures, the mean time until passage of flatus was postoperative day 3, the mean postoperative hospitalization was 10.7 days, and there were eight complications in seven patients (25%). Two patients were reoperated 2 and 3 months later for adhesion and ischemic stenosis of the colon above the anastomosis. There was one death in the laparoscopic-assisted group (3.2%). The length of operative specimen was 19.6 cm, and the mean number of resected lymph nodes was six. The authors concluded that in contrast to laparoscopic biliary surgery, the benefits of laparoscopic-assisted colorectal surgery are not obvious. Fleshman et al. (109) compared laparoscopy with minilaparotomy approaches to colorectal diseases. Minilaparotomy was performed in 35 patients to achieve right colectomy (14), left colectomy (eight), total colectomy (two), low anterior resection (six), abdominoperineal resection (two), colostomy (one), and ileal resection (one). Laparoscopic techniques were used in 52 patients to perform right colectomy (20), left colectomy (11), low anterior resection (five), abdominoperineal resection (seven), total colectomy (three), ileal resection (one), colostomy (three), transverse colectomy (one), and colostomy closure (one). Mean operative times were 69 minutes for minilaparotomy (range, 33 to 180 minutes) and 173 minutes for laparoscopy (range, 60 to 300 minutes). Mean incision lengths were 12 cm (range, 8 to 18 cm) and 8 cm (range, 0 to 25 cm); mean time to bowel movement was 4 days (range, 1 to 7 days) and 3.9 days (range, 0 to 8 days); mean day of discharge was 6.9 days (range, 3 to 15 days), and 6 days (range, 1 to 15 days) postoperatively, respectively. Laparoscopy procedures were completed in 39 of 52 patients (75%); the mean time to bowel movement was 3.5 days (range, 0 to 6 days), and the mean day of discharge was 5.3 days (range, 1 to 14 days). The authors concluded that the use of a small incision, whether by minilaparotomy or by laparoscopy, results in similar early return of function and discharge. An effort to summarize data from such diverse reports in which authors evaluate different aspects and place emphasis on different elements of this new technology is easier said than done. Notwithstanding this difficulty, values for certain relevant concerns have been published. Of paramount importance is the complication rate, which ranges from 6.9% to 25% (2,46,102–104,108), with an average of approximately 15% (100). Reported conversion rates for laparoscopic colectomy to open colectomy range from 3% to 13.2% (46,102,104,106,107), with most reports in the upper range. The operative time varies with the type of procedure performed and is recorded to be from 60 to 390 minutes, with most in the 175- to 200-minute range (102,104,108). Return of bowel function ranges from 1 to 8 days, with an average of approximately 4 days (102,106,108). Postoperative feeding has been offered as early as the first postoperative day and is believed to be possible sooner than with open colectomy, an average 2.4 to 3.7 days (range, 1 to 5 days) (2,105). The hospital length

370

& PART III: MINIMALLY INVASIVE SURGERY

of stay ranges from 2 to 40 days (2,102,104,106,108,110) but averages 5.6 days (100). The operative mortality rate is noted to be 0% to 3.2% (46,104,107,108). Less postoperative medication is required (105). Patients return to normal activity in approximately 17 days (106). Understandably, laparoscopic colectomy costs are higher than for open colectomy (106). Deguili et al. (111) reviewed 108 patients undergoing laparoscopic colorectal surgery for large bowel disease. Conversion to open was necessary in 4.6% of patients whereas a mandatory conversion was needed in 9.2% of patients because of advanced carcinoma. The overall morbidity rate was 11.9%. There were no anastomotic leaks. In two patients (1.85%) a complication developed that required reoperation. Postoperative mortality was nil. Mean postoperative stay was 7.2 days. No trocar site recurrences were observed in the carcinoma patients. Senagore et al. (112) reviewed the outcome of the standardization of all the intraoperative and postoperative processes used in their department for the performance of laparoscopic sigmoid colectomy for all colonic pathologies. Criteria for exclusion for an attempted laparoscopic sigmoid colectomy were BMI > 35 and prior major abdominal operations (exclusive of hysterectomy, cholecystectomy, appendectomy). Conversion was performed when a sequential step could not be completed in a reasonable time frame. A standard perioperative care plan was used. A total of 207 colectomies were performed with a 12.1% conversion rate. Indications for the laparoscopic sigmoid colectomies were diverticular disease (96), colonic neoplasia (27), prolapse (14), endometriosis (10), and other (10). Mean operative time was 119 minutes. Mean length of stay was 2.9 days for completed cases and 6.4 days for converted cases. Anastomotic leaks occurred in two patients (1.1%), one of whom died of multisystem organ failure yielding an operative mortality of 0.6%. The overall complication rate was 6.6% and the 30-day readmission rate was 8%. Lezoche et al. (113) reported a prospective nonrandomized study based on a series of 469 consecutive patients (73.6% with malignant lesions) for laparoscopic and open approach. There were 166 patients who underwent right hemicolectomy (RHC) and 303 left hemicolectomy (LHC). In the RHC group, 108 patients underwent laparoscopic approach and 58 underwent an open operation (26 vs. 13 for benign lesions and 82 vs. 45 for adenocarcinomas, respectively). Left hemicolectomy was performed by laparoscopy in 202 patients and by laparotomy in 101 (55 vs. 30 for benign lesions and 147 vs. 71 for adenocarcinomas, respectively). There were no conversions to open operation in laparoscopic right hemicolectomy while 10 patients (4.9%) in the laparoscopic LHC group required conversion: three of 34 performed for diverticular disease and seven of 147 performed for malignancy. Mean operative time for laparoscopic surgery was longer than for open surgery (182 vs. 140 minutes for RHC and 222 vs. 190 minutes for LHC, respectively), but with increasing expertise this decreased significantly. Mean hospital stay in patients who underwent laparoscopic procedures was significantly shorter both in right hemicolectomy and left hemicolectomy groups (9.2 vs. 13.2 days and 9.9 vs. 13.2 days, respectively). Similar major complication rates were observed between the two laparoscopic and open groups

(1.8% vs. 1.7% for RHC and 4.1% vs. 4.9% for LHC, respectively). Mean follow-up time was 57.3 months in the RHC group and 57.5 months in the LHC group. The local recurrence rate was lower after laparoscopic surgery in both arms (7% vs. 8.8% for RHC and 3.3% vs. 7% for LHC, respectively), but the differences were not statistically significant. Two port-site recurrences were observed in the laparoscopic group, one after a palliative right hemicolectomy, one after a Dukes C left hemicolectomy converted to open operation (1.7% vs. 0.9%, respectively). Metachronous metastases rates were similar between the laparoscopic and open groups (20.9% vs. 17.6% for RHC and 4.4% vs. 5.3% for LHC, respectively). Cumulative survival probability at 72 months after laparoscopic right hemicolectomy was 0.79 when compared with 0.77 after open operation and 0.96 after laparoscopic left hemicolectomy compared with 0.88 after open operation. Cumulative survival probability for Dukes A, B and C in the laparoscopic right hemicolectomy group was 0.88, 0.85, and 0.73 when compared with 0.9, 0.89, and 0.6 after open surgery, respectively. Cumulative survival probability for Dukes A, B and C in the laparoscopic left hemicolectomy was 0.1, 0.97, and 0.89 when compared with 0.1, 0.94 and 0.7 after open operation, respectively. Their results suggest that laparoscopic hemicolectomy for both benign and malignant lesions can be performed safely. Oncologic outcomes were comparable with those of open operation. Schlachta et al. (114) reviewed the 10-year experience of a surgical group from university teaching hospitals in three centers. There were 750 laparoscopic colon and rectal procedures attempted of which 669 were completed laparoscopically. Malignant disease was the indication for operation in 49.6% of cases. Right hemicolectomy and sigmoid colectomy accounted for 54.5% of procedures performed. Intraoperative complications occurred in 8.3% with 29% of these resulting in conversion to open operation. The overall rate of conversion to open operation was 10.8%, most commonly for oncologic concerns. Median operative time was 175 minutes for all procedures. Postoperative complications occurred in 27.5% of procedures completed laparoscopically but were mostly minor wound complications. Pulmonary complications occurred in only 1%. The anastomotic leak rate was 2.5%. The early reoperation rate was 2.4%. Postoperative mortality was 2.2%. No port-site metastases have yet been detected. The median postoperative length of stay was 5 days. Lauter and Froines (115) reported their experience with 155 procedures. Mean operative time for completed laparoscopic-assisted colectomy, converted procedures, right and sigmoid resections were 164, 203, 121, and 177 minutes, respectively. Twenty-two patients had additional concurrent laparoscopic procedures. Thirty-nine had undergone previous abdominal operation. The conversion rate was 12%. Mean length of stay for all patients was 4.5 days. There were eight major and 16 minor complications. There were no portsite metastases. Major complications and conversion rate decreased from the first 50 cases to the last 50 cases. Seshadri et al. (116) compared the outcomes of laparoscopic total abdominal colectomy and laparoscopic total proctocolectomy with institutional open procedures used as controls. A total of 73 total abdominal colectomies for various disease processes including IBD and neoplasia

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

were evenly distributed between laparoscopic (n ¼ 37) and open (n ¼ 36) approaches. The median operative time was longer with the laparoscopic method (270 vs. 178 minutes) but the median length of hospital stay was significantly shorter (6 vs. 9 days). The short-term postoperative complication rate up to 30 days from operation was not statistically different (25% vs. 44%), although there was a clear trend toward a reduced number of overall complications in the laparoscopic group (9 vs. 24). Wound complications were significantly fewer (0% vs. 19%) and postoperative pneumonia was nonexistent in laparoscopic patients. Long-term complications also were less common in the laparoscopic group (20% vs. 64%) largely because of reduced incidence of impotence, incisional hernia, and ileostomy complications. Total proctocolectomy was performed laparoscopically in 15 patients and with an open procedure in 13 patients over the same period. Median operating time was longer for the laparoscopic patients (400 minutes vs. 235 minutes) whereas the length of hospital stay and morbidity and mortality were not significantly different. The results indicate that laparoscopic total abdominal colectomy can be performed safely with a statistically significant reduction in wound and long-term postoperative complications when compared with its open counterpart. Kockerling et al. (117) investigated the results of laparoscopic colorectal surgery in 500 consecutive patients operated on by unselected surgeons in 18 centers; 269 operations were performed for benign indications and 231 for carcinoma (palliative and curative). An anastomosis was performed in 84% with an overall leakage rate of 5.3% (colon 3.6% and rectum 11.8%), which required operative reintervention in 1.7%. The mean operating time was 176 minutes and showed a decrease in tendency over the period under study. The conversion rate was 7.0% and the overall complication rate 21.4%. The reoperation rate was 6.6; the most common cause was bleeding. There was one ureteral lesion (0.2%), but urinary tract infections were fairly common (4.8%). A postoperative pneumonia was diagnosed in 1.6% of cases. No thromboembolic complications were reported. The 30-day mortality rate was 1.4% and overall hospital mortality 1.8%. Kockerling et al. (118) investigated the safety of laparoscopic colorectal surgery as reflected by the anastomotic insufficiency rates in the various sections of the bowel and compared these rates with those of open colorectal surgery. From this prospective multicenter study, the 24 participating centers treated 1143 patients. In all, 626 operations were performed for benign indications and 517 for carcinoma. Most procedures involved the sigmoid colon and rectum (80.9%). An anastomosis was performed in 83% of the operations. Most of the anastomoses were laparoscopic-assisted using the stapling technique. They observed an overall leakage rate of 4.3% (colon 2.9%; rectum 12.7%) and surgical reintervention was required in 1% of the cases. The rate of conversion to open surgery was 5.6%. Intraoperative complications occurred in 5.9% and reoperation was necessary in 4.1% of cases. The overall morbidity rate was 22.3% and the 30-day mortality rate was 1.6%. Morbidity and mortality rates with this method approximate those seen with conventional colorectal surgery.

371

Schwenk et al. (119) conducted a Cochrane review to compare laparoscopic and conventional colorectal resection with regards to possible benefits of the laparoscopic method in the short-term postoperative period (up to 3 months post-surgery). All randomized controlled trials were included regardless of the language of publication. Operative time was longer in laparoscopic surgery, but intraoperative blood loss was less than in conventional surgery. Intensity of postoperative pain and duration of postoperative ileus were shorter after laparoscopic colorectal resection and pulmonary function was improved after a laparoscopic approach. Total morbidity and local (surgical) morbidity were decreased in the laparoscopic groups. General morbidity and mortality were not different between groups. Until the 30th postoperative day, quality of life was better in the laparoscopic patients. Postoperative hospital stay was less in the laparoscopic patients. These authors concluded that under traditional perioperative treatment, laparoscopic colonic resections show clinically relevant advantages in selected patients. Furthermore, if the long-term oncologic results of laparoscopic and conventional resection of colonic carcinoma show equivalent results, the laparoscopic approach should be preferred in patients suitable for this approach to colectomy.

Carcinoma Since the introduction of laparoscopic colon surgery, the propriety of adopting this technology to carcinoma has been questioned. Are sound oncologic principles of colorectal resection violated by this approach? Is there a different pattern of recurrence? Is disease-free survival compromised? Many publications have advocated the procedure but were often retrospective small series with relatively short follow-up. Patankar and Lee (1) recently conducted a comprehensive review of published data. In their assessment of oncologic equivalency of laparoscopic resection they reviewed 10 publications with respect to the resection margins and lymph node yield and on both accounts found that laparoscopic resection achieves the same oncologic resection as open operation. Several prospective and retrospective reviews have reported survival to be comparable with those of open resections. Retrospective and Prospective Reviews Long-term survival data has been reported by three large studies, two of these are retrospective case reviews and one is a prospective randomized controlled trial (120–122). Lujan et al. (121) reported the oncologic results of 122 consecutive patients undergoing laparoscopic colon resection for carcinoma at one institution that had complete 5-year survival data. The National Cancer Database (NCDB) of the American College of Surgeons was chosen for comparison because the American College of Surgeons uses date of treatment initiation as the starting point, and their 5-year relative survival rates were analyzed separately for each stage of disease. The 5-year survival rates in the laparoscopic resection group compared favorably with the open resections performed in the same institute as well as to the NCDB reported stage-specific rates. Interestingly, there was a 9% to 11% survival benefit between stage III

372

& PART III: MINIMALLY INVASIVE SURGERY

laparoscopic patients and the open and the NCDB groups. The authors stated that the lumping of colon and rectum together may have affected the survival rates and they also raised the concern regarding the confounding effects of adjuvant therapy. Pantankar et al. (122) reported their experience with 172 curative intent, laparoscopic colorectal resections. The open resection group was a computerized case-matched control group at the same institution. The survival data was also compared with the two most authoritative sources of information on carcinoma incidence and survival in the United States, the NCDB of the American College of Surgeons Commission on Cancer and the American Cancer Society, and the Surveillance Epidemiology and End Results (SEER) Program of the National Cancer Institute. The data from SEER reports 5-year overall relative survival rates as ranging from 60% to 62%. The data available from the NCDB reports the overall 5-year relative survival for years 1993–1998 to be 62%. The overall 5-year survival rate in this study was 69% for the laparoscopic group and 64% for the open group and compares favorably with the SEER and NCDB statistics. The 5-year survival rates were also analyzed separately for the colon versus rectum sites, and for each stage. Separate analysis was undertaken to avoid the confounding effect of adjuvant therapy with possible interruptions and/or partial treatments. The results show that, in strictly comparable patients as possible through the limitations of nonrandomized study design, the survival with laparoscopic approach was no different from the open resections. Clinical Trials There are currently several multi-institutional, large-scale prospective, randomized trials comparing laparoscopicassisted to open colon resection in patients with colon carcinoma. The National Institutes of Health trial [Clinical Outcomes of Surgical Therapy (COST) Study Group] begun in 1995, proposed to study 1200 patients randomly assigned to laparoscopic or open colectomy for curable colon carcinoma has recently been published and detailed below (123). The Colon Carcinoma Laparoscopic or Open Resection (COLOR) trial is a European multicenter randomized trial which began in September 1997. The COLOR trial prospectively randomized 1248 patients with colon carcinoma to laparoscopic or open colon resection in 29 participating centers. All patients will be followed for a minimum of 5 years. The primary outcome measure is carcinoma-free survival at 3 years with secondary outcome measures being morbidity, port-site recurrences, quality of life, cost, resection margins, and harvesting of lymph nodes. The short-term outcomes have recently been published (124) (Tables 1 and 5). They concluded that laparoscopic surgery can be used for safe and radical resection for carcinoma in the right, left, and sigmoid colon. The Medical Research Council Conventional versus Laparoscopic-Assisted Surgery in Colon Carcinoma (MRC CLASICC) trial opened recruitment in 1995. It involved 27 hospitals in the United Kingdom. The aim was to recruit 1000 patients, but ultimately 794 patients were recruited. The primary endpoints are pathological resection margins, mortality, local recurrence rates, disease-free survival

TABLE 1 & Bowel Activity Following Laparoscopic Resection for

Colon and Rectal Carcinoma

Author (Ref.)

Year

Type of Study

Ramos et al. (125) Khalili et al. (126) Hong et al. (127) Baker et al. (128) Champault et al. (129) Curet et al. (130) Lacy et al. (120) Leung et al. (131) COLOR (124) CLASICC, Guillou et al. (132)

1997

RC

1998

RC

2001

RC

2002

RC

2002

PNR

2000

PR

2002

PR

2004

PR

2005

PR

2005

PR

Procedure

Oral Intake

Bowel Activity

Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open

– – 3.9 4.9 2.1 4 – – 1.8 4.2 2.7–4.1a 4.4–5.8 2.3 3.5 4.2 4.9 2.9 3.8 6 6

2.5 3.9 – – 1.8 3 1.4 1.2 1.4 3.2 – – – – 2.4–4.0b 3.1–4.6 3.6 4.6 5 6

a

Fluids–regular diet. Gas–bowel movement. Abbreviations: RC, retrospective comparative; PNR, prospective nonrandomized; PR, prospective randomized.

b

and overall survival at 3 years. The secondary outcome measures include quality of life and cost-effectiveness. This trial included patients with carcinoma of both the colon and rectum. The short-term endpoints of this trial have also recently been reported (132). Abraham et al. (154) conducted a meta-analysis of randomized clinical trials comparing the short-term outcomes of laparoscopic with those of open resection for colorectal carcinoma. The outcomes of 2512 procedures from 12 trials were analyzed. Laparoscopic resection took on average 32.9% longer to perform than open resection but was associated with lower morbidity rates. Specifically, wound infection rates were significantly lower (odds ratio 0.47). In patients undergoing laparoscopic resection, the average time to passage of first flatus was reduced by 33.5%, that to tolerance of a solid diet by 23.9% and that to 80% recovery of peak expiratory flow by 44.3%. Early narcotic analgesia requirements were also reduced by 36.9%, pain at rest by 34.8% and during coughing by 33.9%, and hospital stay by 20.6%. There were no significant differences in perioperative mortality or oncological clearance. They concluded laparoscopic resection for colorectal carcinoma is associated with lower morbidity, less pain, a faster recovery and a shorter hospital stay than open resection without compromising oncologic clearance. Prospective Randomized Studies Stage et al. (155) prospectively randomized 34 patients with potentially curable colon carcinoma to laparoscopic (n ¼ 18) or open surgery (n ¼ 16). Conversion rate was 16.7%. Patients in the laparoscopic surgery group experienced significantly longer operative times (150 vs. 95 minutes), less pain, shorter hospital stay (5 vs. 8 days), and more

373

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

TABLE 2 & Long-term Results of Retrospective Noncomparative Studies of Laparoscopic Resection for Colon and Rectal Carcinoma Recurrence (%) Author (Ref.) Melotti et al. (133) Franklin et al. (134) Schiedeck et al. (135) Anderson et al. (136) Lechaux et al. (137) Lujan et al. (121) Lumley et al. (138) Poulin et al. (139)

Scheidbach et al. (140) Yamamoto et al. (141) Morino et al. (142) Tsang et al. (143) Watanabe et al. (144)

n

Year 1999

163

Conver- Operating sion Rate Time (%) (min) 20.4

50b

2000

Morbidity Mortality (%) (%)

LOS (days)

F/U (MOS) a

0

–

6

14

2

–

24

40

37

1.8

14

30

8

166–228d

22

2

8

40.3

15

150

12

2

–

65

23

1

–

399

6.3

2002

93

–

2002

166

11

2002

102

2002

154

7.8

DFS

36

79

69

39

49

0.3

1.5

6.3

–

98–86–89c

1

–

–

100–77–52c

–

11

0.6

–

–

79

61

64

14.7

2

2.9

10.7

64

–

1.1

71

13.6

0.6

2.6

10.3

91–83–74c

2.5

31

–

0

4.3

2002

AR 52 APR 28 292

27 35 5.5

205 210 172

22.3

2.7

2002

70f

2.9

–

18.6

–

8

2003

100e

12

250

36

2

2003

44e

0

180

– 14.6

34

72.1 93–87–53c

6.5 8.0

2.3

Overall

6

80e

130

Distant

0

2002

2003

Local

1.2

15.1

2000

Portsite

Overall

Survival (%)

25.2

11.6

0

3.4

8.2

81

–

23

2.9

0

–

2.9

100

92

12

46

–

1.4

4.2

–

73

81

0

–

15

–

0

4.5

11.4

80

–

0

8

61

–

0

2.3

1.5

97.9

–

a

Stage III only. Includes all resections (RHC-APR—all procedures). c Stage I, II, III. d Varied with type of resection. e Rectal carcinoma. f Tis, T1, T2 only rectal. Abbreviations: LOS, length of stay; F/u, follow-up; MOS, months of survival; DFS, disease-free survival; AR, anterior resection; APR, abdominoperineal resection. b

TABLE 3 & Long-term Results of Retrospective Comparative Studies of Laparoscopic Resection for Colon and Rectal Carcinoma Recurrence (%) Author (Ref.) Ramos et al. (125) Bouvet et al. (145) Khalili et al. (126) Schwandner et al. (146) Hong et al. (127) Baker et al. (128)

Year

Method

n

1997

L O L O L O L O L O L O

18a 18 91b 57 80 90 32a 32 98 219 28a 61

L O

101 334

1998 1998 1999 2001 2002

Anthuber 2003 et al. (147) a

Conversion Rate (%) 10 42 7.5 –

12.2

Operation Time (min)

Morbidity (%)

229 208 240 150 161 163

22.2 16.6 24 21

140 129

31.3 31.3 27 39

218 219

31 65

25

11

Mortality (%)

LOS (days)

0 5.5 2.2

7.4 12.9 6 7

–

–

3.1 1.8 3.6 3.3

6.9 10.9 13 18

0 5

14.4 19.9

F/U (MOS) Overall

26 21 18 33.1 32.1 30.6 21.6 35.6 30.8

Carcinoma rectum – APR All procedures. c Right hemicolectomy. Abbreviations: LOS, length of stay; L, laparoscopic; O, open; MOS, months of survival; DFS, disease-free survival. b

13 18 12.5 15.6 20.6 17.8 32.1 37.7

7

Portsite

Local

Survival (%) Distant

Overall

DFS

0 0 0

5.5 16.5 –

–

–

–

–

–

0 0 0 0 0 1.4 0 0

3 6 0 0 12.2 8.7 14.3 13.1

10 11 12.5 15.6 9.1 9.2 7.1 4.9

88 85 –

93 98 –

2

5

0 0

78.5 78.5 42% at 41 mos 32% at 31 mos

– 43.5 43.5

374

& PART III: MINIMALLY INVASIVE SURGERY

TABLE 4 & Long-term Results of Prospective Nonrandomized Comparative Studies of Laparoscopic Resection for Colon and Rectal

Carcinoma Recurrence (%)

Author (Ref.) Year Franklin et al. (148) Santoro et al. (149) Hartley et al. (150) Champault et al. (129) Feliciotti et al. (151) Lezoche et al. (152)

Feliciotti et al. (153) Patankar et al. (122)

1996 1999 2001 2002 2002 2002

2003 2003

Conversion Rate (%)

n

Method

191a 224 50 50 42b 22 74 83 104 93 55 44 86 63 81b 43 172 172

L O L O L O L O L O RHC L RHC O LHC L LHC O L O L O

33 8.1

Operation Time (min)

Morbidity (%)

180 125 145 125

23.8 17 14 14 28.6 18 13.5 33.7

Mortality (%)

5.7 9.7 0 0 0 0 0 0 1.9 1.1

4.8 0 – 7 – 12.3

190 140 240 190

LOS F/U (days) (MOS) Overall

1.9 2.3 7.5 6.3

13.5 15 8.2 12.3

9.2 13.2 10 13.2 0 0 1.2 2.4

38 52 56 49

20 23 – 24.3 25 13.1 0

42

43.8 52 59

– 1.5 – – 16.3 13.4

Portsite 0 – 2.5 2.3 0 0 2.7 – 1 0 2.7 1.5 – 0 0 0.6 0

Local

– – 5 4.5 – – 1.3 2.7 5.4 9 1.5 7.5 20.8 16.6 3.5 2.9

Survival (%)

Distant Overall DFS 87 81 – – 71 77 63.1 59.1 89 87 86.5 81.8 97.1 85.7 70.9 60.6 69 64

– – – – – – 10.8 10.7 – – – – 18.2 21.2 12.2 10.5

90 92 73.2 70.1 – –

87 87

62.5 60.6 – –

a

All procedures. Rectal carcinoma. Abbreviations: LOS, length of stay; L, laparoscopic; O, open; MOS, months of survival; DFS, disease-free survival; RHC, right hemicolectomy; LHC, left hemicolectomy.

b

in the two groups in surgical margins, stage of disease, and length of stay. On average, 19 lymph nodes were resected in the laparoscopic group compared with 25 in the open surgery group. Fifteen percent of patients in each group experienced complications and mortality was similar in both groups. The laparoscopic group experienced less blood loss (252 vs. 344 mL), longer operative time (200 vs. 125 minutes), quicker return of pulmonary function (3 vs. 6 days), less need for analgesia in the first two days postoperatively (0.78 vs. 0.92 mg/kg of morphine) and faster return of flatus (3 vs. 4 days). There were no local or port-site recurrences in

rapid return to self care. There were no differences between the two groups and size of specimen removed, number of lymph node glands removed, the Dukes’ classification, complications, postoperative reduction in pulmonary function or level of fatigue. Immunodepression as measured by IL-6 and serum C-reactive protein levels, was more pronounced in the laparoscopic group. Milsom et al. (156) prospectively randomized 109 patients with adenocarcinoma or large polyps to laparoscopic surgery (n ¼ 55) or open surgery (n ¼ 54). Conversion was required in 6.8% of patients. There were no differences

TABLE 5 & Long-term Results of Prospective Randomized Comparative Studies of Laparoscopic Resection for Colon and Rectal Carcinoma Recurrence (%)

Author (Ref.)

Year

Method

n

Curet et al. (149) Lacy et al. (139) COST (142)

2000

Leung et al. (150) COLOR, Veldkamp et al. (143)

2004

L O L O L O L O L

25a 18 111 108 435 428 203b 200 627

O

621

L

526

O

268

CLASICC, Guillou et al. (151) a

2002 2004

2005

2005

Conversion Rate (%)

Operation Time (min)

Morbidity (%)

Mortality (%)

28

210 138 142 118 150 95 190 144 202 (50–540) 170 (45–580) 180 (140–220) 135 (100–175)

4 27.8 10.8 28.7 21.1 19.9 17.2 20.5 21

4 0 0.9 2.8 0.5 0.9 2 1 1

5.2 7.3 5.2 7.9 5 6 8.2 8.7 8.2

20

2

9.3

13 C-8, R-18 11 C-8, R-14

5

C-9 R-11 C-9 R-13

11 – 21 23 17

29

4

LOS (days)

F/U (MOS) 59 44 43 53 39 35

Overall

Portsite

5.8 – 17 27 17.5 19.6 22 18

0 0 0.9 0 0.5 0.2 0 0

All procedures. Rectosigmoid carcinoma. c Gas, bowel movement. Abbreviations: C, colon; R, rectum; LOS, length of stay; L, laparoscopic; O, open; MOS, months of survival; DFS, disease-free survival. b

Local – – 6.6 6.6

6.6 4.1

Survival (%) Distant (%) Overall DFS – – 8.8 13.7

18 15

68 39 82 74 86 85 76 73

91 79

}

3 yr 75 78

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

the laparoscopic group compared with the two wound recurrences in the open group. Carcinoma-related deaths occurred in three laparoscopic surgery patients and four open surgery patients. Lacy et al. (120) randomized 219 patients with colon carcinoma (111 laparoscopic group, 108 open group) over 5 years. Median length of follow-up was 43 months. Conversion rate was 11%. Number of lymph nodes resected was similar in the two groups (11.1). Operative time was longer (142 vs. 118 minutes) and intraoperative blood loss lower (105 vs. 193 mL) in the laparoscopic group. The patients in the laparoscopic group had quicker return of gastrointestinal function (54 vs. 85 hours), shorter hospital stay (5.2 vs. 7.9 days), and less morbidity (10.8% vs. 28.7%). One patient developed a port-site metastasis. Overall, mortality was similar between the two groups but carcinoma-related mortality was significantly lower in the laparoscopic group (9% vs. 21%). The probability of overall survival and the probability of carcinoma-related survival were significantly higher in the laparoscopic group. The Cox model showed that laparoscopic-assisted colectomy was independently associated with reduced risk of carcinoma relapse (hazard ratio 0.39), death from any cause (0.48), and death from carcinoma-related causes (0.38) compared with open colectomy. This superiority of laparoscopic-assisted colectomy was due to differences in patients with stage III carcinoma. The differences were so pronounced that data on the stage III patients approached outcomes for stage II patients in the laparoscopic group. The authors postulate that decreased stress and improved immune function associated with a minimally invasive approach may account for these differences. Delgado et al. (92) prospectively randomized patients with colorectal carcinoma to receive either laparoscopicassisted colectomy (n ¼ 129) or open colectomy (n ¼ 126). They analyzed data by whether patients were younger than 70 years of age or 70 years and older thus ending up with four groups of comparison. There was a higher incidence of co-morbid factors, right colon lesions, and right colectomies in the older group but the distribution was similar in the laparoscopic-assisted and the open group. The conversion rate in the laparoscopic-assisted group was 11.4% in the younger groups and 17% in the older groups. Mean operating time was significantly longer and blood loss significantly lower in the laparoscopic-assisted colectomy than the open colectomy group with no difference in the two age groups. Return of gastrointestinal function was earlier and length of stay was shorter in the laparoscopicassisted group compared with the open colectomy group with these differences seen in both age groups. Complication rates in the younger group were similar for the laparoscopic-assisted and open colectomy patients. However, in the older age group, complication rates for laparoscopic-assisted patients were significantly less than for the open patients (10.1% vs. 31.3%) because of differences in wound infections (5% vs. 12%). Hasegawa et al. (157) randomized 59 patients with T2 or T3 colorectal carcinomas who underwent curative resection to laparoscopic-assisted colectomy (n ¼ 29) or open colectomy (n ¼ 30). The patients in the laparoscopic group experienced longer operative times (275 vs. 188 minutes), less blood loss (58 vs. 137 mL), smaller incisions (5.9 vs.

375

17.8 cm), faster progression to oral intake (1.6 vs. 3.2 days), less analgesic requirement (1.7 vs. 3.4 days), and shorter hospital stay (7.1 vs. 12.7 days). No differences in morbidity rates were found between the two groups. C-reactive protein levels were lower in the laparoscopic-assisted group on postoperative days 1 and 4 while there were no differences between the two groups with respect to IL-6 and leukocyte count and natural killer cell activity. Weeks et al. (158) reported the results of short-term quality of life based on data obtained from the National Institute of Health Trial based on 449 patients from 37 centers with clinically resectable colon carcinoma. In an intention-to-treat analysis, the global rating scale score at 2 weeks postoperative was higher for the laparoscopic group than the open group (76.9 vs. 74.4). These differences disappeared by the two month time analysis period. The laparoscopic surgery patients required fewer days of parenteral analgesics (3.2 vs. 4.0 days) and oral analgesics (1.9 vs. 2.2 days) and were hospitalized for a shorter period of time (5.6 vs. 6.4 days). The conversion rate was 25.7% and these patients reported slightly poorer quality of life results than patients who were not converted. These surprising results could be due to the fact that the instruments used to measure quality of life were not sensitive enough to identity true differences. Also, sample size could have been a factor. The analysis on intention-to-treat resulted in patients who were converted being analyzed in the laparoscopic group, which could also have resulted in skewing of the results. Finally, it is possible that the overriding factor in quality of life in patients who have malignancy is the malignancy and their worries about survival, thus obscuring differences seen with operative approach. Schwenk et al. (159) prospectively randomized 60 patients with primary colorectal carcinomas to laparoscopic (n ¼ 30) or open (n ¼ 30) surgery. Length of stay was similar because no patient was discharged before postoperative day 7 to collect complete data. The laparoscopic group required less postoperative analgesics (0.78 vs. 1.38 mg/kg), had lower pain score (507 vs. 755), and lower postoperative fatigue scores (322 vs. 531). Inflammatory Response The inflammatory and immune responses to laparoscopic colectomy have been studied in both animal and human trials (160). Carcinoma patients may already have a suppressed immune system and therefore any further attenuation may be potentially harmful. Immunosuppression is mediated by a number of factors, and multiple serological factors have been measured, including C-reactive protein, IL-6, IL-8, and tumor necrosis factor. IL6 is a major indicator of tissue damage and a major mediator of acute phase protein response. Several authors have found that serum levels of IL-6 reveal the greatest differences between laparoscopic-assisted and open colectomy groups concluding that the acute phase response is attenuated in the laparoscopic-assisted colectomy patients. Similarly, several authors have found a significant decrease in the systemic inflammatory response, specifically IL-6, C-reactive protein, and granulocyte elastase, leading them to conclude that laparoscopic colectomy is less traumatic than open colectomy. Although the blood levels of the cytokines are lower in the laparoscopic group, peritoneal fluid

376

& PART III: MINIMALLY INVASIVE SURGERY

levels are the same, suggesting that the peritoneal trauma is the same despite the lower systemic response. The shortterm benefits of laparoscopy may be due at least in part to a lower systemic inflammatory response. In parallel with the MRC CLASICC trial, Tang et al. (161) evaluated systemic immune response in patients having laparoscopic-assisted colectomy (n ¼ 118) compared with those undergoing conventional open operation (n ¼ 118) for colorectal carcinoma. Operative time was 18 minutes longer for the laparoscopic-assisted patients (88 vs. 70 minutes) and the incision length was shorter (9 cm vs. 15 cm). There were no significant differences in morbidity rates between the two groups. There were no differences in T-cell, B-cell CD4:CD8 ratios, immunoglobulin (Ig)G, IgM, IgA, natural killer cell, or nitroblue tetrazolium test between the two groups. There were marginal differences in C3 and C4 complement levels, which did not reach statistical significance. Duke’s classification, size of carcinoma, wound length, and operating time had little influence on the postoperative immune parameters. Low Anterior Resection Delgado et al. (162) reported their results of laparoscopic techniques in 220 patients with rectal carcinoma. Neoadjuvant chemoradiotherapy was used in 59% of their patients. In more than 75% of the patients, a surgical procedure with sphincter preservation was performed. The rate of conversion to the open approach was 20%. Ten patients had intraoperative complications. Postoperative complications developed in 26.3% of patients. The length of hospital stay was 6.8 days. The distribution of stages was as follows: stage I, 16.8%; stage II, 33.6%; stage III, 26.4%; stage IV, 19.1%. The mean number of lymph nodes was 13.8. The incidence of local relapse was 5.3% with a follow-up of 18 months. They concluded, laparoscopic surgery can be safely performed in patients with adenocarcinoma of the rectum with good short-term results. Yamamoto et al. (141) examined the short-term results of laparoscopy in the treatment of 70 curative cases of rectosigmoid and rectal carcinoma. Mesorectal transection was performed at least 5 cm below the lesion for rectosigmoid and upper rectal carcinomas and a TME was performed for lower carcinomas. Primary anastomosis was performed by a double-stapling technique, or a handsewn coloanal anastomosis was performed. The median follow-up was 23 months. An anastomosis was performed in 93% of the operations. Oral intake was started on median postoperative day 1, and the median length of hospitalization was 8 days. Two patients needed conversion to conventional open operation. A total of 15 postoperative complications occurred in 13 patients (18.6%), including anastomotic leakage in six (8.6%) and bowel obstruction in three (4.3%). Reoperation was required in six patients. Two patients developed recurrence of their carcinoma at the anastomotic site. The expected 5-year survival and disease-free survival rates were 100% and 92%, respectively. Morino et al. (142) conducted a prospective consecutive series of 100 laparoscopic TMEs for low mid-rectal carcinomas. All patients had a sphincter-saving procedure. The distal limit of the rectal neoplasm was on average 6.1 cm from the anal verge. The mean operative time was 250 minutes. The conversion rate was 12%. The mean

postoperative stay was 12 days. The 30-day mortality was 2% and the overall postoperative morbidity was 36% including 17 anastomotic leaks. With a median follow-up of 45.7 months, 18.5% died of carcinoma and 8.5% are alive with metastatic disease. The port-site metastases rate was 1.4%. The local regional pelvic recurrence was 4.2%. This series confirms the safety of the procedure while oncologic results are comparable with the open published series. Law et al. (163) reported on the early outcomes of 100 patients with laparoscopic resection for rectal carcinoma with TME. Operations included 91 anterior resections, eight abdomino-perineal resections, and one Hartmann’s procedure. Conversion was required in 15 patients but none with APR. Postoperative complications occurred in 31%—anastomotic leak in three patients and one patient required reoperation for small bowel obstruction. Zhou et al. (164) conducted a prospective randomized controlled trial to assess the feasibility and efficacy of the laparoscopic approach of TME with anal sphincter preservation in 82 patients by the laparoscopic procedure and 89 patients by the open technique. In the laparoscopic group, 30 patients in whom the low anterior resection was performed had the anastomosis below the peritoneal reflection and more than 2 cm above the dentate line, 27 patients in whom ultra-low anterior resection was performed had anastomotic height within 2 cm of the dentate line, and 25 patients in whom coloanal anastomosis was performed had the anastomosis at or below the dentate line. In the open group, the numbers were 35, 27, and 27, respectively. There was no statistical difference in operation time, administration or parenteral analgesics, start of food intake, and mortality rate between the two groups. However, blood loss was less, bowel function recovered earlier and hospitalization time was shorter in the laparoscopic group. Leroy et al. (165) assessed the feasibility, safety and long-term outcome of laparoscopic rectal carcinoma resections following the principles of TME in 102 consecutive unselected patients. Conversion to an open approach was required in 3%. The overall morbidity and mortality rates were 27% and 2%, respectively, with an overall anastomotic leak of 17%. In 91.8% the resection was considered curative. Mean follow-up was 36 months. There were no trocar-site recurrences. The local recurrence rate was 6%, and the carcinoma-specific survival of all curatively resected patients was 75% at 5 years. The overall survival rate of all curatively resected patients was 65% at 5 years. Rullier et al. (166) reported on the feasibility of laparoscopic rectal resection in patients with mid- or low-rectal carcinoma with regard to quality of mesorectal excision, autonomic pelvic nerve preservation, and anal sphincter preservation. Laparoscopic rectal excision was performed in 32 patients (21 men) with rectal carcinoma located 5 cm from the anal verge. The operative procedure was performed 6 weeks after radiotherapy and included TME, intersphincteric resection, transanal coloanal anastomosis with coloplasty and loop ileostomy. Three patients needed conversion to a laparotomy. Postoperative morbidity occurred in 10 patients, related mainly to coloplasty. Macroscopic evaluation showed an intact mesorectal excision in 29 of 32 excised specimens; microscopically, 30 of the 32 resections were R0. Sphincter preservation was achieved in 31 patients. The hypogastric nerves and pelvic plexuses were

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

identified and preserved in 24 of the 32 patients. Sexual function was preserved in 10 of 18 evaluable men. They concluded a laparoscopic approach can be considered in most patients with mid- or low-rectal carcinoma. Tsang et al. (143) examined outcomes after patients with mid- or low-rectal carcinomas underwent laparoscopic TME with construction of a colonic J-pouch in 44 patients. There was no conversion to an open procedure. The median distance of the anastomosis from the anal verge was 4 cm. No procedure-related death occurred. Four patients developed significant complications that required reoperation. With a median follow-up period of 15 months, no port-site recurrence was noted. Five patients developed distant metastases and two had local recurrence in the pelvis. Bowel function was satisfactory at 6, 12, and 18 months after ileostomy closure. Jayne et al. (167) assessed bladder and sexual function in patients who had undergone laparoscopic rectal, open rectal or laparoscopic colonic resection as part of the UK MRC CLASICC trial, using the International Prostatic Symptom Score, the International Index of Erectile Function and the Female Sexual Function Index. Sexual and bladder function data from the European Organization for Research and Treatment of Cancer QLQ-CR38 collected in the CLASICC trial were used for comparison. Questionnaires were completed by 71.2% of the 347 patients. Bladder function was similar after laparoscopic and open rectal operations for rectal carcinoma. Overall, sexual function and erectile function tended to be worse in men after laparoscopic rectal surgery than after open rectal surgery (overall function: difference – 11.18; erectile function: difference –5.84). Total mesorectal excision was more commonly performed in the laparoscopic rectal group than in the open rectal group. TME (odds ratio 6.38) and conversion to open operation (odds ratio 2.86) were independent predictors of postoperative male sexual dysfunction. No differences were detected in female sexual function. Laparoscopic rectal resection did not adversely affect bladder function, but there was a trend toward worse male sexual function. This may be explained by the higher rate of TME in the laparoscopic rectal resection group. Advanced Colorectal Carcinoma Milsom et al. (168) reported a prospective analysis of 30 patients with incurable colorectal carcinoma who underwent laparoscopic surgery for palliative purposes. Resection of a single segment of the bowel was performed in 15 patients. One patient underwent both right colectomy and sigmoidectomy because of double lesions. Stoma creation only was performed in 11 patients. Three patients were converted to an open procedure. For resection, median operative time was 170 minutes. For stoma creation median operative time was 60 minutes. There were two postoperative deaths. Median time to passage of flatus was 2 days and of stool 5 days after resection and 2 days for both flatus and stool after stoma creation. Median time to discharge was 8 days after resection and 7 days after stoma creation. They concluded the laparoscopic approach for patients with incurable colorectal carcinoma can provide effective palliation with avoidance of a major laparotomy in the majority of cases. Gonzalez et al. (25) also evaluated the feasibility and outcomes of the laparoscopic approach for the

377

palliation of advanced complicated colorectal carcinoma. They reviewed 21 laparoscopic palliative procedures for emergent complications of advanced colorectal carcinoma. Indications for operation included perforation (n ¼ 10), bleeding (n ¼ 7), and obstruction (n ¼ 4). A proximal diverting procedure was performed in all patients, and a concomitant colon resection was performed in 18 patients (86%). The mean operative time was 181 minutes. The average length of hospital stay was 8.6 days and time to first bowel movement was 61 hours. The complication rate and the 30-day mortality rates were 33% and 0%, respectively. Leung et al. (131) reported on their detailed comprehensive study in which 403 patients with rectosigmoid carcinoma were randomized to receive either laparoscopicassisted (n ¼ 203) or conventional open (n ¼ 200) resection. Survival and disease-free interval were the main endpoints. The demographic data of the two groups were similar after curative resection, the probabilities of survival at 5 years of the laparoscopic and open resection groups were 76.1% and 72.9%, respectively. The probabilities of being disease-free at 5 years were 75.3% and 78.3%, respectively. The operative time for the laparoscopic group was significantly longer whereas postoperative recovery was significantly better than for the open resection, but these benefits were at the expense of higher direct costs. The distal margin, the number of lymph nodes found in the resected specimen, overall morbidity and overall mortality did not differ between groups. They concluded, laparoscopic resection of rectosigmoid carcinoma does not jeopardize survival and disease control of patients. In Summary In their comprehensive review of the literature, Kieran and Curet (160) found in most reports length of stay has been 5 to 6 days for laparoscopic cases compared to 7 to 8 days for open cases. Operative times ranged from 88 to 275 minutes (with most in the 170 to 190 range) compared with 95 to 201 minutes (with most in the 120 to 150 range) for open cases. In general, morbidity rates have been similar between the two groups ranging from 2% to 46% (average 17.6%) for laparoscopic cases and 2% to 60% (average 22.9%) for open cases. Of note is that small bowel perforation, ureteral injury, trocar site injury, and mesenteric and epigastric vessel injury have been reported with laparoscopic-associated complications. Lacy (120) found a higher complication rate in open vs. the laparoscopic group (30% vs. 10.5%) primarily because of a higher incidence of wound complications (16.7% vs. 7.2%). Delgado found similar results in patients < 70 or > 70 years of age (169). Anastomotic leaks were also similar in the two groups (169). Conversion varied widely from 0% to 40% with most reports in the 15% to 20% range. Cited reasons for conversion include bulky disease, invasion of carcinoma, obesity, adhesions, inability to mobilize the splenic flexure, bleeding, ureteral injury, unclear anatomy, prolonged operative time, stapler malfunction, and anastomotic defects (160). Conversion rates decrease with increasing experience. The cost issue is unclear. Intraoperative costs are certainly higher for the laparoscopic approach but some argue that the total cost is less because of the shorter hospital stay. However, hospital stay may not be substantially shorter or enough to compensate for the increased instrument

378

& PART III: MINIMALLY INVASIVE SURGERY

costs. Lymph node harvests are similar with average range 6–23 laparoscopically and 6–26 open. Resection margins are also similar with the two techniques. Lacy et al. (120) reported a randomized study with an unforeseen better long-term survival for node positive patients treated by laparoscopic colectomy. Similarly, Capussotti et al. (170) conducted a prospective nonrandomized trial in which they compared short- and long-term results of laparoscopic and open curative resection for adenocarcinoma of the left colon or rectum in 255 consecutive patients. A total of 74 patients underwent a laparoscopic resection and 181 an open resection. The carcinoma was in the descending colon in 32 cases, the sigmoid colon in 98 cases, and the rectum in 125 cases including 87 mid–low-rectal carcinomas. Ten laparoscopic resectionprocedures (13.5%) were converted to open surgery. The hospital mortality was 0.08%, and in hospital morbidity was 16.2% for laparoscopic resection and 13.3% for open resection. The median postoperative stay was 1 day shorter for laparoscopic resection (9 days) than for open resection (10 days). The mean number of lymph nodes retrieved were 13.8 for open resection and 12.7 for laparoscopy resection. Age exceeding 70 years, T stage, N stage, grading, mid–low-rectal site, and laparoscopy were found by multivariate analysis to be significant prognostic factors for disease-free and carcinoma-related survival. When patients were stratified by stage, a trend toward a better diseasefree and carcinoma-related survival was identified in stage III patients undergoing laparoscopic resection. Overall, disease-free and carcinoma-related 5-year actuarial survivals were 48.5%, 50.5%, and 56.5%, respectively, for the open and 68.2%, 65.2%, and 76.4% for the laparoscopic group. Node positive patients who completed the laparoscopic resection had a longer 5-year disease-free survival (73% vs. 47.8%), carcinoma-related (82.1% vs. 54.5%) survival. There was a trend toward a reduction in the laparoscopic group, a finding similar to that of Lacy et al. (120). Capussotti also noted a trend toward a decreased incidence of distant metastases (170). COST Trial and Expert Consensus The long awaited results of the landmark COST trial have recently been published and at the time of writing considered the definitive study (123). Minimally invasive laparoscopic-assisted surgery was first considered in 1990 for patients undergoing colectomy for carcinoma. Concern that this approach would compromise survival by failing to achieve proper oncologic resection or adequate staging or by altering patterns of recurrence (based on frequent reports of recurrences within surgical wounds) prompted a controlled trial evaluation. Members of the COST study group conducted a trial with 48 institutions and randomly assigned 872 patients with carcinoma of the colon to undergo open or laparoscopic-assisted colectomy performed by credentialed surgeons. The median followup was 4.4 years. The primary endpoint was the time to recurrence of the carcinoma. Operating times were significantly longer in the laparoscopic surgery group than in the open colectomy group (150 vs. 95 minutes). Bowel margins were less than 5 cm in 6% of patients in the open colectomy group and 5% in the laparoscopic surgery group. In each group, the median number of lymph nodes

examined was 12. Perioperative recovery was faster in the laparoscopic surgery group than in the open colectomy group as reflected by a shorter median hospital stay (5 vs. 6 days) and briefer use of parenteral narcotics (3 vs. 4 days) and oral analgesics (1 vs. 2 days). The rates of intraoperative complications (2% in open colectomy group and 4% laparoscopic group), 30-day postoperative mortality (1% open vs. 0.5% laparoscopic), complications at discharge and 60 days (open 20% vs. laparoscopic 21%), hospital readmission (open 10% vs. laparoscopic 12%), and reoperation (both < 2%) were very similar between groups. The percentage of patients receiving chemotherapy did not differ significantly between groups and paralleled the rate of stage III disease. At 3 years, the rates of recurrence were similar in the two groups—16% among patients in the group that underwent laparoscopic-assisted surgery and 18% among patients in the open colectomy group (hazard ratio for recurrence 0.86). Recurrence rates in surgical wounds were less than 1% in both groups. The overall survival rate at 3 years was also very similar in the two groups (86% in the laparoscopic group and 85% in the open colectomy group), hazard ratio for death in the laparoscopic group, 0.91; with no significant difference between the groups in the time to recurrence or overall survival for patients with any stage of carcinoma. They concluded in this multi-institutional study, the rates of recurrent carcinoma were similar after laparoscopic-assisted colectomy and open colectomy suggesting that the laparoscopic approach is an acceptable alternative to open surgery for colon carcinoma. Based on this study, The American Society of Colon & Rectal Surgeons Executive Council held a special meeting May 12, 2004 to develop a position statement on ‘‘Laparoscopic Colectomy for Curable Cancer.’’ The following statement was unanimously approved by the ASCRS Executive Council and endorsed by The Society of American Gastrointestinal Endoscopic Surgeons (SAGES). Laparoscopic colectomy for curable cancer results in equivalent cancer-related survival to open colectomy when performed by experienced surgeons. Adherence to standard cancer resection techniques including but not limited to complete exploration of the abdomen, adequate proximal and distal margins, ligation of the major vessels at their respective origins, containment and careful tissue handling, and en bloc resection with negative tumor margins using the laparoscopic approach will result in acceptable outcomes. Based upon the COST trial, prerequisite experience should include at least 20 laparoscopic colorectal resections with anastomosis for benign disease or metastatic colon cancer before using the technique to treat curable cancer. Hospitals may base credentialing for laparoscopic colectomy for cancer on experience gained by formal graduate medical education training or advanced laparoscopic experience, participation in hands on training courses and outcomes. The European Association of Endoscopic Surgery (EAES) initiated a consensus development conference on the laparoscopic resection of colon carcinoma (78). A systematic review of the current literature was combined with the opinions of experts in the field of colon carcinoma surgery to formulate evidence-based statements and recommendations on the laparoscopic resection of colon carcinoma. Advanced age, obesity, previous abdominal

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

operations, were not considered absolute contraindications for laparoscopic colon surgery. The most common cause for conversion is the presence of a bulky or invasive carcinoma. Laparoscopic operation takes longer to perform than the open counterpart but the outcome is similar in terms of specimen size and pathological examination. Immediate postoperative morbidity and mortality are comparable for laparoscopic and open colonic surgery for carcinoma. The laparoscopic-operated patients had less postoperative pain, better preserved pulmonary function, earlier restoration of gastrointestinal function, and earlier discharge from the hospital. The postoperative stress response is lower after laparoscopic colectomy. The incidence of port-site metastases is < 1%. Survival after laparoscopic resection of colon carcinoma appears to be at least equal to survival after open resection. The costs of laparoscopic surgery for colon carcinoma are higher than those for open surgery. They concluded, laparoscopic resection of colon carcinoma is a safe and feasible procedure that improved short-term outcome. Results regarding the long-term survival of patients enrolled in large multicenter trials will determine its ultimate role. A panel of worldwide recognized experts in laparoscopic colectomy was assembled and a consensus document was published (78). In brief, their findings are documented as follows. Reported morbidity with laparoscopic colon resection for carcinoma has varied widely but has not differed from that after open colectomy (78). Rates range from 1.5% to 28% with most in the 10% to 15% range (78). In a review of 11 studies, rates of wound infection were 5.7%, respiratory problems 3.1%, cardiac problems 2.9%, hemorrhage 1.9%, anastomotic leak 1.5%, urinary tract injury and small bowel perforations 0.6%, port-site herniation and hematoma 0.4%, and septicemia, peritonitis, anastomotic stricture, anastomotic edema, hypoxia, acute renal failure, renal insufficiency, urinary retention, deep venous thrombosis, small bowel obstruction, phlebitis, intra-abdominal abscess, each 0.2%. Mortality rates after laparoscopic colectomy were also similar to that of open colectomy (78). In most but not all reports, length of stay after laparoscopic colectomy is shorter than after open colectomy (78). Duration ranged from 5.7 to 18.7 days with differences in hospital stay between laparoscopic and open colectomy groups varying from 1 to 7 days. Analgesia needs have been measured by the number of doses per day required or the number of days analgesics are needed. Less analgesia is needed for laparoscopic colectomy. Gastrointestinal function has been measured in several ways, mainly, time to first bowel movement, time to passage of flatus, or time to resume oral intake. By all parameters, recovery is quicker with laparoscopic colorectal operations. Time to flatus ranged 1.5 to 4.5 days; time to bowel movement ranged 1.5 to 4.7 days, and time to oral intake 2.1 to 6.9 days (Table 1). Postoperative pulmonary function is less impaired after laparoscopic resection for colon carcinoma. Overall and disease-free survival following laparoscopic resections are equivalent to open resection (Tables 2–5). Attempts were made to quote the most recent work from an institution with multiple publications and reports with substantial numbers of patients. With respect to quality of life, effects are most pronounced in the early days after operation. The operative

379

costs for laparoscopic resection of colon carcinoma are higher because of longer operating times and the use of more expensive disposable devices. Furthermore, the stress response after laparoscopic colectomy is lower. Abdominoperineal Resection Darzi et al. (54) compared the adequacy of excision of the first 12 patients undergoing laparoscopic abdominoperineal resection of the rectum to the previous 16 patients undergoing open abdominoperineal resection. In all patients, the procedure was carried out with curative intent. The data demonstrate similar nodal harvest in both groups as well as extent of radial excision. Chindasub et al. (57) reported on 10 patients who underwent laparoscopic abdominoperineal resection with only two complications: postoperative bleeding that required open operation and one case of colostomy ischemia from a tight aperture. Kockerling et al. (171) analyzed the results of 116 patients who underwent laparoscopic abdominoperineal resections, 84.5% of which were performed with curative intent. The mean operating time was 226 minutes. Intraoperative complications occurred in 6% which in more than half of the cases was a vascular injury involving the presacral venous plexus; the conversion rate was 3.4%. Postoperatively, the overall morbidity rate was 34.4%. Reoperation was necessary in 5.2% of patients performed for bleeding in one half of the cases and ileus in the other half. Postoperative mortality was 1.7%. In most of the curative resections an oncologically radical operation with high transection of the inferior mesenteric artery and a complete dissection of the pelvis down to the floor was performed. The median number of lymph nodes investigated was 11.5. At least one follow-up examination was performed in 81% of patients with a mean follow-up period being 491 days. Local recurrence developed in 7% of patients and distant metastases developed in 6% of patients. The recurrence-free survival rate was 71%.

Stomas Fuhrman and Ota (60) reported their early experience with 17 patients who had successfully undergone laparoscopic intestinal diversion. The mean follow-up of this group was 24.3 weeks. Seven patients had their stomas created as part of a laparoscopic abdominoperineal resection, six patients underwent palliative laparoscopic colostomy for an obstructing carcinoma of the rectum or a colorectal-genitourinary fistula, and four patients treated by local excision of a carcinoma were believed to require a proximal protecting fecal diversion. All six patients treated with palliative intent experienced successful relief from their obstruction or fistula and were able to resume enteral nutrition. Complications included prolapse in one patient and paracolostomy hernia in another. Neither the complication required further operative correction. Oliveira et al. (174) reviewed 32 patients who underwent laparoscopic intestinal diversion (25 loop ileostomies, four loop colostomies, and three end colostomies). Indications for fecal diversion were fecal incontinence (11), Crohn’s disease (six) unresectable rectal carcinoma (four), pouch-vaginal fistula (three), rectovaginal fistula (two),

380

& PART III: MINIMALLY INVASIVE SURGERY

colonic inertia (two), radiation proctitis (one), anal stenosis (one), Kaposi’s sarcoma of the rectum (one), and tuberculous fistula (one). Conversion was required in five patients (15.6%) because of the presence of adhesions (three), enterotomy (one), or colotomy (one). Major postoperative complications occurred in two patients (6%) and both cases consisted of stoma outlet obstruction after construction of a loop ileostomy. One patient underwent reoperation, at which time a rotation of the terminal ileum at the stoma site was found. The other patient had narrow fascial opening that was successfully managed with 2 weeks of self-intubation of the stoma. The mean operative time was 76 minutes (range, 30 to 210 minutes), the mean length of hospitalization was 6.2 days (range, 2 to 13 days), and stoma function started after a mean 3.1 days (range, 1 to 6 days). The authors concluded that the laparoscopic creation of intestinal stomas is safe, feasible, and effective. Care must be taken to ensure adequate fascial opening and correct limb orientation. Ludwig et al. (172) reported their initial experience in assessing the safety and efficacy of laparoscopic stoma procedures. A simple two-cannula technique was used in 24 such procedures (16 loop ileostomies, six end-sigmoid colostomies, one transverse colostomy, and one loop-sigmoid colostomy). Indications for diversion were rectovaginal fistula (seven), perianal sepsis (seven), incontinence (four), advanced rectal or colon carcinoma (four), and complicated pelvic infection (two). There were 15 women and nine men with a median age of 44 years (range, 25 to 88 years). The median operative time was 60 minutes (range, 20 to 120 minutes) and the median blood loss was 50 mL (range, 0 to 150 mL). There were no intraoperative complications. One case was converted to a laparotomy because of dense adhesions. The median time to passage of both flatus and stool was 1 day (range, 1 to 3 days) for ileostomy patients, 2 days (range, 2 to 4 days) for flatus, and 3 days (range, 2 to 6 days) for stool after colostomy. The median time to discharge was 6 days (range, 2 to 28 days) and was often delayed by the primary disease process or ostomy teaching. One major postoperative complication, a pulmonary embolism, occurred 8 days after operation in a patient with near obstructing, widely metastatic colon carcinoma. This patient later died of pulmonary failure. All stomas functioned well, with no revisions required. The authors concluded that laparoscopic fecal diversion procedures can be performed safely, simply, and effectively. Swain and Ellis (175) conducted a retrospective review of the medical records of 53 consecutive patients who underwent laparoscopic-assisted creation of a loop ileostomy as an adjunct to anorectal or perineal surgery. The average duration of operation for laparoscopy-assisted creation of a loop ileostomy was 47 minutes with no conversion to laparotomy. All patients were able to tolerate a regular diet on the first postoperative day. Closure was accomplished 69 days later with an average operative time of 52 minutes. One patient developed an ileus after takedown of his stoma. The other 52 patients were able to tolerate a regular diet by the second postoperative day. They concluded laparoscopy-assisted creation of loop ileostomy is an effective method for temporary fecal diversion in patients undergoing anorectal surgery. Liu et al. (176) assessed the results of laparoscopic stoma creation for fecal diversion in 80 patients. The most

common indications were unresectable advanced colorectal carcinoma (n ¼ 20), pelvic malignancy (e.g., ovarian, cervix and prostate, n ¼ 16), and perianal Crohn’s disease with complex fistulas (n ¼ 16). Only in one female patient with pelvic malignant disease the procedure was converted to laparotomy due to obesity (conversion rate 1.3%). Laparosscopic stoma creation included loop ileostomy (n ¼ 30), loop sigmoid colostomy (n ¼ 40), and sigmoid colostomy (n ¼ 9). Postoperative complications were documented in nine patients (overall morbidity rate, 11.4%), including four minor complications treated conservatively (two cases of prolonged atonia and one case each of pneumonia and urinary tract infection) and five major complications requiring reoperation (reoperation rate, 6.3%): one parastomal abscess (drainage), one stoma retraction following rod dislocation (laparoscopic stoma recreation), small bowel obstruction in two patients (small bowel resection), one port-site hernia (fascial closure), and hemorrhage (managed by relaparoscopy). Mean operative time was 74 minutes. Patients were discharged from hospital after a mean of 10.3 days. They believe laparoscopic stoma creation is the method of choice for fecal diversion.

& HAND-ASSISTED TECHNIQUES Kang (20) compared the perioperative parameters and outcomes achieved with hand-assisted laproscopic (HAL) colectomy (30) versus open colectomy (30) for the management of benign and malignant colorectal disease, including carcinoma patients treated with curative intent. The HAL colectomy patients had significantly shorter hospital stays and incision lengths, faster recovery of gastrointestinal function, less analgesic use and blood loss, and lower pain scores on postoperative days 1, 3, and 14. There were no significant differences in operative time, complications, or time to return to normal activity. Cobb et al. (180) examined the results of HAL colon surgery for benign disease in 37 patients. Indications for operation were: polyp (thirteen), uncomplicated diverticular disease (eight), complicated diverticular disease (i.e. colovesical fistula, phlegmon, etc.) (seven), chronic constipation (four), rectal prolapse (two), ulcerative colitis (one), endometriosis (one), and fecal incontinence (one). Procedures performed were sigmoidectomy (fourteen), right colectomy (nine), lower anterior resection (seven), subtotal colectomy (five), cecectomy (one), and transverse colectomy (one). There were no deaths. One case was converted to celiotomy (unable to rule out malignancy). The median operative time was 122 minutes. Return of flatus was noted (median) at postoperative day 3 and the median length of stay after operation was 4 days. One patient developed a superficial wound infection and there was one pelvic abscess (drained percutaneously) and one patient developed urinary retention. There were no reoperations. Nakajima et al. (181) compared the outcomes in 23 patients who underwent total proctocolectomy or total abdominal colectomy using either a HAL technique or laparoscopic-assisted operation. There were 12 handassisted (five total proctocolectomy and seven total abdominal colectomy) and 11 laparoscopic-assisted (seven total proctocolectomy and four total abdominal colectomy) for ulcerative colitis (n ¼ 17), familial polyposis (n ¼ 5),

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

and colonic inertia (n ¼ 1). One laparoscopic case was converted (9.1%). The operative time was shorter for the hand-assisted cases (210 vs. 273 minutes). Blood loss and incision length were similar. Postoperative recovery and morbidity rates were comparable. They concluded HAL surgery reduces the operative time but patient morbidity rates and recovery are similar to laparoscopic cases. The hand-assisted technique may be preferable for extensive colorectal procedures such as total proctocolectomy and total abdominal colectomy. Targarona et al. (182) performed a prospective randomized trial comparing laparoscopic-assisted colectomy and hand-assisted colectomy. A total of 54 patients were enrolled in this study, 27 laparoscopic and 27 hand-assisted. The operative times were similar but hand-assisted was associated with a far lower conversion rate, 7% vs. 23%. Immediate clinical outcomes, oncologic features, and costs were similar for the two procedures but hand-assisted was associated with a significantly greater increase in IL-6 and C-reactive protein than the conventional laparoscopic procedure. This comparative study shows that hand-assisted simplifies difficult intraoperative situations reducing the need for conversion.

& OBESITY Senagore et al. (183) compared the outcome of laparoscopic bowel resection in obese and nonobese patients for all patients who underwent a segmental colectomy for any pathologic condition. Patients with a body mass index (BMI) above 30 were defined as obese and patients with a body mass index below 30 were defined as nonobese. A total of 260 patients were evaluated, 77.3% in the nonobese group and 22.7% in the obese group. The obese group had significantly more conversions to an open procedure (23.7% vs. 10.9%), a longer operative duration (109 vs. 94 minutes), a higher morbidity rate (22% vs. 13%), and a higher anastomotic leakage rate (5.1% vs. 1.2%). Despite higher conversion rates and increased risk of pulmonary complications and anastomotic leakages in obese patients, they parallel those of open operation and laparoscopic colectomy can be performed safely in both obese and nonobese patients with the similar benefit of a shorter hospital stay in both groups. Using the same definition of obesity, Schwandner et al. (184) compared the outcome of laparoscopic colorectal surgery in obese and non-obese patients. All patients who underwent laparoscopic operation for both benign and malignant disease and were laparoscopically completed were analyzed. A total of 589 patients were evaluated, including 95 patients in the obese group and 494 in the nonobese group. There was no significant difference in conversion rate (7.3% in the obese group vs. 9.5% in the non-obese group). No significant differences were observed with respect to age, diagnosis, procedure, duration of operation, and transfusion requirements. In terms of morbidity there were no significant differences related to overall complication rates with respect to BMI (23.3% in the obese group vs. 24.5% in the non-obese group). Major complications were more common in the obese group without showing statistical significance (12.8% in the obese group vs. 6.6% in the non-obese group). Conversely, minor com-

381

plications were more frequently documented in the nonobese group (8.1% in the obese group vs. 15.5% in the non-obese group). In the postoperative course, no differences were documented in terms of return of bowel function, duration of analgesics required, oral feeding and length of hospitalization. They concluded these data indicate that laparoscopic colorectal surgery is feasible and effective in both obese and non-obese patients. Delaney et al. (185) conducted a case-matched study comparing outcomes after open and laparoscopic colectomy in 94 patients with a BMI > 30. By using intention-to-treattype analysis, there was no difference in median operating time (100 vs. 110 minutes), complication (21% vs. 24%), readmission (17% vs. 10.6%), reoperation rates (6.4% vs. 4.3%), or direct costs (median US $3368 vs. US $3552; mean US $4003 vs. US $4037) between laparoscopic colectomy or open colectomy; however, a median length of stay (3 vs. 5.5 days) was significantly shorter after laparoscopic colectomy. Twenty-eight patients required conversion for adhesions (n ¼ 11), bleeding (n ¼ 3), obesity hindering vision or dissection (n ¼ 9), large phlegmon or neoplasm (n ¼ 4), and ureteric injury (n ¼ 1). The mean operating time for conversions was 142 minutes and length of stay was 6.4 days. Compared with laparoscopically completed cases, the median length of stay (5 vs. 2 days) and median operating times (150 vs. 95 minutes) were significantly higher in the converted group, but there was no difference in the complication, readmission, or reoperation rates. Compared with open colectomy, the operating time was significantly higher in the converted group but there was no significant difference in the length of stay, complication, readmission, or reoperation rates. They concluded, laparoscopic colectomy can be performed safely in obese patients although obesity is associated with a high conversion rate.

& QUALITY OF LIFE Adachi et al. (186) evaluated the quality of life of patients who had undergone laparoscopic or open colonic resection for carcinoma. The study included 26 patients with laparoscopic colectomy and 87 with conventional open colectomy for cure. Quality of life was estimated by the 9-item questionnaire with scoring system of 1 (high), 2 (fair), and 3 (low). Laparoscopic colectomy was significantly different from open colectomy with regard to C-reactive protein level (6.34 vs. 11.15 mg/dL) on postoperative day 1, albumin level (3.54 vs. 3.36 g/dL) and lymphocyte count (1354/mm3 vs. 995/mm3) on postoperative day 7 and weight loss on postoperative day 14 (3.95% vs. 5.45%). Total score of the quality-of-life questionnaire was not significantly different between the two groups (10.95 vs. 11.81). Both laparoscopic and open colonic resections were similarly accepted by the patients as a good operation that they would recommend to others (1.105 vs. 1.206). These results indicate that although laparoscopic colonic resection for carcinoma was less invasive than conventional open colectomy, both laparoscopic and open colonic resections were favorably accepted by the patients and quality of life after operation was not significantly different between the two procedures. Thaler et al. (187) compared laparoscopic colectomy and open colectomy in terms of outcome and health-related quality of life. Forty-nine patients who underwent laparoscopic right hemicolectomy or sigmoid resection for

382

& PART III: MINIMALLY INVASIVE SURGERY

benign polyps or uncomplicated diverticular disease were evaluated. Health-related quality of life was assessed by the SF-36 Physical and Mental Component Summary Score (PCS, MCS) and by the SF-36 Health Survey, which measures eight different health-quality domains, including physical and social function (PF, SF), general health perception (GH), physical and emotional limitations (RP, RE), body pain (BP), vitality (VT), and mental health (MH). There were significant differences between the two groups in resection type (26 right hemicolectomy:23 sigmoid resection in laparoscopic colectomy vs. 16 right hemicolectomy: 34 sigmoid resection in open colectomy) and length of follow-up (median 39, and 53.5 months, respectively) but neither parameters were predictive of the main SF-36 scores (PCS and MCS). There were no differences between the groups in recurrence rates (8% in laparoscopic colectomy vs. 11% open colectomy) or surgery-related complications including incisional hernias (16.3% in laparoscopic colectomy vs. 17% in open colectomy) and small bowel obstructions (2% in laparoscopic colectomy vs. 10.5% in open colectomy). None of the eight SF-36 Health Survey domains or the PCS or MCS scores showed significant differences between laparoscopic colectomy and open colectomy patients in health-related quality of life. However, occurrence of hernia after operation was predictive of lower SF-36 scores, specifically in physical functioning, general health perception, social functioning, mental health, and mental component summary score. In addition, small bowel obstruction was significantly associated with lower scores in body pain, general health perception, social functioning, emotional role limitations, mental health, and mental component summary score. Laparoscopic colectomy was not different from open colectomy for selected indications that measure long-term outcome and healthrelated quality of life. SF-36 appears to be an appropriate instrument to measure postoperative health-related quality of life, showing responsiveness to changes in objective outcome measures. Weeks et al. (158) compared short-term quality of life outcomes after laparoscopic-assisted colectomy vs. open colectomy for colon carcinoma in a multicenter randomized controlled trial (COST). Of 449 patients, 428 provided quality of life data. In an intention-to-treat analysis comparing Symptoms Distress Scale (SDS) pain intensity, SDS summary, Qualify of Life Index summary, and global rating scale scores, the only statistically significant difference observed between groups was the global rating scale score for 2 weeks postoperatively. The mean (median) global rating scale scores for 2 weeks postoperatively were 76.9 for laparoscopicassisted colectomy vs. 74.4% for open colectomy. While in the hospital patients assigned to laparoscopic-assisted colectomy required fewer days of both parenteral analgesics compared with patients assigned to open colectomy (mean 3.2 vs. 4.0 days) and oral analgesics (mean 1.9 vs. 2.2 days). Only minimal short-term quality of life benefits were found with laparoscopic-assisted colectomy for colon carcinoma compared with standard open colectomy. Dunker et al. (188) assessed the functional outcome of quality of life of laparoscopic-assisted ileal pouch-anal anastomosis compared with conventional ileal pouch-anal anastomosis. Sixteen patients who underwent laparoscopicassisted ileal pouch-anal anastomosis were matched with

19 patients who had a conventional ileal pouch-anal anastomosis. Patients were matched for time period after operation, distribution of FAP/ulcerative colitis and one-/two-stage procedure. Quality of life was measured with the SF-36 Health Survey questionnaire and the Gastrointestinal Quality of Life Index. The Body Image questionnaire was used to measure patients’ perceptions of and satisfaction with their own body and their attitude toward their bodily appearance (body image) and the degree of satisfaction of patients with respect to physical appearance of the scar (cosmesis). Patients in the conventional group were older than the patients in the laparoscopic-assisted group (mean 39.2 vs. 30.6 years). No differences were found in functional outcome and quality of life. Satisfaction with the cosmetic result of the scar was significantly higher in the laparoscopic-assisted group compared with the conventional group. Body image score was higher in the laparoscopic-assisted group compared with the conventional group although not significant. In the long term, better cosmesis is the most important advantage after laparoscopic surgery.

& COST ISSUES Cost issues have been a growing reality in the practice of medicine but there has been no randomized clinical trial of the costs of laparoscopic colon resection compared with those of open colon resection in the treatment of colonic carcinoma. A subset of Swedish patients included in the COLOR trial was included in a prospective analysis; costs were calculated up to 12 weeks after operation (189). All relevant costs to society were included. Two hundred and ten patients were included in the primary analysis, 98 of whom had laparoscopic colon resection and 112 open resection. Total cost to society did not differ significantly between the two groups (difference in means for laparoscopic colon resection vs. open colon resection, 41846). The cost of operation was significantly higher for laparoscopic colon resection than for open colon resection (difference in means 41171), as was the cost of the first admission (difference in means 41556) and the total cost to the healthcare system (difference in means 42244). Within 12 weeks of surgery for colonic carcinoma there was no difference in total cost to society incurred by laparoscopic colon resection and open colon resection. The laparoscopic colon resection procedure, however, was more costly to the healthcare system. Delaney et al. (190) performed a comparison of outcome and costs after laparoscopic and open colectomy with 150 patients in each group. American Society of Anesthesiologists classification, body mass index, diagnosis, complications, and rate of readmission within 30 days were similar for both groups. Operating room costs were significantly higher after laparoscopic colectomy but length of hospital stay was significantly lower. This resulted in significantly lower total costs owing to lower pharmacy, laboratory, and ward nursing costs. They concluded laparoscopic colectomy results in significantly lower direct costs compared with open colectomy for carefully matched patients. Senagore et al. (177) compared the direct cost structure of elective open (n ¼ 71) and laparoscopic resection (n ¼ 61) for sigmoid diverticulitis. Indirect costs and total

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

costs were not addressed. Operating time was similar (109 minutes for the laparoscopic procedure vs. 101 minutes for open procedures). The laparoscopic group had a significantly shorter length of stay (3.1 vs. 6.8 days), fewer pulmonary complications (1.6% vs. 5.6%) and fewer wound infections (0% vs. 7%). Conversion to open colectomy was required in 6.6% of patients. Readmission occurred in 4.9% of laparoscopic colectomy patients and 5.6% of open colectomy patients. Operative death occurred in 1.6% of the laparoscopic patients and there were no deaths in the open group. Total direct cost per case was significantly less for laparoscopic procedures (US $3458) than for open colectomies (US $4321). Their data demonstrates that laparoscopic colectomy is a cost-effective means of electively managing sigmoid diverticular disease. In a similar study, Dwivedi (178) found that although the mean operating room charges were greater in the laparoscopic sigmoid colectomy patients (US $9566), the mean hospital charges were less (US $13,958). In the laparoscopic-assisted approach to ileocecal Crohn’s disease, Duepree (173) found that the direct cost per case was significantly lower for the laparoscopic resection group (US $2547 vs. $2985) than for open operation and thus proved to be economically advantageous. Young-Fadok et al. (205) also compared the costs of laparoscopic ileocolic resection for Crohn’s disease to open operation. The laparoscopic group had significantly lower direct costs (US $8684 vs. $11,373) and indirect costs (US $1358 vs. $2349) than the open group. This resulted in total costs of US $9895 for laparoscopic versus US $13,268 for open procedures. There were significant postoperative benefits in terms of resolution of ileus, narcotic use, and hospital stay. In their study the laparoscopic approach translated into cost savings of more than US $3300 for laparoscopic patients. In a study comparing laparoscopic versus open sigmoid colectomy for diverticulitis, Lawrence et al. (179) found that the average total hospital charges were US $25,700 for open sigmoid colectomy and US $17,414 for laparoscopic colectomy.

& LAPAROSCOPIC COMPLICATIONS AND THEIR PREVENTION As minimally invasive procedures become more popular, it is inevitable that complications will arise. This is certainly true of operations on the colon. This section alerts the reader to the types of complications that might be encountered and provides suggestions about how the frequency of these complications might be minimized. Larach et al. (191) reviewed whether the techniques learned during their early learning experience proved to be effective in reducing the complications related to laparoscopic colorectal surgery. They divided 195 laparoscopic operations into ‘‘early’’ and ‘‘latter’’ groups. The incidence of conversions required because of iatrogenic injuries showed a decline from 7.3% in the early group to 1.4% in the latter group. Postoperative complications were observed in 30.3% of patients. Complications specifically related to the technique of laparoscopic surgery occurred in nine (4.6%) of patients. These were postoperative bleed-

383

ing in three patients, port-site hernias in five patients and left ureteric stricture in one patient. Eight (6.5%) of these complications occurred in the early group whereas only 1.1 (1.4%) occurred in the latter group. Analyzing the conversions caused by intraoperative iatrogenic injuries and the specific postoperative complications together, revealed that the incidence of 13.8% in the early group was reduced to 2.8% in the latter group.

& TROCAR COMPLICATIONS Many complications of laparoscopic procedures occur during trocar insertion. Instrumentation has been designed in an effort to eliminate this complication but no device nor entry technique is entirely safe. Shielded blades, radialexpandable and optical trocars have been recommended. Even the open technique with the Hassan-type cannula may be associated with small bowel or vascular injury. Procedure-based surveys of laparoscopic entry access injury show a reassuringly low incidence varying from 5 per 10,000 to 3 per 1000 (192). Chandler et al. (192) used existing injury-based reporting systems to access a uniquely large number of entry injuries to define the nature and outcomes of such events. Between 1980 and 1999, 594 structural organs were injured in 506 patients, resulting in 65 deaths (13%). Bowel and retroperitoneal vascular injuries comprised 76% of all injuries incurred in the process of establishing a primary port. Nearly 50% of both small and large bowel injuries were unrecognized for 24 hours or longer. Delayed recognition along with age > 59 years and major visceral vascular injuries were each independent significant predictors of death. In a recent review of the literature, Van der Voort (193) calculated the incidence of laparoscopy-induced gastrointestinal injury to be 0.13% (430 of 329,935) and of bowel perforation 0.22% (66 of 29,532). The small intestine was most frequently injured (55.8%), followed by the large intestine (38.6%). In at least 66.8% of bowel injuries, the diagnosis was made during the laparoscopy or within 24 hours thereafter. A trocar or Veress needle caused the most bowel injuries (41.8%), followed by a coagulator or laser (25.6%). In 68.9% of instances of bowel injury, adhesions or a previous laparotomy were noted. Management was mainly by laparotomy (78.6%). The mortality rate associated with laparoscopy-induced bowel injury was 3.6%. Trocar injuries to abdominal viscera occur if the viscera are unusually close to the point of trocar insertion. Distances between parietal peritoneum and underlying viscera can be increased by lifting the abdominal wall at the umbilicus with towel clips. Levrant et al. (194) reported that 59% of patients with previous midline incisions and 28% of patients with previous suprapubic and transverse incisions have anterior wall adhesions. Omentum was involved in 96% of adhesions while bowel was included in 29%. In addition, a full bladder or stomach can be injured if not emptied prior to Veress needle or trocar insertion. The Swiss Association for Laparoscopic and Thoracoscopic Surgery (SALTS) prospectively collected the data on 14,243 patients undergoing various standard laparoscopic procedures between 1995 and 1997 (195). They found 22 trocar and four needle injuries (incidence, 0.18%). Nineteen

384

& PART III: MINIMALLY INVASIVE SURGERY

lesions involved visceral organs; the remaining seven were vessel injuries. The small bowel was the single-most affected organ (six cases), followed by the large bowel and the liver (three cases each). All vascular lesions except for one laceration of the right iliac artery, occurred as venous bleeding of either the greater omentum or the mesentery. Nineteen trocar injuries were recognized intraoperatively; diagnosis of two small bowel and one bladder injuries were made postoperatively. Only five injuries could be repaired laparoscopically; there was one death among the 26 injuries (4.0%). In the Swiss review, 331 patients (2.3%) had intraoperative bleeding complications (196). Whereas 44 patients suffered from an external bleed of the abdominal wall, the bleeding was internal in the remaining 287. Thirty-three patients with internal bleeding required blood transfusion with a mean blood loss of 1630 mL. Surgical hemostasis was necessary in 68% of external and 91% of internal bleeds. There were 250 patients (1.8%) with postoperative bleeding complications. External bleeding occurred in 143 patients and 107 patients developed internal bleeding. External bleeding was mainly treated conservatively (92%), whereas 50% of internal bleeds required further surgical intervention. Major vascular injuries occurred in 12 patients (incidence 0.08%) with open treatment being necessary in all cases. The key to minimizing morbidity in cases of access injury is immediate recognition. Upon recognition, a standard repair of visceral injuries should be carried out, and in some cases this can be accomplished laparoscopically but there should be no hesitation to convert to open. In cases of vascular injury, conversion to open is usually necessary. The epigastric vessels are the most commonly injured, followed by injuries to the greater omentum and mesenteric vessels, and least common are injuries to the retroperitoneal vessels (195). Injury to a major visceral vessel (i.e. portal vein, hepatic artery, gastroduodenal artery) can carry a mortality rate of up to 44% (192). Bhoyrul et al. (197) analyzed risk factors associated with injuries resulting from use of disposable trocars with safety shields as reported to the Food and Drug Administration. They analyzed the 629 trocar injuries reported from 1993 through 1996. There were three types of injury: 408 injuries of major blood vessels, 182 other visceral injuries (mainly bowel injuries), and 30 abdominal wall hematomas. Of the 32 deaths, 81% resulted from vascular injuries and 19% resulted from bowel injuries. Eighty-seven percent of deaths from vascular injuries involved the use of disposable trocars with safety shields and 9% involved disposable trocars with a direct viewing feature. The aorta (23%) and inferior vena cava (15%) were the vessels most commonly traumatized in the fatal vascular injuries. Ninety-one percent of bowel injuries involve trocars with safety shields and 7% involve direct view trocars. The diagnosis of an enterotomy was delayed in 10% of cases and the mortality rate in this group was 21%. In 10% of cases, the surgeon initially thought the trocar had malfunctioned but in only one instance was malfunction subsequently found when the device was examined. These data show that safety shields and direct view trocars cannot prevent serious injuries. Bowel injuries often went unrecognized in which case they were highly lethal. Device malfunction was rarely a cause of trocar injuries.

& VISCERAL INJURY Small bowel injury, including the duodenum, may occur because of a grasping or cauterizing instrument. Of course, any viscera or vessel may be injured during trocar insertion. Such injuries might be minimized by cannula insertion using the open technique and subsequent cannula insertions under vision of the laparoscope. Splenic injury may be minimized by exercising caution during traction in the left upper quadrant to obtain exposure of the splenic flexure of the colon. & VASCULAR INJURY Injury to the mesenteric, iliac, or epigastric vessels may be encountered during trocar insertion. Again, insertion under direct vision of the laparoscope will hopefully minimize these injuries. Also, insertion of cannulas lateral to the rectus muscle will avoid epigastric vessels. Visualization of the undersurface of the anterior abdominal wall with the laparoscope will help avoid unnamed vessels. A case of superior mesenteric artery and portal vein thrombosis following laparoscopic-assisted right hemicolectomy has been reported (198). The patient presented with severe abdominal pain 10 days after a seemingly uneventful operation. After 10 days of abdominal pain, a CT scan revealed the diagnosis, and the patient underwent thrombolytic therapy and treatment with heparin with ultimate resolution of the thrombosis. Factors possibly contributing to this unique complication included decreased splanchnic flow with carbon dioxide pneumoperitoneum or mechanical compression of the colon and mesentery with vascular compression when the area was exteriorized for anastomosis. & URINARY TRACT INJURY Ureteral injury (either from direct transection or diathermy) has been observed with laparoscopic colectomy. In the presence of a phlegmonous mass, ureters may be drawn up into the mass and injured during dissection. Sackier (199) suggested the insertion of an illuminated ureteral catheter preoperatively to permit visualization of the illuminated ureter in its entirety. Bladder injury has also been reported during abdominoperineal resection (200). This may be encountered because the perineal portion of dissection is usually greater in extent in procedures performed laparoscopically. & ABDOMINAL WALL (PORT-SITE) RECURRENCE One of the concerns of adopting the laparoscopic approach to colorectal malignancy is the reported development of recurrence noted at a port site and not necessarily the one through which the specimen was retrieved or even one created during the operation in question (201–203). An initial review reported the incidence to vary from 1.5% to 21.0% (202), although most authors report laparoscopic colectomies for carcinoma without any evidence of port-site implantations. At least 30 port-site recurrences have been described, but the number of procedures performed is unknown, and hence the exact incidence of this complication is unknown but was initially estimated to be 4% (202). Another concern is that this development has occurred in patients with Dukes’ A, B, and C lesions, not necessarily those undergoing

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

palliative resection. Furthermore, most of these recurrences present within the first year after operation. The exact reason for these abdominal wall metastases has not definitely been elucidated and is probably multifactorial. Some offered explanations include seeding of malignant cells from intraoperative manipulation and instrumental contamination. Gas flow generated during pneumoperitoneum may spread malignant cells. Gas leaking around a port may lead to malignant cells lodging in wounds. The chimney effect, operative technique, and excessive manipulation may be contributing factors. Experimentally, pneumoperitoneum with carbon dioxide stimulates the growth of malignant colonic cells (204). In the hamster model, pneumoperitoneum increased implantation in mesenteric and midline incisions (205). Over the past decade the incidence of port-site recurrence has diminished suggesting that the incidence may be related in part to operative technique. The use of bags to remove the specimen and irrigation of the ports with different solutions has been advocated. Irrigation of the peritoneal cavity with various solutions has been suggested. Amongst those listed are povidone-iodine, heparin, methotrexate, cyclophosphamide, taurolidine, distilled water, saline, heparin þ 5-FU. Much of this information is only animal tested (78). The use of wound protectors is currently recommended. Balli et al. (206) suggested a number of techniques to prevent port-site metastases. In addition to the general oncologic principles after resection, consider irrigation of trocar sites with 5% Betadine before removal, bag the specimen, protect extraction site and irrigation of trocar sites with Betadine and water. In a Medline search, Curet (207) identified over 100 articles published during the last 15 years regarding the history incidence, etiology, and prevention of port-site metastases. The incidence of port-site metastases, initially thought to be as high as 21% is now thought to be closer to the incidence of wound metastases after open operation. There are four reported prospective randomized trials examining the incidence of wound recurrence following laparoscopic versus open resections for carcinoma. Stage et al. (155) have followed up for a median of 14 months, 18 patients who underwent laparoscopic-assisted colectomy versus 16 who were treated with standard open colectomy and found no incisional wound or port-site recurrences. Lacy et al. (208) reported no incisional wound or port-site recurrences after a median follow-up period of 21 months in a randomized study of 91 segmental resections, 44 of which were performed laparoscopically. Milsom et al. reported two cases of abdominal wall wound recurrences associated with widespread disease of 42 patients followed up for a median of 1.7 years after undergoing open colectomy. Conversely, no wound ecurrences were found in any of 38 patients treated with laparoscopic-assisted colectomy after a median follow-up period of 1.5 years (156). These studies have not been seen as providing definitive answer to the issue because of the small sample size. However, Lacy (120) reported the results of a large randomized controlled trial that had 106 patients in the laparoscopy arm versus 102 in the open colectomy group. The strengths of the study include power analysis to identify sample size necessary to show statistical significance, clearly defined inclusion and

385

exclusion criteria, and endpoints of the study. The median length of follow-up was 43 months. No statistically significant difference in the occurrence of port-site recurrences was identified in this study. Silecchia et al. (209) reported the results of the Italian Prospective Registry of Laparoscopic Colorectal Surgery confirm that the incidence of abdominal wall recurrences is similar to that reported in open studies, i.e. < 1% in 1753 cases of carcinoma. Ziprin reviewed 27 studies, each with a minimum of 50 cases from 1993 to 2001 and found an overall incidence of only 0.715 with an incidence as the only evidence of metastatic disease of 0.33% (210). Ziprin’s study suggests that the incidence of port-site metastases after laparoscopic surgery is similar to that seen after open surgery. Two of the prospective randomized trials found no port-site metastases at follow-up ranging from 1.5 to 2 years (156,208). In a recent review of 29 publications involving 5305 patients, the incidence of port-site metastases was 0.72% (78). In another recent review by Patankar and Lee (1) collated information from 23 reports showed wound recurrence rates following laparoscopic colectomy to be low—between 0% and 2.4% and compared favorably to their review of eight reports following open resection with rates ranging from 0.9% to 3.3%. Specific reference to the incidence of wound implantation following open colectomy is seldom made. Hughes et al. (211) described a 1% incidence of metastases to the surgical access wound. Most of these patients had advanced disease at the time of the initial operation. Reilly et al. (212) reviewed 1711 patients with primary carcinoma treated for cure and found that 0.6% of patients had documented incisional recurrences. Nine of 11 patients were found to have multiple sites of recurrence, suggesting that it is usually a harbinger of diffuse abdominal disease. From the registry initiated under the auspices of The American Society of Colon and Rectal Surgeons, the American College of Surgeons Commission on Cancer, and the Society of American Gastrointestinal Endoscopic Surgeons, recurrences were evaluated by the primary surgeon and reported to the registry (213). A minimum follow-up of 1 year was obtained for 480 of 493 evaluable patients (97.4%). Wound recurrence was identified in five patients (1.1%).

& BLEEDING Bleeding, either intraoperatively or postoperatively, is not a complication unique to laparoscopic procedures. Bleeding may be mesenteric in origin or may originate in other locations. Anastomotic bleeding may arise whether the anastomosis is created with sutures or staples or whether it is performed extracorporeally or intracorporeally. & ANASTOMOTIC DEHISCENCE Anastomotic complications following bowel operations are a dreaded but well-recognized occurrence. Fistula formation may also arise. There is no documented evidence to suggest that the incidence of these complications has increased with the use of laparoscopic procedures. Rotation of a loop involved in an anastomosis is a rare but known complication after open resection. Rotation of the terminal ileum after a laparoscopic-assisted right

386

& PART III: MINIMALLY INVASIVE SURGERY

hemicolectomy has been described (214). To prevent this complication, inspection of the terminal ileum before exteriorization and again after completion of the anastomosis is recommended (214). If the anastomosis is performed extracorporeally closure of the mesentery prior to the anastomosis will also obviate this complication.

& MISSED LESIONS A frustrating situation for the surgeon is the inability to identify the pathology at the time of laparoscopic colectomy. Where serosal involvement is absent, definitive recognition of the pathology is difficult or may be impossible. This limitation has resulted in the resection of a wrong segment of colon. If in doubt, intraoperative colonoscopy with transillumination should identify the location of the pathology. The preoperative injection of carbon particles into the normal colonic wall around the lesion may permit accurate delineation of the lesion and avoid timeconsuming colonoscopy (205). Unrecognized synchronous carcinomas have also been reported (216). Reported success rates for the detection of neoplasms after tattooing vary between 78.6% and 98% (78). Nizam et al. (217) reviewed the world literature. A total of 734 citations on India ink alone were present. Nine major studies were identified and reviewed. Various India ink preparations were used. Ink was unsterilized in 57%, autoclaved 42%, and gas sterilized in 1% of cases. Prophylactic antibiotics were used in 1% of cases. Dilution of India ink varied from undiluted to 1:100 (with 0.9% saline). The volume injected range from 0.1 to 2 mL per site injected, commonly with tangential needle insertion and delivery of ink into the submucosa in the majority of cases. Intraoperative localization was easier with multiple tattoo injections. Five reports of complications have been made. In only one instance did overt clinical complications develop. Risk of a clinical complication with colonic tattooing with India ink is 0.22%. Injection into the peritoneal space has been reported in 0.5% to 8% (218,219). & ELECTROSURGICAL INJURY Electrosurgical injuries occur during laparoscopic operations and are potentially serious. The overall incidence of recognized injuries is between one and two patients per 1000 operations. The majority of such injuries go unrecognized at the time of the electrical insult and commonly present 3 to 7 days afterward with fever and pain in the abdomen. Because of the delayed presentation they often carry a high mortality. Since these injuries appear late, the pathophysiology remains speculative. Nduka et al. (220) conducted a comprehensive review of the physics of electrosurgery and provided the surgeon with an insight to the mechanisms responsible in each type of injury. The authors summarized the main causes of electrosurgical injuries as inadvertent touching or grasping of tissue during current application, direct coupling between a portion of intestine and a metal probe that is touching the activated probe, insulation breaks in the electrodes, direct sparking to the intestine from the diathermy probe, and current passage to the intestine from recently coagulated electrically isolated tissue. The majority of injuries, not surprisingly, are caused by monopolar diathermy. Bipolar diathermy

is safer and should be used in preference to monopolar diathermy, especially in anatomically crowded areas. An awareness of the hazards of diathermy together, with an understanding of the mechanisms of injury should enable the surgeon to dissect tissue and achieve hemostasis, and at the same time, decrease the risk of serious complications for the patient.

& ULTRASONIC ENERGY INJURY Ultrasonic energy can be used instead of electrosurgery and its utilization can replace mechanical surgical clips and scissors. Ultrasonically activated devices can cut and coagulate tissues (Harmonic Scalpel1, Ethicon Endosurgery Inc.; Autosonix1, US Surgical Corporation, Norwalk, Connecticut, U.S.; Sonosurg1, Olympus Corp.). These devices can coagulate vessels up to 5 mm in diameter (221). & WOUND As with any incision, wound infection is always a possibility. This may occur in a port site or through the incision created for specimen retrieval. The antibiotic prophylaxis used for a colon operation should minimize this complication. Richter’s hernia developing at a port site is possible when using cannulas 10 mm or larger in size. It is therefore important to make every effort to close the fascia through which a 10-mm or larger cannula has been inserted. This does not appear necessary for 5-mm cannulas. Patients may, indeed, require laparotomy for correction of such a hernia. Winslow et al. (222) randomized 37 patients to laparoscopic colectomy and 46 to open colonic resection. Seven patients in the laparoscopic group were converted to open. Laparoscopic colectomy was performed using a limited midline incision for anastomosis and specimen extraction. Incision length was significantly greater in the open group (19.4 cm) compared with the laparoscopic extraction site (6.3 cm). Wound infections occurred in 13.5% of patients after laparoscopic (2.7% trocar, 10.8% extraction sites) and in 10.9% of patients after open colectomy. Over a mean follow-up period of 30.1 months, incisional hernias developed in 24.3% of patients after laparoscopic colectomy and 17.4% after open colectomy. In the laparoscopic group, extraction sites accounted for 85.7% of all wound complications. They concluded the extraction site for laparoscopic colectomy is associated with a high incidence of complications comparable to open colectomy. In their review Hackert et al. (223) noted there is no standardized retrieval technique for the different procedures. Specimen retrieval after colonic resection is difficult due to the large size of the specimen usually resected. Three major complications are described in the literature: wound infection (0.9%), hernias (0–2%), and incision site recurrence (0–1.3%). In laparoscopic surgery for malignant disease retrieval is usually performed using a plastic bag, whereas retrieval can be performed handassisted without a bag during operations for benign diseases. Wound-edge protectors are recommended by several authors yet there is no standard system which is accepted broadly. The morbidity rate for specimen retrieval complications ranges between 0% and 9%.

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

Kercher et al. (224) reviewed the outcomes for patients with and without the use of a wound protector in 141 patients who underwent laparoscopic-assisted colectomy (98 for benign/malignant neoplasms, 35 for diverticular disease, and 8 for Crohn’s disease). There were no differences between the wound protector group (n ¼ 84) and the no-wound protector group (n ¼ 57) with respect to mean age, average body mass index, gender, indication for operation, co-morbidities, antibiotics used, or mean operative time. Nine patients in the wound protector group and eight in the no-wound protector group developed a wound infection at the colon extraction site. Patients undergoing resection for Crohn’s disease or diverticulitis had a higher infection rate (18.6%) than patients undergoing resection for polyps or carcinoma (9.2%). No-wound recurrence of carcinoma was observed in either group at a mean follow-up of 23 months. They concluded that the wound protector, although useful for mechanical retraction of small wounds, does not significantly diminish the rate of wound infection.

& MISCELLANEOUS Brachial plexus injury attributed to the Trendelenburg position during prolonged laparoscopic procedures has been reported (109). Care should be exercised to avoid excessive hyperextension of the patient’s shoulder. & NONSPECIFIC COMPLICATIONS A host of nonspecific complications have been reported to be associated with laparoscopic colectomy. These may include cardiac arrhythmias such as atrial fibrillation, urinary tract infection, urinary retention, pulmonary embolus, pneumonia, atelectasis, cerebral vascular accident, and stress gastritis. Pneumoperitoneum may cause mechanical pressure on the inferior vena cava which reduces cardiac filling and potentially reduces blood pressure. Pulmonary effects may be mechanical with direct pressure on the diaphragm or metabolic due to increased CO2 that needs to be disposed via ventilation. Patients should be monitored for hypercarbia and the risk of CO2 embolus exists. Physiologic changes associated with pneumoperitoneum are often controlled by the anesthesiologist. Frequently, cervical emphysema, pneumothorax, and pneumomediastinum are attributed to the passage of insufflating gas through defects in the diaphragm. The anesthesiologist must maintain a high index of suspicion for these potential complications and must undertake appropriate monitoring. If there is clinical evidence of a tension pneumothorax, immediate chest tube decompression is indicated. Cottin et al. (225) reported seven cases of carbon dioxide embolism during laparoscopic surgery. None of the cases involved colorectal operations, but this complication is not procedure specific. In the seven cases cited, gas embolism occurred during insufflation or a few minutes later. All the patients had a previous abdominal or pelvic surgical history. Five patients presented with cardiac bradycardia or arrhythmia. Cardiovascular collapse or cyanosis was the first manifestation in three cases. Sudden bilateral mydriasis was the earliest neurologic sign and was present in five cases. The gas embolism complication was lethal in two cases.

387

Baixauli et al. (226) reported a case of portal vein thrombosis in a patient with no other demonstrable hypercoagulable states or risk factors who underwent an uneventful laparoscopic sigmoid colectomy. They suggest that heparin prophylaxis may be advisable to avoid these kinds of complications, especially if a past history of coagulable disorders is present.

& REFERENCES 1. Patankar SK, Lee W. Current status of laparoscopic resection for colorectal cancer. J Clin Gastroenterol 2004; 38:621–627. 2. Jacobs M, Verdeja JC, Goldstein HS. Minimally invasive colon resection (laparoscopic colectomy). Surg Laparosc Endosc 1991; 1:144–150. 3. Fowler DL, White SA. Laparoscopic-assisted sigmoid resection. Surg Laparosc Endosc 1991; 1:183–188. 4. Schlinkert RT. Laparoscopic-assisted right hemicolectomy. Dis Colon Rectum 1991; 34:1030–1031. 5. Saclarides TJ, Ko ST, Airan M, et al. Laparoscopic removal of a large colonic lipoma. Report of a case. Dis Colon Rectum 1991; 34:1027–1029. 6. Cooperman AM, Katz V, Zimmon D, et al. Laparoscopic colon resection: a case report. J Laparoendosc Surg 1991; 1:221–224. 7. Nezhat C, Pennington E, Nezhat F, et al. Laparoscopically assisted anterior rectal wall resection and reanastomosis for deeply infiltrating endometriosis. Surg Laparosc Endosc 1991; 1:106–108. 8. Lange V, Meyer G, Schardey HM, et al. Laparoscopic creation of a loop colostomy. J Laparoendosc Surg 1991; 1:307–312. 9. Ortiz H, Armendariz P, Yarnoz C. Early postoperative feeding after elective colorectal surgery is not a benefit unique to laparoscopic-assisted procedures. Int J Colorectal Dis 1996; 11:246–249. 10. Hotokezaka M, Dix J, Mentis EP, et al. Gastrointestinal recovery following laparoscopic vs. open colon surgery. Surg Endosc 1996; 10:485–489. 11. Hildebrandt U, Kessler K, Plusczyk T, Pistorius G, Vollmar B, Menger MD. Comparison of surgical stress between laparoscopic and open colonic resections. Surg Endosc 2003; 17:242–246. 12. Duepree HJ, Senagore AJ, Delaney CP, Fazio VW. Does means of access affect the incidence of small bowel obstruction and ventral hernia after bowel resection? Laparoscopy versus laparotomy. J Am Coll Surg 2003; 197: 177–181. 13. Simons AJ, Anthone GH, Ortega AK. Laparoscopic-assisted colectomy learning curve. Dis Colon Rectum 1995; 38:600–603. 14. Wishner JD, Baker JW Jr., Hoffman GC, et al. Laparoscopic assisted colectomy. The learning curve. Surg Endosc 1995; 9:1179–1183. 15. Agachan F, Joo JS, Weiss EG, et al. Intraoperative laparoscopic complications: Are we getting better? Dis Colon Rectum 1996; 39:S14–S19. 16. Romanelli JR, Litwin DEM. Hand-assisted laparoscopic surgery. Prob Gen Surg 2001; 18:45–51. 17. Southern Surgeons Club Study Group. Handoscopic surgery. A prospective, multicenter trial of minimally invasive techniques for complex abdominal surgery. Arch Surg 1999; 34:477–486. 18. HALS Study Group. Hand-assisted laparoscopic surgery vs standard laparoscopic surgery for colorectal disease. Surg Endosc 2000; 14:896–901. 19. Taragona EM, Gracia E, Garriga J, et al. Prospective randomized trial comparing conventional laparoscopic colectomy with hand-assisted laparoscopic colectomy. Surg Endosc 2002; 16:234–239. 20. Kang JC, Chung MH, Chao PC, et al. Hand-assisted laparoscopic colectomy vs open colectomy: a prospective randomized study. Surg Endosc 2004; 18: 577–581. 21. Senagore AJ, Madbouly KM, Fazio VW, Duepree HJ, Brady KM, Delaney CP. Advantages of laparoscopic colectomy in older patients. Arch Surg 2003; 138:252–256. 22. Schmitt SL, Cohen SM, Wexner SD, et al. Does laparoscopic-assisted ileal pouch anal anastomosis reduce the length of hospitalization? Int I Colorectal Dis 1994; 9:134–137. 23. Moore JWE, Bokey EL, Newland RC, et al. Lymphovascular clearance in laparoscopically assisted right hemicolectomy is similar to open surgery. Aust N Z J Surg 1996; 66:605–607. 24. Jacquet P, Sugarbaker PH. Wound recurrence after laparoscopic colectomy for cancer. New rationale for intraoperative intraperitoneal chemotherapy. Surg Endosc 1996; 16:295–296. 25. Gonzalez R, Smith CD, Ritter EM, et al. Laparoscopic palliative surgery for complicated colorectal cancer. Surg Endosc 2005; 19:43–46. 26. Sackier JM. Laparoscopic equipment and basic techniques. Perspect Colon Rectal Surg 1993; 6:175–189. 27. Veress J. Neues instrument zur ausfahrung von brust- oder bauchpunktionen und pneumothoraxbehandlung. Dtsch Med Wochenschr 1938; 41:1480–1481.

388

& PART III: MINIMALLY INVASIVE SURGERY

28. Darzi A. Hand-assisted laparoscopic colorectal surgery. Surg Endosc 2000; 14:999–1004. 29. Ballantyne GH, Leahy PR. Hand-assisted laparoscopic colectomy: evolution to a clinically useful technique. Dis Colon Rectum 2004; 47:753–765. 30. Nakajima K, Milsom JW, Margolin DA, Szilagy EJ. Use of the surgical towel in colorectal hand-assisted laparoscopic surgery (HALS). Surg Endosc 2004; 18:552–553. 31. Nakajima K, Lee SW, Cocilovo C, et al. Hand-assisted laparoscopic colorectal surgery using GelPort. Surg Endosc 2004; 18:102–105. 32. Chang YJ, Marcelle PW, Rusin LC, Roberts PL, Sohoetz DJ. Hand-assisted laparoscopic sigmoid colectomy: helping hand or hindrance? Surg Endosc 2005; 19:655–661. 33. Hasson HM. Modified instrument and method for laparoscopy. Am J Obstet Gynecol 1971; 110:886–887. 34. Nisii H, Hirai T, Ohara H, et al. Laparoscopic-assisted colon surgery by abdominal wall lifting with newly developed lifting bars. Surg Endosc 1997; 11:754–757. 35. Monson JRT, Hill ADK, Darzi A. Laparoscopic colonic surgery. Br J Surg 1995; 82:150–157. 36. Munegato G, De Min V, Schiano di Visconte M, Salemi S, Barbaresco S, Mazzarolo G. The diagnosis of non-palpable lesions in laparoscopic surgery of the colon. Chir Ital 2003; 55:657–661. 37. Feingold DL, Addona T, Forde KA, et al. Safety and reliability of tattooing colorectal neoplasms prior to laparoscopic resection. J Gastrointest Surg 2004; 8:543–546. 38. Zmora O, Dinnewitzer AJ, Pikarsky AJ, et al. Intraoperative endoscopy in laparoscopic colectomy. Surg Endosc 2002; 16:808–811. 39. Sackier JM, Berci G, Paz-Partlow M. Elective diagnostic laparoscope Am J Surg 1992; 161:326–331. 40. Reissman P, Teoh TA, Piccirillo M, et al. Colonoscopic-assisted laparoscopic colectomy. Surg Endosc 1994; 8:1352–1353. 41. John TG, Greig RD, Crosbie JL, et al. Superior staging of liver tumors with laparoscopy and laparoscopic ultrasound. Ann Surg 1994; 220:711–719. 42. Franklin ME, Ramos R, Rosenthal D, et al. Laparoscopic colonic procedures. World J Surg 1993; 17:51–53. 43. Van Ye TM, Cattery RP, Henry LG. Laparoscopically assisted colon resections compare favorably with open technique. Surg Laparosc Endosc 1994; 4:25–31. 44. Scoggin SD, Frazee RC, Snyder SK, et al. Laparoscopic-assisted bowel surgery. Dis Colon Rectum 1993; 36:747–750. 45. Quattlebaum JK, Flanders HD, Usher CH, et al. Laparoscopically assisted colectomy. Surg Laparosc Endosc 1993; 3:81–87. 46. Zucker KA, Pitcher DE, Martin DT, et al. Laparoscopic-assisted colon resection. Surg Endosc 1994; 8:12–18. 47. Puente I, Sosa JL, Sleeman D, et al. Laparoscopic assisted colorectal surgery. J Laparoendosc Surg 1944; 4:1–7. 48. Steichen FM. The use of staplers in anatomical side-to-side and functional end-to-end enteroanastomoses. Surgery 1968; 64:948–952. 49. Bernstein MA, Dawson JW, Reissman P, et al. Is complete laparoscopic colectomy superior to laparoscopic assisted colectomy? Am Surg 1996; 62:507–511. 50. Larson DW, Nelson H. Laparoscopic colectomy for cancer. J Gastrointest Surg 2004; 8:636–642. 51. Liang JT, Lai HS, Lee PH. Laparoscopic medial-to-lateral approach for the curative left hemicolectomy. Dis Colon Rectum 2005 Sept 16 [Epub ahead of print]. 52. Tanaka J, Ito M, Shindo Y, et al. Laparoscopically assisted resection of the lower rectum. Surg Endosc 1996; 10:338–340. 53. Loungnorath R, Fleshman J. Hand-assisted laparoscopic colectomy techniques. Semin Laparosc Surg 2003; 10:219–230. 54. Darzi A, Lewis C, Mengies-Gow N, et al. Laparoscopic abdominoperineal excision of the rectum. Surg Endosc 1995; 9:414–417. 55. Decanini C, Milsom JW, Bohm B, et al. Laparoscopic oncologic abdominoperineal resection. Dis colon Rectum 1994; 37:552–558. 56. Larach SW, Salomon MC, Williamson PR, et al. Laparoscopic-assisted abdominoperineal resection. Surg Laparosc Endosc 1993; 3:115–118. 57. Chindasub S, Charntaracharmnong C, Nimitvanit C, et al. Laparoscopic abdominoperineal resection. J Laparoendosc Surg 1994; 4:17–21. 58. Lange V, Meyer G, Schardey HM, et al. Laparoscopic creation of a loop colostomy. J Laparoendosc Surg 1991; 1:217–220. 59. Khoo RE, Montrey J, Cohen MM. Laparoscopic loop ileostomy for temporary fecal diversion. Dis Colon Rectum 1993; 36:966–968. 60. Fuhrman GM, Ota DM. Laparoscopic intestinal stomas. Dis Colon Rectum 1994; 37:444–449. 61. Lyerly HK, Mault JR. Laparoscopic ileostomy and colostomy. Ann Surg 1994; 219:317–322. 62. Mattingly M, Wasvary H, Sacksner J, Deshmukh G, Kadro O. Minimally invasive, endoscopically assisted colostomy can be performed without general anesthesia or laparotomy. Dis Colon Rectum 2003; 46:271–273. 63. Sosa JL, Sleeman D, Puente I, et al. Laparoscopic-assisted colostomy closure after Hartmann’s procedure. Dis Colon Rectum 1994; 37:149–152. 64. Constantino GN, Mukalian GG. Laparoscopic reversal of Hartmann procedure. J Laparoendosc Surg 1994; 4:429–433. 65. Vernava AM, Liebscher G, Longo WE, et al. Laparoscopic restoration of intestinal continuity after Hartmann procedure. Surg Laparosc Endosc 1995; 5:129–132. 66. Anderson CA, Fowler DL, White S, et al. Laparoscopic colostomy closure. Surg Laparosc Endosc 1993; 3:69–72. 67. Khan AL, Ah-See AK, Crofts TJ, et al. Reversal of Hartmann’s colostomy. J R Coll Surg Edinb 1994; 39:239–242.

68. Wexner SD, Johansen OB, Nogueras JJ, et al. Laparoscopic total abdominal colectomy. A prospective trial. Dis Colon Rectum 1992; 35:651–655. 69. Nakajima K, Lee SW, Cocilovo C, et al. Laparoscopic total colectomy. Handassisted vs standard technique, Surg Endosc 2004; 18:582–586. 70. Weber PA, Merola S, Wasielewski A, Ballantyne GH. Telerobotic-assisted laparoscopic right and sigmoid colectomies for benign disease. Dis Colon Rectum 2002; 45:1689–1694. 71. Merola S, Weber P, Wasielewski A, Ballantyne GH. Comparison of laparoscopic colectomy with and without the aid of a robotic camera holder. Surg Laparosc Endosc Percutan Tech 2002; 12:46–51. 72. Corcione F, Esposito C, Cuccurulo D, et al. Advantages and limits of robotassisted laparoscopic surgery. Surg Endosc 2005; 19:117–119. 73. Delaney C, Lynch A, Senagore A, et al. Comparison of robotically performed and traditional laparoscopic colorectal surgery. Dis Colon Rectum 2003; 46:1633–1639. 74. Munz Y, Moorthy K, Kudchadkar R, et al. Robotic assisted rectopexy. Am J Surg 2004; 187:88–92. 75. D’Annibale A, Morpurgo E, Fiscon V, et al. Robotic and laparoscopic surgery for treatment of colorectal disease. Dis Colon Rectum 2004; 47:2162–2168. 76. Hanly EJ, Talamini MA. Robotic abdominal surgery. Am J Surg 2004; 188(4A suppl):19S–26S. 77. Gutt CN, Oniu T, Mehrabi A, Kashfi A, Schemmer P, Buchler MW. Robotassisted abdominal surgery. Br J Surg 2004; 91:1390–1397. 78. Veldkamp R, Gholghesaei M, Bonjer HJ, et al. European Association of Endoscopic Surgery (EAES). Laparoscopic resection of colon cancer: consensus of the European Association of Endoscopic Surgery (EAES). Surg Endosc 2004; 18:1163–1185. 79. Slim K, Pezet D, Riff D, et al. High morbidity rate after converted laparoscopic colorectal surgery. Br J Surg 1995; 82:1406–1408. 80. Pandya S, Murray JJ, Coller JA, Rusin LC. Laparoscopic colectomy: indications for conversion to laparotomy. Arch Surg 1999; 134:471–475. 81. Marusch F, Gastinger I, Schneider C, et al. Laparoscopic Colorectal Surgery Study Group (LCSSG). Importance of conversion for results obtained with laparoscopic colorectal surgery. Dis Colon Rectum 2001; 44:207–214. 82. Gervaz P, Pikarsky A, Utech M, et al. Converted laparoscopic colorectal surgery. Surg Endosc 2001; 15:827–832. 83. Schlachta CM, Mamazza J, Seshadri PA, Cadeddu MO, Poulin EC. Predicting conversion to open surgery in laparoscopic colorectal resections. A simple clinical model. Surg Endosc 2000; 14:1114–1117. 84. Tekkis PP, Senagore AJ, Delaney CP. Conversion rates in laparoscopic colorectal surgery: a predictive model with 1253 patients. Surg Endosc 2005; 19:47–54. 85. Casillas S, Delaney CP, Senagore AJ, Brady K, Fazio VW. Does conversion of a laparoscopic colectomy adversely affect patient outcome? Dis Colon Rectum 2004; 47:1680–1685. 86. Basse L, Madsen JL, Billesbolle P, Bardram L, Kehlet H. Gastrointestinal transit after laparoscopic versus open colonic resection. Surg Endosc 2003; 17:1919–1922. 87. Kasparek MS. Muller MH, Glatzle J, et al. Postoperative colonic motility in patients following laparoscopic-assisted and open sigmoid colectomy. J Gastrointest Surg 2003; 7:1073–1081. 88. Asao T, Kuwano H, Nakamura J, Morinaga N, Hirayama I, Ide M. Gum chewing enhances early recovery from postoperative ileus after laparoscopic colectomy. J Am Coll Surg 2002; 195:30–32. 89. Raue W, Haase O, Junghans T, Scharfanberg M, Muller JM, Schwenk W. ’Fast-track’ multimodal rehabilitation program improves outcome after laparoscopic sigmoidectomy: a controlled prospective evaluation. Surg Endosc 2004; 18:1463–1468. 90. Reissman P, Agachan F, Wexner SD. Outcome of laparoscopic colorectal surgery in older patients. Am Surg 1996; 62:1060–1063. 91. Peters WR, Fleshman JW. Minimally invasive colectomy in elderly patients. Surg Laparosc Endosc 1995; 5:477–479. 92. Delgado S, Lacy AM, Garcia-Valdecasas JC, et al. Could age be an indication for laparoscopic colectomy in colorectal cancer? Surg Endosc 2000; 14:22. 93. Sklow B, Read T, Birnbaum E, Fry R, Fleshman J. Age and type of procedure influence the choice of patients for laparoscopic colectomy. Surg Endosc 2003; 17:923–929. 94. Law WL, Chu KW, Tung PHM. Laparoscopic colorectal resection: A safe option for elderly patients. J Am Coll Surg 2002; 195:768–773. 95. Seshadri PA, Mamazza J, Schlachta CM, Cadeddu MO, Poulin EC. Laparoscopic colorectal resection in octogenarians. Surg Endosc 2001; 15:802–805. 96. Vignali A, Di Palo S, Tamburini A, Radaelli G, Orsenigo E, Staudacher C. Laparoscopic vs. open colectomies in octogenarians: a case-matched control study. Dis Colon Rectum 2005 Aug 3; [Epub ahead of print]. 97. Schlachta CM, Mamazza J, Seshadri PA, Cadeddu M, Gregoire R, Poulin EC. Defining a learning curve for laparoscopic colorectal resections. Dis Colon Rectum 2001; 44:217–222. 98. Bennett CL, Stryker SJ, Ferreira MR, Adams J, Beart RW Jr. The learning curve for laparoscopic colorectal surgery. Preliminary results from a prospective analysis of 1194 laparoscopic-assisted colectomies. Arch Surg 1997; 132:41–44. 99. Dincler S, Koller MT, Steurer J, Bachmann LM, Christen D, Buchmann P. Multidimensional analysis of learning curves in laparoscopic sigmoid resection: eighth-year results. Dis Colon Rectum 2003; 46:1371–1378. 100. Ortega AK, Beart RW Jr, Steel GD Jr, et al. Laparoscopic bowel surgery registry: preliminary results. Dis Colon Rectum 1995; 38:681–686. 101. Kuhry E, Bonjer HJ, Haglind E, et al. COLOR Study Group. Impact of hospital case volume on short-term outcome alter laparoscopic operation for colonic cancer. Surg Endosc 2005; 19:687–692.

CHAPTER 7: LAPAROSCOPIC COLON AND RECTAL SURGERY &

102. Wexner SD, Reissman P, Pfeifer J, et al. Laparoscopic colorectal surgery: an analysis of 140 cases. Surg Endosc 1996; 10:133–136. 103. Ramos JM, Beart RW Jr, Goes R, et al. Role of laparoscopy in colorectal surgery. A prospective evaluation of 200 cases. Dis Colon Rectum 1995; 38:494–501. 104. Huscher C, Silecchia G, Croce E, et al. Laparoscopic colorectal resection. A multicentre Italian study. Surg Endosc 1996; 10:875–879. 105. Begos DG, Arsenault J, Ballantyne GH. Laparoscopic colon and rectal surgery at a VA hospital. Analysis of the first 50 cases. Surg Endosc 1996; 10:1050–1056. 106. Tucker JG, Ambroze WL, Orangio GR, et al. Laparoscopically assisted bowel surgery. Analysis of 114 cases. Surg Endosc 1995; 9:297–300. 107. Lumley JW, Fielding GA, Rhodes M, et al. Laparoscopic-assisted colorectal surgery: lessons learned from 240 consecutive patients. Dis Colon Rectum 1996; 39:155–159. 108. Slim K, Pezet D, Stenci J, et al. Prospective analysis of 40 initial laparoscopic colorectal resections: a plea for a randomized trial. J Laparoendosc Surg 1994; 4:241–245. 109. Fleshman JW, Fry RD, Birnbaum EH, et al. Laparoscopic-assisted and minilaparotomy approaches to colorectal diseases are similar in early outcome. Dis Colon Rectum 1996; 39:15–22. 110. Gagnon J, Poulin EC. Beware of the Trendelenburg position during prolonged laparoscopic procedure. Can J Surg 1993; 36:505–506. 111. Degiuli M, Mineccia M, Bortone A, Arrigoni A, Pennazio M, Spandre M, Cavallero M, Calvo F. Outcome of laparoscopic colorectal resection. Surg Endosc 2004; 18:427–432. 112. Senagore AJ, Duepree HJ, Delaney CP, Brady KM, Fazio VW. Results of a standardized technique and postoperative care plan for laparoscopic sigmoid colectomy: a 30-month experience. Dis Colon Rectum 2003; 46:503–509. 113. Lezoche E, Feliciotti F, Guerrieri M, et al. Laparoscopic versus open hemicolectomy. Minerva Chir 2003; 58:491–502, 502–507. 114. Schlachta CM, Mamazza J, Gregoire R, Burpee SE, Poulin EC. Could laparoscopic colon and rectal surgery become the standard of care? A review and experience with 750 procedures. Can J Surg 2003; 46:432–440. 115. Lauter DM, Froines EJ. Initial experience with 150 cases of laparoscopicassisted colectomy. Am J Surg 2001; 181:398–403. 116. Seshadri PA, Poulin EC, Schlachta CM, Cadeddu MO, Mamazza J. Does a laparoscopic approach to total abdominal colectomy and proctocolectomy offer advantages? Surg Endosc 2001; 15:837–842. 117. Kockerling F, Schneider C, Reymond MA, et al. Early results of a prospective multicenter study on 500 consecutive cases of laparoscopic colorectal surgery. Laparoscopic Colorectal Surgery Study Group (LCSSG). Surg Endosc 1998; 12:37–41. 118. Kockerling F, Rose J, Schneider C, et al. Laparoscopic colorectal anastomosis: risk of postoperative leakage. Results of a multicenter study. Laparoscopic Colorectal Surgery Study Group (LCSSG). Surg Endosc 1999; 13:639–644. 119. Schwenk W, Haase O, Neudecker J, Muller JM. Short term benefits for laparoscopic colorectal resection. Cochrane Database Syst Rev 2005; July 20(3):CD003145. 120. Lacy AM, Garcia-Valdecasas JC, Delgado S, et al. Laparoscopic-assisted colectomy vs. open colectomy for treatment of non-metastatic colon cancer: a randomized trial. Lancet 2002; 359:2224–2229. 121. Lujan HJ, Plasencia G, Jacobs M, et al. Long-term survival after laparoscopic colon resection for cancer. Dis Colon Rectum 2002; 45:491–501. 122. Patankar S, Larach S, Ferrara A, et al. Prospective comparison of laparoscopic vs. open resections for colorectal adenocarcinoma over a ten-year period. Dis Colon Rectum 2003; 46:601–611. 123. The Clinical Outcomes of Surgical Therapy Study Group. A comparison of laparoscopically assisted and open colectomy for colon cancer. NEJM 350:2050–2059. 124. Veldkamp R, Kuhry E, Hop WC, et al., Colon Cancer Laparoscopic or Open Resection Study Group (COLOR). Laparoscopic surgery versus open surgery for colon cancer: short-term outcomes of a randomized trial. Lancet Oncol 2005; 6:477–484. 125. Ramos JR, Petrosemolo RH, Valory EA, Polania FC, Pecanha R. Abdominoperineal resection: laparoscopic versus conventional. Surg Laparosc Endosc 1977; 7:148–152. 126. Khalili TM, Fleshner PR, Hiatt JR, et al. Colorectal cancer: comparison of laparoscopic with open approaches. Dis Colon Rectum 1998; 41:832–838. 127. Hong D, Tabet J, Anvari M. Laparoscopic vs. open resection for colorectal adenocarcinoma. Dis Colon Rectum 2001; 44:10–18. 128. Baker RP, White EE, Titu L, Duthie GS, Lee PW, Monson JR. Does laparoscopic abdominoperineal resection of the rectum compromise long-term survival? Dis Colon Rectum 2002; 45:1481–1485. 129. Champault GG, Barrat C, Raselli R, Elizalde A, Catheline JM. Laparoscopic versus open surgery for colorectal carcinoma: a prospective clinical trial involving 157 cases with a mean follow-up of 5 years. Surg Laparosc Endosc Percutan Tech 2002; 12:88–95. 130. Curet MJ, Putrakul K, Pitcher DE, Josloff RK, Zucker KA. Laparoscopically assisted colon resection for colon carcinoma: perioperative results and longterm outcome. Surg Endosc 2000; 14:1062–1066. 131. Leung KL, Kwok SP, Lam SC, et al. Laparoscopic resection of rectosigmoid carcinoma: prospective randomised trial. Lancet 2004; 10;363(9416):1187–1192. 132. Guillou PJ, Quirke P, Thorpe H, et al., MRC CLASICC trial group. Short-term endpoints of conventional versus laparoscopic-assisted surgery in patients with colorectal cancer (MRC CLASSIC trial): multicentre, randomized controlled trial. Lancet 2005; 365:1718–1726.

389

133. Melotti G, Tamborrino E, Lazzaretti MG, Bonilauri S, Mecheri F, Piccoli M. Laparoscopic surgery for colorectal cancer. Semin surg Oncol 1999; 16:332–336. 134. Franklin ME, Kazantsev GB, Abrego D, Diaz-E JA, Balli J, Glass JL. Laparoscopic surgery for stage III colon cancer: long-term follow-up. Surg Endosc 2000; 14:612–616. 135. Schiedeck TH, Schwandner O, Baca I, et al. Laparoscopic surgery for the cure of colorectal cancer: results of a German five-center study. Dis Colon Rectum 2000; 43:1–8. 136. Anderson CA, Kennedy FR, Potter M, et al. Results of laparoscopically assisted colon resection for carcinoma. Surg Endosc 2002; 16:607–610. 137. Lechaux D, Trebuchet G, Le Calve JL. Five-year results of 206 laparoscopic left colectomies for cancer. Surg Endosc 2002; 16:1409–1412. 138. Lumley J, Stitz R, Stevenson A, Fielding G, Luck A. Laparoscopic colorectal surgery for cancer: intermediate to long-term outcomes. Dis Colon Rectum 2002; 45:867–872. 139. Poulin EC, Schlachta CM, Gregoire R, Seshadri P, Cadeddu MO, Mamazzsa J. Local recurrence and survival after laparoscopic mesorectal resection for rectal adenocarcinoma. Surg Endosc 2002; 16:989–995. 140. Scheidbach H, Schneider C, Huegel O, et al. Laparoscopic sigmoid resection for cancer: curative resection and preliminary medium-term results. Dis Colon Rectum 2002; 45:1641–1647. 141. Yamamoto S, Watanabe M, Hasegawa H, Kitajima M. Prospective evaluation of laparoscopic surgery for rectosigmoidal and rectal carcinoma. Dis Colon Rectum 2002; 45:1648–1654. 142. Morino M, Parini U, Giraudo G, Salval M, Contul BR, Garrone C. Laparoscopic total mesorectal excision: a consecutive series of 100 patients. Ann Surg 2003; 237:335–342. 143. Tsang WW, Chung CC, Li MK. Prospective evaluation of laparoscopic total mesorectal excision with colonic J-pouch reconstruction for mid and low rectal cancers. Br J Surg 2003; 90:867–871. 144. Watanabe M, Hasegawa H, Yamamoto S, Baba H, Kitajima M. Laparoscopic surgery for stage I colorectal cancer. Surg Endosc 2003; 17:1274–1277. 145. Bouvet M, Mansfield PF, Skibber JM, et al. Clinical, pathologic, and economic parameters of laparoscopic colon resection for cancer. Am J Surg 1998; 176(6):554–558. 146. Schwandner O, Schiedeck TH, Killaitis C, Bruch HP. A case-control-study comparing laparoscopic versus open surgery for rectosigmoidal and rectal cancer. Int J Colorectal Dis 1999; 14:158–163. 147. Anthuber M, Fuerst A, Elser F, Berger R, Jauch KW. Outcome of laparoscopic surgery for rectal cancer in 101 patients. Dis Colon Rectum 2003; 46:1047–1053. 148. Franklin ME Jr, Rosenthal D, Abrego-Medina D, et al. A prospective comparison of open vs laparoscopic colon surgery for carcinoma: five-year results. Dis Colon Rectum 1996; 39:S35–S46. 149. Santoro E, Carlini M, Carboni F, Feroce A. Colorectal carcinoma: laparoscopic versus traditional open surgery. A clinical trial. Hepatogastroenterology 1999; 46:900–904. 150. Hartley JE, Mehigan BJ, Qureshi AE, Duthie GS, Lee PW, Monson JR. Total mesorectal excision: assessment of the laparoscopic approach. Dis Colon Rectum 2001; 44:315–321. 151. Feliciotti F, Guerrieri M, Paganini AM, Sanctis A, Campagnacci R, Lezoche E. Results of laparoscopic vs open resections for colon cancer in patients with a minimum follow-up of 3 years. Surg Endosc 2002; 16:1158–1161. 152. Lezoche E, Feliciotti F, Paganini AM, et al. Laparoscopic vs open hemicolectomy for colon cancer. Surg Endosc 2002; 16:596–602. 153. Felicitti F, Guerrieri m, Paganini AM, et al. Long-term results of laparoscopic versus open resections for rectal cancer for 124 unselected patients. Surg Endosc 2003; 17:1530–1535. 154. Abraham NS, Young JM, Solomon MJ. Meta-analysis of short-term outcomes after laparoscopic resection for colorectal cancer. Br J Surg 2004; 91:1111–1124. 155. Stage JG, Schulze S, Moller P, et al. Prospective randomized study of laparoscopic versus open colonic resection for adenocarcinoma. Br J Surg 1997; 84:391. 156. Milsom JW, Bo¨hm B, Hammerhofer KA, Fazio V, Steiger E, Elson PA. Prospective, randomized trial comparing laparoscopic versus conventional techniques in colorectal cancer surgery: a preliminary report. J Am Coll Surg 1998; 187:46. 157. Hasegawa H, Kabeshima Y, Watanabe M, Yamamoto S, Kitajima M. Randomized controlled trial of laparoscopic versus open colectomy for advanced colorectal cancer. Surg Endosc 2003; 17:636. 158. Weeks JC, Nelson H, Gelber S, Sargentfor AD, The Clinical Outcomes of Surgical Therapy (COST) Study Group, Schroeder G. Short-term quality-of-life outcomes following laparoscopic-assisted colectomy vs open colectomy for colon cancer: a randomized trial. JAMA 2002; 287:321. 159. Schwenk W, Bo¨hm B, Mu¨ller JM. Postoperative pain and fatigue after laparoscopic or conventional colorectal resections. Surg Endosc 1998; 12:1131. 160. Kieran JA, Curet MJ. Laparoscopic colon resection for colon cancer. J Surg Res 2004; 117:79–91. 161. Tang CL, Eu KW, Tai BC, Soh JG, MacHin D, Seow-Choen F. Randomized clinical trial of the effect of open versus laparoscopically assisted colectomy on systemic immunity in patients with colorectal cancer. Br J Surg 2001; 88:801. 162. Delgado S, Momblan D, Salvador L, et al. Laparoscopic-assisted approach in rectal cancer patients: lessons learned from > 200 patients. Surg Endosc 2004; 18:1457–1462. 163. Law WL, Chu KW, Tung HM. Early outcome of 100 patients with laparoscopic resection for rectal neoplasm. Surg Endosc 2004; 18:1592–1596.

390

& PART III: MINIMALLY INVASIVE SURGERY

164. Zhou ZG, Hu M, Li Y, et al. Laparoscopic versus open total mesorectal excision with anal sphincter preservation for low rectal cancer. Surg Endosc 2004; 18:1211–1215. 165. Leroy J, Jamali F, Forbes L, Smith M, Rubino F, Mutter D, Marescaux J. Laparoscopic total mesorectal excision (TME) for rectal cancer surgery: long-term outcomes. Surg Endosc 2004; 18:281–289. 166. Rullier E, Sa Cunha A, Couderc P, Rullier A, Gontier R, Saric J. Laparoscopic intersphincteric resection with coloplasty and coloanal anastomosis for mid and low rectal cancer. Br J Surg 2003; 90:445–451. 167. Jayne DG, Brown JM, Thorpe H, Walker J, Quirke P, Guillou PJ. Bladder and sexual function following resection for rectal cancer in a randomized clinical trial of laparoscopic versus open technique. Br J Surg 2005; 92:1124–1132. 168. Milsom JW, Kim SH, Hammerhofer KA, Fazio VW. Laparoscopic colorectal cancer surgery for palliation. Dis Colon Rectum 2000; 43:1512–1516. 169. Delgado S, Lacy AM, Garcia Valdecasas JC, Balague C, Pera M, Salvador L, et al. Could age be an indication for laparoscopic colectomy in colorectal cancer? Surg Endosc 2000; 14:22–26. 170. Capussotti L, Massucco P, Murator A, Amisano M,. Bima C, Zorzi D. Laparoscopic as a prognostic factor in curative resection for node positive colorectal cancer: results for a single-institution nonrandomized prospective trial. Surg Endosc 2004; 18:1130–1135. 171. Kockerling F, Scheidbach H, Schneider C, et al. Laparoscopic abdominoperineal resection: early postoperative results of a prospective study involving 116 patients. The Laparoscopic Colorectal Surgery Study Group. Dis Colon Rectum 2000; 43:1503–1511. 172. Ludwig KA, Milsom JW, Garcia-Ruiz A, et al. Laparoscopic techniques for fecal diversion. Dis Colon Rectum 1996; 39:285–288. 173. Duepree HJ, Senagore AJ, Delaney CP, Brady KM, Fazio VW. Advantages of laparoscopic resection for ileocecal Crohn’s disease. Dis Colon Rectum 2002; 45:605–610. 174. Oliveira L, Reissman P, Nogueras J, et al. Laparoscopic creation of stomas. Surg Endost 1997; 11:19–23. 175. Swain BT, Ellis CN Jr. Laparoscopy-assisted loop ileostomy: an acceptable option for temporary fecal diversion after anorectal surgery. Dis Colon Rectum 2002; 45:705–707. 176. Liu J, Bruch HP, Farke S, Nolde J, Schwnadner O. Stoma formation for fecal diversion: a plea for the laparoscopic approach. Tech Coloproctol 2005; 9: 9–14. 177. Senagore AJ, Duepree HJ, Delaney CP, Dissanaike S, Brady KM, Fazio VW. Cost structure of laparoscopic and open sigmoid colectomy for diverticular disease: similarities and difference. Dis Colon Rectum 2002; 45:485–490. 178. Dwivedi A, Chahin F, Agrawal S, et al. Laparoscopic colectomy vs. open colectomy for sigmoid diverticular disease. Dis Colon Rectum 2002; 45:1309–1314. 179. Lawrence DM, Pasquakle MD, Wasser TE. Laparoscopic versus open sigmoid colectomy for diverticulitis. Am Surg 2003; 69:499–503. 180. Cobb WS, Lokey JS, Schwab DP, Crockett JA, Rex JC, Robbins JA. Handassisted laparoscopic colectomy: a single-institution experience. Am Surg 2003; 69:578–580. 181. Nakajima K, Lee SW, Cocilovo C, Foglia C, Kim K, Sonoda T, Milsom JW. Laparoscopic total colectomy: hand-assisted vs standard technique. Surg Endosc 2004; 18:582–586. 182. Targarona EM, Gracia E, Garriga J, et al. Prospective randomized trial comparing conventional laparoscopic colectomy with hand-assisted laparoscopic colectomy: applicability, immediate clinical outcome, inflammatory response, and cost. Surg Endosc 2002; 16:234–239. 183. Senagore AJ, Delaney CP, Madboulay K, Brady KM, Fazio VW. Laparoscopic colectomy in obese and nonobese patients. J Gastrointest Surg 2003; 7:558–561. 184. Schwandner O, Farke S, Schiedeck TH, Bruch HP. Laparoscopic colorectal surgery in obese and nonobese patients: do difference in body mass indices lead to different outcomes? Surg Endosc 2004; 18:1452–1456. 185. Delaney CP, Pokala N, Senagore AJ, et al. Is laparoscopic colectomy applicable to patients with body mass index > 30? A case-matched comparative study with open colectomy. Dis Colon Rectum 2005; 48:975–981. 186. Adachi Y, Sato K, Kakisako K, et al. Quality of life after laparoscopic or open colonic resection for cancer. Hepatogastroenterology 2003;. 50: 1348–1351. 187. Thaler K, Dinnewitzer A, Mascha E, et al. Long-term outcome and healthrelated quality of life after laparoscopic and open colectomy for benign disease. Surg Endosc 2003; 17:1404–1408. 188. Dunker MS, Bemelman WA, Slors JF, van Duijvendijk P, Gouma DJ. Functional outcome, quality of life, body image, and cosmesis in patients after laparoscopic-assisted and conventional restorative proctocolectomy: a comparative study. Dis Colon Rectum 2001; 44:1800–1807. 189. Janson M, Bjorholt I, Carlsson P, et al. Randomized clinical trial of the costs of open and laparoscopic surgery for colonic cancer. Br J Surg 2004; 91: 409–417. 190. Delaney CP, Kiran RP, Senagore AJ, Brady K, Fazio VW. Case-matched comparison of clinical and financial outcome after laparoscopic or open colorectal surgery. Ann Surg 2003; 238:67–72. 191. Larach SW, Patankar SK, Ferrara A, Williamson PR, Perozo SE, Lord AS. Complications of laparoscopic colorectal surgery. Analysis and comparison of early vs. later experience. Dis Colon Rectum 1997; 40:592–596.

192. Chandler JG, Corson SL, Way LW. Three spectra of laparoscopic entry access injuries. J Am Coll Surg 2001;. 192:478–490. 193. Van der Voort M, Heijnsdijk EA, Gouma DJ. Bowel injury as a complication of laparoscopy. Br J Surg 2004; 91:1253–1258. 194. Levrant SG, Bieber EJ, Barnes RB. Anterior abdominal wall adhesions after laparotomy or laparoscopy. J Am Assoc Gynecol Laparosc 1997; 4:353–356. 195. Schafer M, Lauper M, Krahenbuhl L. Trocar and Veress needle injuries during laparoscopy. Surg Endosc 2001; 15:275–280. 196. Schafer M, Lauper M, Krahenbuhl L. A nation’s experience of bleeding complications during laparoscopy. Am J Surg 2000; 180:73–77. 197. Bhoyrul S, Vierra MA, Nezhat CR, Krummel TM, Way LW. Trocar injuries in laparoscopic surgery. J Am Coll Surg 2001; 192:677–683. 198. Millikan KW, Szczerba SM, Dominguez JM, et al. Superior mesenteric portal vein thrombosis following laparoscopic-assisted right hemicolectomy. Dis Colon Rectum 1996; 39:1171–1175. 199. Sackier JM. Visualization of the ureter during laparoscopic colonic resection. Br J Surg 1993; 80:1332. 200. Williams IM, Haray PN, Lloyd-Davies E, et al. Bladder injury during laparoscopic abdominoperineal resection. Br J Surg 1995; 82:1207. 201. Fusco MA, Paluzi MW. Abdominal wall recurrence after laparoscopic-assisted colectomy for adenocarcinoma of the colon. Dis Colon Rectum 1993; 36:858–861. 202. Wexner SD, Cohen SM. Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 1995; 82:295–298. 203. Ugarte F. Laparoscopic cholecystectomy port seeding from a colon carcinoma. Am Surg 1995; 61:820–821. 204. Jacobi CA, Sabat R, Bo¨hm B, et al. Pneumoperitoneum with carbon dioxide stimulates growth of malignant colonic cells. Surgery 1997; 121:72–78. 205. Jones DB, Guo LW, Reinhard MK, et al. Impact of pneumoperitoneum on trocar site implantation of colon cancer in hamster model. Dis Colon Rectum 1995; 38:1182–1188. 206. Balli JE, Franklin ME, Alameida A, et al. How to prevent port-site metastases in laparoscopic colorectal surgery. Surg Endosc 2000; 14:1034. 207. Curet MJ. Port site metastases. Am J Surg 2004; 187:705–712. 208. Lacy AM, Delgado S, Garcı´a-Valdecasas JC, et al. Port site metastases and recurrence after laparoscopic colectomy. A randomized trial. Surg Endosc 1998; 12:1039. 209. Silecchia G, Perrotta N, Giraudo G, et al., For the Italian Registry of Laparoscopic Colorectal Surgery. Abdominal wall recurrence after colorectal resection for cancer: results of the Italian registry of laparoscopic colorectal surgery. Dis Colon Rectum 2002; 45:1172–1177. 210. Ziprin P, Ridgway PF, Peck DH, Darzi AW. The theories and realities of portsite metastases: a critical appraisal. J Am Coll Surg 2002; 195:395. 211. Hughes ESR, McDermott FT, Polglase AL, et al. Tumor recurrence in the abdominal wall scar tissue after large bowel cancer surgery. Dis Colon Rectum 1983; 26:571–572. 212. Reilly WT, Nelson H, Schroeder G, et al. Wound recurrence following conventional treatment of colorectal cancer. A rare but perhaps underestimated problem. Dis Colon Rectum 1996; 39:200–207. 213. Vukasin P, Ortega AK, Greene FL, et al. Wound recurrence following laparoscopic colon cancer resection: results of The American Society of Colon and Rectal Surgeons Laparoscopic Registry. Dis Colon Rectum 1996; 39:S20–S23. 214. Bonjer HJ, Lange JF. Rotation of the terminal ileum in laparoscopic right hemicolectomy. Surg Endosc 1993; 7:534. 215. Kitamura K, Yamane T, Oyama T, et al. Rapid and accurate method for delineating cancer lesions in laparoscopic colectomy using activated carbon injection. J Surg Oncol 1995; 58:31–34. 216. McDermott JP, Devereaux DA, Caushaj PF. Pitfall of laparoscopic colectomy: an unrecognized synchronous cancer. Dis Colon Rectum 1994; 37:602–603. 217. Nizam R, Siddiqi N, Landas SK, Kaplan DS, Holtzapple PG. Colonic tattooing with India ink: benefits, risks, and alternatives. Am J Gastroenterol 1996; 91:1804–1808. 218. Coman E, Brandt LS, Brenner S, et al. Fat necrosis and inflammatory pseuduotumor due to endoscopic tattooing of the colon with India ink. Gastrointest Endosc 1991; 37:65–68. 219. Park SI, Genta RS, Romeo DP, Weesner RE. Colonic abscess and focal peritonitis secondary to India ink tattooing of the colon. Gastrointest Endosc 1991; 37:68–71. 220. Nduka CC, Sager PA, Monson JRT, et al. Cause and prevention of electrosurgical injuries in laparoscopy. J Am Coll Surg 1994; 179:161–170. 221. Hoenig DM, Chrostek CA, Amaral JF. Laparosonic coagulating shears: alternative method of hemostatic control of unsupported tissue. J Endourol 1996; 10:431–433. 222. Winslow ER, Fleshman JW, Birnbaum EH, Brunt LM. Wound complications of laparoscopic vs open colectomy. Surg Endosc 2002; 16:1420–1425. 223. Hackert T, Uhl W, Buchler MW. Specimen retrieval in laparoscopic colon surgery. Dig Surg 2002; 19:502–506. 224. Kercher KW, Nguyen TH, Harold KL, et al. Plastic wound protectors do not affect wound infection rates following laparoscopic-assisted colectomy. Surg Endosc 2004; 18:148–151. 225. Cottin V, Delafosse B, Viale JP. Gas embolism during laparoscopy. A report of seven cases in patients with previous abdominal surgical history. Surg Endosc 1996; 10:166–169. 226. Baixauli J, Delaney CP, Senagore AJ, Remzi FH, Fazio VW. Portal vein thrombosis after laparoscopic sigmoid colectomy for diverticulitis: report of a case. Dis Colon Rectum 2003; 46:550–553.

Index Abdominal colectomy, 70 Abdominal wall metastases, 168 Abdominoperineal resection endorectal ultrasound, 277 for hemangiomas, 30 for rectal adenocarcinoma abdominal portion of, 215 excised tissue amounts, 215 high ligation of inferior mesenteric artery, 265 indications, 215 perineal defects after, 288 perineal portion of, 228 recurrence, 281 sacral defects after, 286 for squamous cell carcinoma of anal canal, 317 total mesorectal excision (TME), 254 Abdominosacral approach, for rectal adenocarcinoma, 283 Aberrant crypt foci (ACF), 72 Acanthosis nigricans, 95 Adenocarcinoma sequence description of, 3 discovery of, 3 genetic model of, 4 molecular genetic studies, 4 survival rates, 322 treatment modes, 322 Adenocarcinoma, of colon of anal glands, 321 in anorectal fistula, 322 in children and young adults, 147 clinical features, 321 complications of, 321 diagnostic investigations of anoscopy, 98 barium enema, 98 carcinoembryonic antigen, 102 computed tomography, 100 cytology, 102 hydrocolonic sonography, 102 intravenous pyelography, 99 magnetic resonance imaging, 100 occult blood testing, 98 positron emission tomography, 101 radioimmunodetection, 101 sigmoidoscopy, 98 ultrasonography, 99 epidemiology of, 63–64 etiologic factors alcohol ingestion, 76 bile acids, 72 cholecystectomy, 78 diet, 72 diverticular disease, 78 exercise levels, 78 genetics, 55 inflammatory bowel disease, 55 intestinal microflora, 80 iron stores, 80 irradiation, 77 menopause, 79 oxygen radical production, 80 smoking, 76 ureteric implantation, 77 incidence of, 54, 55 metastasis direct continuity, 88

[Adenocarcinoma] hematogenous, 89 implantation, 89 to liver, 135 to lung, 137 lymphatic, 89 modes, 88 to ovary, 140 palliative resection, 133 to peritoneum, 161 sites of, 89 staging, 89 transperitoneal, 89 pathology, 82 polypoid classification of, 7 description of, 82 pathology of, 82 prevalence of, 52 treatment of, 321 ulcerative, 82 venous invasion, 160 Adenocarcinoma, of rectum abdominoperineal resection excised tissue amounts, 315 preoperative adjuvant radiotherapy and, 317 adjuvant therapy chemoradiotherapy, 317 complications of, 321 radiotherapy, 317 clinical features, 321 complications of, 321 local therapeutic methods external beam radiotherapy, 319 laser therapy, 315 rationale for, 319 low anterior resection rectum marking, 323 recurrence rates, 322 recurrence of classification, 283 clinical features, 282 contributing factors, 281 follow up, 281 operative treatment, 283 Adenoma(s) benign, management of, 5–6 description of, 2 flat, 5 polypoid, 5 serrated, 72 sessile, 5 traditional, 72 tubular, 2 tubulovillous, 17 villous, 2 Adenomatous polyp with colorectal carcinoma, 42 lifetime risk of colon carcinoma, 43 Adenomatous polyposis coli (APC) gene, 56 Adenosquamous carcinoma, 184 Adjuvant X-ray and 5-FU infusion study (AXIS), 121 Alcohol intake, 76 Allele-specific oligonucleotide analysis, 70 Amsterdam criteria, 56 Anal canal adenocarcinoma of anal glands, 321

391

[Anal canal] in anorectal fistula, 322 of rectum, 322 treatment of, 321 diagnosis and treatment, 316 excision, 311 small cell carcinoma, 322 squamous cell carcinoma clinical manifestations, 316 description of, 316 diagnosis of, 316 metastases, 317 sites for, 320 workup for, 316 treatment abdominoperineal resection, 317 chemotherapy, 317 local excision, 317 radiation therapy, 317 undifferentiated carcinoma, 322 Anal canal manometry, 231 Anal carcinoma colorectal surgeon involvement, 310 high-risk groups, 306 incidence of, 306 screening, cytology test, 307 staging of, 307 Anal cytology, 307 Anal dysplasia, 309 Anal glands, adenocarcinoma of, anatomy of, 321 Anal intercourse, 307 Anal intraepithelial neoplasia (AIN) biopsy in, 309 and Bowen’s disease, 309 diagnosis, 309 high-grade, 309 high-resolution anoscopy in, 310 treatment options, 309 application of imiquimod, 310 cautery ablation, 310 excision, 311 topical 5-fluorouracil, 310 Anal mapping biopsy, 310 Anal squamous cell carcinoma, 306 molecular biology of, 306 multifactorial interaction of, 307 progression, 306 Anal squamous intraepithelial lesions, screening, 307 Anaphylaxis, 88 Anastomosis with circular stapler, 236 intracorporeal, 355, 367 leak, 96 for low anterior resection, 216 sutureless, 227 Angiosarcoma diagnosis, 186 histomorphology, 186 locations, 186 risk factors, 186 Anorectal adenocarcinoma, 313 Anorectal fistula, adenocarcinoma in, 322 Anorectal physiology, 214 Anterior resection syndrome, 230 Antiemetics, 176 Appendectomy, effect on adenocarcinoma survival rates, 180

392

& INDEX

Ashkenazi heritage, 56 Aspirin, 81–82 Attenuated familial adenomatous polyposis (AFAP) abdominal surgery for, 21 carcinomas in, 17 clinical features, 17 colonoscopic surveillance, 18 definition of, 17 diagnosis of, 17 duodenal adenomas, 19 genetic testing, 17 baseline colonoscopy, 18 esophagoduodenoscopy, 18 germ-line mutations, 18 protein truncation test (PTT), 18 surgical management, 18 Autosonix1, 386

Bannayan-Ruvalcaba-Riley syndrome, 15 Barium enema double-contrast, 43 screening use, 43 Basal cell carcinoma of perianal region clinical features, 315 description of, 314 prognosis of, 315 treatment of, 315 B-cell lymphomas, 182 Beta-carotene intakes, 74 Bevacizumab, 173 evaluation, 123 in neoplastic angiogenesis inhibition, 123 Bilateral salpingo-oophorectomy, for adenocarcinoma, 145 Bile acid metabolism, 80 Biopsy rectal carcinoma, 328 description of, 327 indications, 335 mucosa, 328 Bipolar diathermy, 386 Bisexual men, 307 Bladder carcinoma, 53 dysfunction of, 259 involvement, 94 reconstruction, 132 Bleomycin, 321 Boari flap, 132 Body mass index, 157 Bone metastases diagnosis, 145 incidence sites, 145 treatment, 145 Bowel carcinoma, 145 Bowen’s disease diagnosis, 309 5-fluorouracil (5-FU) therapy in, 309 incidence, 309 management, 311 perianal, 309 treatment of, 309 Brachial plexus injury, 387 Brachytherapy, 319 catheters, 283 perioperative high-dose-rate, 287 Brain neoplasms and colorectal adenomas, 23 genetic abnormalities, 23 molecular diagnosis, 23 in patients with familial adenomatous polyposis, Turcot’s syndrome, 23 Budding-positive lesions, 161

Calcium chemopreventive action pathway, 74 colorectal lesions, effect in, 75 deficiency, adenocarcinoma pathogenesis and, 74 sensing receptor, 74 supplementation in adenoma, 74 Cancer and Leukemia Group B (CALGB) Protocol, 88 Carbon dioxide (CO2), embolism, 387 Carcinoembryonic antigen (CEA), 102, 139 adenocarcinoma detection using, 103 metastasis detection using, 46 preoperative levels, effect on prognosis, 103 testing kits for, 103 Carcinogenesis alternate pathway of, 72 oxygen radicals, role of, 80 Carcinoids of colon chemical activity, 179 clinical features, 178 incidence of, 178 metastatic disease, 180 pathology of, 178 treatment of, 179 imaging procedures for, 179 of rectum clinical presentation, 291 description of, 291 investigations of, 291 pathology of, 291 treatment of, 291 Carcinoma adenosquamous, 184 anal colorectal surgeon involvement, 310 high-risk groups, 306 incidence of, 306 screening test, 307 staging of, 307 basal cell, of perianal region clinical features, 315 description of, 314 prognosis of, 315 treatment of, 315 bladder, 53 chromosomal instability of, 59 chronic pain causes arthopathies and musculoskeletal pain, 174 chronic postsurgical incisional pain, 174 peripheral neuropathies, 174 phantom pain, 174 radiation fibrosis, 174 visceral pain, 174 of colon, 307 colorectal annual incidence of, 42 complicated, 123 development factors, 54 early, 39, 40 family syndrome, 65 gastric, 63 hereditary nonpolyposis colorectal carcinoma, 56 history of, 53 and intestinal microflora association, 80 intramucosal, 9 invading, 86 invasive, 8 laparoscopy for, 344 metachronous, 134 microsatellite instability of, 59 mucosal, 94 obstructive, 11 ovarian, 63 postoperative follow-up of, 46

[Carcinoma] radiation therapy for, 77 rectal electrocoagulation for, 337 and rectosigmoid junction, 79 recurrence, 137 risk factors, 41 scirrhous, 161 screening of average risk individuals, 41 definition of, 41 serosal, 94 sporadic, 9 submucosal, 94 undifferentiated, 322 of urinary tract, 63 Carcinoma-susceptibility genes, 71 Carcinomatosis, peritoneal, 140 Carcinosarcoma, 186 Cautery ablation, 310 Cecal diverticulitis, right colectomy for, 363 Celecoxib (Celebrex1), 20 Chemoradiation complications, 320 regimen, 317 therapy postoperative, 267 preoperative, 260, 290 Chemotherapy for adenocarcinoma, 171 5-fluorouracil, 118 leucovorin, 119 of rectum, 266 recurrence, 118 antisarcoma, 22 postoperative, 282 for squamous cell carcinoma of anal canal clinical studies of, 318 failure sites after, 320 regimen, 317 toxicity, 320 Cholecystectomy adenocarcinoma pathogenesis and, 78 and colorectal carcinoma, 78 laparoscopic, 342 Choriocarcinoma, 186 Chromogranin positivity, 186 Cigarette smoking, 76 Cisplatin, 320, 171 Cisplatinum, 319 Colectomy advantages, 342–343 contraindications, 343 disadvantages, 343 hand-assisted laparoscopic (HAL), 346 indications, 343 laparoscopic, 342 description of, 342–343 left-sided, 355 results of, 364 right-sided, 353–355 total abdominal, with ileorectal anastomosis, 362–363 operating room setup in, 348–349 positioning, 357 procedures, 353 right-sided, 353–355 robot-assisted, 363 trocar positionings of, 353 Colitis, obstructive, 128 Collagenous colitis, 327 Coloanal anastomosis colorectal resection, 232 fecal continence, 231 low anterior rectal resection with, 234 mortality and morbidity, 232 preoperative radiotherapy, 232 for rectal adenocarcinoma, 252

INDEX &

Colon carcinoids of chemical activity, 179 imaging procedures for, 179 pathology of, 178 results, 180 serotonin secretions, 178 treatment of, 178 choriocarcinoma of, 186 leukemic infiltration of, 185 medullary carcinoma of, 185–186 Colon, adenocarcinoma of clinical features bleeding, 94 general examination, 95 pain, 94 symptoms, 94 colloid, 82 complications of, 96 adjacent viscera invasions, 129 bleeding, 97 infections, 97 perforation, 97 depressed description of, 85 pathologic features of, 85 diagnosis of, 53 diagnostic methods anoscopy, 98 carcinoembryonic antigen, 102 cytology, 102 hydrocolonic sonography, 102 intravenous pyelography, 99 magnetic resonance imaging, 100 occult blood testing, 98 positron emission tomography, 101 radioimmunodection, 101 sigmoidoscopy, 98 ultrasonography, 99 diffusely infiltrating, 82 epidemiology of geographic predilection, 54 occupational factors, 54 patient age, 53 racial predilection, 54 religious predilection, 54 etiologic factors alcohol ingestion, 76 bile acids, 72 cholecystectomy, 78 diet, 72 exercise levels, 78 genetics, 55 inflammatory bowel disease, 55 intestinal microflora, 80 iron stores, 80 irradiation, 77 menopause, 79 oxygen radical production, 80 examinations for abdominal, 95 digital rectal, 95 growth of, 92 juvenile, 80 metastasis implantation, 89 to liver, 135 to lung, 137 lymphatic, 89 modes, 88 to ovary, 140 to peritoneum, 161 sites, 89 staging, 89 transperitoneal, 89 mucinous, 81 operative treatment, 170 pathology, 82

[Colon] perineural invasion, 161 polypoid, 82 description of, 323 pathology of, 82 during pregnancy, 177 prevention approaches, 81 prognostic factors, 170 adjacent organ involvement, 156 appendectomy, 160 lymph node metastasis, 157 metastatic disease, 157 obstruction, 156 pathologic features, 160 perforation, 156 perineural invasion, 161 perioperative blood transfusion, 160 resection technique, 157 sex, 155 systemic manifestations, 157 recurrent contributing factors, 166 endoscopic laser therapy, 170 follow-up recommendations, 166 incidence of, 166 operative, 170 patterns of, 167 positron emission tomography, 101 radiotherapy, 133 survival rates, 166 SEER data regarding, 54 symptoms of, 94 synchronous carcinomas, 134 treatment of chemotherapy, 118 palliative resection, 133 radiotherapy, 133 ulcerative, 82 venous invasion, 163 Colon Cancer Laparoscopic or Open Resection (COLOR) trial, 368, 372, 382 Colon carcinoma and coffee intake, 76 and fiber diet, 73 and folate intake, 77 routine preoperative CT scanning in, 100 screening, 70 Colonic choriocarcinomas, 186 Colonic endoscopy, 70 Colonic J-pouch, 132 Colonic lavage, 125 Colonic lymphoma, 181 Colonic polyps, 343 Colonic splenic flexure, 356 Colonic stents, 126 Colonoscopy and biopsy, 42 characteristics of, 42 dye spraying, 43 fecal occult blood test (FOBT), 42 flat/depressed lesions, 5 mortality from colorectal carcinoma, 43 screening, 42 Colon resection, 342, 356, 369 Colon surgery, postoperative care, 364 Coloplasty and colonic J-pouch, 236 colorectal/coloanal anastomosis, 236 with distal rectal carcinoma, 235 low anterior rectal resection, 235 neo-rectal sensitivity, 235 quality of life assessment, 236 Colorectal adenoma aspirin, effects of, 81 prevention, 75 recurrence of, 75 selenium, effect of, 75 Colorectal anastomosis, 157

393

Colorectal carcinoma aggregation, 63 and alcohol intake, 76 and aspirin application, 81 beta-carotene intake and, 74 Bethesda Guidelines for testing, 61, 62 bile acid metabolism in, 80 causes of, 40 chemotherapy study, 118 cholecystectomy and, 78 and cigarette smoking, 76 in cirrhotic patients, 150 clinical features, 152 complications, 96 curative resection for, 104 detection and prevention of, 11 disease recurrence in, 166 early, diagnosis of, 40 early, effect on prognosis, 40 familial, 65 fiber intake in, 73 genetic testing for, 65 growth speed estimation, 94 history of, 40, 43 iron deficiency anemia in, 94 and leisure physical activity, 79 lymphatic mapping to, 87 magnesium and, 75 and malakoplakia, 177 metachronous, 134 micrometastases in, 87 for microsatellite instability, 56 molecular events to, 58, 59 multiple primary, 65 multivariate-adjusted relative rate of, 76 mutation for, 59 NSAIDs, effects of, 81 nutritional interventions in, 75 obstruction, 96 personal and family feature risks for, 66 and polyps relation, 95 and post-menopausal hormone therapy, 79 postoperative complications, 147 postoperative follow-up of, 46 progression pathway, 59 recurrent, 131 risks, 63 screening, 13 of average-risk individuals, 41 definition of, 41 individuals eligible for, 40 and selenium, 75 submucosal invasion of, 86 treatment agents fluorouracil, 174 irinotecan, 174 oxaliplatin, 172 treatment algorithm for, 87 urinary tract involvement in, 131 vitamin C intake and, 74 Colorectal disorders, 343 Colorectal hepatic metastases, 138 Colorectal liver metastases, biopsy of, 135 Colorectal lymphoma signs and symptoms, 182 treatment, 182 Colorectal neoplasms, 80 Colorectal obstruction, treatment, 126 Colorectal polyps, 6 Colorectal procedures robotics, 363–364 Colorectal resections, 365, 366, 367 benefits of colectomy in, 366 conversion rate results of, 364–365 gastrointestinal transit normalization in, 365 hepatic and, 137 Herculean tasks in, 364 laparoscopic procedures in elderly, 366

394

& INDEX

[Colorectal resections] learning curve for, 366–368 morbidity and mortality, 368 nonspecific complications, 387 postoperative care in, 364 postoperative ileus after, 365 quality of life after, 381–382 Colorectal schwannomas, 186 Colorectal screening program, 152 Colostomy for abdominoperineal resection, 229 intraperitoneal, 226 maturation of, 226 for unresectable rectal carcinoma, 256 Computed tomography (CT) diagnostic applications adenocarcinoma, 100 Computed tomography colonography (CTC) for colorectal carcinoma screening, 44 fecal DNA testing detection of neoplasia, 44 DNA mutations, 44 methylation, 44 stool-based testing, 44 for mass screening test, 44 Congenital hypertrophy of retinal pigment epithelium (CHRPE), 63 in patients with familial adenomatous polyposis, 22, 23 Cowden’s disease, 14 diagnosis of, 14 familial syndrome, 14 genetic alterations, 14 Lhermitte-Duclos syndrome, 14 management and surveillance, 14 neoplastic risk, 14 C-reactive protein, 342 Crohn’s disease adenocarcinoma and, 110 Cronkhite-Canada syndrome, 14 Cryoablation, 142 Cul-de-sac involvement, 177 Curative resection, for adenocarcinomas abdomen exploration, 104 bilateral salpingo-oophorectomy, 116 bowel preparation, 104 of descending colon, 104, 105 end-to-end anastomosis, 108, 110 follow-up after, 46 incisions, 104 intraoperative considerations, 105 principles of, 104 for right hemicolectomy, 105 of sigmoid colon, 105, 106 stapled, 110 suturing, 110 of transverse colon, 104 Cystoureterectomy, 132 Cysts, cutaneous, 63 Cytette brush, 307 Cytokeratin, 145 immunohistochemistry, 87 Cytologic test for anal carcinoma, 307 screening, 307 of urine, 44 Cytoreduction, 171

Da Vinci telerobotic surgical system, 363 de Gramont fluorouracil regimen, 174 Dermatomyositis, 95 Descending colon, adenocarcinoma of, curative resection for, 104 Desmoid(s) asymptomatic, 22 familial adenomatous polyposis (FAP), 21 symptomatic, 22

[Desmoid(s)] tumors abdominal surgery, 21 classification of, 22 definition of, 21 fibroaponeurotic tissue, 21 intra-abdominal, 21 management of, 22 mesodermal abnormalities, 21 mortality causes, 21 signs and symptoms of, 21 treatment approaches, 22 Diathermy dissection, 242 Diet in adenocarcinoma pathogenesis calcium deficiency, 74 chemical inhibitors, 75 fiber, 73 micronutrients, 75 Disease-specific survival (DSS), 319 DNA chip technology, 165 double-strand break repair, 59 methylation, 165 mismatch repair defects, 58 genes, 56 polymerase, 56 precursors, 76 repair genes, 72 sequencing, 65 synthesis and repair, 76 Double-contrast barium enema (DCBEs), 43 detection of colonic neoplasms, 43 and flexible sigmoidoscopy, 42 incidence/mortality from colorectal carcinoma, 43 surveillance population, 43 Double-stapling technique, 224, 376 Double stent insertion, 133 Doxorubicin, 321, 180 Doxorubicin-based regimens, 186 Duodenal adenomas, treatment of, 20 Duodenal papilla, 20 Duodenal polyp and carcinoma, 20 classification of, 20 endoscopic features and histology of, 20 macroscopic appearance of, 19 pathogenesis of, 20 Duodenectomy, 20 Dye spray technique, 86 Dysplasia-carcinoma cascade, 55

Elastic fiber stain, 163 Electrocautery, for carcinoma of low rectum, 332 Electrocoagulation clinical applications, 338 rectal polyps, 328 failure of, 338 monopolar, 346 technique, 338 Endocavitary radiation, 252 abdominoperineal resection, 252 disease-specific survival rate, 253 Endoluminal colorectal stenting, 126 Endorectal ultrasonography accuracy of, 211 with rectal carcinoma/villous adenoma, 211 Endoscopic ultrasonography clinical consequences, 213 staging of colonic neoplasms, 213 Endoscopy laser surgery for adenocarcinoma of colon, 98 of rectal adenocarcinoma, 209 Enema barium, double-contrast, 43

Ethiodized oil, 142 Excision for carcinoma of low rectum, per anal description of, 328 preparation, 328 for rectal adenoma, per anal circumferential villous or tubulovillous adenoma, 330 sessile adenoma, 330 Exercise, adenocarcinoma pathogenesis and, 78 Exonuclease 1 (EXO1), 56 Exophytic carcinomas, 160, 248 Exophytic lesions, 160 External beam pelvic radiation abdominoperineal resection, 288 anorectal carcinoma, 288 perineal wound complications, 288 External beam radiation (EBR) therapy, 319, 253 Extracolonic carcinomas, 63 Extrahepatic disease, 169

Familial adenomatous polyposis (FAP), 82 clinical manifestations of, 30 colonoscopic polypectomy for, 328 definition of, 16 diagnosis of, 17, 66 discovery of, 3 ectodermal abnormalities, 22 extracolonic findings, 19 gastric, 14 genetic counseling, 26 genetic model of, 4 genetic testing, 65 consequences of, 22 decisions regarding, 22 description of, 16 time points for, 17 illustration of, 3 incidence of, 16 interpretation results, 69–70 management approaches anastomosis, 11 anoretal mucosectomy, 23 with continent ileostomy, 20 with ileostomy, 23 mesodermal abnormalities, desmoid tumors, 21 molecular genetic studies, 4 mutations detection in, 65 natural history of, 16 neoplastic, 2 polyps distribution, 17 prevalence of, 3 rectal, electrocoagulation of, 327 screening technique, 65 size variations, 2 treatment of, 20 tubular, 2 tubulovillous, 17 variant manifestations in, 69 Familial colorectal carcinoma, 65 Febrile neutropenia, 120 Fecal occult blood test (FOBT), 42 Fiber diet and colon carcinoma, 73 fruit, 73 grain, 73 intake, 73 source, 73 vegetable, 73 Fistula, 132 anorectal, 322 Flat adenoma definition, 5 management of, 5 molecular analysis of, 5 mucosal elevations, 5

INDEX &

Flexible sigmoidoscopy, 74 advantages of, 42 for colorectal carcinoma screening, 43 complications of, 42 fecal occult blood test combined use, 42 comparisons between, 43 18-Fluorodeoxyglucose (18FDG), 168 Fluoropyrimidine, 119, 143 5-Fluorouracil (5-FU), for adenocarcinomas, 250 of colon, 118 leucovorin, 118 therapy, in Bowen’s disease, 310 Fluorouracil-based chemotherapy, 118, 173 Folate, 76 Folinic acid, 172 regimen, 174 Friut fiber, 73

Gall bladder, adenocarcinoma pathogenesis and, 78 Gardner’s syndrome, 17 diagnosis of, 17 gastric, 14 polyps distribution, 17 Gas embolism complication, 387 Gastric carcinoma, 63 Gastric polyps gastric biopsy, 19 prevalence of, 19 Gastrointestinal malakoplakia, 177 Gastrointestinal malignancy, 95 Gastrointestinal stromal tumor (GIST), 32 diagnostic immunohistochemical staining, 293 on neoplasms, 293 radical surgical resections, 293 rectum, 286 Gelport, 347 Genetic counseling, 66–67 dynamic communication process, 27 prevention of carcinoma, 27 Genetic information, 70 definition, 72 misuse of, 71 Genetic syndrome, characteristic determination, 66 Genetic testing, 65 advantages of, 23 algorithm, 65 applications of, 23 benefits, 66 for colorectal carcinoma, 65 DNA sequencing, 28 drawbacks in, 66 for FAP families, 66–67 genetic discrimination threat in, 70 for HNPCC, 66 indications, 66–67 interpretation of APC, 70 of FAP, 69 of HNPCC, 70 limitations of, 66 linkage analysis, 28 mutation detection by, 28, 66–67 mutation-positive test, 28 screening techniques, 28 techniques, 65 Genetics, in adenocarcinoma of colon clinical relevance of, 59 description of, 55 effect on prognosis, 55 mismatch repair genes, 56 molecular pathway, 57 oncogenes, 56 testing, 56 tumor suppressor genes, 56 Genitourinary symptoms, 132

Germline DNA, testing techniques, 65 Germline testing, 63 Grain fiber, 73 Granulocyte elastase, 342 Graspers, for laparoscopy, 347 Groin, metastasis, 320 Gum-chewing aids, 366 Guy rope suture technique, for end ileostomy, 56

Hand-assisted device, as wound protector, 353 Hand-assisted laparoscopic (HAL) colectomy, 346 sigmoid resection, 347 Hand-assisted laparoscopic surgery (HALS) colorectal procedures, 347 hand ports in, 347 Hand-assisted ports, 347 Hand-assisted techniques, 380 Handport, 347 Harmonic Scalpel1, 346, 386 Harmonic shears, 346 Hartmann’s procedure, 368, 376 Hassan-type cannula, 383 Health-quality domains, 382 Hemangiomas, of large bowel, 30 capillary, 30 classification of, 30 clinical manifestations of, 30–31 definition of, 30 diagnosis of, 31 treatment of, 31 Hematoxylin-eosin staining, 87 Hemicolectomy, 104 laparoscopic-assisted right, 384 suturing, 108 Hemoglobin, 94 Hemorrhage rectal, 287 Heparin prophylaxis, 387 Hepatectomy, 136 clinical data, 141 for colorectal liver metastases, 139 recurrence rates, 141 Hepatic arterial chemotherapy, 140 Hepatic artery infusion pump, 174 Hepatic metastases, 80 bilobar, 134 neoadjuvant chemotherapy for, 138 PET in detection of, 135 Hepatic resection, for metastatic colorectal carcinoma, 135 Hereditary flat adenoma syndrome (HFAS), 59 Hereditary nonpolyposis colorectal carcinoma (HNPCC), 9, 56 associated carcinoma, 61 clinical criteria, 61 colorectal carcinoma screening and, 41 diagnosis, 61 genetic patterns, 60 genetic testing, 66 algorithm, 69 impacts, 70 interpretation, 70 incidence, 61 Lynch syndrome, 63 malignancies in, 63 molecular genetic diagnosis of, 63 ovarian carcinoma in, 63 predisposition to, 61, 70 protein truncation in, 63 survival rates of, 64 Hernia, Richter’s, 386 High-energy ultrasonic vibrations, 346 Highly active antiretroviral therapy (HAART), 309 Histamine, blockers, 88 hMLH1 (chromosome 3p21), 56 hMSH2 (chromosome 2p16), 56 hMSH3 (chromosome 5q11.2-q13.2), 56

395

Homosexual men, 306 hPMS1 (chromosome 2q31-33), 56 hPMS2 (chromosome 7q11), 56 Human immunodeficiency virus (HIV)-negative homosexual men, 307 Human immunodeficiency virus (HIV)-positive men, 307 Human papilloma virus (HPV) anal carcinoma and, 306 Human papilloma virus type 16 (HPV-16) diagnosis and treatment, 308–309 infection, 307 virus-like-particle vaccine, 308 Hydrocolonic sonography, 102 Hydronephrosis, 133 Hydroxyl radicals generation, 80 Hyperplastic polyposis, 6, 72 colorectal carcinoma in, 16 randomized colorectal chemoprevention trials for, 16 recurrence, 16 surveillance, 16 Hyperthermic intraperitoneal chemotherapy, 171

IFL plus bevacizumab, 173 Ileal interposition, 132 Ileal pouch–anal anastomosis (IPAA), 11 Ileocolic vessels, 354, 363 Ileorectal anastomosis (IRA), 11 colectomy with, 21 colon resection, 21 periodic surveillance, 21 Imiquimod (Aldera) application of, 310 cream with fluorouracil, 310 formulation, 310 reactions of, 310 in treatment for high-grade AIN, 310 Immune response modifier, 310 Immunoscintigraphy, 102 Immunotherapy, 279 for adenocarcinoma, 123 of rectum, 279 Inflammatory bowel disease (IBD), 55 surveillance in, 45 Inflammatory polyps, 15 Inguinal lymph node, 320 Inguinal metastasis, 320 Injection therapy, for rectal carcinoma, 263 Insufflation needles, 346 Insufflators, 345 Interferon, 119, 123 Interleukin (IL), 342 Intestinal stenting, 126 Intra-abdominal carcinoma, 168 Intra-abdominal desmoids magnetic resonance imaging (MRI), 21 symptom in patients, 21 Intra-arterial hepatic chemotherapy, 174 Intracavitary radiotherapy, for rectal adenocarcinoma, 252 Intramucosal carcinoma, 9 Intraoperative radiotherapy, 273, 274 Intravenous pyelography (IVP) for adenocarcinoma, 99 Intromit, 347 Invasive carcinoma with pedunculated polyp Haggitt level, 3 sessile lesion, 3 transanal excision, 3 sessile polyp Haggitt classification, 8 histopathologic parameters, 8 malignant lesion, 8 risk factors, 8 transanal excision, 8

396

& INDEX

In vitro–synthesized protein assay (IVSP), 27 screening with, 69 Ipsilateral ureteroureterostomy, 132 Irinotecan, 172 Irinotecan and oxaliplatin (IROX), 172 Irinotecan, fluorouracil, and leucovorin (IFL), 123 Irinotecan, leucovorin, and fluorouracil regimen (FOLFIRI), 172 Iron deficiency anemia in colorectal carcinoma, 94 criteria, 94 Isolated melanotic mucocutaneous pigmentation (IMMP), 12 development of carcinoma in, 13 pigment identification, 13 Isosulfan blue (Lymphazurin) adverse reactions to, 88 in lymphatic mapping, 87

J-pouch adjuvant chemoradiation therapy, 234 advantages of, 232 anal anastomosis, 232, 233, 234 coloanal anastomosis, 232, 233 colonic, 232, 233, 234 description of, 232 rectal sensory testing, 234 ultra-low anterior resection, 234

Kaposi’s sarcoma in child and adult, 10 for colorectal neoplasm, 11 definition of, 11 epithelial dysplasia, 11 gastrointestinal (GI) tract, 10 genetic alterations of, 11 management and surveillance, 11 mechanism of polyp-cancer sequence in, 11 types of, 10 versus solitary juvenile polyp, 10 Kerato-acanthomas, 63

Laceration, in stoma patients for rectal adenocarcinoma, 252 for rectal adenoma, 335 Laparoscopic-assisted bowel resection, 342 Laparoscopic-assisted colectomies, 343 Laparoscopy abdominoperineal resection, 359, 379 advantages of, 342 colectomy, total abdominal, with ileorectal anastomosis, 362–363 in colorectal surgery, results of, 364 complications of, 383 anastomotic dehiscence, 385–386 bleeding, 385 electrosurgical injuries, 386 lesions, 370, 386 nonspecific, 387 wound healing, 386 conversion rates for, 364–365 diagnostic colectomy, 353 left-sided, 355 disadvantages of, 343 in elderly patients, 366 equipment and instrumentation for cannulas, 346 clip appliers, 347 graspers, 347 insufflators, 345 needles, 348 power instruments, 346 recording media, 345

[Laparoscopy] retractors, 347 staplers, 348 sutures, 348 trocars, 346 video camera, 344 gasless, 350 indications benign neoplasms, 344 carcinoma, 344 description of, 343 laparoscope, 344 minilaparotomy and, 369 morbidity and mortality, 368 operative procedure gasless methods, 350 operating room setup, 348 ports, 351 preoperative preparations, 348 operative uses of abdominal wall recurrence, 385 abdominoperineal resection, 359, 379 pneumoperitoneum, 349 stomas, 379–380 ureteral injury, 384 vascular injury, 384 visceral injury, 384 patient information, 343 postoperative care, 364 specimen localization, 351 right-sided, 353 trocar-cannula insertion, 350 Lapdisc, 347 Laser ablation, 263 Laser surgery, adenocarcinoma recurrence, 171 Lavage on-table, for obstruction of colon, 124, 125 Left hemicolectomy (LHC), 370 Leiomyomas of large bowel clinical manifestations of, 32 definition of, 32 origin of, 182 pathology of, 182 treatment of, 182 smooth muscle neoplasms, 32 Leiomyosarcoma, of colon, 182 Leucovorin, 118, 119, 123, 172 for adenocarcinomas, 121, 173 Leukemia infiltration, in colon, 185 Leukopenia, 122 Levamisole, 118 Ligasure, 346 Linitis plastica, 78 Lipomas, of large bowel clinical manifestations of, 181 description of, 180 diagnosis of, 181 treatment of, 182 Liver colorectal metastases, 100 intraoperative liver ultrasonography to detect, 100 prognostic factors, 137 treatment approaches, 135 metastasectomy from colorectal carcinoma, 143 resection, 143 risk factors, 143 treatment, 143 resection for metastatic colorectal carcinoma, 142 outcome, 142 reactions, 142 Locoregional radiation, 250 Low anterior resection, for rectal adenocarcinoma excised tissue amounts, 265 incision, 216 indications, 216

[Low anterior resection] patient position, 215 rectal perforations, 221 rectum marking, 265 rectum mobilization, 216 anterior, 216 posterior, 216 sigmoid colon mobilization, 216 Lugol’s solution, 309 Lung colorectal metastases to prognostic factors, 143 resectability criteria, 140 resections, for metastatic colorectal carcinoma, 143 Lymphatic basin, 90 Lymphatic mapping, feasibility and accuracy of, 87 Lymphatic spreading, 161 Lymphedema, chronic, 186 Lymph node(s) adenocarcinoma metastasis to, 78 metastasis, 320 clinical predictors of, 246 drug treatment for, 321 frequency determination of, 245 local excision, 245 risk factors for, 245 univariate analysis of, 246 sentinel, 87 strata of, 163 Lymphoid polyps, 15, 21 Lymphoma of colon incidence of, 180 pathology of, 180 prognosis for, 182 survival rates, 182 treatment of, 182 large bowel, 181 of rectum clinical presentation, 292 results of, 292 treatment of, 292 Lymphosarcoma, 292 Lynch syndromes, 64 molecular screening of, 64

Magnesium, and colorectal carcinoma, 75 Magnetic resonance imaging (MRI) for adenocarcinoma, of colon, 100 clinical applications, 213 endorectal ultrasonography, 212 lymph node invasion, 213 preoperative neoadjuvant therapy, 212 in rectal carcinoma, 212 soft tissue resolution, 212 Mainz pouch, 132 Malakoplakia associated diseases, 177 and colorectal carcinoma, 177 gastrointestinal, 177 lymph node involvement by, 177 Malignant large bowel obstruction, relieving methods balloon dilatation, 176 Nd:YAG laser, 176 Malignant melanoma, 323 Mammography, 15 Meat consumption, 72 and fish, 72 Medullary carcinoma, of colon, 185–186 Melanoblasts, 184 Melanoma, 57 of colon clinical features, 184

INDEX &

[Melanoma] clinical studies, 184 description of, 184 diagnosis of, 184 etiology, 323 examination, 184 incidence, 184 intestinal, 184 malignant, 323 metastasis, 323 symptoms, 184 treatment, 184 Memorial Sloan-Kettering Cancer Center, 322 Mental Component Summary Score (MCS), 382 Merkel-cell carcinoma of anal canal, 322 diagnosis, 322 Mesenteric vessel ligation, 354 Metachronous carcinoma, 46, 47 incidence, 134 Metastasis of adenocarcinoma of colon direct continuity, 88 hematogenous, 89 implantation, 89 to liver, 135 to lung, 142 lymphatic, 89 modes, 88 to ovary, 145 palliative resection, 133 to peritoneum, 161 sites of, 89 staging, 89 transperitoneal, 89 clinical features, 94 detection of, 100, 103, 46 examinations for abdominal, 95 digital rectal, 95 to liver, 137 sites for, 321 treatment for, 321 prognostic factors adjacent organ involvement, 156 appendectomy, 160 carcinoembryonic antigen levels, 169 lymph node metastasis, 86 obstruction, 156 pathologic features, 160 perforation, 156 perineural invasion, 164 perioperative blood transfusion, 160 resection technique, 157 sex, 155 systemic manifestations, 157 recurrent endoscopic laser therapy, 170 follow-up recommendations, 166 patterns of, 167 positron emission tomography, 101 survival rates, 168 of squamous cell carcinoma, to anal canal, 321 synchronous carcinomas, 134 Metoclopramide, 176 Microdissection genotyping, 294 Microsatellite instablility (MSI) analysis outcomes, 67 in neoplastic tissue, 70 phenotype, 67 testing, 59 Minilaparotomy laparoscopy and, comparisons of, 369 laparotomy incision, 346 Mini micrometastases, 163 Mismatch repair-deficient cells, 59 Mitomycin C, 122, 133, 143, 171, 174

Monopolar electrocoagulation, 346 Mucinous lesions, 155 Mucosa, of rectum, biopsy technique, 328 Mucosal carcinoma, 94 Muir-Torre syndrome, 61 Multivisceral resection outcome, 130 therapeutic benefit of, 131 Mural penetration, 162 Muscularis propria carcinomas, 94 Mydriasis, bilateral, 387

Neoplasms anal age of onset, 306 etiology of, 306 incidence of, 306 pathogenesis of, 306 brain, in patients with familial adenomatous polyposis, 21 intraepithelial, 307 perianal, 309 Neoplastic angiogenesis, 123 Neuroendocrine carcinoma of colon and rectum, 185 pathology, 185 Neuropathic pain, 174 Neurosensory toxicity, 172 Neutropenia, 172 Nigro chemoradiation regime, 322 Nodal involvement, 87, 161, 162 Node-positive rectal carcinoma, 163 Nonexophytic carcinomas, for local excision, 248 Nonexophytic lesions, 160 Non-FAP hereditary colorectal carcinoma, 59 Non-HNPCC hereditary colorectal carcinoma, 59 Non-Hodgkin’s lymphoma barium enema studies for, 181 Nonsignet cell mucin carcinomas, 161 Nonsignet ring cell mucinous carcinomas, 161 Nonsteroidal anti-inflammatory drugs (NSAID), 26 for adenocarcinoma prevention, 76 chemoprevention by, 76 for colorectal carcinoma, 76 nonaspirin, 81, 82 Nordic bolus schedule, 172

Obesity, 157 complications, 381 laparoscopic colorectal surgery in, 381 Obstruction in adenocarcinoma, of colon, 124, 125 of colon description of, 185 effect on survival rates, 149 Hartmann’s procedure, 124 primary resection, 126 subtotal colectomy, 124 on-table lavage, 125–126 three-stage, 124 Obstructive colitis complications, 129 diagnostic and therapeutic problems, 129 incidence, 129 pathogenic mechanism of, 129 signs and symptoms, 129 Occult blood test, of stool for colorectal carcinoma screening, of average-risk individuals, 43 clinical efficacy studies, 41 cost-effectiveness, 40 patient considerations, 40 rationale, 42 sensitivity, 42 flexible sigmoidoscopy and, 43

397

[Occult blood test] combined use, 42 comparisons between, 43 Octogenarians, 366 Omniport, 347 Oncogenes, in adenocarcinoma etiology, 56 On-table lavage, for obstruction of colon, 125–126 Oophorectomy, 116, 145 for adenocarcinoma, 117 Opioids, 176 Osteomas, 22, 69 in patients with familial adenomatous polyposis, 22 Ovarian carcinoma, 63, 177 colon involvement by, 177 Ovarian metastases, 84, 116, 117, 145 Ovary, colorectal metastases to, 145 Oxaliplatin, 143, 172 Oxaliplatin, leucovorin, and fluorouracil regimen (FOLFOX4), 172

Pagetoid, 313 Paget’s disease, perianal case study, 313 circumferential, 313 clinical features, 312 description of, 312 diagnosis of, 313 histogenesis of, 312 history of, 312 prognosis of, 313 treatment of, 313 Pain adenocarcinoma of colon, 94 of rectum, 94 Palliative resection, for adenocarcinoma of colon, 133 metastasis, 135 of rectum, 261 Palliative therapy electrocoagulation, 262 laser ablation, 263 palliative abdominoperineal resection, 262 radical surgery, 262 for rectal carcinoma endorectal ultrasound, 261 locoregional disease, 261 magnetic resonance imaging (MRI), 261 surgical treatment, 262 Pancreas, metastases to, 147 Pancreatoduodenectomy, 20 Papanicoloau staining, 307 Paucisymptomatic microperforation, 351 Pedigree analysis, 62, 66 Pelvic arterial infusion, 133 Pelvic en bloc surgery, 177 Pelvic exenteration, 132 Pelvic floor, reconstruction of, for abdominoperineal resection, 226 Pelvic resections, 131 Pemphigoid, 95 Per anal excision for carcinoma of low rectum, 328 for rectal adenoma circumferential villous or tubulovillous adenoma, 330–331 description of, 328 equipment, 329 sessile adenoma, 330 Perforation of colon, in adenocarcinoma of colon, 97, 128, 156 description of, 97 effect on survival rates, 139, 140, 149 prevalence of, 128

398

& INDEX

[Perforation] treatment approaches, 103 of rectum, 221 Perianal lesions, verrucous carcinoma, 315 Perianal neoplasms basal cell carcinoma clinical features, 315 prognosis of, 315 treatment of, 315 Bowen’s disease clinical features, 309 description of, 309 treatment of, 309 squamous cell carcinoma clinical features, 311 description of, 311 treatment of, 312 Perianal Paget’s disease and anorectal adenocarcinoma, 313 case study, 313 circumferential, 313 clinical features, 312 description of, 312 diagnosis of, 313 histogenesis of, 312 history of, 312 and perianal Bowen’s disease, 313 prognosis, 313 radiation therapy for, 313 treatment of, 313 Perianal region basal cell carcinoma of clinical features, 315 description of, 314 prognosis of, 315 treatment of, 315 Perianal skin, 305 Perianal wound pain, 313 Perineural invasion, 164 Perioperative mortality, 284, 285, 288 Peritoneal carcinomatosis drugs in, 146 prognostic indicators, 146 resection, 147 survival rate, 147 treatment, 147 Peritoneal seeding, 161 Peutz-Jeghers syndrome, 11 carcinoma screening for, 12–13 clinical criteria, 11 diagnosis of, 11 endoscopic resection of, 13 genetic testing, 12 hamartomatous polyps, 12 management of, 13 mucocutaneous hyperpigmentation, 12 phenotypic characteristics of, 12 screening protocols, 13 testicular/cervical malignancy, 12 p53 gene, in adrenocarcinoma pathogenesis, 58 Physical Component Summary Score (PCS), 382 Placebo, doses of, 308 Plasma C-reactive protein concentrations, 80 Plasmacytoma, of colon, 184 Pneumaturia, 132 Pneumoperitoneum, for laparoscopic procedures, 349 Pneumosleeve, 347 Polymerase chain reaction (PCR), 67 Polypectomy, colonoscopic snare for pedunculated polyp, 328 Polypoid adenomas, 5 Polyp prevention study, 75 Polyps classification of, 3 definition of, 3

[Polyps] flat, 5 gastric, 19–20 hamartomatous Cowden’s disease, 14 Cronkhite-Canada syndrome, 14 definition of, 10 juvenile polyposis, 10 inflammatory, 15 invasive carcinoma of, management of, 3 lymphoid, 15 neoplastic definition of, 2 prevalence of, 3 recurrence of, 5 size variations, 3 pedunculated colonoscopic polypectomy for, 3, 4 invasive carcinoma of, 7 per anal excision for of lower rectum, 329 of middle rectum, 330 removal of reasons for, 5 sessile invasive carcinoma of, 7 transanal endoscopic microsurgery, 337 small bowel, 21 untreated, natural history of, 16–17 Positron emission tomography (PET) for adenocarcinoma, 101 Post-menopausal hormone therapy, 79 Postoperative chemoradiation therapy, 275 death rate, 280 Pouch of Douglas, 294 Pregnancy adenocarcinoma presentation during, 177 incidence of, 176, 277 operative treatment, 177 prognosis of, 177 staging of, 177 Preoperative chemoradiation therapy, 277 clinical benefit, 277 death rate, 280 with mesorectal excision, 276 Preoperative radiotherapy, 267, 268, 271 morbidity and mortality, 280 perineal wound infection, 280 Procarcinogens, and bile pigments, 80 Proctocolectomy with continent ileostomy, 23 diagnosis of neoplasia, 26 for familial adenomatous polyposis (FAP), 26 with ileal pouch-anal anastomosis fecal incontinence, 26 patient satisfaction, 23 S-pouch construction, 26 with ileostomy description of, 128 patient management, 26 pouch adenomas, 27 endoscopic treatment, 26 Prophylactic colectomy, 26 Prostate, lung, colorectal, ovarian (PLCO) cancer screening trial, 74 Protein truncation test, 67, 68 Pulmonary metastases, 142 from colorectal carcinoma, 142 prognosis, 142 treatment, 142 Pulmonary resections for colorectal carcinoma, 143 predictors, 143 prognostic factors, 143 Pyelography, intravenous for adenocarcinoma, 99

Quality of life, 215

Radiation-associated rectal carcinoma, 77 Radical extirpative operations, for rectal adenocarcinoma abdominoperineal resection abdominal portion of, 226–227 excised tissue amounts, 290 high ligation of inferior mesenteric artery, 265 indications, 244 perineal portion of, 215, 226–227 preoperative adjuvant radiotherapy and, 266 recurrence, 212 survival rates after, 240 coloanal anastomosis description of, 214 survival rates, 230 double dynamic graciloplasty, 240 J pouch, 232 low anterior resection anastomosis with circular stapler, 221–222 dividing of lateral ligaments, 216 double-stapling technique, 232 excised tissue amounts, 214 incision, 216 indications, 214 patient position, 214 preoperative adjuvant radiotherapy and, 266 rectum mobilization, 216 sigmoid colon mobilization, 216 postoperative care, 236 resectability assessments, 214 selection criteria for, 214 survival rates, 237, 238 Radical lymphadenectomy, for rectal adenocarcinoma, 252 Radioimmunodetection, 101 Radionuclide studies, 101 for adenocarcinoma, 179 Radiotherapy for adenocarcinoma of colon, 133 recurrence, 102 adjuvant, 46 advantages of, 244 invasive distal rectal carcinoma, 251 local excision, 251 local recurrence rate, 251, 259 external beam, 263 intracavitary, 252 for lymphoma of rectum, 292 for squamous cell carcinoma of anal canal, 317 Rectal adenocarcinoma, 208 abdominoperineal resection abdominal portion of, 226 description of, 226 high ligation of inferior mesenteric artery, 265 perineal defects after, 288 preoperative adjuvant radiotherapy and, 266 sacral defects after, 288 survival rates after, 244 adjuvant therapy chemoradiotherapy, 317 radiotherapy, 317 clinical features, symptoms, 321 coloanal anastomosis description of, 214 survival rates, 230 complications of, 321 diagnostic investigations of barium enema, 209 blood tests, 247 chest radiography, 209 cystoscopy, 247 endoscopy, 247 intravenous pyelography, 247 magnetic resonance imaging, 212 ultrasonography, 212

INDEX &

[Rectal adenocarcinoma] local therapeutic methods concomitant pelvic organ excision, 252 cryotherapy, 290 electrocoagulation, 289 external beam radiotherapy, 319 intracavitary radiotherapy, 252–253 laser therapy, 315 metal stents, 264 palliative resection, 244 radical lymphadenectomy, 258 rationale for, 319 low anterior resection anastomosis with circular stapler, 221–222 double-stapling technique, 232 excised tissue amounts, 214 incision, 216 indications, 214 patient position, 214 preoperative adjuvant radiotherapy and, 266 rectum mobilization, 216 sigmoid colon mobilization, 216 preoperative preparations, 213–214 radical extirpative operations double dynamic graciloplasty, 236 J pouch, 232 Kraskes approach, 236 postoperative care, 236 resectability assessments, 214 selection criteria for, 214 survival rates, 240 recurrence of chemotherapy, 266 circumferential, effect of resection margins on, 215 classification, 292 clinical features, 283 hyperthermia, 289 incidence of, 265 investigations of, 282–283 pain control for, 282 patterns of, 283 radiotherapy, 266 total mesorectal excision to prevent, 256–257 spread of direct extension, 208 implantation, 208 lymphatic, 208 transperitoneal, 208 venous, 208 Rectal biopsy description of, 327 of rectal carcinoma, 328 of rectal mucosa, 328 Rectal carcinoma, 53, 73, 76, 79, 132 central ulceration, 209 coloanal anastomosis, 230 depth and invasion of, 209 diagnosis of, 240 electrocoagulation for, 337 hard protuberant mass, 209 imaging technology, 209 injection therapy of, 263 local excision for, 244 local recurrence rate and survival for, 158, 159 meta-analysis for, 247 morphology of, 247 node-positive, 163 oncologic resection group, 247 optimizing surgical techniques, 233 palliative therapy for, 261 pelvic autonomic nerve preservation, 258 pelvic surgery for, 244 postoperative morbidity, 215 with preoperative chemoradiotherapy, 279 radical surgery for, 262 radiotherapy in, 132 registry, 242

[Rectal carcinoma] sexual dysfunction, 244 sphincter-sparing resection for, 230 surgical treatment for, 278 total mesorectal excision (TME), 247 transanal excision, 249 ultra-low anterior resection, 226 ultrasonography for, 247 Rectal hemorrhage, 288 Rectal mesentery abdominoperineal resection, 242 coloanal anastomosis, 242 Rectal mobilization anterior, 220 massive presacral bleeding, 220 muscle welding method, 220 advantages of, 220 control of presacral bleeding, 220 pelvis autonomic nerves, 221 presacral venous plexus, 220 rectovaginal reflection, 221 vertebral venous system, 220 Rectal vascularization, 255 Rectosigmoid resection, for hemangiomas, 177 Rectourinary fistulation, 132 Rectum adenocarcinoma of, 321 levels of, 214 Recurrence adenocarcinoma, of colon carcinoembryonic antigen for detection of, 169 clinical features, 168 colonoscopic evaluations, 99 contributing factors, 166 follow-up recommendations, 166 incidence of, 120, 160, 166 with malignant ureteral obstruction, 171 patterns of, 138 positron emission tomography and, 101 risk factors, 157 treatment of, 285–286 adenocarcinoma, of rectum effect of resection margins on, 215 nonoperative treatment, 282 operative treatment, 283–285 pain control for, 282 patterns of, 283–284 total mesorectal excision to prevent, 255 detection of, 46 Renal function, abnormal, 133 Replication error pathway, 60 Resection procedures abdominoperineal for hemangiomas, 25 for sarcoma of rectum, 292 for squamous cell carcinoma of anal canal, 305 curative bowel preparation, 104 follow-up after, 46 principles of, 104 right hemicolectomy technique, 105 of transverse colon, 112 low anterior dividing of lateral ligaments, 232 double-stapling technique, 232 incision, 216 indications, 216 patient position, 216 rectum mobilization, 216 sigmoid colon mobilization, 216 for rectal adenocarcinoma abdominal portion of, 226 description of, 226 excised tissue amounts, 266 indications, 290 mesenteric artery, 265

399

[Resection procedures] perineal defects after, 288–289 perineal portion of, 226 preoperative adjuvant radiotherapy and, 266 sacral defects after, 288–289 Retractors, for laparoscopy, 347 Retropubic prostatectomy, 131 Reversed transcription (RT)-PCR, 67 Richter’s hernia, 386 Right hemicolectomy (RHC), 343, 368, 370 Riley–Smith syndrome, 15 Robot-assisted laparoscopic operations, 363 Robotic-assisted suture rectopexy, 363 Robotics advantages and disadvantages, 363 techniques of, 363 Ruvalcaba-Myhre-Smith syndrome, 15

Sacral nerves, effect on anal continence, 45 Sacral resection, 283 Salpingo-oophorectomy, for adenocarcinoma, 116 Sarcoma of colon, 182 of rectum, 286 Schwannomas of colon and rectum, 186 location, 186 occurrence, 186 spindle-cell, 186 symptoms, 186 Scintigraphy clinical uses carcinoids of colon, 179 Scirrhous carcinoma, pathologic diagnosis of, 161 Screening of anal carcinoma, 307 of colorectal carcinoma average-risk individuals, 42–43 increased risk individuals, 43 of polyps, 42 Selenium broccoli, effect in carcinomas, 75 Sensory neuropathy, 172 Sentinel lymph node technique, 320 Septicemia, 97, 129 Serosal carcinomas, 94 Serrated adenoma clinical importance of, 9 epithelial polyps, 9 genetic alterations, 9 hyperplastic polyps, 9 neoplastic polyps, 10 Sexual dysfunction pelvic surgery, 244 for rectal carcinoma, 244 Sialyl Lewis antigen expression, 166 Sigmoid colon adenocarcinoma of, curative resection for, 132 Sigmoidoscopy clinical uses, 98 adenocarcinoma, 98 flexible, 98 adenocarcinoma, 98 advantages of, 42 characteristics of, 45 for colorectal carcinoma screening, 42 complications of, 42 fecal occult blood test and, 43 polyp excision, 42 proctosigmoidoscopy, 114 proximal neoplasia, 42 randomized controlled trial, 42 Sigmoid resection, 343 Signet-ring cell adenocardnomas, 161 Single-stranded conformation analysis (SSCA), 67, 68, 69 Slaughter-Snowe bill, 71 Small cell carcinoma, 322

400

& INDEX

Smoking, adenocarcinoma pathogenesis and, 75 Sonosurg1, 386 Sphincter preservation of, 159 strength assessments, 314 Spindle-cell schwannoma, 186 Splenic flexure, ligament, 113 Splenic injury, 384 Sporadic carcinomas, 9 Sporadic colorectal carcinoma, 165 S pouch, 26 Squamous cell carcinoma of anal canal, 316 characteristics of, 316 clinical manifestations, 316 description of, 316 diagnosis of, 316 metastases, 316, 317 workup for, 316 of colon, 183 control trials, 318 disease-specific survival (DSS) in, 319 high dose-rate (HDR) boost of, 319 locoregional control (LRC) of, 319 patient malignancies in, 319 of perianal region clinical features, 311 description of, 311 treatment of, 312 treatment, 319 abdominoperineal resection, 317 chemotherapy, 317–319 local excision, 317 radiation therapy, 317 Staging, anal carcinoma, 307 Staplers, for laparoscopy, 348 Stenting in colorectal obstruction treatment, 126 endoluminal colorectal, 126 intestinal, 126 Stent insertion, 127 Stent migration, 127 Stent placement, 127 Stent-related mortality, 128 Stent reobstruction, 127 Stoma laparoscopy use for stoma closure, 361 stoma creation, 360 Stool, dysfunctions, 227 Streptococcus bovis, 97 Submucosal carcinomas, 94 Submucosal invasion, 86

Subtotal colectomy, for obstruction of colon, 124 Supraclavicular lymph nodes, 209 Surgical anal canal, 305 Surveillance after adenomatous polyp removal, 45 colonoscopy for, 45 definition of, 45 for inflammatory bowel disease, 45 for ureterosigmoidostomy, 46 Surveillance, epidemiology, and end results (SEER) cancer registry, 317 registry data, 77 Sutures, for laparoscopy, 348 Swiss association for laparoscopic and Thoracoscopic surgery (SALTS), 383 Symptomatic desmoids, 22 Symptoms distress scale (SDS), 382 Synaptophysin reactivity, 186 Synchronous neoplasms, 46 colorectal, characteristics, 125 Synchronous polyps, and synchronous carcinoma, 134

Tactile sensation loss, 343 Telerobotic-assisted laparoscopic colectomies, 363 Total mesorectal excision (TME), 357, 377 abdominoperineal resection, 226 adjuvant therapy, 266 for biopsy-proven rectal carcinoma, 213 classification, 232 definition, 256 Denonvilliers fascia, 256 intersphincteric resection, 230 lateral/circumferential margins, 254 local recurrence rate, 232 optimizing surgical techniques, 233 pelvic autonomic nerve preservation, 258 preoperative radiotherapy, 242 symptomatic anastomotic leakage, 258 Transanal endoscopic microsurgery (TEM) local recurrence rate for, 337 randomized controlled trial, 337 for sessile adenoma, 335–337 survival rate of complications, 337 versus direct local excision, 337 Transverse colon, adenocarcinoma of, curative resection for, 112 Trocars complications, 383

[Trocars] injuries, 383 for laparoscopic stoma creation, 360 risk factors, 384 Truncal vagotomy, 80 Tumor budding, 161 Tumor suppressor genes, in adenocarcinoma etiology, 56 Turcot’s syndrome, 61, 63

Ulcerative colitis, 55, 98 Ulceroinflammatory condition, 128 Ultrasonic energy devices Autosonix1, 386 Harmonic Scalpel1, 386 Ultrasonography clinical uses adenocarcinoma, 99 rectal adenocarcinoma, 244 endoscopic, 211 Ureter injury, from laparoscopy, 354 Ureterosigmoidostomy adenocarcinoma pathogenesis and, 77 surveillance after, 45

Vagina, metastases to, 147 Venous invasion, 163 Ventral hernia, incidence of, 342 Veress needle, 349, 383 Verocay bodies, 186 Verrucous carcinoma treatment of, 315 Vessel-sealing device, 346 Vincristine, 321 Virtual colonoscopy, 44 Vitamin C intakes, 74, 77 Voice-controlled robotic camera holder, 363

Waterjet dissection, 346 White line of Toldt, 354, 355 Whole-brain radiotherapy, 145 Wound healing of, after laparoscopy, 386 infection, 386 Wound-edge protectors, 386

York-Mason’s transsphincteric approach, for rectal adenoma, 335