Palliative Gastroenterology, An Issue of Gastroenterology Clinics Vol 35 Issue 1

Gastroenterol Clin N Am 35 (2006) xiii–xiv

GASTROENTEROLOGY CLINICS OF NORTH AMERICA PREFACE

Palliative Gastroenterology

Todd H. Baron, MD, FACP, Geoffrey P. Dunn, MD, FACS Guest Editors

G

astroenterologists can expect to encounter a substantial number of patients who have advanced and terminal illnesses. Most of the literature on palliative care medicine is not directed specifically toward gastroenterology. This issue of Gastroenterology Clinics of North America acknowledges cognitive and psychologic barriers that may hamper the practitioner in addressing the needs of this patient population. Although it is not an exhaustive summary of current approaches to palliative care, this issue is a point of reference for practicing gastroenterologists seeking guidance for the management of salient problems encountered in palliative care, the goal of which is to relieve suffering and improve quality of life. Palliative care may be the sole aim of care or offered simultaneously with all other medical treatment. The first article introduces the historic and philosophical background of palliative care, addresses palliative care’s current status in the field of medicine, and concludes by affirming it as a durable philosophy of care that is applicable across a wide spectrum of illnesses encountered in the practice of gastroenterology. Ensuing articles cover basic palliative care skills, such as palliative care assessment, palliative endoscopy, and chronic pain management. The issue concludes with an article on problems encountered in individuals with nontransplantable liver disease. Although much of the material addresses oncologic illness, the principles and many of the interventions covered are applicable to a much wider spectrum of illnesses, such as liver and intestinal failure. Although the field of palliative care has been developed previously outside the field of gastroenterology, gastroenterologists face many of the same 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2006.01.004

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

xiv

PREFACE

problems and some problems that are unique to the specialty. There is little doubt that the future will witness a more clearly defined consensus about palliative care by gastroenterologists. It is hoped that this issue will stimulate interest by gastroenterologists in accomplishing this. Palliative care challenges some of our most basic assumptions about the meaning of illness and leads us to ask new questions and discover new problems. As practitioners, we are creating a tradition of service in the relief of suffering that is equally welcome, with our recent accomplishments in eliminating and preventing disease. For a start, we must at least acknowledge the worthiness of a philosophy of care that represents a shift from the elimination of disease to the unconditional relief of suffering and the affirmation of our patients’ lives. Todd H. Baron, MD, FACP Division of Gastroenterology & Hepatology Mayo Clinic College of Medicine Rochester, MN 55905, USA E-mail address: [email protected] Geoffrey P. Dunn, MD, FACS Department of Surgery and Palliative Care Consultation Service Hamot Medical Center 2050 South Shore Drive Erie, PA 16505, USA E-mail address: [email protected]

Gastroenterol Clin N Am 35 (2006) 1–21

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Palliative Care: A Promising Philosophical Framework for Gastroenterology Geoffrey P. Dunn, MD Department of Surgery and Palliative Care Consultation Service, Hamot Medical Center, 2050 South Shore Drive, Erie, PA 16505, USA

T

he decade of the 1990s witnessed the introduction of comprehensive and systematic palliative care in hospital and outpatient settings in North America. Palliative care is defined as interdisciplinary care that aims to relieve suffering and improve the quality of life (QOL) for patients who have advanced illness and their families. It can be offered as the sole aim of care or simultaneously with all other medical treatment. The recognition for the need to this approach reflects a fundamental shift in priorities within the culture of medicine. It was presaged by the widespread and growing popularity of home-based hospice care introduced in the mid 1970s. Public awareness of end-of-life issues was further enhanced by judicial decisions that ruled on an individual’s right to deny medical treatment [1], the debate on physician-assisted suicide, and adverse experiences with end-of-life care [2] that contrasted so sharply with progress in other venues of medical care. Despite this reversal of what seemed to be a conspiracy of silence about death and dying in a highly materialistic culture, ambivalence remains about the role of palliative care in medical treatment. The length of stay in Medicare-certified hospices averages approximately 20 days of a defined benefit period of 6 months, despite increasing acceptance by physicians and consistently favorable experiences of patients and families [3]; and the notion of death as a natural and potentially positive experience in life’s trajectory is only hinted at in a recent policy statement by the National Cancer Institute (NCI), whose stated goal is ‘‘to eliminate suffering and death due to cancer by 2015’’ [4]. Despite this, evidence is accumulating that what patients and their families want is consistent with the primary aims of palliative care. In a study [5] of 475 families members within 2 years of a loss, concerns that mattered included honoring the deceased’s wishes, assistance and support in the home, honesty and directness in communication, continuous access to services, privacy, practical and

E-mail address: [email protected] 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.012

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

2

DUNN

personal care assistance, and to be remembered through ongoing contact after the death. Despite the proliferation of palliative and hospice care programs, patients still give eloquent testimony [6] about their fear of being abandoned by their treating physicians when their disease has progressed to the point at which no further treatment can be given. During this interval of 20 years, much has happened to establish palliative care as a health philosophy that will enrich the practice of gastroenterology, and conversely, developments in gastroenterology have already improved palliative care. Several recent concepts from the field of gastroenterology such as ‘‘the second brain’’ [7] and ‘‘intestinal failure’’ [8] fit well within the conceptual framework of palliative care. This type of synergy will ultimately encourage the application of this philosophy to a much broader spectrum of patients than those who are deemed to be at ‘‘end of life.’’ This broader spectrum of care can be described as palliative gastroenterology. HISTORIC BACKGROUND The term ‘‘palliative care,’’ introduced by Canadian surgeon Balfour Mount, comes from the Latin pallium, meaning ‘‘cloak’’ or ‘‘cover,’’ in the sense of using one to restore dignity and comfort in a situation in which the underlying disease process cannot be altered. Balfour used this description of the hospice concept when he introduced it to the Royal Victoria Hospital, an acute care multispecialty hospital in Montreal. Mount can be credited as the salient pioneer of hospice care in North America as well as being one of the first to bring hospice services into an acute care hospital. The origins of palliative care as now understood in many countries around the world are seen most clearly in the long and productive career of the late Dr. Cicely Saunders [9]. She is acknowledged generally as the founder of the modern hospice movement [9] and mentor to many of the pioneers of palliative care, such as Mount. Dr. Saunders was blessed with robust health, intelligence, a good education, and fate—all for which she more than paid thanks. Like so many notable contributors to the welfare of society, Saunders’ career path began inauspiciously. She was born to a well to do family and studied politics, philosophy, and economics at Oxford University until World War II, when she felt called to care for the ill and wounded as a nurse. She competed training as a nurse, and after the war, she encountered her index patient, Tasma, a Polish refugee who suffered from advanced malignancy. Her encounter with him triggered her life-long work, which ultimately developed a framework of care that now daily restores dignity and comfort to thousands of dying patients. The clinical foundation of her work was her demonstration that terminally ill patients suffering from cancer pain enjoyed improved quality of life with scheduled dosing of morphine. This foundation was based on careful observations and meticulous record keeping during the late 1950s. Now, a half century later, this innovation is still percolating slowly through the medical system, despite the documented efficacy and safety of the practice. It is the patients, families, and health care workers who harbor exaggerated or misplaced fears of

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

3

addiction and respiratory depression and a reluctance to acknowledge progressing illness that account partially for this hesitant acceptance. Saunders’ other seminal contributions to palliative care include her model of ‘‘total pain’’ (discussed below) and the opening of St. Christopher’s Hospice in 1967, which has served as an institutional model of patient care, teaching, and research. St. Christopher’s Hospice was the institutional expression of Saunders’ strong belief that the science of hospice and palliative medicine is matched by the importance of psychologic and spiritual care. Her efforts led to the early assimilation of hospice care into the health care systems of Great Britain and Ireland [10]. By 1987, palliative medicine had become recognized as a medical specialty by the Royal Academy of Medicine. To paraphrase her fellow countryman, Winston Churchill, it would be hard to find another individual in the field of medicine whose personal contributions have had so wide and profound an impact on the care of so many. By the end of the 1970s, hospice was rapidly establishing itself in the United States mostly as an outpatient-based service that offered an alternative to what many believed was an increasingly burdensome, overly medicalized, and institutionalized way of dying. In 1982, the Tax Equity and Fiscal Responsibility Act established the Medicare Hospice Benefit (1983) less than 10 years after the establishment of the first hospice program in the United States. This legislative achievement was a response to popular pressure not an initiative from any medical institution. The grass roots initiative and the lay appeal of hospice in its earlier days may have led to its isolation from mainstream medicine during the 1980s, a period in which there was scant reference to the concept in medical textbooks or in medical school and residency training. Only in the late 1990s, following the study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT) in 1995 [2], the American Board of Internal Medicine’s publication Caring for the Dying in 1996 [11], and the inauguration of the American Medical Association’s Education of Physicians on End-of-Life Care (EPEC) curriculum in 1999 [12], did the concept of hospice and palliative care assume an established role across the continuum of medicine in the United States. In 1996 the American Board of Internal Medicine published its education resource document Caring for the Dying, specifically for the use of program directors and faculty of residency and subspecialty training. In 1997, the Institute of Medicine issued its report Approaching Death: Improving Care at the End of Life [13]. This report concluded that suffering from serious pain and other burdensome symptoms in many dying patients has persisted despite the availability of effective treatments. A much-needed remedy for this state of affairs followed in 1999 with the inauguration of the American Medical Association’s sponsorship of the Education for Physicians on End of Life Care (EPEC) project funded by the Robert Wood Johnson Foundation. Since its initiation in 1999, EPEC has been used to train approximately 1000 physicians, including surgeons, to become instructors of other physicians and allied health care workers on the principles and practices of compassionate and competent palliative and end-of-life care. During the late 1990s,

4

DUNN

philanthropic foundations funded a variety of initiatives that brought the concept of palliative care to the care of selected disease populations (eg, end-stage renal disease [14]) and specializations (surgery [15]). The concepts of palliative care are beginning to be taught in medical schools [16] and postgraduate curricula [17]. The number of articles specifically addressing palliative care in the general medical literature has increased markedly since the time of the SUPPORT in 1995. Kathleen Foley [18], a neurologist and preeminent exponent of pain management and palliative care, observed, ‘‘It is innately human to comfort and provide care to those suffering from cancer, particularly those close to death. Yet what seems self-evident at an individual, personal level has, by and large, not guided policy at the level of institutions in this country’’ [18]. Despite this gloomy assessment, there are indications of change at the institutional level, one of which is reflected in a consensus report co-authored by Dr. Foley. In 1999, a consensus on principles for guiding care at the end of life was endorsed by 17 medical organizations and specializations [19]. As of 2002, the American Board of Surgery identified palliative care as a domain of expected competence for candidates seeking certification. The American College of Surgeons has issued two statements of principles on care at the end of life and palliative care [20,21], in addition to forming a task force for the palliative care education of surgeons [16]. BASIC PRINCIPLES AND MODELS OF SUFFERING Much of what is emerging in the practice and philosophy of palliative care has been strongly influenced by several models of suffering, which is the immediate target of palliative treatment. Complicating this orientation is the belief in some cultures of the positive and redemptive value of suffering. Although the willingness and ability to relieve suffering comprise the initial orientation of palliative caring, equally important affirmative qualities have been identified, such as the restoration and development of an individual’s social, psychologic, and spiritual function. The recognition and relief of suffering are the primary aims of palliative care. Several models of suffering have influenced the philosophy and practice of palliative care. The foremost model was that of Victor Frankl [22], a concentration camp survivor who was a psychiatrist at the time of his captivity. His experience in captivity led ultimately to the development of logotherapy, a form of existential analysis. Frankl’s account of his experiences [22] bears witness that survival is related to the capacity to make sense of suffering, to give suffering a sense of meaning. This is the existential crisis that can be observed in one form or another in all patients encountered in palliative care. Frankl’s belief describes a basic assumption of palliative care: that there must be a purpose to suffering and dying if one believes there is any purpose to existence itself and that choosing one’s attitude in these situations is one freedom that cannot be taken away under any circumstances.

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

5

Frankl’s experience addressed the relationship between suffering and existential issues. Since then, several models of suffering proposed by physicians have made palliative care a more clinically specific construct. In the early 1980s, anticipating the subsequent decades of social and legal debate about autonomy and relief of suffering at end-of-life, Cassell pointed out in a landmark article that physicians tend to deconstruct suffering to its physical dimension while ignoring the psychologic and spiritual pain that accompanies dying. This approach not only fails to relieve suffering but actually compounds it [23]. Cassell [24] defined suffering as arising from the threat to the integrity (wholeness) of the person. He pointed out that ‘‘bodies do not suffer, only persons do.’’ Persons are unique and experience disease differently. A person suffers when the intactness of his or her persona (integrity) is threatened. The elements of an individual’s persona include the individual’s past, present, and future, social role, private life, and a transcendent (spiritual) dimension. It is the presence of the person’s transcendent dimension that permits the separation of death from failure. In Cassell’s framework, suffering is not relieved until the threat to personhood has passed or is diminished. The degree of suffering in Cassell’s and others’ model correlates with the degree of existential threat. According to these models, the intactness of personhood, not the body, is the barometer of health. This conceptual framework resonates with the current popular interest in ‘‘holistic’’ health care and ‘‘treating the whole person.’’ Although the public seems genuinely interested in the holistic model of health care and supports a multimillion-dollar market for ‘‘alternative’’ medicine, the public also is an avid (and generally contented) consumer of health care organized according to the biophysical model. MacDonald [25] points out that the public never fully accepted the biophysical model of cancer because to most nonphysicians of all cultures cancer has always been regarded as an illness with social, psychologic, and spiritual connotations. The biophysical model addresses physical threats to well being or relies heavily on derivatives of it, such as pharmacology to address nonphysical problems. Shifting the goal of care from controlling disease to ameliorating suffering will inevitably present our society with the same task of introspection that has been asked of physicians and surgeons, to allow them to see ‘‘the whole person.’’ Dame Cicely Saunders’ model of ‘‘total pain’’ [26] outlines four cardinal dimensions of pain (physical, socioeconomic, psychologic, and spiritual) that contribute to suffering. Her theory brings a sharper clinical focus to Cassell’s model of suffering by dissecting out specific areas in which threats to personhood may lurk. It was more than coincidence that Saunders had previous professional experience as a nurse, social worker, and physician prior to her description of total pain. The models of Saunders and Cassell could be seen respectively as the anatomy and physiology of palliative care. Brody [27] has described suffering as the anguish felt by the individual as his life, a unique story, is fragmented, preventing its completion and transformation into a transcendent legacy. As the end of life draws nearer, the greater is the potential

DUNN

6

anguish. The implication for the physician is that patients who have progressive, life-limiting illness may be more likely to seek guidance in fixing their ‘‘broken stories’’ than fixing their broken bodies. A more recent and positive conceptualization of suffering and end of life is Byock’s [28] view that dying is a natural phase of life consisting of a series of developmental tasks (see Appendix), with abundant opportunity for rewarding personal growth reminiscent of Erik H. Erikson’s [29] theory of development in which the individual grows as a result of resolving a series of conflicts normally encountered in life. Byock’s four developmental tasks of the dying [28] include:    

Renew a sense of personhood and meaning Bring closure to personal and community relationships Bring closure to worldly affairs Accept the finality of life and the transcendent

Byock’s framework was one of a number from this era that illustrates what previously had seemed so counterintuitive; dying could be conceived as the last phase of a succession of periods in life, each with its own characteristics of health [28]. The wide spectrum of these developmental tasks suggests that the applicability of this model reaches much farther back in the trajectory of an individual’s life than its very last phases. This is particularly important for those living with chronic illnesses punctuated by life-threatening exacerbations, which, if survived, can initiate an irreversible and prolonged decline. PALLIATIVE CARE: OLD HUMANITIES AND NEW SCIENCE All of the models of suffering suggest that a purely biophysical approach to its relief would be a much too limited construct, frustrating to the patient as well as the clinician, unless the patient’s difficulty is an uncomplicated, reversible, physical problem such as a hemorrhoid or nonpathologic fracture. Is it a coincidence that the palliative model of care has arisen at the end of a century during which insights in theoretical physics of the first two decades subsequently found their expression in the natural sciences and then the social and applied sciences? As an analogy, Newtonian physics provided a workable scientific framework for almost all observable, everyday phenomena sufficient to land a man on the moon. However, the Newtonian framework was inadequate to explain phenomena on a subatomic and galactic scale. The extremes of birthing and dying and the extremes of the deeply personal and the broadly social that are encountered in the world of clinical medicine also demand a different guidance than the reductive, compartmentalized, and mechanistic approach that may, nonetheless, solve many problems in daily practice. In addition to the social, psychologic, and spiritual arguments favoring palliative care, there are economic reasons to consider substituting a physical model (disease-directed therapy) with a palliative model of care earlier than in the last 2 months of life. In an increasingly aging population, 28% of all Medicare costs are incurred in the last 2 years of life, of which half is spent

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

7

during the last 2 months of life [30]. The SUPPORT study [2] yields abundant data to show that seriously ill hospitalized patients in the last 6 months of life, in addition to never getting better, were frequently undertreated for pain and their preferences for care were often unknown or even ignored. The average length of stay in hospice during this time was 48 days, during which a third died within 7 days. The difference between the total Medicare expenditures during the last 2 years of life, the increasingly shorter stay in hospice, and the incidence of failed symptom control identified in SUPPORT paints a picture that cries out for a model somewhere between traditional hospice care and curative care. The association of hospice with palliative care has stigmatized palliative care for many health care professionals and laypersons who believe that palliative care is simply a semantic twist on ‘‘hospice,’’ which they may already have oversimplified as only ‘‘preparation for death.’’ In some cases, advocates for hospice care also have indulged in this oversimplification. Hospice care can be classified as a subset of palliative care for patients in the last phases of illness. The physical model of medicine practiced in our death-denying culture has so equated physical death with failure that any acceptance of death as an appropriate outcome of care will be exceedingly difficult. Kass [31] summarizes the limitations of the biophysical model with the following counsel: ‘‘If medicine takes aim at death prevention, rather than at health and relief of suffering, if it regards every death as premature, as a failure of today’s medicine— but avoidable by tomorrow’s—then it is tacitly asserting that its true goal is bodily immortality.. . .Physicians should try to keep their eyes on the main business, restoring and correcting what can be corrected and restored, always acknowledging that death will and must come, that health is a mortal good, and that as embodied beings we are fragile beings that must stop sooner or later, medicine or no medicine’’[31]. It will be some time before leading popular magazines are ready to publish a list of the ‘‘100 Best Hospitals to Die In,’’ despite the fact that the majority of individuals dying natural deaths do so in the hospital setting! A review of the historical evolution of hospice to palliative care demonstrates that both of these philosophies, instead of narrowly focusing on the loss of life, have evolved as forms of social, psychologic, and spiritual health rehabilitation and growth that can occur in the face of ongoing physical decline. Doyle and colleagues [32] summarize this by asserting that ‘‘palliative medicine is concerned with three things: the quality of life, the value of life, and the meaning of life.’’ Seen in this light, it is existence, not death, that is the focus of palliative care. DELIVERY MODELS OF PALLIATIVE CARE There are currently two delivery models of palliative care in the United States medical system: palliative care consultation and hospice services with or without the Medicare Hospice Benefit. Palliative care consultation in hospitals and outpatient settings has became increasingly available since the late 1990s

8

DUNN

because of increasing interest in improved end-of-life care and the favorable impact on patient care made by hospice programs in this country since its introduction in the late 1970s. Palliative care and hospice care share the following features: both provide active total care of patients whose disease is not responsive to curative therapy. Their treatment goal is the best possible quality of life for the patient and family through the relief of pain and other distressing symptoms, the provision of a support system to help patients live as actively as possible, and the provision of support to the patient’s family during the illness and into bereavement. Palliative care can be given concurrently with all other appropriate medical care. A hospice program provides palliative care for terminally ill patients and their families. All hospice programs are palliative care programs, but palliative care is not limited to terminally ill patients. A recent survey [33] has shown that in-hospital palliative care programs in the United States have proliferated rapidly since 2000, numbering approximately 1,027 (25% of hospitals) in 2003. These programs were defined by the American Hospital Association in its annual survey as the provision of specialized medical care, drugs, or therapies for the management of acute or chronic pain and the control of symptoms, which are administered by specially trained physicians and other clinicians, and support services, such as counseling on advanced directives, spiritual care, and social services, to patients who have advanced disease and their families. Palliative care programs meeting this definition were more likely to be offered in larger hospitals, teaching and veterans hospitals, and hospitals designated as American College of Surgeons’ Commission on Cancer-approved cancer hospitals. Hospitals owning their own hospice programs and Catholic-affiliated hospitals were also more likely to report having a palliative care program. In some cases ‘‘palliative care’’ describes programs that specialize primarily in pain management or ethics consultation but do not necessarily offer the full continuum of specialty services (eg, bereavement counseling and pastoral care) offered in Medicare-certified hospice programs. Multiple reviews [34–38] have documented improved pain and nonpain symptom management, improved patient and family satisfaction, reduced critical care unit and hospital stay, decreased in-hospital deaths, and significant cost reductions. Despite this record, many programs have relied heavily on philanthropic donations, grants, cost sharing, and genuine volunteerism by medically trained and laypersons. The criteria for palliative care referral can be generated automatically (ie, any patient staying in an intensive care setting over a defined period of time and any patient with a Karnofsky status of less than 50) or by clinical judgement for any given patient (indications for palliative care consultation are listed in Box 1). Some hospitals have designated units for palliative care, whereas others use a consultative model with or without a designated unit. A high rate of implementation of recommendations from palliative care teams has been noted [39].

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

9

Box 1: Indications for palliative care consultation 

Patient has an illness typified by progressive deterioration and worsening symptoms, often ending fatally.



Patient has limiting or threatening conditions with declining functional status or mental or cognitive function.



Suboptimal control of pain or other distressing symptoms.



Patient and family would benefit from clarification of goals and plan of care or resolution of ethical dilemmas.



Patient or surrogate declines further invasive or curative procedures, preferring comfort-oriented symptom management only.



Patients on medical, surgical, or critical care units who are expected to die imminently or shortly following hospital discharge.



Bereavement support of hospital workers, particularly after the death of a colleague.

Courtesy of Robert A. Milch, MD, Buffalo, NY.

For any individual whose treatment is palliative and no longer life-prolonging, hospice referral from the hospital or outpatient setting should be considered for any individual with a physician-certified prognosis of less than 6 months if the disease pursues its usual course. Any terminal diagnosis is appropriate (Box 2 lists primary hospice services). A common mistaken belief is that a do-not-resuscitate (DNR) order is required for enrollment into a hospice

Box 2: Primary hospice services 

Physical symptom control, such as pain, nausea, dyspnea, and other symptoms



HHA services, such as bathing, dressing, and feeding



Psychosocial counseling for patient and family



Spiritual support for patient and family



Completion of advance directives, wills, and funeral planning



Volunteers



Bereavement services



All drugs related to terminal illness



All durable medical equipment



Therapies such as occupational, physical, nutritional, musica, and massagea



Other services as approved in plan of care: radiation, surgery, chemotherapya, TPNa, transfusiona, and intravenous hydrationa

a

Varies per program. Courtesy of Robert A. Milch, MD, Buffalo, NY.

DUNN

10

program. A DNR requirement is forbidden by Medicare-certified programs because of concerns about coercion and conflict of interest. The presence of a DNR order has been used mistakenly as a marker for other preferences about end of life care [40]. As a practical matter, most patients or surrogates who have had appropriate counseling prior to the election of hospice care are ready to authorize a DNR order. Another inaccurate belief about hospice is the degree of therapeutic nihilism exercised during hospice care, such as the refusal to initiate intravenous (IV) therapy. No treatments are excluded automatically from consideration as long as the criterion for a treatment rests on its ability to achieve the symptom control, regardless of its impact on the underlying disease process. Depending on the resources and size of the hospice program, services will vary and may include the options of IV and total parenteral nutrition (TPN) therapy, chemotherapeutic agents, transfusion, invasive procedures such as gastrointestinal tract stenting, video-assisted thoracoscopy, and open fixation of pathologic fractures. The Medicare-certified hospice team is mandated to include a physician medical director, skilled nursing, a social worker, a pastoral care worker, home health aides, volunteers, and a bereavement program. Box 3 lists Medicare hospice benefits. Although the gastroenterologist may not be in a position to make a referral to palliative care or hospice services, his or her interest and support for these interventions on behalf of a patient will be appreciated deeply by the patient and family, especially if the relationship has been a positive and trusting one. QUALITY OF LIFE MEASUREMENT: THE NEW METRIC OF SUCCESS In any serious illness, questions relating to the impact of medical care on survival and the nature of survival cannot avoid touching on the hopes, fears, and other existential considerations of the individual. These subjective elements shape the individual’s evaluation of the QOL. Because of the importance of

Box 3: Medicare hospice benefit 

Patient elects hospice benefit for terminal illness, signs off Medicare A (hospital)



Primary physician may remain as primary, bills Part B for services



Consultants (eg, medical oncologist or surgeon) bill hospice directly



Patient is recertified as eligible for hospice at the beginning of each benefit period



Benefit periods of 90 days (first benefit period)/90 days (second benefit period)/60 days (each subsequent benefit period) indefinitely, with unlimited length of stay as long as the prognosis continues to meet the Medicare criterion of 6 months or less survival



Patient may revoke hospice benefit at any time

Courtesy of Robert A. Milch, MD, Buffalo, NY.

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

11

the patient’s perceptions in connecting the outer world intervention with the inner reality of the individual, the concepts of palliative care and QOL, although not synonymous, are inseparable. Strong arguments have been made [41] for the scientific reliability of QOL measurement that should reassure skeptics in the medical community that QOL outcomes measurement does not represent a ‘‘soft’’ standard but the ultimate challenge to scientific ability, insight, and compassion. QOL is the endpoint when measuring outcomes in palliative care. The twin goals of palliative care, symptom control and promotion of quality of life, are not measured adequately by the traditional oncologic outcome measurements of mortality and morbidity. Because the patient report is the ‘‘gold standard’’ for measuring the efficacy of palliative intervention, a new science of QOL measurement with psychometric validation is developing rapidly to fill this vacuum (see Rosenfeld elsewhere in this issue). PALLIATIVE CARE AND GASTROENTEROLOGIC ILLNESS: A PERFECT MATCH? Considering the variety of chronic, life-limiting, and highly morbid illnesses encountered in gastroenterologic practice, one wonders how gastroenterology ‘‘slipped by’’ the effect palliative care has made on other specializations [42], which include neurology, nephrology, and surgery. Increased survival from nongastrointestinal comorbidities such as congestive heart failure and dementia will undoubtedly expand the pool of patients appropriate for palliative care presenting to the gastroenterologist. Table 1 [43–51] contains a selection of disease or clinical scenarios likely to be seen by a gastroenterologist, in which the application of the palliative care model would seem appropriate, although as yet untested. Many other entities that are not necessarily imminently lethal (morbid obesity and irritable bowel syndrome) could be added to the list as well. In each of the diseases listed, the salient physical symptoms or distressing signs (eg, bleeding) are listed as possible entry points for more global intervention. The success of total pain management is predicated on treating the most compelling symptom first, usually a physical one. The ultimate, more important existential issues related to closure, purpose in life, and connection beyond the self are addressed more thoroughly if the immediate distractions of physical discomfort are well controlled. Any of the symptoms listed can be seen as the first opportunity to begin the process during which the physician will help the individual and family to hope for the best and plan for the worst. WHERE DOES THE GASTROENTEROLOGIST FIT INTO THE CONTINUUM OF PALLIATIVE CARE? The association of death with failure, culpability, or loss of control poses a potential psychologic barrier to the successful incorporation of a palliative philosophy into the clinical practice of gastroenterology, as it has for other medical disciplines. Other barriers to successful palliative care include denial or fear

DUNN

12

Table 1 Selected clinical problems appropriate for palliative gastroenterology Disease or syndrome Common GI tract symptoms Advanced GI cancer Anorexia/cachexia, bleeding, dysphagia, fistula, nausea, pain, vomiting

Palliative care implications

Interdisciplinary management required for management of malignant GI tract obstruction (surgery, pharmacotherapy, stenting) Total parenteral nutrition of no established benefit for improved survival or quality of life in patients with advanced gastrointestinal malignancy [43] Counseling for limited prognosis, goal framing, and prospective care planning Liver tumors, primary Anorexia/cachexia, ascites, High symptom burden requiring and metastatic bleeding, fever, nausea, interdisciplinary management in vomiting, pain the setting of limited prognosis and often late in a previously existing chronic illness (cirrhosis, previously diagnosed nonhepatic malignancy) Chemotherapy Altered liver function, bleeding, Nausea and vomiting are the constipation, diarrhea, emesis, most distressing side effects of fatigue, nausea, neutropenic chemotherapy [44] enterocolitis, pancreatitis (eg, Poor symptom control and vinca alkaloids), stomatitis diminished quality of life can adversely affect hope and compliance with chemotherapy regimen resulting in decreased survival [45] Severe complications may require discontinuation of diseasedirected therapy and consider hospice referral. Neurologic and Aspiration, cachexia, Significant number of patients myopathic disease constipation, diarrhea, predictably die from swallowing dysphagia dysfunction as a result of neurodegenerative diseases (amotrophic lateralizing sclerosis, Parkinson’s disease) and some forms of muscular dystrophy, usually after placement of a gastrostomy feeding tube. Benefit of gastrostomy tube feeding not established in patients with advanced dementia [46]

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

13

Table 1 (continued) Disease or syndrome

Common GI tract symptoms

Palliative care implications

Chronic pancreatitis

Diabetes, diarrhea, malnutrition, pain, weight loss

Liver failure

Ascites, bleeding, cachexia, encephalopathy, jaundice

Intestinal failure

Depression, diarrhea, fatigue, nutritional support, complications, weight loss

Complicated long-term pain management, often with opioids. Questions of physical dependence versus true addiction and invasive pain management techniques including surgery need to be addressed using an interdisciplinary approach High impact of illness on psychosocial functioning, especially when related to alcoholism Significant psychologic, socioeconomic, and spiritual distress associated with causes of liver failure (ethanol abuse, IV drug abuse) For Childs ‘‘C,’’ prognosis is limited to months, making hospice referral an option Good control of symptoms may up-grade functional status allowing more latitude for therapies Palliative care consultation may facilitate process of liver transplantation [47] Scant literature in reported QOL outcomes for patients eceiving home parenteral nutrition for malignant disease, one study for benign disease reported loss of friends, employment, and depression in 66% of families surveyed [48] Multiple losses stemming from loss of bowel autonomy: loss of normal eating patterns and independence, fear of loss of socioeconomic status, loss of bowel function control. (continued on next page)

DUNN

14

Table 1 (continued) Disease or syndrome

Common GI tract symptoms

Palliative care implications

Inflammatory bowel disease

Anorexia, rectal bleeding, diarrhea, body image, fistula, growth retardation, malaise, malnutrition, nausea, surgical complications, pain, vomiting, weight loss

QOL outcomes are increasingly measured in clinical trials Patients frequently resort to alternative and unconventional therapies [49] and are more likely to with increasing duration of disease [50]. This can be a manifestation of low level of trust or poor symptom control. Occasional patient will have the additional burden of IBD-associated malignancy. Psychosocial impact of disease affects health care utilization [51] Interdisciplinary model of palliative care is patient centered and specific, helpful for selecting among invasive intervention alternatives.

Abbreviations: GI, gastrointestinal; IBD, inflammatory bowel disease.

of death, lack of training in the techniques of communication and symptom control, and, in some cases, indifference. Another potential barrier to the inclusion of a comprehensive philosophy of palliation in the framework of medicine is deeply rooted prejudice directed against the ‘‘soft’’ behavioral and social sciences (not to mention the humanities), which are deemed ‘‘less valid’’ than the physical sciences. In the face of these barriers, it would represent a substantial leap for a gastroenterologist to accept palliation as an affirmative process of ameliorating suffering that takes an equal if not greater priority to the intent to cure. Perhaps gastroenterologists can be inspired to accept the construct of palliative care by the recent progress of their professional cousins, the surgeons! In October 2003, the American College of Surgeons’ Task Force on Professionalism published the American College of Surgeons’ Code of Professional Conduct [52]. In its commentary, the Task Force wrote, "We singled out terminally ill patients as worthy of specific mention. Most surgeons are uncomfortable with death; it is an outcome that might be equate with defeat. Many surgeons are also uncomfortable with the transition from curative to palliative care. Effective palliation at these difficult times obligates sensitive discussion with patients and their families. Surgeons must accept a pivotal role in facilitating this therapeutic transition, for both patients and the healthcare team’’ [53].

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

15

Palliative gastroenterology, similar to surgical palliative care, is more a matter of rediscovery or renaming much of the daily experience most gastroenterologists have with patients suffering from advanced and critical illness than the assimilation of radically new concepts. The pragmatic temperament of those physicians who perform procedures facilitates their endorsement of any new technique or concept that effectively yields results, in this case, the relief of suffering. There is a wide spectrum of opportunity for the gastroenterologist to participate in palliative care, depending on his or her time, abilities, and degree of interest. Deciding ‘‘how involved should I be in the palliative care of a patient?’’ is a personal one, but at the very least the gastroenterologist should be aware of the indications for referral and be willing to facilitate referral to the appropriate individuals or agencies in a timely and supportive manner. Years of faithful service by a physician can be quickly forgotten by a patient and family if there is footdragging by the physician about an appropriate hospice or palliative care referral. There are several levels of involvement in palliative care beyond the primary level of engagement expected of all physicians (Box 4).

EDUCATIONAL RESOURCES The palliative care movement in the United States during the late 1990s and early 2000s has been catalyzed by the availability of several well-designed and continuously updated Web sites, among which is the End of Life/Palliative Education Resource Center (EPERC) Web site (www.eperc.mcw.edu). The purpose of EPERC is to share educational resource material among the community of health professional educators involved in palliative care education. The site offers ‘‘Fast Facts,’’ which consists of one-page peer-reviewed outlines of key information on important end-of-life clinical topics for end-oflife educators and clinicians, down-loadable educational materials, suggested articles, and links to clinical and educational resource centers. The Center for Advance Palliative Care (CAPC) has a useful Web site (www.capc.org) for surgeons who may be involved in organizing hospital-based palliative care programs. The CAPC, an initiative supported by the Robert Wood Johnson Foundation and assisted by the Mount Sinai School of Medicine in New York, provides health care professionals with the tools, training, and technical assistance necessary to start and sustain successful palliative care programs in hospitals and other health care settings. Through its Web site, the CAPC provides management training seminars, audio conferences, and other offerings, and a core curriculum for programs in planning and early stages of development. The CAPC also provides information on palliative care to hospitals, clinicians, policymakers, payers, and researchers. Other Web sites and organizations of interest include the National Hospice and Palliative Care Organization (www.nhpco.org), the Robert Wood Johnson Office of Promoting Excellence in End-of-Life Care (www.promotingexcellence.org), and the American Academy of Hospice and Palliative Medicine (www.aahpm.org).

16

DUNN

Box 4: Levels of the gastroenterologist’s involvement in palliative care Level 1: the gastroenterologist Is familiar with basic principles of palliative care as endorsed by specialty organizations and examining boards and has met palliative and end-of-life care CME requirements where mandated (eg, California). Is competent in the management of acute and uncomplicated chronic pain, major nonpain symptoms, and the communication of bad news. Can recognize the syndrome of imminent demise. Refers appropriately to palliative care and hospice services. Level 2 (level 1 plus. . .): the gastroenterologist Participates in palliative care indirectly or directly by participation in ethics, patient care, critical care, and pharmacy committees. Promotes interest in palliative care by inviting grand rounds speakers, supports educational initiatives, and supports palliative care consultants and services. Level 3 (level 1 plus. . .): the gastroenterologist Active or retired, participates in hospice or palliative care interdisciplinary teams, as either paid staff or a volunteer. Is certified by the American Academy of Hospice and Palliative Medicine (or American Board of Medical Specialties when available). Level 4 (level 1 plus. . .): the gastroenterologist Is active in the practice of gastroenterology, the primary focus of which is palliative care education, research, or direct patient care (eg, palliative gastroenterology specialist).

CERTIFICATION IN PALLIATIVE AND HOSPICE CARE For gastroenterologists who wish to pursue palliative care in more depth, either as a primary focus of practice or as an adjunct to their primary specialization, the option off pursuing certification by the American Board of Hospice and Palliative Medicine (ABHPM) has been available since 1996. Candidates are required to hold an American Board of Medical Specialties (ABMS)-approved primary Board Certification, certified status for at least 2 years in a primary medical specialty, have had 2 years’ practice experience beyond residency or fellowship, 2 years’ experience of interdisciplinary team work with a hospice or palliative care program, physician and nonphysician professional references, and evidence of direct participation in the care of at least fifty terminally ill patients [54]. The ABHPM is not yet affiliated with the American Board of Medical Specialties, although an application to the ABMS to recognize palliative medicine as a subspecialty has been submitted by the American Board of Internal Medicine, and a decision regarding

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

17

the application will be made in September 2006. To date, approximately 1900 physicians have been certified by the ABHPM, including specialists in family medicine, geriatrics, radiation oncology, surgery, anesthesia, psychiatry, and gastroenterology, among others. If palliative medicine becomes an ABMSaccredited subspecialty, after the initial period of ‘‘grandfathering,’’ only individuals who have completed an Accreditation Council for Graduate Medical Education-accredited fellowship in Hospice and Palliative Medicine will be able to sit for the certifying examination. Currently, there are 17 accredited palliative medicine fellowships in the United States [55]. Information about the ABHPM and the status of ABMS specialty certification in hospice and palliative care can be obtained at www.abhpm.org. FUTURE OF PALLIATIVE GASTROENTEROLOGY The future of palliative care research lies in the consensus of physicians and their leadership that the concept of palliative care is valid and essential to all practitioners. The conversion of rapidly fatal illnesses to chronic illnesses in an already aging population should provide some impetus to the acceptance of functional and QOL outcomes as equally valid as survival and diseasefree intervals. The existing barriers to this acceptance are formidable and include economic and political obstacles such as access to research funding. Only 0.9% of the 1999 NCI budget went to palliative and hospice care research [56], despite the fact that a significant number of those diagnosed with cancer will succumb to it, still have it, or have permanent sequelae of its treatment at the time of death. The concept of palliative gastroenterology is still too new to identify research funding for nononcologic illnesses such as inflammatory bowel disease or liver failure. Other barriers include the unique ethical concerns related to research in a highly vulnerable population and little collaborative experience between gastroenterologists and palliative care professionals. SUMMARY Martin Adson, a surgeon, completes the Introduction to the book Liver Metastases: Biology, Diagnosis, and Treatment [57] with an eloquent moral charge to all physicians and surgeons that summarizes the reason, the method, and the reward for the palliation of suffering caused by metastatic cancer. His words apply to all other incurable conditions and those treating them: ‘‘The common denominators of all metastatic cancers are pain, fear, and loss and the common denominator of involved physicians is concern. Neither these terrifying components of cancer nor the efforts of defenders can be measured or controlled scientifically, but disregard of this reciprocal relationship by physicians and surgeons is a denial of moral obligation. Abandonment in the context of medical practice is a form of spiritual malpractice. There is no specific formula for the relationship required when science is found wanting, but palliation of suffering of this kind must begin with acknowledgement. This

18

DUNN

involves. . .a capacity for affection and willingness to share one’s importance and strength with the person whose importance is threatened by the cancer that you have been trained to treat.’’ APPENDIX: FAST FACT CONCEPT NO. 084: SWALLOW STUDIES, TUBE FEEDING, AND THE DEATH SPIRALa The reflex by families and doctors to provide nutrition for the patient who cannot swallow is overwhelming. It is now common practice for such patients to undergo a swallowing evaluation and, if the patient ‘‘fails,’’ to move forward with feeding tube placement (nasogastric, jejeunogastric). Data suggest that placement of feeding tubes has an in-hospital mortality of 15%–25%, and one-year mortality of 60%. Not surprisingly, predictors of early mortality include: high age, CNS [central nervous system] pathology (CVA [cardiovascular accident], dementia), cancer, except early stage head/neck cancer, disorientation, and low albumin. The clinical scenario, the tube feeding death spiral, typically goes like this: 1. Hospital admission for complication of ‘‘brain failure’’ or other predictable end organ failure due to primary illnesses (eg, urosepsis in setting of advanced dementia) 2. Inability to swallow and/or direct evidence of aspiration and/or weight loss with little oral intake 3. Swallowing evaluation followed by a recommendation for nonoral feeding due to either aspiration or inadequate intake 4. Feeding tube placed leading to increasing ‘‘agitation’’ leading to patientremoval or dislodgement of feeding tube 5. Re-insertion of feeding tube; hand and/or chest restraints placed 6. Aspiration pneumonia 7. Intravenous antibiotics and pulse oximetry 8. Repeat of [steps] 4 through 6 one or more times 9. Family conference 10. Death

Note: at [the Web author’s] institution, the finding of a dying patient with a feeding tube, restraints and pulse oximetry, is known as Weissman’s triad. Suggestions include: 1. Recognize that the inability to maintain nutrition through the oral route, in the setting of a chronic life-limiting illness and declining function, is usually a marker of the dying process. Discuss this with families as a means to a larger discussion of overall end of life goals. 2. Ensure that your colleagues are aware of the key data and recommendations on tube feedings (see below; also see Fast Fact no. 10: Tube Feed or Not Tube Feed).

a

Adapted from Weissman, DE. Swallow studies, tube feeding and the death spiral. Fast Fact and Concept no. 85. 2003. End-of-Life Physician Education Resource Center [Web site]. Available at: www.eperc.mcw.edu.; with permission.

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

19

3. Ensure there is true informed consent prior to feeding tube insertion; families must be given alternatives (eg, hand feeding, comfort measures) along with discussion of goals and prognosis. 4. Assist families by providing information and a clear recommendation for or against the use of a feeding tube. Families who decide against feeding tube placement can be expected to second-guess their decision and will need continued team support. 5. If a feeding tube is placed, establish clear goals (eg, improved function) and establish a timeline for re-evaluation to determine if goals are being met (typically 2–4 weeks).

Bibliography Finucane TE, et al. Tube feeding in patients with advanced dementia. J Gen Int Med 1999;282: 1365–9. Finucane TE, Bynum JP. Use of tube feeding to prevent aspiration pneumonia. Lancet 1996;348: 1421–4. Cowen ME, et al. Survival estimates for patients with abnormal swallowing studies. JGIM 1997;12:88–94. Rabeneck L, et al. Long term outcomes of patients receiving percutaneous endoscopic gastrostomy tubes. JGIM 1996;11:287–93. Grant MD, et al. Gastrostomy placement and mortality among hospitalized Medicare beneficiaries. JAMA 1998;279:1973–6.

References [1] Cruzan v Director, Missouri Department of Health, 497 DS 261 (1990). [2] The SUPPORT Clinical Investigators. A controlled trial to improve care for seriously ill hospitalized patients: the study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). JAMA 1995;274(20):1591–8. [3] NHPCO data. Available at: www.hhpco.org/files/public/facts_figures_for2004data.pdf. Accessed January 21, 2006. [4] von Eschenbach AC. Realistic goal: end suffering, death from cancer by 2015. USA Today January 21, 2004;A15 [5] Hickman SE, Tolle S, Tilden V, et al. The Oregon report card: improving care of the dying. 1999. Available at: www.ohsu.edu/ethics. Accessed January 21, 2006. [6] Brody JE. A doctor’s duty, when death is inevitable. The New York Times; section E: August 10, 2004. p. 7. [7] Gershon MD. The second brain. New York: Harper Collins; 1998. [8] Nightingale J, editor. Intestinal failure. London: Greenwich Medical Media; 2001. p. xxix–xxxx. [9] Saxon W. Cicely Saunders dies at 87; reshaped end-of-life care. New York Times; section A: August 4, 2005. p. 17. [10] Schonwetter RS, editor. Hospice and palliative medicine: core curriculum and review syllabus. Dubuque (IA): Kendall/Hunt Publishing Company; 1999. p. 9. [11] American Board of Internal Medicine Committee on Evaluation of Clinical Competence: Caring for the dying: identification and promotion of physician competency. Philadelphia: American Board of Internal Medicine; 1996 [educational resource document]. [12] Emmanuel LL, von Gunten CF, Ferris FD, editors. The EPEC curriculum: education for physicians on end -of-life care. The EPEC project. 1999. Available at: www.EPEC.net. Accessed December 22, 2005. [13] Field MJ, Cassel CK, editors. Approaching death: improving care at the end of life: report by the Committee on Care at the End of Life. Washington (DC): National Academy Press; 1997.

20

DUNN

[14] Moss AH, Campbell B, Cohen LM, et al. End Stage renal disease workgroup. Available at: www.promotingexcellence.org/esrd/indexv2.html. Accessed December 22, 2005. [15] Surgeons’ Palliative Care Workgroup. Report to the field. J Am Coll Surg 2003;197(4): 661–86. [16] Billings JA, Block S. Palliative care in undergraduate medical education: status report and future directions. JAMA 1997;278(9):233–8. [17] Weissman DE, Mullan PB, Ambuel B, et al. End-of-life curriculum reform: outcomes and impact in a follow-up study of internal medicine residency programs. J Palliat Med 2002;5(4): 497–506. [18] Foley KM. Preface. In: Foley KM, Gelband H, editors. Improving palliative care for cancersummary and recommendations. Washington (DC): National Academy Press; 2001. p. ix. [19] Cassel C, Foley K. Principles for care at the end of life: an emerging consensus among the specialties of medicine. Report sponsored by the Milbank Memorial Fund. December 1999. Available at: http://www.milbank.org. Accessed December 22, 2005. [20] American College of Surgeons. Principles guiding care at end of life. Bull Am Coll Surg 1998;83:46. [21] American College of Surgeons. Principles of palliative care. Bull Am Coll Surg 2005;90(8): 34–5. [22] Frankl VE. Man’s search for meaning. New York: Washington Square Press; 1985. [23] Cassell EJ. The nature of suffering and the goals of medicine. New Engl J Med 1982;306(11):639–45. [24] Cassell EJ. The nature of suffering and the goals of medicine. New York: Oxford University Press; 1991. p. 33–4. [25] MacDonald N. Palliative medicine and modern cancer care. In: Doyle D, Hanks G, Cherny NI, et al, editors. Oxford textbook of palliative medicine. 3rd edition. Oxford (UK): Oxford University Press; 2004. p. 24. [26] Saunders CM. The challenge of terminal care. In: Symington T, Carter RL, editors. Scientific foundations of oncology. London: Heinemann; 1976. p. 673–9. [27] Brody H. My story is broken; can you help me fix it? medical ethics and joint construction of narrative. Lit Med 1994;13(1):79–92. [28] Byock IR. The nature of suffering and the nature of opportunity at the end of life. Clin Geriatr Med 1996;12(2):237–52. [29] Erikson EH. Identity and the life cycle. New York: WW Norton & Co, Inc.; 1980. p. 1–187. [30] National Hospice Organization. Hospice fact sheet. Arlington (VA): National Hospice Organization; 2001. [31] Kass LR. Ethical dilemmas in the care of the ill. What is the patient’s good? JAMA 1980;244: 1947. [32] Doyle D, Hanks G, Cherny NI, et al, editors. Introduction. In: Doyle D, Hanks G, Cherny NI, et al, editors. Oxford textbook of palliative medicine. 3rd edition. Oxford (UK): Oxford University Press; 2004. p. 4. [33] Morrison SR, Maroney-Galin C, Kralovec PD, et al. The growth of palliative care programs in United States hospitals. J Palliat Med 2005;8(6):1127–34. [34] Morrison S, Meier DE. Palliative care. N Engl J Med 2004;350:2582–90. [35] Elsayem A, Swint K, Fisch M, et al. Palliative care inpatient service in a comprehensive cancer center: clinical and financial outcomes. J Clin Oncol 2004;22:2008–14. [36] Meier DE. Palliative care in hospitals: making the case. New York: Center to Advance Palliative Care; 2002. [37] Higginson IJ, Finlay IG, Goodwin DM, et al. Do hospital-based palliative teams improve care for patients or families at the end of life? J Pain Symptom Manage 2002;23: 96–106. [38] Campbell ML, Guzman JA. Impact of a proactive approach to improve end-of-life care in a medical ICU. Chest 2003;123:266–71.

PALLIATIVE CARE: A FRAMEWORK FOR GASTROENTEROLOGY

21

[39] Manfredi PL, Morrison RS, Morris J, et al. Palliative care consultations: how do they impact the care of hospitalized patients? J Pain Symptom Manage 2000;20(3):166–73. [40] Beach MC, Morrison RS. The effect of do-not-resuscitate orders on physician decisionmaking. J Am Geriatr Soc 2002;50:2057–61. [41] Koller M, Nies C, Lorenz W. Quality of life issues in palliative surgery. In: Dunn GP, Johnson AG, editors. Surgical palliative care. Oxford (UK): Oxford University Press; 2004. p. 94–111. [42] Robert Wood Johnson Foundation, Office of Promoting Excellence in End-of-Life Care. Available at: http://www.promotingexcellence.org/esrd/indexv2.html. Accessed December 22, 2005. [43] Jeejeebhoy KN. Parenteral nutrition in gastrointestinal disease. In: Irvine EJ, Hunt RH, editors. Evidence-based gastroenterology. Hamilton (Ont, Canada): BC Decker Inc.; 2001. p. 358. [44] Coates A, Abraham S, Kaye SB, et al. On the receiving end: patient perception of the sideeffects of cancer chemotherapy. European Journal of Cancer and Clinical Oncology 1983;19:203–8. [45] Fischbach W. Side-effects of chemotherapy on the gastrointestinal tract: symptoms, grade and classification. In: Staritz M, Adler G, Schmiegel W, et al, editors. Proceedings of the Falk Workshop, Oct. 11, 2002: side-effects of cancer chemotherapy on the gastrointestinal tract: physiology, prophylaxis, and therapy. Dordrecht (Netherlands): Kluwer Academic Publishers; 2003. [46] Finucane TE, Christmas C, Travis K. Tube feeding in patients with advanced dementia: a review of the evidence. JAMA 1999;282(14):1365–70. [47] Molmenti EP, Dunn GP. Transplantation and palliative care: the convergence of two seemingly opposite realities. Surg Clin North Am 2005;85:373–82. [48] Smith C. Quality of life in long term total parenteral nutrition patients and their family caregivers. JPEN J Parenter Enteral Nutr 1993;17:501–6. [49] Hilsden RJ, Verhoef MJ. Complementary medicine use by patients with inflammatory bowel disease. Am J Gastroenterol 1998;93:697–701. [50] Moser G, Tillinger W, Sachs G, et al. Relationship between the use of unconventional therapies and disease-related concerns: a study of patients with inflammatory bowel disease. J Psychosom Res 1996;4:318–23. [51] Turnbull GK, Vallis TM. Quality of life in inflammatory bowel disease: the interaction of disease with psychosocial function. Am J Gastroenterol 1995;90:1450–4. [52] American College of Surgeons Task Force on Professionalism. Code of professional conduct. J Am Coll Surg 2003;197(4):603–4. [53] American College of Surgeons Task Force on Professionalism. Professionalism in surgery. J Am Coll Surg 2003;197(4):605–8. [54] American Board of Hospice and Palliative Medicine. Certification Eligibility Requirements 2006. Available at: www.abhpm.org/gfxc_13.aspx. Accessed January 22, 2006. [55] American Board of Hospice and Palliative Medicine [Web site]. Available at: www.abhpm. org. Accessed December 20, 2005. [56] Foley KM. Improving palliative care for cancer- summary and recommendations. Washington (DC): National Academy Press; 2001 [pamphlet]. [57] Garen OJ, Geraghty JG, Nagorney DM. Liver metastases: biology, diagnosis, and treatment. London: Springer-Verlag, Ltd; 1998. p. 1–5.

Gastroenterol Clin N Am 35 (2006) 23–39

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Palliative Care Assessment: What Are We Looking For? Kenneth Rosenfeld, MD Veterans Integrated Palliative Program, Greater Los Angeles Veterans Affairs Healthcare System, 11301 Wilshire Boulevard, Los Angeles, CA 90037, USA

G

astroenterologists are providing both specialty and primary care to an increasing number of patients who have advanced, progressive illness, who are near the end of life. Dramatic increases in palliative endoscopic procedures, including placement of percutaneous endoscopic gastrostomy tubes and stenting of obstructions of the upper and lower digestive tract, have fueled growth in the palliative gastrointestinal (GI) population, making the gastroenterologist a frequently visited or even the primary physician for many patients. These patients present unique challenges to decision making, symptom management, and psychosocial assessment, yet few gastroenterologists have received training in core competencies required to provide highquality care to patients who have palliative needs. To fulfill one’s professional responsibility, the GI specialist needs to gain, at a minimum, a set of core competencies in palliative assessment and management directed to patients who have advanced illness. This article provides an overview of palliative care assessment directed toward the gastroenterologic specialist. First, it presents an evidence-based conceptual model of high-quality end-of-life care to serve as the basis for a palliative assessment strategy. An approach to assessment is then presented, recognizing that comprehensive palliative assessment ideally requires participation of an interdisciplinary team (historically a physician nurse, social worker, and chaplain). The physician’s role as team leader often includes screening for problems in each palliative domain, referring to appropriate team members for a more detailed assessment, and integrating data to collaboratively (with the team, patient and caregivers) develop a management plan. Because the relief of suffering and the cure of disease are dual obligations of the medical profession [1], palliative care provides possibilities for benefiting patients and caregivers when cure is no longer realistic.

E-mail address: [email protected] 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.002

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

24

ROSENFELD

CONCEPTUALIZING THE PALLIATIVE CARE ASSESSMENT Palliative care is defined by the World Health Organization (WHO) as ‘‘the active total care of patients whose disease is not responsive to curative treatment. Control of pain, other symptoms, and psychologic, social, and spiritual problems is paramount. The goal of palliative care is achievement of the best quality of life for patients and their families’’ [2]. The wide scope of the WHO definition suggests that palliative care addresses a broader range of dimensions than the traditional ‘‘curative’’ model, in which maximizing survival duration is the primary goal of assessment and treatment burden, venue of care, and other quality of life concerns are addressed secondarily, if at all. Over the past decade, research and expert opinion have clarified the highly subjective experience of approaching life’s end and have defined areas important to patients and families that should be included in the palliative assessment. In a focus group study of chronically ill individuals, Singer and colleagues [3] identified five end-of-life quality domains, including adequate pain and symptom management, avoidance of inappropriate prolongation of dying, achievement of a sense of control, relief of burden, and strengthening relationships with loved ones. This and other qualitative [4] research as well as the opinions of expert groups [5,6] suggest that palliative assessment address the following components (Box 1): (1) physical symptoms and functioning; (2) emotional and spiritual concerns; (3) personal history; (4) social support and caregiver needs; (5) goals of care and advance care planning; and (6) anticipatory issues, including preferred terminal care setting and after death concerns. Adding to this framework a few standard assessment elements (clinical summary, medications, physical examination, and laboratory evaluation) provides the physician with a comprehensive and patient-centered strategy. CLINICAL SUMMARY The first step in palliative assessment entails a review of the patient’s illness history, including the primary life-limiting illness and other concurrent illnesses Box 1: A suggested palliative assessment approach 1. Clinical summary 2. Physical symptoms and functioning 3. Psychological and spiritual functioning 4. Goals of care/advance care planning 5. Social support and caregiver needs 6. Anticipatory concerns (care setting and afterdeath) 7. Medications 8. Physical examination 9. Laboratory 10. Summary and recommendations

PALLIATIVE CARE ASSESSMENT

25

with relevance to the primary illness or its treatment. The clinical summary includes a review of the patient’s treatment course, including the response to treatment, side effects and toxicities, and treatment complications. The clinical summary provides important information (for survival and functioning) which will assist the physician in discussing goals of care and planning treatment. It also gives the physician clues as to the types of symptoms to be assessed, based on illness characteristics, clinical course, and treatment toxicities. Information regarding issues such as the recency of the diagnosis or treatment successes or setbacks also helps the physician to anticipate the types of emotional and spiritual challenges the patient may be dealing with and the factors the patient may be weighing in treatment decision making. For example, a patient who has undergone and survived intubation for respiratory failure is likely to view a discussion regarding future use of a ventilator very differently than one who has never been on a ventilator. Entering the assessment with as detailed an understanding as possible of the patient’s previous medical care facilitates sensitive and comprehensive history taking and promotes patient-centered decision making. PHYSICAL SYMPTOMS AND FUNCTIONING Physical problems, in the form of symptoms and functional decline, are virtually universal in patients who have advanced illness. In a study [7] of 275 consecutive patients who had cancer, pain, fatigue, anorexia, nausea, constipation, and confusion was each reported in 60% to 90% of patients. These and other symptoms are sources of distress for both patients and caregivers and have a major impact on quality of life. Thorough evaluation of physical symptoms falls squarely within the responsibility of the physician. Although, often thought of as an expected and unavoidable part of advanced illness, prompt recognition of pain and other distressing symptoms can lead to effective management in the vast majority of cases [8]. Symptoms may result from the disease process itself or from toxicities of disease treatment; diseases may cause symptoms directly through damage to organs (as in cirrhosis), by compression or invasion of adjacent structures (eg, cancer), through the release of compounds by a diseased organ (eg, paraneoplastic syndromes), from the body’s own defenses (eg, tumor necrosis factor and interleukins in infection), or from metabolic derangements caused by the failure of diseased organs (eg, hyperammonia in hepatic failure). Alternatively, symptoms can be caused by adverse effects of treatments used to treat the underlying disease, such as radiation therapy, chemotherapy, and medications. Importantly, symptoms in advanced disease usually reflect a complex interaction of pathophysiologic processes, emotional factors, and spiritual concerns aptly labeled ‘‘total pain’’ [9]. The concept of total pain has important implications for the provider because optimal symptom management often requires attention to each contributing factor. A list of common symptoms is provided in Box 2.

26

ROSENFELD

Box 2: Common symptoms near the end of life Pain Fatigue Anorexia Cachexia/weight loss Nausea/vomiting Constipation Edema Dyspnea Pressure sores Insomnia Confusion

When a symptom is present, a more detailed history permits the physician to localize its likely cause and to provide a baseline of symptom severity through which treatment effectiveness may be monitored. Clarifying a symptom’s underlying cause is important because symptom management differs based on its cause. For instance, nausea from partial gastric outlet obstruction caused by an enlarged liver is managed differently from nausea experienced as a side effect of opioids. Finally, symptom assessment should also elicit all necessary information to develop an immediate management plan. Even in situations in which further diagnostic testing is indicated, prompt treatment of symptoms should always be prioritized. The evaluation of pain provides a good example of how to approach symptom assessment in general. As in other symptoms, prompt and thorough assessment of pain is essential because of the prevalence of pain (60%–90% in those who have advanced cancer [7]), its impact on quality of life [10], and its ready response in most cases to treatment based on easily applied guidelines [9]. Nevertheless, studies reproducibly demonstrate that pain is undermanaged, even in established oncology settings where pain should be anticipated [11]. Pain assessment should elicit information necessary to initiate management in the immediate-term and to identify likely causative factors, which can help with longer term management. The WHO cancer pain management ‘‘ladder’’ proposes initial management based on the severity of pain rather than its precise cause, so immediate management-directed evaluation includes assessment of pain severity, frequency, and response to previously attempted treatments. Pain severity can be assessed through a number of approaches [12], most commonly by the 0-to-10 point scale now adopted widely as the ‘‘fifth vital sign.’’ Pain frequency, whether continuous or intermittent throughout the day, aids in

PALLIATIVE CARE ASSESSMENT

27

decisions about whether to prescribe short- or long-acting analgesics and in round-the-clock versus as-needed dosing decisions. In assessing the response to treatment, it is helpful to identify the patient’s pain level and intensity before the analgesic is taken, once the medication has reached its full effect (typically after 1 hour for immediate-release medications). Additional questions that can help to localize the cause of pain include its location, quality (eg, aching, stabbing, or sharp, which is characteristic of nociceptive pain, or burning, tingling, or lancinating, which is characteristic of neuropathic pain) radiation, and precipitating and relieving factors. It is helpful to inquire about the pain’s impact on patient mood and functioning, which provides additional information relevant to the patient’s perspective on the pain. The approach described above for pain assessment can be adapted to the full range of common symptoms, recognizing that assessment of any symptom requires a basic understanding of symptom-specific pathophysiology and its relationship to symptom manifestations and management. Numerous screening tools have been developed to identify the presence and intensity of a range of common symptoms. The Edmonton Symptom Assessment Scale [13] and the Condensed Memorial Symptom Assessment Scale [14], are particularly easy to use. An assessment of physical functioning is integral to palliative management because function affects prognosis [15], the appropriateness of certain treatments such as chemotherapy, and the patient’s care needs and social interactions. Functional assessment may elicit the patient’s ability to perform basic activities of daily living such as those defined by Katz and colleagues [16] (toileting, bathing, dressing, continence, transfers, and feeding) and also more complex ‘‘instrumental’’ activities of daily living, believed to be required for independent living (eg, preparing meals, managing finances, shopping for groceries, using a telephone, and performing housework) [17]. In addition to task-based function ratings, observer-rated performance scales are also used widely, most notably the Eastern Cooperative Oncology Group Performance Scale [18], a 6-point scale (scored 0 to 5) in which higher numbers indicate more severe functional impairment, and the Karnofsky Performance Scale [19], an 11-point scale (scored from 100 to 0) with higher numbers signifying better performance status (Table 1). The Karnofsky scale has recently been modified to the Palliative Performance Scale [20], which includes five palliative-focused domains, including activity level, ambulation, self-care, intake, and level of consciousness. Serial application of these scales can help the provider to identify and anticipate a patient’s likely care needs as the disease progresses. PSYCHOLOGIC AND SPIRITUAL FUNCTIONING Life-limiting illnesses create challenges that tax an individual’s inner strength but also provides an opportunity for personal growth near the end of life [21]. In contrast with contemporary medicine which focuses its attention on the molecular mechanisms, genetics, and pathophysiology of disease, palliative care inquires into the patient’s experience of illness, which is an amalgam of the

ROSENFELD

28

Table 1 Commonly-used observer-rated performance status measures ECOG 0 1 2 3 4 5

Fully active, no restrictions on activity Restricted in strenuous activity but able to carry on less vigorous activity Ambulatory, able to perform self-care; out of bed >50% of day Limited self-care; in bed >50% of waking hours Disabled; no self-care; confined to bed or chair Dead

Karnofsky 100 90 80 70 60 50 40 30 20 10

Normal, no evidence of disease Normal activity, minor symptoms of disease Normal activity with effort: some symptoms of disease Cares of self: unable to carry on normal activity or active work Requires occasional assistance but is able to care for needs Requires considerable assistance and frequent medical care Disabled: requires special care and assistance Severely disabled: hospitalization is indicated, death not imminent Very sick, hospitalization necessary: active treatment necessary Moribund, fatal processes progressing rapidly

Data from Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist 1969;9(3):179–86; Verger E, Salamero M, Conill C. Can Karnofsky performance status be transformed to the Eastern Cooperative Oncology Group scoring scale and vice versa? Eur J Cancer 1992;28A(8-9):1328–30; with permission.

disease state and the patient’s psychologic and spiritual makeup. The illness experience includes a patient’s education, religious or spiritual grounding, employment and hobbies, family relationships, experiences with illness in self and loved ones, and other elements that shape the ‘‘person’’ behind the ‘‘patient.’’ Given the illness experience’s influence on a patients’ goals of care and treatment preferences, symptoms, and coping abilities, the physician’s grasp of these elements can facilitate treatment decision making, symptom management, and planning for terminal care strengthening the alliance between physician and patient. Although psychologic and spiritual assessment may not be the physician’s sole responsibility within the interdisciplinary team, the physician can, during the initial palliative evaluation, gain a basic understanding of the core elements of the patient’s personal history. An astute physician can implicitly weave an assessment of the patient’s cognitive state (ie, presence of delirium or dementia), emotional functioning, and spiritual well being into the patient interview by attending to clues of attention, affect, and distress. Such implicit evaluation can prompt a more thorough evaluation if areas of concern are identified. Given the intense personal and interpersonal challenges patients who have advanced illness face [1], the contributions of these elements to patients’ overall care cannot be overstated.

PALLIATIVE CARE ASSESSMENT

29

Cognitive dysfunction occurs at a high rate during the advanced phase of illness [22] and is often caused by remediable factors. Screening for cognitive function is an important part of the palliative evaluation with the dual goals of detecting cognitive dysfunction, if present, and forming a baseline in case cognitive impairment develops later in the patient’s clinical course. The screening tool most commonly used in clinical practice is the Mini-Mental State Examination (MMSE) [23], an 11-item instrument (scored 0 to 30) that assesses a range of cognitive abilities. Marked impairment on the MMSE may reflect either delirium or dementia, which would lead to a more thorough history of the impairment, including its onset, course (ie, fluctuating versus continuous), features (eg, attention and level of consciousness), and possible precipitating factors. If delirium is suspected, the Confusion Assessment Method [24], a 9-item observer-rated tool, has been shown to be a highly sensitive and specific diagnostic method. Both depression and anxiety are encountered frequently in the palliative care population. Although a detailed evaluation for depression and anxiety may be performed best by a mental health professional, research has found that easily performed single-item screens have a high diagnostic yield and can be incorporated readily into the palliative assessment. For example, Chochinov and colleagues [25] have reported that the single question ‘‘Are you depressed?’’ outperformed longer assessment tools in identifying depression in 197 cancer patients, and Jacobsen and colleagues [26] found that a single-item ‘‘distress thermometer’’ compared favorably with longer assessment tools in identifying anxiety and depression in 380 ambulatory oncology patients. The physician should inquire about the patient’s emotional response to illness, including coping strategies and commonly expressed fears about burden, dependency, indignity, and abandonment. A sympathetic inquiry about such fears can help the physician direct appropriate resources to them (eg, social work or psychologist support) and can be therapeutic in its own right. Examples of questions to assess emotional symptoms and coping are provided in Box 3. Patients facing advanced illness, regardless of their religious background, typically experience a broad range of spiritual and existential concerns. Studies have shown that patients’ spirituality is particularly important in the setting of advanced illness [27] and that patients value physicians who are willing to explore the spiritual aspects of their lives [28]. Patients’ religious or spiritual beliefs also may influence their preferences for life-sustaining treatments in important ways, making it incumbent on physicians to understand those beliefs as part of treatment decision making and advance care planning. Several frameworks have been offered to guide the physician in exploring patients’ religious and spiritual perspectives [29]. A popular spiritual assessment tool proposed by Puchalski and Romer [30] uses the acronym FICA, which includes assessment for the presence of faith (F) or belief or for sources of meaning in the patient’s life; the importance (I) that the patient’s spiritual beliefs (or sources of meaning) play in dealing with the illness; the patient’s spiritual community (C) or other community that serves as a source of support during

ROSENFELD

30

Box 3: Examples of inquiry into psychological and spiritual functioning Questions to solicit personal history Where do you originally come from? Who is your family? Who are your close friends? What type of work do (did) you do? What do you enjoy doing with your free time? Have you done anything in your life that you’ve regretted? What experiences have you had with serious illness, in yourself or those close to you? How has this illness affected your life? Questions to assess patient emotional concerns What is the hardest part of this illness for you? Where do you turn for comfort when things are difficult? What things about your illness worry you? What do you see as your biggest challenge now? Do you often feel depressed? Do you often feel anxious? Questions to assess spiritual orientation Do you consider yourself a religious person? Do you consider yourself a spiritual person? What do you find most meaningful in your life now? Do you feel peaceful, or are you having difficulty finding peace? What are you hoping for in the future? How might I be of assistance to you in supporting your spiritual beliefs?

difficult times; and how the patient would like these issues to be addressed (A) as part of his or her health care. Once spiritual concerns have been identified, or if the physician feels uncomfortable inquiring about spiritual concerns, involvement of a hospital chaplain as a spiritual expert is indicated. In addition to the general spiritual assessment, the physician may wish to explore with the patient a set of ‘‘developmental tasks’’ near the end of life, as proposed by Byock (Table 2) [21]. These tasks should be seen as a positive opportunity for growth near the end of life. The physician can present these tasks as a means of permitting the patient some element of hope, control, and goal orientation in situations in which the patient (or physician) fears that ‘‘nothing more can be done.’’

PALLIATIVE CARE ASSESSMENT

31

Table 2 Developmental landmarks and taskwork for the end of life [20,31] Landmarks

Taskwork

Sense of completion with worldly affairs

Transfer of fiscal, legal and formal social responsibilities Closure of multiple social relationships (employment, commerce, organizational, congregational) Components include: expressions of regret, expressions of forgiveness, acceptance of gratitude and appreciation Leave taking; the saying of goodbye Life review The telling of ‘‘one’s stories’’ Transmission of knowledge and wisdom Self-acknowledgment Self-forgiveness Acceptance of worthiness Reconciliation, fullness of communication and closure in each of one’s important relationships. Component tasks include: expressions of regret, expressions of forgiveness and acceptance, expressions of gratitude and appreciation, acceptance of gratitude and appreciation, expressions of affection Leave-taking; the saying of goodbye Acknowledgment of the totality of personal loss represented by one’s dying and experience of personal pain of existential loss Expression of the depth of personal tragedy that dying represents Decathexis (emotional withdrawal) from worldly affairs and cathexis (emotional connection) with an enduring construct Acceptance of dependency Developing self-awareness in the present

Sense of completion in relationships with community

Sense of meaning about ones’ individual life Experienced love of self Experienced love of others Sense of completion in relationships with family and friends

Acceptance of the finality of life-of one’s existence as an individual

Sense of a new self (personhood) beyond personal loss Sense of meaning about life in general

Surrender to the transcendent, to the unknown - ‘‘letting go’’

Achieving a sense a awe Recognition of a transcendent realm Developing/achieving a sense of comfort with chaos In pursuit of this landmark, the doer and ‘‘taskwork’’ are one. Here, little remains of the ego except the volition to surrender.

32

ROSENFELD

GOALS OF CARE Assisting in clarifying goals of care and identifying a treatment plan that best adheres to those goals are among the most frequent requests of the palliative care specialist. Goal setting is fundamental to all medical specialties, but in patients who have advanced illness, determining goals of care is both particularly important and difficult. First, in advanced illness, the ability of medical interventions to cure or even effectively manage the patient’s disease is often limited. As a result, assumptions cannot be made that a patient prefers maximal attempts to treat disease, particularly as the burdens of medical interventions required to serve that goal increase. Second, patients vary widely in the way they weigh the tradeoffs inherent in choosing among goals. One patient may value the chance to live longer over any other consideration, whereas another patient may feel that other factors such as family burden, independence, or physical comfort may carry more weight than the desire to live as long as possible. These deeply held values depend on many personal and cultural factors but ultimately present challenges as doctors and patients negotiate treatment decisions. Finally, discussing goals of care is emotionally difficult for both the patient and physician because it often involves giving bad news, addressing issues of hope, frustration, and despair, and the physician confronting the physician’s professional limitations. Nevertheless, wellconducted and documented discussions enhance the patient-physician relationship, reassure patients that care will be matched optimally to their goals, and facilitate medical care by informing providers about patients’ priorities. Before initiating a goals-of-care discussion, the physician must first identify the appropriate individuals to act as decision makers. A patient who has decisional capacity (ie, can understand the situation, including the benefits and burdens of alternative approaches to care, can manipulate information to arrive at a preference that is rationally grounded according to his or her values, and can communicate that preference) should typically be the primary decision maker. However, some patients prefer that goals be set as part of a family decision making process or even that family members take charge of decision making [31,32]. When a patient lacks the capacity to make decisions, the appropriate surrogate needs to be identified. Some patients have formally identified a surrogate by previously completing a durable power of attorney for health care document. If no surrogate has previously been designated, a surrogate can be selected either through stipulations set forth by state law, when such law exists (typically in the order of spouse, adult child, sibling, more distant relative), or informally, by identifying the most involved family member or loved one. Once appropriate decision makers have been identified, arriving at goals of care involves a series of steps that includes clarifying prognosis, identifying achievable and desired outcomes and tradeoffs among them, soliciting the value the patient places on those outcomes and tradeoffs, and deciding on a treatment plan that fits those values. A suggested framework for such a discussion is given in Table 3.

PALLIATIVE CARE ASSESSMENT

33

Table 3 Goals of care discussion framework Domain

Sample intro question

Topics to discuss

Condition and prognosis

‘‘What have the doctors told you about what to expect in the future with regard to your illness?

1. How much the patient knows 2. How much he/she wants to know 3. Information about the condition 4. Addressing emotions

Achievable goals/ tradeoffs

‘‘In situations like this there are a number of approaches one can take, depending on what’s most important to a person. . .’’ Patient values ‘‘What things are you weighing in your mind as we discuss our treatment options? What things are important to you?’’ Recommended plan ‘‘Based on our discussion it of care sounds as if you want your care to focus on________.’’ Proxy decision maker ‘‘Whom would you want to represent your wishes in medical decisions if you were unable to speak for yourself?’’ Unacceptable health ‘‘Can you imagine any states situation of poor quality of life in which you wouldn’t want the doctors to help you live longer?’’ End of life priorities ‘‘If it became clear that you were close to the end of your life, what would be important to you at that time?’’

Different approaches to care based on focus – survival duration, comfort, being at home, and so on Attitudes towards maximizing survival, comfort, being at home, burden, etc. Patient concerns about recommendations Encourage discussion between patient and proxy about end of life preferences

Patient values (eg, towards cognitive/physical function or burden that influence treatment preferences) Views towards comfort versus survival duration when death is near; recommended limits on rx of little utility in dying patients (eg, CPR, intubation)

The first step in discussing goals of care requires that the physician understand the patient’s prognosis for survival as well as functioning, symptoms, and other outcomes, contingent on a given treatment approach. For instance, patients diagnosed with advanced or metastatic esophageal cancer who are functionally well enough to receive chemotherapy have a median survival of approximately 7 to 9 months [33,34], with 1-year survival depending on several measurable factors, including performance status, location of metastases, and serum alkaline phosphatase level [35]. Although the ability to predict a prognosis varies widely across providers and diseases [36,37], a command of published data on prognosis permits the physician insight into the patient’s possible futures based on identified goals of care. Because discussing prognosis often involves breaking difficult news to the patient, the framework for delivering bad news also applies [38] to the goals of care discussion.

34

ROSENFELD

During the next steps in the discussion the physician reviews the range of treatment approaches that may reasonably be pursued, and elicit’s the patient values regarding the outcomes and tradeoffs associated with each approach to care. As discussed above, patients approaching the end of life value a variety of outcomes, including but not limited to survival duration, comfort, being at home, and limiting family burden. Patients also differ widely in the weight they place on these outcomes [39], making it essential to individualize goal setting and treatment decision making. Although the full result of a given set of treatment goals on the patient’s future may be difficult to predict at the outset, some goals of care are associated with a high likelihood of certain outcomes. For instance, a patient whose goals are to prolong survival at all costs is likely to spend his or her final days of life in an intensive care unit, whereas a focus on remaining comfortable at home may well involve foregoing treatment of a disease complication that might prolong a patient’s life for a period of time. Many tradeoffs may not be readily apparent to the patient or even to the physician. For instance, although whole-brain irradiation is considered the standard of care for multiple, symptomatic brain metastases, it generally requires a 2week treatment course involving daily trips to the hospital. In situations in which life expectancy is short or the patient is debilitated, the decision to pursue brain irradiation should entail frank discussions of which valued outcomes may be enhanced and which may be adversely affected by alternative approaches to care. After eliciting the patient’s values toward treatment outcomes, the physician can recommend a plan of care. Experts have argued that given the physician’s expertise in the technical aspects of care, including prognosis and treatment effectiveness, the physician should make treatment recommendations rather than simply providing options to the patient [40]. Recommending a care plan includes reaffirming the primary outcomes on which care will focus, along with how the tradeoffs inherent in choosing those outcomes will be managed. The physician should recommend which treatments should be provided and which should be withheld (often including cardiopulmonary resuscitation or mechanical ventilation) because they are unlikely to serve the overall care goals. The physician should offer opportunities for the patient to express concern or disagreement and educate the patient that he or she is free to alter the treatment plan as clinical circumstances change. The final step in the goals of care discussion entails advance care planning. The present author’s suggestions for advance care planning include a minimum of three elements: (1) the designation of a surrogate decision maker in case of future mental incapacity; (2) the identification of unacceptable health states; and (3) the identification of priorities in cases in which survival duration is expected to be short and life-prolonging medical treatment is required. The latter two elements are useful even when the patient has previously completed a living will because they help to clarify the vague general statements included in living will documents [41]. Examples of questions to introduce these discussion elements are listed in Table 2.

PALLIATIVE CARE ASSESSMENT

35

SOCIAL SUPPORT AND CAREGIVER NEEDS In contrast to most disciplines of medicine, palliative care defines the unit of care to include both the patient and his or her caregivers. Patients facing advanced illness are typically part of a social network of family or other loved ones who provide physical care as well as emotional and spiritual support for the dying person [21]. Terminal decline is accompanied by increasing physical dependence; anticipating and planning for dependence can help mobilize the resources and supports for caregivers that can allow individuals to remain at home until death. Alternatively, unmet physical care needs can lead to physical safety hazards within the home and ultimately to the need for institutional care near the time of death. Screening for social supports can help the physician identify interpersonal issues that may affect treatment decision making, adherence to treatment, and clinical course. Moreover, it will enrich the patient-physician relationship by communicating to the patient the physician’s concern about the disease process and about the patient’s overall illness experience. Screening commonly entails identifying the patient’s living situation, including current and potential caregivers, family dynamics, including discord or stressors on family relationships, financial concerns, and current community supports (eg, church groups and support groups). More detailed assessment by the team social worker is almost always indicated. The social worker can educate patients and caregivers about available benefits and community resources, provide emotional counseling, and facilitate transitions into hospice and, if additional care is needed, from the home-based to an institution-based care setting. Caregivers are affected financially, emotionally, and physically, during a loved one’s illness. In a large study [42] of the seriously ill, one third of patients required active caregiving assistance from a family member. In 20% of cases, a family member had to quit work or make a major life change to accommodate the needs of the patient, and one third of families lost most or all of their life savings providing care for the ill patient. Caregivers of patients who have advanced illness have also been shown to be at high risk for emotional problems, including depression [43], and coping during the bereavement period affects long-term depressive symptoms and quality of life. An assessment of caregiver needs includes screening for financial difficulties, caregiving responsibilities in terms of time spent and perceived physical and emotional burden, emotional adjustment, including changes in the patient-caregiver relationship and roles, and sources of caregiver support. Involvement of the team social worker is usually indicated to assess caregiver needs further and to initiate appropriate support. ANTICIPATORY CONCERNS Patients’ concerns about the future represent a major source of stress, with regard to emotional and existential issues and with regard to practical concerns, including their preferred setting for their end-of-life care and their care after death. In a study of patients, families, and health care providers, Steinhauser

36

ROSENFELD

and colleagues [39] found that dying at home was ranked among the least important overall of nine attributes of a good end-of-life experience, yet there was wide variability across respondents, with some patients feeling a strong preference to die at home and others having a strong desire to die outside of the home. Such variability highlights the importance of active inquiry into patients’ preferred location, if any, of their terminal care. In the same study, the authors found that other practical concerns in preparing for death, for example having funeral arrangements planned, weighed heavily on patients’ minds and should be discussed openly as part of the palliative evaluation. Finally, the physician should be prepared to discuss patient preferences for care of their body after death, including the preference for donation of organs or of the body for scientific purposes. This discussion helps to promote self-determination and facilitates patients’ sense of making a contribution even in death. MEDICATIONS Palliative assessment requires thorough attention to medications, including those prescribed to relieve symptoms, improve quality of life, or extend survival (if desired) and to those that should be discontinued because of toxicity or lack of benefit to the patient. Physical frailty, organ dysfunction, and cognitive decline accompanying advanced illness, often complicated by advanced age or other comorbidities, increase patients’ sensitivity to medication side effects [44]. Polypharmacy also increases with advanced illness complexity and the number of symptoms to be treated. This combination of sensitivity to medication side effects and risk of polypharmacy places the seriously ill patient at high risk for medication toxicity [45] and provides an opportunity to minimize that risk through judicious discontinuation of harmful or unnecessary medications, substitution of better tolerated alternatives, and prescription of lowest risk options as new medications are needed. As part of the medication review, the physician should inquire about symptoms that may indicate medication toxicity, including neuropsychiatric (fatigue and sedation, hallucinations, agitation, and confusion), gastrointestinal (nausea, constipation, diarrhea, and anorexia), and cardiovascular symptoms (orthostasis), recognizing that medication toxicity can often be difficult to distinguish from symptoms of the underlying disease. The physician should pay special attention to medications with anticholinergic (eg, tricyclics for neuropathic pain and certain antiemetics) and sedative (eg, opiates, muscle relaxants, tricyclics, anticonvulsants, and neuroleptics) properties, which risk exacerbating common symptoms, including confusion, dry mouth, constipation, and fatigue. Finally, the physician should seriously consider discontinuing medications being used for primary or secondary disease prevention, which offer little or no benefit over the course of the patient’s remaining life. The benefits of these medications, including statins for hyperlipidemia, antihypertensives for mild or moderate hypertension, and aspirin for cardioprotection, are measured over populations of patients with years of continuous use. Although such medications may carry symbolic meaning for the patient or family, continuing them

PALLIATIVE CARE ASSESSMENT

37

until death carries added expense and risk of toxicity, often without accompanying medical benefit. PHYSICAL EXAMINATION A directed physical examination of the palliative care patient can provide both diagnostic and psychologic benefit. Diagnostically, the examination can help the physician identify the cause of specific physical symptoms. For example, abdominal pain in the patient who has advanced cancer may result from any number of causes, including tumor infiltration of the bowel or adjacent structures, adynamic ileus from opioids, opioid-induced constipation, urinary tract infection, or other causes that may readily be distinguishable through a careful physical examination. The physician should of course be judicious in limiting uncomfortable examination elements to the minimum required to achieve the examination’s overall palliative goals. In addition to its diagnostic value, the physical examination can offer psychologic benefit to the patient and family. At a time in the illness course when cure can no longer be offered, even as the physician makes recommendations to reduce the intensity of medical interventions, the physical examination’s ‘‘laying on of hands’’ offers a potent symbol of nonabandonment. LABORATORY EVALUATION Although a detailed discussion of laboratory and other testing for diagnostic and monitoring purposes is beyond the scope of this article, it should be noted that the decision to perform such testing should be determined largely by the goals of care developed with the patient and family. Patients with palliative goals often suffer from overly zealous testing at the hands of the medical care system. Such testing may be considered ‘‘standard of care,’’ it may provide reassurance to a physician who is ambivalent about the goals of care, or it may be performed simply out of habit. On the other hand, testing is appropriate and, in fact, necessary in situations in which it will help to achieve outcomes that are important to the dying patient and family. Because tests create additional burdens on the patient in terms of comfort, time, and cost, the physician should remain vigilant to how testing will influence patient management. Such influence depends on several factors, including sensitivity and specificity of the test as well as the pretest probability of findings that would lead to a change in clinical management. SUMMARY Patients who have advanced illness present unique challenges to the specialist physician. Gastroenterologic care of this group of patients has changed substantially over the past 2 decades, with diagnostic and therapeutic procedures often supplanting the more communication-based aspects of care. At the same time, given the range of patients’ palliative needs and the increasingly important role of the gastroenterologist in their overall care, the physician has a responsibility to develop basic competencies in palliative medical practice. Skill in palliative

38

ROSENFELD

assessment can help the physician assure that care remains finely attuned to the goals, needs, and priorities of the patient and family. Acquiring and applying palliative care skills will, as a result, lead to both better outcomes for patients and families and a fully satisfying professional experience for the physician. References [1] Cassel EJ. The nature of suffering and the goals of medicine. N Engl J Med 1982;306(11): 639–45. [2] World Health Organization. Cancer pain relief and palliative care. Geneva (Switzerland): World Health Organization; 1990. [3] Singer PA, Martin DK, Kelner M. Quality end-of-life care: patients’ perspectives. JAMA 1999;281:163–8. [4] Steinhauser KE, Clipp EC, McNeilly M, et al. In search of a good death: observations of patients, families, and providers. Ann Intern Med 2000;132:825–32. [5] AGS Ethics Committee. The care of dying patients: a position statement from the American Geriatrics Society. J Am Geriatr Soc 1995;43(5):577–8. [6] Field MJ, Cassel CD, editors. Approaching death: improving care at the end of life. Washington (DC): National Academy Press; 1997. [7] Neuenschwander H, Bruera E. Asthenia. In: Oxford textbook of palliative care. 2nd edition. New York: Oxford University Press; 1998. p. 575. [8] Management of Cancer Pain Guideline Panel. Management of cancer pain: clinical practice guideline no. 9. AHCPR publication no. 94–0592. Rockville (MD): AHPCR; 1994. [9] Saunders C. Care of patients suffering from terminal illness at St. Joseph’s Hospice, Hackney, London. Nursing Mirror 1964;14:vii–x. [10] Marks RM, Sachar EJ. Undertreatment of medical inpatients with narcotic analgesics. Ann Intern Med 1973;78(2):173–81. [11] Cleeland CS, Gonin R, Baez L, et al. Pain and treatment of pain in minority patients with cancer. The Eastern Cooperative Oncology Group Minority Outpatient Pain Study. Ann Intern Med 1997;127:813–6. [12] de Wit R, van Dam F, Abu-Saad HH, et al. Empirical comparison of commonly used measures to evaluate pain treatment in cancer patients with chronic pain. J Clin Oncol 1999;17(4):1280. [13] Bruera E, Kuehn N, Miller MJ, et al. The Edmonton Symptom Assessment System (ESAS): a simple method for the assessment of palliative care patients. J Palliat Care 1991;7(2):6–9. [14] Chang VT, Hwang SS, Kasimis B, et al. Shorter symptom assessment instruments: the Condensed Memorial Symptom Assessment Scale (CMSAS). Cancer Invest 2004;22(4):526–36. [15] Hwang SS, Scott CB, Chang VT, et al. Prediction of survival for advanced cancer patients by recursive partitioning analysis: role of Karnofsky performance status, quality of life, and symptom distress. Cancer Invest 2004;22(5):678–87. [16] Katz S, Ford AB, Moskowitz RW, et al. Studies of illness in the aged: the index of ADL: a standardized measure of biological and psychosocial function. JAMA 1963;185:914–9. [17] Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist 1969;9(3):179–86. [18] Verger E, Salamero M, Conill C. Can Karnofsky performance status be transformed to the Eastern Cooperative Oncology Group scoring scale and vice versa? Eur J Cancer 1992;28A(8–9):1328–30. [19] Karnofsky DA, Burchenal JH. The clinical evaluation of chemotherapeutic agents in cancer. In: MacLeod CM, editor. Evaluation of chemotherapeutic agents. New York: Columbia University Press; 1949. p. 196. [20] Anderson F, Downing GM, Hill J, et al. Palliative performance scale (PPS): a new tool. J Palliat Care 1996;12(1):5–11. [21] Byock IR. The nature of suffering and the nature of opportunity at the end of life. Clin Geriatr Med 1996;12(2):237–52.

PALLIATIVE CARE ASSESSMENT

39

[22] Bruera E, Miller L, McCallion J, et al. Cognitive failure in patients with terminal cancer: a prospective study. J Pain Symptom Manage 1992;7(4):192–5. [23] Folstein M, Folstein S, McHugh S. Mini-mental state: a practical method for grading the cognitive status of patients for the clinician. J Psychiatr Res 1975;12:189–98. [24] Inouye SK, van Dyck CH, Alessi CA, et al. Clarifying confusion: the confusion assessment method: a new method for detection of delirium. Ann Intern Med 1990;113(12):941–8. [25] Chochinov HM, Wilson KG, Enns M, et al. ‘‘Are you depressed?’’: screening for depression in the terminally ill. Am J Psychiatry 1997;154:674–6. [26] Jacobsen PB, Donovan KA, Trask PC, et al. Screening for psychologic distress in ambulatory cancer patients. Cancer 2005;103(7):1494–502. [27] Emanuel LL, Alpert HR, Baldwin DC, et al. What terminally ill patients care about: toward a validated construct of patients’ perspectives. J Palliat Med 2000;3(4):419–31. [28] Curtis JR, Wenrich MD, Carline JD, et al. Understanding physicians’ skills at providing endof-life care: perspectives of patients, families, and health care workers. J Gen Intern Med 2001;16:41–9. [29] Lo B, Ruston D, Kates LW, et al. Discussing religious and spiritual issues at the end of life: a practical guide for physicians. JAMA 2002;287:749–54. [30] Puchalski CM, Romer AL. Taking a spiritual history allows clinicians to understand patients more fully. J Palliat Med 2000;3:129–37. [31] High DM. All in the family: extended autonomy and expectations in surrogate health care decision-making. Gerontologist 1988;28(Suppl 3):S46–51. [32] Available at: http://www.dyingwell.com/landmarks.htm#dev. [33] Kroep JR, Pinedo HM, Giaccone G, et al. Phase II study of cisplatin preceding gemcitabine in patients with advanced oesophageal cancer. Ann Oncol 2004;15(2):230–5. [34] Conroy T, Etienne PL, Adenis A, et al. Vinorelbine and cisplatin in metastatic squamous cell carcinoma of the oesophagus: response, toxicity, quality of life and survival. Ann Oncol 2002;13(5):721–9. [35] Chau I, Norman AR, Cunningham D, et al. Multivariate prognostic factor analysis in locally advanced and metastatic esophago-gastric cancer: pooled analysis from three multicenter, randomized, controlled trials using individual patient data. J Clin Oncol 2004;22(12): 2395–403. [36] Lamont EB, Christakis NA. Physician factors in the timing of cancer patient referral to hospice palliative care. Cancer 2002;94(10):2733–7. [37] Lynn J, Harrell F Jr, Cohn F, et al. Prognoses of seriously ill hospitalized patients on the days before death: implications for patient care and public policy. New Horiz 1997;5(1): 56–61. [38] Baile WF, Buckman R, Lenzi R, et al. SPIKES: a six-step protocol for delivering bad news: application to the patient with cancer. Oncologist 2000;5(4):302–11. [39] Steinhauser KE, Christakis NA, Clipp EC, et al. In search of a good death: observations of patients, families, and providers. Ann Intern Med 2000;132(10):825–32. [40] Emanuel EJ, Emanuel LL. Four models of the doctor-patient relationship. JAMA 1992;267: 2221–6. [41] Brett AS. Limitations of listing specific medical interventions in advance directives. JAMA 1991;266:826. [42] The SUPPORT Principal Investigators. A controlled trial to improve care for seriously ill hospitalized patients: the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT). JAMA 1995;274:1591–8. [43] Schulz R, Mendelsohn AB, Haley WE, et al. End-of-life care and the effects of bereavement on family caregivers of persons with dementia. N Engl J Med 2003;349(20):1936–42. [44] Balducci L, Yates J. General guidelines for the management of older patients with cancer. Oncology 2000;14(11A):221–7. [45] Field TS, Gurwitz JH, Avorn J, et al. Risk factors for adverse drug events among nursing home residents. Arch Intern Med 2001;161(13):1629–34.

Gastroenterol Clin N Am 35 (2006) 41–52

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Conveying Adverse News in End-of-Life Situations M. Bruce Dratler, MDa,b,*, Mary Katherine Burns, RNb, Heather L. Dratlerc a

Northwest Physicians Associates, 1012 Water Street, Meadville, PA 16335, USA West County Hospice of Crawford County, 448 Pine Street, Meadville, PA 16335, USA c Cornell University, College of Agriculture and Life Sciences, 274 Roberts Hall, Ithaca, NY 14853-4203, USA b

T

here is no easy way to convey adverse news to patients and their families in end-of-life situations. However, differences in attitudes and customs during end-of-life situations have been observed across cultures and religions. This article examines such differences in attitudes as they relate to the cultural and religious backgrounds of patients and their families. Once physicians grasp an understanding of cultural differences in end-of-life situations, they can strive to convey adverse news in the most sensitive and respectful way possible. The discussion also considers various approaches to the actual communication of adverse news and incorporates the authors’ personal experiences with different patients to illustrate a broad range of situations. If physicians can learn to convey adverse news to patients and their families in the most sensitive and honest way possible they can help to reduce the fear and uncertainty that is undoubtedly encountered in end-of-life situations.

UNDERSTANDING DIFFERENCES IN ATTITUDES ABOUT DEATH BASED ON CULTURE The population of the United States has come to include a large variety of ethnic groups. Although it is important to avoid stereotypes and generalizations, it is still possible to identify some basic values across these cultural dimensions. Such values have developed over the years as products of the histories, origins, opportunities, standards, and principles of various cultures. Although specific predictions cannot be made for particular patients and families, knowledge of general cultural values can help physicians choose appropriate approaches for communicating with patients and families about their situations.

*Corresponding author. E-mail address: [email protected] (M.B. Dratler). 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.001

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

42

DRATLER, BURNS, & DRATLER

AFRICAN AMERICAN VALUES According to the United States Census of 2000, African Americans account for 12.3% of the American population [1]. However, African Americans account for only 8% of hospice patients [2]. This percentage may derive from the idea that African Americans’ attitudes toward health care stem from their long history of struggle and discrimination in American society [3]. For example, the Tuskegee experiments conducted between 1932 and 1972 explicitly denied access to penicillin treatment for black males who had syphilis. The men were not notified of the existence of a treatment, nor were they informed that the government, in fact, was not administering any treatment at all [4]. It has been suggested that knowledge of this experiment, along with other discriminatory factors, has led African Americans to a general distrust of the health care system that is led by whites. Studies have shown that only 16% of elderly African Americans complete advance directives, compared with 40% of elderly whites [5]. It has been indicated that such directives and donot-resuscitate orders are perceived by African Americans as denying treatment to patients, which is also reminiscent of the Tuskegee experiments [6]. Less use of invasive procedures for heart and lung problems and less access to transplantation procedures have been reported for African Americans, perhaps supporting the idea that they actually cannot completely trust the health care system [7]. It has also been found that African Americans are more interested in prolonging life than easing pain, in comparison with whites [3]. This finding correlates with the general value that African Americans place on suffering as a spiritually meaningful experience [8]. African Americans are typically very religious and family oriented. In times of death, they commonly place great importance on the presence of their families and the expression of emotions [6]. Likewise, African Americans are more likely to value shared decision making between the patient and the family members, as opposed to pure patient autonomy in end-of-life situations [8]. Interestingly, the combination of distrust for the health care system and reluctance of patients to make decisions about their care on their own has led to a lack of communication between terminally ill African American patients and their physicians [3]. Unfortunately, this lack of communication between patient and physician results in the physician choosing the direction of treatment, without patient input. A pattern has been discovered among African Americans that identifies a preference for more aggressive treatments for terminal illnesses [7]. Without communicating such preferences to the physician, African Americans may be left with a treatment that does not fully satisfy them. Bridging the gap between African Americans and their physicians may involve greater physician understanding of the distrust, subsequent reluctance to communicate, and reliance on family members within the African American community. Understanding these factors within African American culture can help physicians to shape their approaches to conveying adverse news.

CONVEYING ADVERSE NEWS IN END-OF-LIFE SITUATIONS

43

ASIAN VALUES Historically, Asian cultures differ from Western cultures along the dimension of collectivism versus individualism [9]. This disparity may become extremely important in making health care decisions. A collectivist society places more emphasis on community and family interdependence, with less importance placed on the well being, advancement, and independence of the individual [9]. Western societies focus more on individual rights and personal gain. Thus, the American medical system generally supports autonomy in patient decision making, truth-telling about diagnoses, and letting the patient make decisions, unless they are declared medically incompetent to do so [3]. This attitude differs greatly from the Asian perspective, in which the family is affected just as much as the individual is by illnesses and therefore should have control over any medical decisions. Asian cultures treat their elders with the utmost respect and reverence [8]. Hence, they often believe that the terminally ill should be protected from the knowledge of their condition. Asian families want their elders to maintain hope and happiness at the end of their lives, so the patients are often excluded from decisions made about their treatment [6], which explains why Asians tend not to sign advance directives [3]. Researchers have found that Asians sign far fewer advance directives than whites do [8]. American doctors need to become aware of the value placed on patient welfare in Asian cultures to be able to better communicate with Asian families about end-of-life situations. Aside from protecting the terminally ill, Asian families also tend to prefer aggressive treatments for their loved ones [8]. They feel an obligation to their elders and thus may feel unfulfilled if every possible treatment is not exhausted. Likewise, elderly Asians have been reported to feel obligated to stay alive, to protect the welfare of the family unit [8]. These basic tendencies of Asian families in end-of-life situations stem from the Asian tradition of honoring elders. However, there are also other tendencies arising from other Asian beliefs that are not necessarily always consistent. In some Chinese cultures, a ‘‘good death’’ is considered one in which the dying patient is surrounded by family members and has a full stomach [6]. Some Chinese families consider it improper to talk about death at all, believing it will bring bad luck [3]. Likewise, in some Filipino cultures, words are considered so powerful that they believe talking about something will make it happen. Thus, they may refuse to discuss the patient’s condition or any medical options [8]. Regardless of these tendencies and attitudes, it has been found that many Asians have assimilated their customs with American culture throughout the generations. Japanese Americans, especially, have demonstrated a great level of heterogeneity in their attitudes toward death [3]. Thus, it is important not to overlook these individual differences among families. With the general knowledge of Asian culture and attitudes, a physician should also interact with the individual families to determine the best method for treatment and care.

44

DRATLER, BURNS, & DRATLER

HISPANIC VALUES The Hispanic culture in America stems from many different Latin nations. Culture has developed from origins in Mexico, Puerto Rico, Cuba, Central America, and South America, to name a few [3]. Each culture, depending on its origins, has a different attitude toward end-of-life care. However, all Hispanic origins tend to share some basic similarities in attitudes. Generally, Hispanic families tend to be very tightly knit units [6,8]. Thus, Hispanics tend to believe that the family should be responsible for making health care decisions [10]. Furthermore, Hispanic families may choose not to inform the patient of his or her terminal diagnosis because of the belief that doing so may eliminate all hope and cause the patient unnecessary distress [11]. Hispanics also tend to disapprove of signing advance directives, powers of attorney, or living wills [11]. This may be because of the patient’s belief that assigning responsibility to one family member may cause a rift between that member and the rest of the family. The patient also may be afraid of offending other family members. Hence, the patient and the family usually choose to make collective family decisions [8]. There is a low rate of Hispanic attendance in hospice programs throughout North America. Some researchers believe this is the result of distrust in the health care system. In addition, language barriers and lack of knowledge about hospice may contribute to this trend [3]. Physicians who speak Spanish may be more effective in communicating with Hispanic families about hospice programs and other issues in end-of-life situations. Many Hispanics are very religious. They may believe in both good and evil spirits that influence the health of family members [3]. Faith may become extremely important during times of illness and death. Furthermore, Hispanics tend to believe death is a very natural part of life. They celebrate the anniversaries of the death of their loved ones every year [3]. It is important to be sensitive to these religious beliefs during treatment and conveying of information. As with all minorities, the degree to which Hispanics will assimilate with the characteristics mentioned above depends on how much they have adopted American culture into their original heritage [11]. It would be wise for physicians to ask Hispanic families questions about treatment preferences. It may also be beneficial for the physician to ask the patients whether they desire the family to handle their care or whether they, in fact, want autonomy.

UNDERSTANDING THE ROLE OF RELIGION IN END-OF-LIFE SITUATIONS The United States has certainly become a melting pot for different religious beliefs. The problem encountered, however, when examining differences in religions, is that there are many differences in attitudes and customs even within the same religion. Generally, however, the basic tenets of most religions involve the belief in some sort of afterlife, the attempt to atone for one’s transgressions, and the endeavor to obey and serve some higher being. In fact, one study

CONVEYING ADVERSE NEWS IN END-OF-LIFE SITUATIONS

45

has shown that patients with strong religious beliefs tend to have less fear and anxiety about their deaths, although, this finding has not been found universally [12]. Even so, this observed effect may be because of the fact that the belief in an afterlife provides people with a definitive answer to what happens after death. It has also been suggested that the fear of death may motivate stronger religious beliefs [12]. One hospice nurse in the authors’ department commented that many people enter the hospice program with weak beliefs and as they progress toward their deaths they strengthen their faiths. This development of beliefs helps patients to cope with the inevitability of their deaths. Furthermore, she suggested that religious leaders can sometimes offer patients better counseling for their situation than the physician or nurse. Many patients, especially of the Christian faith, believe they should confess their sins before death to ensure themselves a place in heaven [13]. The nurse pointed out that religious leaders are often more successful in eliciting fears, feelings, and confessions from patients than are nurses and physicians. In fact, her own father was a member of the hospice program before his death. She related how he had been born with one arm, and it was not until the months approaching his death that he confessed to his minister that he had always felt inadequate because of his disability. He admitted feeling regret that he could not fight alongside his brothers in World War II. After confessing these feelings to the minister, he was able to let his insecurities go and found himself more at peace at the time of his death. Similar to the Christian desire to confess sins before death, Judaism also believes in a form of confession. Judaism maintains that before death, one should ask God for forgiveness for the sins committed against Him and ask for forgiveness from one’s fellow man. Most Jews associate repentance with the high holy days. The 10-day period from the start of Rosh Hashanah to the end of Yom Kippur is known as the Ten Days of Repentance. However, even when accompanied by sincere repentance, one is forgiven only for offenses committed against God, not for sins against man, unless the injured party has been appeased [14]. Islam also encourages a form of meditation before death. Muslims tend to believe that struggling against death is to go against Allah’s will. They often choose to spend their final days praying and contemplating [13]. Thus, as patients of any religion approaches their death, their religious beliefs may play a strong role in the way they cope with adverse news. As noted above, however, not all members of each faith believe in the same specific aspects and traditions that are conveyed in the religion’s law. Physicians, nurses, and religious leaders must get to know the individual patients to determine the best course of action for that patient’s level of faith.

CONVEYING ADVERSE NEWS TO PATIENTS AND THEIR FAMILIES: STUDIES Although each physician undoubtedly will come to develop his or her own style for delivering adverse news, suggestions have been made as to the

46

DRATLER, BURNS, & DRATLER

particular methods, resulting from a variety of research. These suggestions may prove useful because physicians generally have had little training in communication skills [15]. One study, by Lutfey and Maynard [16], introduces the idea of a perspective display series. This method involves the physician approaching the topic of the patient’s diagnosis and impending death in a stepby-step manner, thus enabling the physician to elicit the patient’s perspective and level of understanding. It allows the patient and physician to build on one another’s thoughts until the final topic is encountered. One example given in the study was of a physician attempting to relate adverse news to a patient who had cancer of the gall bladder. The physician approached the subject by asking the patient what his plans were on discharge from the hospital and then asking the patient if anyone had talked to him about the hospice program. The physician here attempted to broach the subject of death by introducing the idea of hospice as an alternative to consider in case the patient did not recover [16]. Posing a hypothetical situation (nonrecovery) allows the patient to analyze the situation and start to plan for it before actually discovering that the situation is real. This advance preparation can make the actual news much less shocking because the patient will be, to some degree, prepared already [16]. The study furthermore suggests personal contact between the physician and patient. For example, the physician in this particular instance told the patient that he was concerned for him while at the same time holding the patient’s hand. This nonverbal contact aids in conveying to the patient that something serious lies ahead [16]. Another study, by Parker and colleagues [17], has examined patients of various ages, genders, and levels of education who have different types and stages of cancer. This study found that the three most important dimensions of conveying adverse news from the patient’s perspective consist of the content of the message, support provided by the physician, and the facilitation and setting of the delivery. This study concluded that the most important aspect of delivery to patients is that their physician possesses an expertise in the subject matter and that they receive detailed information about their diagnosis and treatment options. The authors suggest this may be because knowledge about available treatments provides the patient with hope that something can be done. It is important to note, however, that although the content of the message was rated as most important, the other two elements were still considered very significant [17]. This study also noted demographic differences between patients, such as women and those with more formal education place more emphasis on receiving detailed information, and women place a greater emphasis on emotional support than males do. The study suggests that physicians ask questions to discern the patients’ attitudes about each of the three dimensions to determine the best course of action [17]. Ptacek and Ptacek [18] conducted a study on conveying adverse news in end-of-life situations that correlated well with the study by Parker and colleagues [17]. Ptacek and Ptacek note that three important aspects of conveying news involve what is said, how it is said, and where it is said. They suggest that

CONVEYING ADVERSE NEWS IN END-OF-LIFE SITUATIONS

47

news be delivered in a comfortable setting without interruptions. They recommend that physicians turn off their beepers, sit close to the patient, and empathize with the patient’s emotions. They also suggest letting the patient know that bad news lies ahead, to prepare the patient [18]. This differs slightly from the Lutfey and Maynard study [16], which suggests physicians use nonverbal cues to relate that bad news is imminent. Ptacek and Ptacek recommend using verbal cues instead [18]. Examples of such verbal indicators of bad news include, ‘‘I’m sorry to have to tell you this. . .’’ or ‘‘Unfortunately, I have some bad news. . .’’ [19]. Ptacek and Ptacek furthermore explain that bad news is usually delivered poorly when a physician rushes through the information or struggles to find the right words. This poor delivery can have an upsetting impact on the patient [18]. Another method for delivering adverse news has been developed at the University of Texas. This method, entitled SPIKES, lists a protocol for physicians to deliver bad news to patients [19]. The ‘‘S’’ stands for ‘‘setting up the interview.’’ This portion of the protocol recommends a mental rehearsal for physicians before delivering the news. It suggests privacy, sitting down, involving significant others of the patient’s choice, connecting with the patient by eye or bodily contact, and removing interruptions [19]. A few of these aspects are those mentioned in the study by Ptacek and Ptacek, thus supporting the evidence further. The ‘‘P’’ in SPIKES stands for ‘‘assessing the patient’s perception.’’ This portion of the protocol encourages the physician to use open-ended question to assess how much the patient knows and understands about his or her condition before breaking the news. This allows the physician to clarify any misunderstandings and tailor the information specifically to the patient’s level of knowledge [19]. The ‘‘I’’ in SPIKES stands for ‘‘obtaining the patient’s invitation.’’ This step involves asking the patient at the time of testing how they would like the results to be explained [19]. Knowing ahead of time just how much information the patient wants divulged can help physicians to deliver news in a way that is sensitive to individual patient’s needs. The ‘‘K’’ in SPIKES stands for ‘‘giving knowledge and information to the patient.’’ The authors suggest warning the patient that bad news is coming, as is mentioned in both Lutfey and Maynard’s and Ptacek and Ptacek’s studies. The authors of SPIKES furthermore suggest using nontechnical words and vocabulary that is consistent with the patient’s level of understanding. They warn against delivering all of the information at once and suggest, instead, that it be delivered in portions and checking periodically whether the patient understands everything he or she is being told [19]. The fifth step in SPIKES, ‘‘E,’’ stands for ‘‘addressing the patient’s emotions with empathetic responses.’’ The authors identify four steps for creating an empathetic, supportive response to patients. The first step involves observing any emotion the patient may exhibit, such as crying, a look of sadness, silence, or shock [19]. Then the physician is asked to identify the emotion by naming it to himself or herself and to use open-ended questions to find out exactly what the

48

DRATLER, BURNS, & DRATLER

patient is feeling. The physician must then identify the reason for the emotion, which is usually the bad news. Finally, the physician should connect the emotion with the reason and make a statement showing the patient that he or she is also concerned and understands the patient’s feelings [19]. The final step in SPIKES, ‘‘S,’’ stands for ‘‘strategy and summary.’’ The authors suggest that having a plan for the future reduces the patient’s anxiety and uncertainty surrounding impending death. They warn that physicians should first ask the patient if he or she is ready to discuss such options at that time. Sometimes, patients may wish to have some time to sort out emotions before discussing treatment options [19]. Note that three of the steps in SPIKES, S, K, and E, correlate directly with the three dimensions identified in the study by Parker and colleagues [17]. The S dimension corresponds with facilitation, the K with content, and the E with support. This lends further evidence to the effectiveness of these techniques. Another study, conducted by Wenrich and colleagues [20], recommends that the most important aspect of communication for physicians involves being honest and candid with patients. Information should be delivered in an organized manner, in language that the patient can understand; note that the use of nontechnical language has been suggested in previously mentioned studies, as well. Wenrich and colleagues maintain furthermore that physicians should listen with interest to the concerns and questions of the patient and the family by asking open-ended questions. These questions demonstrate concern for the patient [20]. Patients and their families should be given the opportunity to ask questions in a nonthreatening and unhurried environment, as illustrated also in both the SPIKES method and the Ptacek and Ptacek study [20]. Most importantly, physicians should always maintain an element of hope in their message, even if that hope involves only maximizing the quality of remaining life through the use of a hospice program [20]. The study by Wenrich and colleagues also mentions that the goal of delivering adverse news is to balance honesty with sensitivity so that it is not too blunt or too vague. Finally, it acknowledges that every patient is different and physicians must determine from previous conversations with patients how much information they will want to hear, how quickly, and when [20]. Dias and colleagues [21] used an interview with a woman who had been living with low-grade cancer for many years to discern different characteristics physicians should possess when delivering adverse news. The authors suggest reflection, which means repeating important words the patient uses to give the patient the opportunity to elaborate or ask further questions. They furthermore recommend clarification, to make sure the patient correctly understands the information being relayed to them. They advocate summarizing the information to ensure full comprehension and empathizing with the patient to lend emotional support [21]. Even though these methods and suggestions differ slightly, from one to the next, each study offers information that basically correlates to help physicians develop their own unique methods for conveying adverse news. Another study

CONVEYING ADVERSE NEWS IN END-OF-LIFE SITUATIONS

49

[22] has indicated an approach to teaching physicians these communication skills involving. The study involved pediatric intensive care fellows in a 1-day workshop that enabled them to practice conveying adverse news to mock parents of ill children. The results of the study showed an improvement in the fellows’ performance in conveying such news by the end of the workshop [22]. This approach is just one of the ways in which physicians can work to better deliver bad news to their patients.

CONVEYING ADVERSE NEWS TO PATIENTS AND THEIR FAMILIES: OUR OWN EXPERIENCE Through the personal experiences of the present authors, methods have been developed for conveying adverse news to patients enrolled in the authors’ hospice program. The cornerstone of relaying adverse news is to make sure the patient knows there is always something that can be done and that they are not alone. The goal is to strive to meet patients’ spiritual, emotional, and physical needs. There is always hope. A nurse who helped in the formation of the authors’ local hospice related that this cornerstone belief came about after a patient was referred from a physician from another city. The patient was a woman in her 40s who had terminal colon cancer. Her physician had told her there was nothing he or anyone else could do to help her. He even told her he could not increase her dosage of morphine to help the pain because her tolerance would need to be spared for later when her pain became unbearable. Her condition was deteriorating and, because of her extreme pain, she was unable to care for her children at home. When she was referred, she was assured that there is always something that can be done. She joined the authors’ hospice program, and her pain was able to be controlled. The next weekend, she went with her children to the mall located 30 minutes away. The woman survived for another 9 months. This experience taught that there is always hope to improve a patient’s quality of life, even if there is no cure. The end of her life was made comfortable enough for her to enjoy it with her family. There is never a reason to give up hope, even in the face of adverse news. It is also important to consider options other than hospice. Some patients, when faced with adverse news, are not ready to accept strictly palliative care. One patient who had a diagnosis of metastatic pancreatic carcinoma seemed to the senior author (M.B.D.) to be a perfect candidate for the hospice program, but after speaking with her and her family, hospice was not the appropriate path for her to take at the time. From talking with patients and their families, physicians can get a feel for their attitudes and wishes about their conditions. It was apparent that this person wanted to seek all available treatment options. When various options, including chemotherapy, were offered, she was relieved and said, ‘‘I was afraid you were going to say there was nothing more you could do.’’ This statement illustrates that physicians need the ability to extract patients’ and families’ attitudes and wishes about their treatments and that

50

DRATLER, BURNS, & DRATLER

providing a patient with all available options is always better than saying there is nothing that can be done. Another method used at the authors’ hospice is to begin to prepare patients for adverse news and situations with the advance directive entitled ‘‘Five Wishes’’ [23]. The Five Wishes method covers two medical aspects of patient care as well as three emotional and spiritual aspects [23]. This advance directive has been found to be an excellent tool for introducing the topic of death into the conversation with patients and their families. In addition to broaching subjects of health care, such as appointing a health care agent, the type of health care the patient desires, and other life support issues, the Five Wishes also aids the staff in meeting the emotional needs of patients. It discusses how comfortable the patient wants to be, how the patient wants to be treated, and what the patient wants loved ones to know about him or her [23]. The emotional aspects of this advanced directive have made it a truly useful tool to gracefully introduce the difficult subject of death. Another facet of conveying adverse news involves younger family members. Sometimes families include young children who may not fully understand the concept of death. Many times, these children are overlooked as the patient and older family members deal with their own decisions and emotions. The present authors’ hospice nurses have realized the importance of children’s understanding. When there are younger children involved, the nurses use children’s books that illustrate the concepts of illness and death. These books help the children to comprehend what is happening [24–28]. Through continued use of these children’s books and time spent talking to the children about their questions and fears, the nurses have been able to successfully convey adverse news to young children in the healthiest, least frightening, and most complete way possible. The children benefit greatly from understanding what is going on. Leaving them uninformed about the situation will only lead to confusion and mixed emotions. This approach also helps the adult family members and the patients to know that they can talk openly with the children and even improve communication within their own families. Together, they can face the fears, stresses, and challenges that lie ahead and can find each other to be a warm source of support. SUMMARY It truly is an art to be able to delve into the minds and attitudes of individual patients to find a sensitive yet honest way to deliver adverse news. By taking into account some basic cultural and religious differences and observing individual differences in attitudes between different patients and families, the use of studied techniques for providing adverse news can be tailored to fit each patient’s specific needs. Conveying adverse news will never be simple. End-of-life situations will never be easy. Telling someone something that they do not want to hear will always be stressful. Optimistically, this information and perhaps training programs like the one used with the pediatric intensive care fellows

CONVEYING ADVERSE NEWS IN END-OF-LIFE SITUATIONS

51

can help to reduce the stress and anxiety, for the patient, the family, and the physician, when dealing with the uncomfortable possibility of death. Acknowledgments The authors thank the staff of West County Hospice of Crawford County, the nurses, social workers, aids, and ministers, without whom this report would not have been possible. References [1] US Census Bureau. US Census, 2000. Washington (DC): 2001. [2] National Hospice and Palliative Care Organization. Facts and figures on hospice care in America; 2001. [3] Braun K, Pietsch J, Blanchette P, editors. Cultural issues in end-of-life decision making. Thousand Oaks (CA): Sage Publications Inc; 2000. p. 71–125. [4] Caplan AL. Twenty years after: the legacy of the Tuskegee syphilis study. In: Teays W, Purdy LM, editors. Bioethics, justice and health care. Belmont (CA): Wadsworth-Thomson Learning; 2001. p. 231–5. [5] Hopp FP, Duffy SA. Racial variations in end-of-life care. J Am Geriatr Soc 2000;48:658–63. [6] Ersek M, editor. Core curriculum for the hospice and palliative nursing assistant. Dubuque (IA): Kendall/Hunt; 2003. p. 71–98. [7] Kagawa-Singer M, Blackhall LJ. Negotiating cross-cultural issues at the end-of-life. JAMA 2001;286(23):2993–3001. [8] Searight RH, Gafford J. Cultural diversity at the end of life: issues and guidelines for family physicians. Am Fam Physician 2005;71(3):515–23. [9] Green EGT, Deschamps JC, Paez D. Variation of individualism and collectivism within and between 20 countries: a typological analysis. Journal of Cross Cultural Psychology 2005;36(3):321–39. [10] Blackhall LJ, Murphy ST, Frank G, et al. Ethnicity and attitudes toward patient autonomy. JAMA 1995;274(10):820–5. [11] Berger JT. Cultural discrimination in mechanisms for health decisions: a view from New York. J Clin Ethics 1998;9(2):127–31. [12] Kahoe RD, Dunn RF. The fear of death and religious attitudes and behavior. J Sci Study Relig 1975;14(4):379–82. [13] Doyle D, Hanks GWC, MacDonald N, editors. Oxford textbook of palliative medicine. New York: Oxford University Press; 2001. p. 777–85. [14] Telushkin J. Jewish literacy: the most important things to know about the Jewish religion, its people and its history. New York: William Morrow and Co; 1991. [15] Back AL, Arnold RM, Baile WF, et al. Approaching difficult communication tasks in oncology. CA Cancer J Clin 2005;55:164–77. [16] Lutfey K, Maynard DW. Bad news in oncology: how physician and patient talk about death and dying without using those words. Social Psychology Quarterly 1998;61(4):321–41. [17] Parker PA, Baile WF, de Moor C, et al. Breaking bad news about cancer: patients’ preferences for communication. J Clin Oncol 2001;19(7):2049–56. [18] Ptacek JT, Ptacek JJ. Patients’ perceptions of receiving bad news about cancer. J Clin Oncol 2001;19(21):4160–4. [19] Baile WF, Buckman R, Lenzi R, et al. SPIKES: a six-step protocol for delivering bad news: application to the patient with cancer. Oncologist 2000;5:302–11. [20] Wenrich MD, Curtis JR, Shannon SE, et al. Communicating with dying patients within the spectrum of medical care from terminal diagnosis to death. Arch Intern Med 2001;161: 868–74. [21] Dias L, Chabner BA, Lynch TJ, et al. Breaking bad news: a patient’s perspective. Oncologist 2003;8:587–96.

52

DRATLER, BURNS, & DRATLER

[22] Vaidya VU, Greenberg LW, Patel KM, et al. Teaching physicians how to break bad news. Arch Pediatr Adolesc Med 1999;153:419–22. [23] ‘‘Five wishes’’: Aging with Dignity. Tallahassee (FL); 2001. [24] Boulden J, Boulden J. Someone special is very sick. Weaverville (CA): Boulden Publishing; 1995. [25] Boulden J, Boulden J. Saying goodbye. Santa Rosa (CA): Boulden Publishing; 1992. [26] Channing L. Sad hug, mad hug, happy hug, a children’s story about death. South Deerfield (MA): Channing L Bete Co, Inc; 1994. [27] Klicker RL, Ligammari JJ. Kolie and the funeral. Madison (AL): Guideline Publications; 1988. [28] Doka KJ, editor. Living with grief, children, adolescents, and loss. Washington (DC): Hospice Foundation of America; 2000.

Gastroenterol Clin N Am 35 (2006) 53–64

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Assessment of Quality of Life Outcomes in the Treatment of Advanced Colorectal Malignancies Imran Hassan, MDa, Robert C. Cima, MDa,*, Jeff A. Sloan, PhDb a

Division of Colorectal Surgery, Mayo Clinic, Rochester, MN 55905, USA Division of Health Sciences Research, Mayo Clinic, Rochester, MN 55905, USA

b

T

reatment interventions in patients who have cancer are directed at either curing disease, prolonging life, or controlling symptoms. As with every intervention, however, there are certain associated risks and potential side effects that can adversely affect the patient and his or her quality of life (QOL). When cure is the intent and extending life is the main objective, most patients who have potentially curable malignancies will accept a temporary and even significant decrease in QOL caused by treatment side effects for a potential increase in survival [1]. In situations in which the disease is metastatic, recurrent or locally advanced to the extent that it is considered ‘‘incurable,’’ survival is not an issue. Patients are faced with a completely different dilemma in which there is no hope for cure, and disease-related symptoms or their treatment limit physical and social activities. The realization of incurability can also have a significant psychologic and mental impact [1,2]. In this situation, preserving the quality of the patient’s remaining life is the fundamental objective. As stated by the World Health Organization (WHO), ‘‘Control of pain, of other symptoms, and of psychological, social, and spiritual problems is paramount. The goal of palliative care is the achievement of the best possible QOL for patients and their families’’ [3]. In the setting of palliative treatment for advanced malignancies, the primary goal is to temporarily control the incurable disease and alleviate the distressing symptoms that interfere with activities of daily living and other aspects of functioning, whereas producing minimal toxicity, thus maintaining or improving patient QOL [4]. Therefore, it follows that, in palliative care, preserving QOL in patients who have advanced malignancy is the primary outcome that should be used to determine which interventions are necessary and relatively successful. As a result, the concept of QOL and the issues regarding its valid assessment are central to the development, delivery, and evaluation

*Corresponding author. E-mail address: [email protected] (R.C. Cima). 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.010

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

54

HASSAN, CIMA, & SLOAN

of palliative treatment strategies in patients who have advanced malignancies [5]. DEFINITION OF QUALITY OF LIFE IN THE PALLIATIVE CARE SETTING QOL is a multidimensional construct that is influenced by several factors [4]. The QOL dimensions relevant to palliative care, as suggested by the Institute of Medicine, include the patient’s physical, psychosocial, and spiritual well being and functioning, the patient’s perception of care, and the family’s well being and perceptions [5]. Cella [6] defines QOL as ‘‘the extent to which one’s usual or expected physical, emotional, social, and spiritual well being are effected by a medical condition or its treatment.’’ As such, QOL is considered a strictly subjective perception that is best assessed by the patient. Several studies have shown significant disparities between QOL assessments made by health care providers and those made by patients, in which professional health care providers usually underestimate the actual decrease in patient QOL [7]. This subjective evaluation does not imply that QOL assessment is a ‘‘soft’’ outcome [8]. In fact, the methodological rigor involved in QOL assessment is just as rigorous, if not more so, as it is for objective outcomes such as survival and tumor response [1]. QOL also has been defined as the difference between the patient’s expectations and his or her actual experience. This difference has been supported by studies that have shown that among patients who pragmatically expected adverse effects of their disease or its treatment experienced less suffering when the adverse effects actually occurred. In the palliative setting, this implies that patient QOL can be improved not only by strategies that can change the symptom profile of the disease but also by strategies that can adjust the expectations of the patients to a more pragmatic level. ASSESSMENT OF QUALITY OF LIFE IN PALLIATIVE CARE The good news is that, despite some unique methodological challenges, the assessment of QOL has recently progressed substantially and has had numerous demonstrated successes in the application to cancer generally and colorectal cancer particularly [9–12]. Presently, many guidelines exist for assessing QOL that point clinical oncology researchers to appropriate design and assessments [13–16]. The scientific method applies as equally to QOL assessment as it does to other clinical outcomes, such as tumor response and survival. Several authors have pointed out that the challenges in assessing QOL endpoints are no more difficult, costly, or fraught with measurement error than any other clinical outcome except for survival. The assessment of QOL in palliative care brings additional challenges, albeit surmountable ones [10–12]. Most important is the recognition that patient burden must be considered paramount in such things as the number of questions involved and the timing of the assessment [17]. Sloan and colleagues [17] have demonstrated in repeated applications and settings that the necessary

QOL OUTCOMES IN COLORECTAL MALIGNANCIES

55

information for treatment modification or comparison often can be obtained by the use of simple, brief QOL assessments [17,18]. The key question to be asked in such situations is ‘‘what is needed to know’’ versus ‘‘what would be nice to know.’’ A key component of a QOL assessment is the a priori definition of what will be considered a clinically meaningful benefit to the patient [19]. Much has been written recently on this subject, with considerable advancement made in the available methodology [19–21]. A summary of techniques is available in Guyatt and colleagues’ [22] study, and the evolution of the one-half standard deviation (or 10 points on a 100-point scale) [23] as a reasonable estimate of a meaningful difference makes it now practical and achievable to readily incorporate QOL assessments into virtually any clinical oncology trial. The key to success, as with any other clinical outcome, is to define a meaningful research hypothesis to guide the study design and QOL assessment. As to the statistical methodology for QOL assessment generally and in oncology, although there are specific statistical challenges, such as multiple endpoints and missing data, none requires new statistical methodology [24]. The creation of summary statistics, missing data techniques, and handling of multiple endpoints have all been discussed in detail elsewhere, with satisfying techniques applied successfully [25–27]. POPULAR QUALITY OF LIFE ASSESSMENTS It is estimated that there are now more than 10,000 different QOL assessments available [28]. Hence, the need for designing new QOL assessments or for clinicians to ‘‘make one up’’ for a given study has long since past. The optimal approach today is to define the constructs or domains of QOL that are likely to be affected by a particular disease or treatment and then match the assessment from the pool of 10,000 to the required endpoints. For example, in colorectal cancer, it is typically essential to assess some measure of the patient’s perception of bowel function beyond the number and consistency of stools. There are specific measures, such as the Functional Assessment of Cancer Therapy (FACT)-C [29] for colorectal cancer generally and the FACT-D [30] for diarrhea, the Mayo Bowel Function Questionnaire [31], and the European Organization for Research and Treatment of Cancer (EORTC) colorectal cancer module [32]. All of these measures have been applied successfully in numerous clinical trials. In palliative care, one is often concerned about domains such as fatigue, anxiety, and spiritual well being. There are popular measures for fatigue (eg, Brief Fatigue Inventory [33] and Piper Fatigue Scale [34]). There are more than 100 different measures for anxiety, including the Hospital Anxiety and Depression Scale (HADS) [35], the Profile of Mood States [36], and the simple single-item assessment of anxiety [17]. A review of the various spiritual well being assessments appropriate for cancer patients has been completed recently [37]. One cautionary note is indicated regarding the use of the ubiquitous measures derived from the Medical Outcomes Study, such as the short form (SF)-36 and SF-12 [38,39]. Although these tools have been applied in

56

HASSAN, CIMA, & SLOAN

many cancer-related situations, they have generally been more successful in studies of healthy individuals to detect QOL deficits. In cancer patients, especially those with advanced disease, there are many questions in the SF assessments that are insensitive because of potential irrelevance. For example, there are questions relating to climbing stairs and carrying groceries, which are not likely to be achievable activities of daily living for many cancer patients who have advanced disease. ADVANCED COLON AND RECTAL CANCER Colorectal carcinoma is the second leading cause of cancer-related deaths in the United States, with 130,000 new cases and 57,000 deaths occurring each year. Approximately 20% of the patients who have colorectal cancer present with metastatic disease. Although a small proportion of patients will have potentially resectable hepatic metastases, the majority of patients will have unresectable metastatic disease along with a primary tumor that can be symptomatic or asymptomatic and have a poor prognosis [40,41]. Among the patients who present with localized disease and undergo curative treatment, over half will eventually succumb to metastatic disease [42]. Although the rate of progression of metastatic colorectal cancer is variable, patients who do not undergo any treatment have a median survival of 6 to 9 months from the time of diagnosis, during which they can develop various physical and psychologic symptoms from their disease that can affect and diminish their QOL. PALLIATIVE INTERVENTIONS IN PATIENTS WHO HAVE ADVANCED COLORECTAL CANCER Palliative interventions in patients who have advanced colorectal cancer are directed at disease-related symptoms that can be caused by the primary tumor or the metastatic disease and can be divided into two main groups: 1. Urgent complications, such as abdominal or pelvic pain, obstruction, and perforation from the primary or recurrent disease 2. Chronic constitutional symptoms, such as anemia, fatigue, and weight loss caused by systemic disease burden

Palliative interventions in these patients can therefore be divided into two categories, therapies that treat the systemic disease, such as palliative chemotherapy, and therapies that target the primary or recurrent tumor, such as surgery, endoscopic interventions, and radiotherapy. Determining the most effective palliative strategy in these patients is a difficult and complex task. Although many patients who have advanced colorectal cancer eventually die of disease, the purpose of palliative treatment is to provide relief of both local and systemic symptoms that result from the disease process and to improve the quality of the patient’s remaining life. In this scenario, the side effects or the adverse consequences of any therapy or intervention versus no therapy or intervention at all must be balanced against the reality that the patient will eventually succumb to his or her underlying disease.

QOL OUTCOMES IN COLORECTAL MALIGNANCIES

57

RELEVANCE OF MEASURING QUALITY OF LIFE OUTCOMES IN RELATION TO THERAPY OF ADVANCED COLORECTAL CANCER Patients with advanced colorectal cancer are often severely debilitated because of their underlying disease, and as a result, their tolerance to adverse side effects and complications of therapy are diminished. A large number of these patients are elderly, in their seventh and eighth decades of life, and can have significant comorbidities. However, significant recent advances in the fields of medical and surgical oncology have resulted in the availability of several treatment options for these patients. For instance, chemotherapeutic agents such as oxaliplatin and irinotecans, antiangiogenic agents such as bevacizumab, and oral fluorouracil agents have become available in the management of these patients. Similarly, whereas surgery was once the only palliative option available to deal with the primary disease, the advent of advanced endoscopic techniques has broadened the armamentarium of physicians who take care of these patients. In situations in which different interventions have a similar impact on survival, the outcome to be measured in determining the optimal treatment strategy is patient QOL. However, some of these therapies have been shown to be superior to conventional therapies in terms of their affect on oncologic outcomes. This has given rise to the issue as to which is the best therapy for these patients and what kind of outcomes should be used as a benchmark to determine the benefit or effectiveness of one therapy over the other. In these situations, along with conventional oncologic outcomes, it is essential to assess whether this gain in ‘‘quantity’’ of life is of acceptable ‘‘quality.’’ The following sections discuss the various clinical scenarios that occur commonly in patients who have advanced colorectal cancer, the different palliative interventions that are available, and the available QOL data issues that pertain to these patients. PALLIATIVE CHEMOTHERAPY FOR ADVANCED COLORECTAL CANCER Palliative chemotherapy has been defined as ‘‘the use of antineoplastic medication to affect the cancer and to reduce the adverse signs and symptoms caused either directly or indirectly by the malignant disease process’’ [43]. The objective of palliative chemotherapy in patients who have advanced colorectal disease is to prolong survival and to maintain or improve the patient QOL by systemically treating the effects of the disease. However, chemotherapy in these patients can improve QOL by reducing symptoms, or it can diminish QOL as a result of therapy-related toxic effects [1]. For several years, fluorouracil was the mainstay of palliative chemotherapy regimens. Recently, more potent and effective chemotherapy agents, such as irinotecan and oxaliplatin, oral agents such as capecitabine, and biologic agents such as Bevacizumab, have been developed. Although these agents have resulted in comparable and, in certain situations, even improved median survival compared with traditional fluorouracil-based regimens, they are also associated

58

HASSAN, CIMA, & SLOAN

with much more significant toxicities and side effects, which can have an adverse impact on patient QOL. Best Supportive Care In a randomized trial [44] comparing best supportive care to a combination chemotherapy consisting of 5-fluorouracil, leucovorin, and cisplatin in terms of overall survival and QOL, the median survival was found to be significantly longer for patients receiving chemotherapy (11 months) compared with patients receiving supportive care alone (5 months). Patient QOL was assessed with the functional living index of cancer (FLIC). The FLIC is a 22-item self-report scale developed for patients who have cancer. It provides a single QOL score based on indices of perceived well being, psychologic state, and sociability. Despite mild-to-moderate treatment-related gastrointestinal side effects, QOL was preserved in patients who received chemotherapy, and no differences in QOL scores were seen between the two groups [44]. In another study [45], patients who had metastatic colorectal cancer were randomized to either irinotecan (300–350 mg/m2 every 3 weeks) with supportive care or supportive care alone. Primary endpoints in this trial were overall survival, whereas secondary endpoints included the affect of treatment on performance status, tumor-related symptoms, and patient QOL. QOL was assessed with the EORTC Quality of Life Questionnaire (QLQ-C30) questionnaire (which includes five function scales, one global health status scale, and nine symptom scales). This study showed that treatment with irinotecan and supportive care, compared with supportive care alone, prolonged the survival of patients who had metastatic colorectal cancer (1-year survival probability of 15.4% versus 7.9%), yet the treatment group had fewer tumor-related symptoms and a better QOL. Irinotecan also improved the symptoms of patients and delayed the tumor-related symptoms such as performance status deterioration, weight loss, and pain. Fluorouracil-based Regimens Fluorouracil-based regimens have been the mainstay of chemotherapy treatment of patients who have advanced colorectal cancer for almost 3 decades. In vitro and in vivo evidence has suggested that fluorouracil has a scheduledependent mechanism of action [46]. As a result, various administration regimens were developed that were either bolus-based, such as the North Central Cancer Treatment Group (NCCTG)-Mayo regimen, or infusion-based, such as the de Gramont regimen [47]. In a comparison of these two treatment regimens, the de Gramont regimen had a median survival rate that was similar to the NCCTG regimen but had reduced toxic effects [48]. In a randomized comparison of survival, symptom palliation, and QOL among three chemotherapy regimens, Maughan and colleagues [47] reported that the de Gramont and Lokich regimens were similar in terms of survival, disease response rates, and QOL, but the Lokich regimen was associated with more central line complications and hand-foot syndrome. Raltitrexed therapy showed response rates and overall survival that were similar to other regimens but resulted in greater

QOL OUTCOMES IN COLORECTAL MALIGNANCIES

59

toxicity and inferior QOL. Patient QOL was assessed with the EORTC QLQC30, plus six study-specific questionnaires and the 14-item HAD scale. In a meta-analysis conducted by the Cancer Collaborative Group [41] of palliative chemotherapy trials for advanced colorectal cancer, the majority of which were fluorouracil-based, chemotherapy was shown to prolong the time to disease progression and overall survival; however, the evidence to indicate a benefit on QOL was inadequate. Irinotecan and Oxaliplatin Several phase III trials have now shown that irinotecan with fluorouracil and leucovorin compared with fluorouracil and leucovorin alone improved response rates and progression-free survival, and two of these trials have shown significantly longer overall survival. Although the comparison of fluorouracil and leucovorin alone and in combination with oxaliplatin has demonstrated a higher response rate and progression-free survival, no survival advantage was seen for the combination therapy. The NCCTG trial N9741 [49] was designed to address the question as to which of the two agents, oxaliplatin or irinotecan, should be used as a first-line therapy in metastatic colorectal cancer. In this trial, combination therapies using irinotecan and bolus fluorouracil plus leucovorin (IFL), oxaliplatin and infused fluorouracil plus leucovorin (FOLFOX), and irinotecan and Oxaliplatin (IROX) were compared in patients who had previously untreated metastatic colorectal cancer. The FOLFOX regimen, with a median time to progression of 8.7 months, response rates of 45%, and median survival time of 19.5 months, was found to be superior to the other treatment arms. The use of these agents and even newer agents, such as antiangiogenic agents and oral fluoropyrimidines, are likely to change the management strategies of these patients. Although to date the conventional oncologic endpoints such as response rates, progression-free survival, and overall survival have been evaluated rigorously, QOL outcomes have received much less attention. This outcome is particularly relevant in patients who have advanced disease because these agents have side effects and are associated with toxicities, which can adversely impact patient QOL. Irinotecans are associated with late onset diarrhea, which is dose limiting, whereas the neurotoxicity associated with oxaliplatin limits its use at higher doses and is cumulative. Therefore, although these treatment regimens may offer benefits in terms of oncologic endpoints, the quality of this survival has yet to be assessed. Recently, using data from the N9741 trial [50], it was demonstrated that clinically meaningful changes in simple, single-item patientreported assessments of peripheral neuropathy could identify problems 2 to 3 months before the routinely collected physician-completed National Cancer Institute common toxicity criteria reported an adverse event. SURGICAL VERSUS ENDOSCOPIC INTERVENTIONS FOR OBSTRUCTING COLORECTAL CANCER Patients with metastatic disease that presents with a symptomatic bowel obstruction caused by the primary tumor traditionally underwent either

60

HASSAN, CIMA, & SLOAN

a laparotomy for resection of the primary tumor, an intestinal bypass, or a nonfunctioning stoma to alleviate the obstructing symptoms. This operative approach was associated with significant postoperative morbidity (20%–30%) and mortality (5%–10%). A contemporary surgical series of 120 patients who had advanced colorectal disease reported a 5.8% postoperative mortality rate and a 20% complication rate, in which the most common complication was a wound infection. During the last few years, self-expanding metal stents have been used increasingly in such patients. Most of the reports on metal stents have reported its efficacy in palliating intestinal obstruction, with minimal procedure-related complications. A cohort study that examined outcomes after stent placement and a laparotomy did not show any difference in survival between the two groups [51]. Patients who underwent stent procedures had no postprocedural complications and were discharged earlier from the hospital than patients who underwent a laparotomy. The authors speculated that patients undergoing a stent placement recovered earlier to their premorbid state, although no formal evaluation was performed to measure this outcome. A systemic review of the efficacy and safety of colorectal stents has reported a 16% incidence of obstruction among patients undergoing stent placement for palliation, although Hunberbein and colleagues [52] reported that 18% of their patients who underwent stent placement for malignant rectal obstruction subsequently required surgical palliation because of the failure of stent treatment. Both surgical and endoscopic modalities aim at relieving the symptoms of intestinal obstruction. Clearly, although surgery is a definitive option, it is associated with significant postoperative complications that can potentially influence patient QOL, whereas stenting, although less morbid, has a significant incidence of malfunction. How these factors affect patient QOL has not been assessed. MULTIMODALITY THERAPY FOR RECURRENT RECTAL CANCER Patients undergoing potentially curative resection for rectal cancer have a 4% to 30% risk of developing locoregional pelvic recurrence, with 40% to 50% of these occurring in the absence of systemic disease [53,54]. Without treatment, the median survival of such patients is 3.5 to 6 months and is associated with debilitating pelvic and perineal pain, bowel dysfunction, and presumed impairment of QOL. Recently, with a multimodality approach that includes radical surgery and intraoperative radiotherapy (IORT), 5-year survival rates of 21% to 58% have been reported [53–57]. As a continuation of this benefit, it is assumed that multimodality treatment also improves functional outcomes and restores QOL by definitively treating recurrent pelvic disease. Critics of this approach, however, would argue that the improvement in survival is obtained at a cost of diminished QOL because the significant postoperative morbidity and disability associated with this approach negates any quantitative survival advantage. The potential causes of this impairment in QOL after multimodality treatment can be multifactorial and includes high-dose preoperative radio- and

QOL OUTCOMES IN COLORECTAL MALIGNANCIES

61

chemotherapy, radical surgery, and IORT. The pelvic recurrences usually involve the adjacent viscera or bony structures in the pelvis; therefore, curative surgery often requires the resection of the adjacent organs such as the bladder or neurovascular structures and creation of ostomies and bone resections, all of which can affect functional outcome. Peripheral nerve neuropathy is the main dose-limiting side effect of IORT, whereas External Beam Radiation Therapy is known for its long-term adverse effects on bowel and urogenital function [58]. A review of the literature suggests that there are limited data on the functional outcomes and QOL of patients who undergo therapy for recurrent rectal cancer [59–62]. In fact, little is known about the impact of pelvic recurrence of rectal cancer on patient QOL and the subsequent affect on QOL after multimodality treatment. More work is needed, particularly in this area. SUMMARY QOL assessment in oncology has made great strides in recent years. There was a difficult time initially, during which QOL tools were ‘‘thrown in’’ to many clinical trials as an afterthought, without a pre-specified scientific question. As expected from such a scattershot approach, the results were underwhelming and disappointing. The disappointing results from this period led many practitioners to question the value added by QOL assessment in oncology clinical trials. This healthy skepticism has led to a renaissance period, in which situation-specific and disease-specific QOL assessments have been developed and have contributed substantial information to the cause of the disease, the effects of treatments, and the experiences of cancer patients. Today, there is a dawning recognition that asking the patient directly about their QOL using the same scientific rigor required of other clinical outcomes can provide valuable data for prognosis, treatment, symptom management, and supportive care. With time and further successful experiences like those cited in this article, QOL assessment may eventually become as routinely collected and integrated into oncology clinical practice as pain and blood pressure assessments are today. References [1] Michael M, Tannock IF. Measuring health-related quality of life in clinical trials that evaluate the role of chemotherapy in cancer treatment. CMAJ 1998;158(13):1727–34. [2] Osse BH, Vernooij MJ, Schade E, et al. Towards a new clinical tool for needs assessment in the palliative care of cancer patients: the PNPC instrument. J Pain Symptom Manage 2004;28(4):329–41. [3] World Health Organization. Technical report series 804. In: Cancer Pain and Palliative Care. Geneva: World Health Organization; 1990. p. 11. [4] Jacobsen PB, Weitzner MA. Evaluation of palliative endpoints in oncology clinical trials. Cancer Control 1999;6(5):471–7. [5] Brady MJ, Cella D. Assessing quality of life in palliative care. Cancer Treat Res 1999;100: 203–16. [6] Cella DF. Quality of life: the concept. J Palliat Care 1992;8(3):8–13. [7] Whalen GF, Ferrans CE. Quality of life as an outcome in clinical trials and cancer care: a primer for surgeons. J Surg Oncol 2001;77(4):270–6.

62

HASSAN, CIMA, & SLOAN

[8] Frost MH, Sloan JA. Quality of life measurements: a soft outcome—or is it? Am J Manag Care 2002;8(Suppl 18):S574–9. [9] Cella DF. Quality of life outcomes: measurement and validation. Oncology 1996;10 (Suppl 11):S233–46. [10] Cohen SR. Defining and measuring quality of life in palliative care. Journal of Palliative Care 2001;5:137–56. [11] Dijkers M. Measuring quality of life: methodological issues. Am J Phys Med Rehabil 1999;78(3):286–300. [12] Ferrell BR, Dow KH, Grant M. Measurement of the quality of life in cancer survivors. Qual Life Res 1995;4(6):523–31. [13] National Institutes of Health Consensus Development Program. Draft statement, symptom management in cancer: pain, depression, and fatigue. Statement of the Science Conference: July 15–17, 2002. [14] Chassany O, Sagnier P, Marquis P, et al. Patient reported outcomes: the example of health related quality of life - a european guidance for the improved integration of HRQoL assessment in the drug regulatory process. Drug Inf J 2002;36:209–38. [15] Fayers PM, Hopwood P, Harvey A, et al. Quality of life assessment in clinical trials–guidelines and a checklist for protocol writers: the UK medical research council experience. Eur J Cancer 1997;33(1):20–8. [16] Sloan JA, Cella D, Frost M, et al, for the Clinical Significance Consensus Meeting Group. Assessing clinical significance in measuring oncology patient quality of life. Mayo Clin Proc 2002;77:367–70. [17] Sloan JA, Aaronson N, Cappelleri JC, et al, for the Clinical Significance Consensus Meeting Group. Assessing the clinical significance of single items relative to summated scores. Mayo Clin Proc 2002;77:479–87. [18] Sloan JA, Loprinzi CL, Kuross SA, et al. Randomized comparison of four tools measuring overall quality of life in patients with advanced cancer. J Clin Oncol 1998;16:3662–73. [19] Sloan J, Symonds T, Vargas-Chanes D, et al. Practical guidelines for assessing the clinical significance of health-related quality of life changes within clinical trials. Drug Inf J 2003;37:23–31. [20] Sloan JA. Assessing the minimally clinically significant difference: scientific considerations, challenges and solutions. Journal of Chronic Obstructive Pulmonary Disease 2005;2: 57–62. [21] Sloan JA, Vargas-Chanes D, Kamath CC, et al. Detecting worms, ducks and elephants: a simple approach for defining clinically relevant effects in quality-of-life measures. J Cancer Integrative Medicine 2003;1(1):41–7. [22] Guyatt GH, Osoba D, Wu AW, et al, for the Clinical Significance Consensus Meeting Group. Methods to explain the clinical significance of health status measures. Mayo Clin Proc 2002;77(4):371–83. [23] Norman GR, Sloan JA, Wyrwich KW. The truly remarkable universality of half a standard deviation: confirmation through another look. Expert Review of Pharmacoeconomics and Outcomes Research 2004;4(5):515–9. [24] Sloan JA, Dueck A. Issues for statisticians in conducting analyses and translating results for quality of life end points in clinical trials. J Biopharm Stat 2004;14(1):73–96. [25] Sloan JA, Dueck A, Frost M, et al. Solutions for assessing quality of life in cancer research and clinical practice: part 1. Curr Probl Cancer 2005;29:265–351. [26] Osoba D, Bezjak A, Brundage M, et al, for the Quality of Life Committee of the NCIC CTG. Analysis and interpretation of health-related quality-of-life data from clinical trials: basic approach of The National Cancer Institute of Canada Clinical Trials Group. Eur J Cancer 2005;41(2):280–7. [27] Bottomley A, Vanvoorden V, Flechtner H, et al, for the EORTC Quality of Life Group EORTC Data Center. The challenges and achievements involved in implementing quality of life research in cancer clinical trials. Eur J Cancer 2003;39(3):275–85.

QOL OUTCOMES IN COLORECTAL MALIGNANCIES

63

[28] Osoba DA. Taxonomy of the uses of health-related quality-of-life instruments in cancer care and the clinical meaningfulness of the results. Med Care 2002;40(Suppl 6):SIII-31–III-8. [29] Cella DF, Tulsky DS, Gray G, et al. The functional assessment of cancer therapy scale: development and validation of the general measure. J Clin Oncol 1993;11:570–9. [30] Yellen SB, Cella DF, Webster K, et al. Measuring fatigue and other anemia-related symptoms with the functional assessment of cancer therapy (FACT) measurement system. J Pain Symptom Manage 1997;13(2):63–74. [31] Kozelsky TF, Meyers GE, Sloan JA, et al. Phase 3 double-blind study of glutamine versus placebo for the prevention of acute diarrhea in patients receiving pelvic radiation therapy. J Clin Oncol 2003;21(9):1669–74. [32] De Haes J, Curran D, Young T, et al. Quality of life evaluation in oncological clinical trials: the EORTC model. Eur J Cancer 2000;36:821–5. [33] Mendoza TR, Wang XS, Cleeland CS, et al. The rapid assessment of fatigue severity in cancer patients: use of the brief fatigue inventory. Cancer 1999;85(5):1186–96. [34] Ahlberg K, Ekman T, Gaston-Johansson F, et al. Assessment and management of cancerrelated fatigue in adults. Lancet 2003;362:640–50. [35] Bjelland I, Dahl AA, Tangen Haug T, et al. The validity of the hospital anxiety and depression scale: an updated literature review. J Psychosom Res 2002;52:69–77. [36] McNair DM, Lorr M, Droppleman LF. Edits manual for the profile of mood states [revised 1992]. San Diego (CA): Edits/Educational and Industrial Testing Service; 1992. [37] Sloan JA, et al. Quality of life, spirituality, and mayo franciscan values: a research perspective. Mayo Clin Proc, in press. [38] Ware JE Jr, Gandek V, Kosinski M, et al. The equivalence of SF-36 summary health scores estimated using standard and country-specific algorithms in 10 countries: results from the IQOLA project. J Clin Epidemiol 1998;51(11):1167–70. [39] Ware JE Jr, Kosinski M, Keller SD. A 12-item short-form health survey. Med Care 1996;34: 220–33. [40] Rosen SA, Buell JF, Yoshida A, et al. Initial presentation with stage IV colorectal cancer: how aggressive should we be? Arch Surg 2000;135(5):530–5 [discussion: 534–5]. [41] Amersi F, Stamos MJ, Ko CY. Palliative care for colorectal cancer. Surg Oncol Clin N Am 2004;13(3):467–77. [42] Simmonds PC, for the Colorectal Cancer Collaborative Group. Palliative chemotherapy for advanced colorectal cancer: systematic review and meta-analysis. BMJ 2000;321(7260): 531–5. [43] Ellison NM. Palliative chemotherapy. Am J Hosp Palliat Care 1998;15(2):93–103. [44] Scheithauer W, Rosen H, Kornek GV, et al. Randomised comparison of combination chemotherapy plus supportive care with supportive care alone in patients with metastatic colorectal cancer. BMJ 1993;306(6880):752–5. [45] Cunningham D, Pyrhonen S, James RD, et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998;352(9138):1413–8. [46] Sobrero AF, Aschele C, Bertino JR. Fluorouracil in colorectal cancer–a tale of two drugs: implications for biochemical modulation. J Clin Oncol 1997;15(1):368–81. [47] Maughan TS, James RD, Kerr DJ, et al. Comparison of survival, palliation, and quality of life with three chemotherapy regimens in metastatic colorectal cancer: a multicentre randomised trial. Lancet 2002;359(9317):1555–63. [48] de Gramont A, Bosset JF, Milan C, et al. Randomized trial comparing monthly low-dose leucovorin and fluorouracil bolus with bimonthly high-dose leucovorin and fluorouracil bolus plus continuous infusion for advanced colorectal cancer: a French intergroup study. J Clin Oncol 1997;15(2):808–15. [49] Goldberg RM, Sargent DJ, Morton RF, et al. A randomized control trial of fluorouracil plus leucovorin, irinotecan, and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004;22:23–30.

64

HASSAN, CIMA, & SLOAN

[50] Morton RF, Sloan JA, Grothey A, et al. A comparison of simple single-item measures and the common toxicity criteria in detecting the onset of oxaliplatin-induced peripheral neuropathy in patients with colorectal cancer. ASCO. Orlando, Florida; May 13-17, 2005. Abstract 8087. [51] Carne PW, Frye JN, Robertson GM, et al. Stents or open operation for palliation of colorectal cancer: A retrospective, cohort study of perioperative outcome and long-term survival. Diseases of the Colon & Rectum 2004;47(9):1455–61. [52] Hunerbein M, Krause M, Moesta KT, et al. Palliation of malignant rectal obstruction with selfexpanding metal stents. Surgery 2005;137:42–7. [53] Mannaerts GH, Martijn H, Crommelin MA, et al. Intraoperative electron beam radiation therapy for locally recurrent rectal carcinoma. Int J Radiat Oncol Biol Phys 1999;45(2): 297–308. [54] Miller AR, Cantor SB, Peoples GE, et al. Quality of life and cost effectiveness analysis of therapy for locally recurrent rectal cancer. Dis Colon Rectum 2000;43(12):1695–703. [55] Mannaerts GH, Rutten HJ, Martijn H, et al. Comparison of intraoperative radiation therapycontaining multimodality treatment with historical treatment modalities for locally recurrent rectal cancer. Dis Colon Rectum 2001;44(12):1749–58. [56] Lindel K, Willett CG, Shellito PC, et al. Intraoperative radiation therapy for locally advanced recurrent rectal or rectosigmoid cancer. Radiother Oncol 2001;58(1):83–7. [57] Hahnloser D, Haddock MG, Nelson H. Intraoperative radiotherapy in the multimodality approach to colorectal cancer. Surg Oncol Clin N Am 2003;12(4):993–1013. [58] Shoup M, Guillem JG, Alektiar KM, et al. Predictors of survival in recurrent rectal cancer after resection and intraoperative radiotherapy. Dis Colon Rectum 2002;45(5):585–92. [59] Wiig JN, Poulsen JP, Tveit KM, et al. Intra-operative irradiation (IORT) for primary advanced and recurrent rectal cancer: a need for randomised studies. Eur J Cancer 2000;36(7): 868–74. [60] Mannaerts GH, Rutten HJ, Martijn H, et al. Effects on functional outcome after IORT-containing multimodality treatment for locally advanced primary and locally recurrent rectal cancer. Int J Radiat Oncol Biol Phys 2002;54(4):1082–8. [61] Esnaola NF, Cantor SB, Johnson ML, et al. Pain and quality of life after treatment in patients with locally recurrent rectal cancer. J Clin Oncol 2002;20(21):4361–7. [62] Shibata D, Guillem JG, Lanouette N, et al. Functional and quality-of-life outcomes in patients with rectal cancer after combined modality therapy, intraoperative radiation therapy, and sphincter preservation. Dis Colon Rectum 2000;43(6):752–8.

Gastroenterol Clin N Am 35 (2006) 65–82

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Endoscopic Approaches for Palliation of Luminal Gastrointestinal Obstruction Douglas G. Adler, MD*, Shehzad N. Merwat, MD Department of Internal Medicine, Division of Gastroenterology and Hepatology, University of Texas–Houston Medical School, MSB 4.234, 6431 Fannin, Houston, TX 77030, USA

A

s of this writing, much of the workload of a typical gastroenterologist is devoted to screening patients for gastrointestinal malignancies. Efforts such as colorectal cancer screening via colonoscopy and endoscopic surveillance of patients with Barrett’s esophagus are widespread and widely endorsed. These programs go forward in hopes of diagnosing patients with malignancies while the lesions are still treatable and curable. Despite such efforts, advanced gastrointestinal cancers still are encountered frequently in clinical practice. The malignancies often lead to obstructive symptoms in the gastrointestinal tract, either by direct blockage of the luminal gut (esophageal cancer leading to dysphagia) or by extrinsic compression (pancreatic cancer causing duodenal compression and malignant gastric outlet obstruction [GOO]). In recent years, the armamentarium of endoscopy has broadened considerably and now affords physicians a variety of nonsurgical means to palliate malignant obstruction of the gastrointestinal tract. This article reviews endoscopic techniques to treat malignant esophageal, biliary, small bowel, and colonic obstruction. ENDOSCOPIC PALLIATION OF ESOPHAGEAL OBSTRUCTION Esophageal cancer is the seventh leading cause of cancer-related death in the United States and worldwide. The incidence of adenocarcinoma of the esophagus is increasing in most developed countries [1]. Despite advances in therapy, more than 50% of esophageal cancers are incurable at presentation, and the 5year survival rate is less than 10%. Palliative therapy for dysphagia is a primary concern in the care of these patients. An ideal palliative therapy would result in complete resolution of dysphagia relatively quickly and with minimal complications. Many therapeutic options exist, although all have limitations. The relief of dysphagia often is measured subjectively by the use of a ‘‘dysphagia score.’’ Most studies use a dysphagia score similar to, if not the same as,

*Corresponding author. E-mail address: [email protected] (D.G. Adler). 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.004

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

66

ADLER & MERWAT

the system described by Mellow and Pinkas [2]: A score of 0 denotes the ability to eat a normal diet; 1, the ability to eat some solid food; 2, the ability to eat semisolids only; 3, the ability to swallow liquids only; and 4, complete dysphagia. Endoscopic Methods Alcohol injection Direct injection of pure ethanol into malignant tissue is the simplest and least expensive technique that can be used to recanalize an obstructed esophagus. Alcohol can be injected under endoscopic visualization to cause tumor ulceration and necrosis. It has proved to be an effective modality in the relief of malignant dysphagia. Significant relief of dysphagia was found in two uncontrolled trials [3,4]. In a randomized, controlled trial comparing neodymium:yttrium-aluminum-garnet (Nd:YAG) laser with ethanol injection, the dysphagiafree interval was 37 days and 30 days. An improvement in the dysphagia score of at least 2 points was noted in 88% of the laser group and 78% of the ethanol group, and there was no difference in median survival [5]. Laser therapy Laser therapy involves the use of laser light to coagulate tissue under endoscopic guidance. The most common device used in endoscopy is the Nd:YAG laser. It has proved to be an effective modality in the treatment of malignant esophageal obstruction, and success rates of 75% to 91% have been reported [6–8]. Repeat treatment (sometimes frequent) is necessary in about 50% of all patients [6,7]. Complications include perforation, bleeding, and the formation of tracheoesophageal fistulas and occur in 4% to 20% of patients [9]. Laser therapy involves the use of expensive hardware that requires periodic maintenance, and the overall cost of laser treatment is the same as, if not greater than, treatment with other modalities, including stenting [10,11]. As such, laser therapy is performed less commonly than in years past. Laser therapy can be used, however, for proximal esophageal lesions for which stenting is not feasible. Argon plasma coagulation Argon plasma coagulation is a type of monopolar coagulation based principally on the effects of high-frequency electrical current flowing through a conductive, ionized argon medium. This technique destroys tissue to a depth of 2 to 3 mm and is most useful in superficial lesions. Several studies have been performed to assess its effectiveness in the palliation of malignant dysphagia. In one restrospective study of 32 patients, recanalization was achieved in 89% of patients [12]. A separate report of 83 patients found a similar recanalization rate of 86% [13]. Most of these patients required multiple sessions to maintain patency, averaging five to six sessions per patient, usually at an interval of 3 to 4 weeks [14,15]. Perforation was seen in 1% to 1.8% of procedures, a rate comparable to that seen in other modalities [14,15]. Argon plasma coagulation seems to be a safe and easy alternative to laser treatment; further prospective trials are needed for comparison.

ENDOSCOPIC PALLIATION

67

Photodynamic therapy Photodynamic therapy is an ablative technique using photosensitive agents that are administered to the patient and activated by a specific wavelength of light. As the photosensitive agent is activated, it generates cytotoxic oxygen species, which induce cell necrosis and death. The photosensitive agent has a tropism for faster growing malignant cancer cells or simply may be retained in these tissues owing to poor lymphatic drainage. Activation causes tumor necrosis resulting in a debulking effect that restores luminal patency. Response rates were measured by a subjective improvement in the patient’s dysphagia score in several uncontrolled trials, with patient-reported response rates varying from 70% to 100% [14–17]. The two largest studies narrow this range, however, to 70% to 85% [14,17]. Nineteen percent to 38% of patients required a second photodynamic therapy course, and the dysphagia-free interval ranged from 66 to 80 days [15,17]. In a randomized, controlled trial, 108 patients who underwent Nd:YAG laser therapy were found to have a perforation rate of 7% compared with a 1% perforation rate in 110 patients who underwent treatment with photodynamic therapy (P <.05). Objective tumor response also was equivalent at week 1, but at 1 month was 32% after photodynamic therapy and 20% after Nd:YAG laser (P <.05). Other efficacy end points showed no significant difference between groups. The time to palliation failure was 34 days for photodynamic therapy and 42 days for Nd:YAG laser. The median survival was not significantly different at 123 days for photodynamic therapy and 140 days for Nd:YAG laser [18]. Photodynamic therapy is a relatively expensive treatment modality, and side effects include esophageal stricture (2–5%), Candida esophagitis (2–3%), pleural effusion (3–4%), and photosensitizer-induced skin injury (5–10%). Patients should wear protective clothing and avoid sunlight because most sunscreens block out only UV light, but not damaging infrared light [19,20]. Self-expanding metal stents Several studies have shown the superiority of self-expanding metal stents (SEMS) over rigid plastic stents, and as such rigid plastic stents are now essentially obsolete [21–25]. Metal stents may be uncovered metal meshes or coated by a silicone or polyurethane coating. Uncovered stents face the disadvantage of tumor ingrowth and subsequent obstruction. Both types of stents are at risk for tumor overgrowth at their ends. Some early covered stent designs had relatively high rates of migration because the silicone or polyurethane coating prevented stent integration into the esophageal wall [26,27]. Several methods have been employed to decrease the risk of migration, including adding a proximal enlarged flange, adding barbs, and leaving the proximal and distal ends of the stent uncovered. A more recent prospective, randomized, controlled trial showed a migration rate of 2 of 30 patients with uncovered stents compared with 4 of 32 patients with covered stents (P ¼ .44) [28]. Overall, reintervention rates may be lower for covered stents as opposed to noncovered stents [28,29]. Several different brands and types of SEMS exist, each with unique features aimed at prevention of complications. Comparisons between different brands

68

ADLER & MERWAT

of metallic stents have found similar results in terms of dysphagia relief, technical success, and complications [30–33]. Other variations in stent design have been implemented to prevent gastroesophageal reflux in cases of stents placed at the gastroesophageal junction. The Dua stent (Wilson-Cook Medical Inc., Winston-Salem, North Carolina) contains a windsock-type valve to prevent reflux. Dysphagia scores were reduced from baseline in patients who received this stent, but further studies need to be performed comparing it with standard SEMS to assess its value further [34,35]. Self-expanding plastic stents A self-expanding plastic stent (SEPS) has become available and can be used in the palliation of esophageal obstruction (Polyflex Stent; Boston Scientific, Natick, Massachusetts). This stent is composed of polyester monofilaments covered completely on the inside with a silicone membrane (Fig. 1). This composition produces a smooth inner surface, allowing for an easier passage of food, and a rough outer mesh, which helps to keep the stent in place and potentially prevent migration. The silicone membrane also prevents tumor ingrowth. The proximal end of the stent dilates to a diameter greater than the remainder of the stent, which aids in preventing stent migration. Radiopaque material is used at the proximal, mid, and distal parts of the stents to help identification during fluoroscopy and on subsequent radiography. The Polyflex stents have several features that may be advantageous over metal stents. The stent can be removed, even after it has been in place for several months; this allows it to be used in the management of benign esophageal strictures and tracheoesophageal fistulas. An uncontrolled trial found SEPS placement curative in 17 of 21 patients for various benign esophageal conditions, including refractory peptic, caustic, postradiotherapy, and anastomotic stenoses [36]. The device also has been used to close a benign tracheoesophageal fistula successfully [37]. Disadvantages include a relatively complicated,

Fig. 1. Polyflex SEPS.

ENDOSCOPIC PALLIATION

69

thick, and rigid deployment system. The application device has an outside diameter of 10 mm, and in some cases dilation of the stenosis is required before the stent can be placed, which carries with it the risk of perforation, bleeding, or tumor fracture. Several controlled trials show that SEPS are safe and effective for malignant esophageal strictures and have an acceptable technical success rate [38,39]. Migration rates for the Polyflex stent range from 6% to 25% [38–41]. Repeat interventions were performed in 21% to 25% of cases and were performed most commonly for tumor overgrowth and stent migration [38,39,41]. No major complications were associated with the SEPS. Further prospective trials comparing SEMS with SEPS are warranted. ENDOSCOPIC PALLIATION OF BILIARY OBSTRUCTION The palliation of malignant obstructive jaundice is discussed in more detail in another article in this issue. Obstructive jaundice leads to pruritus, fat malabsorption, and decreased appetite and puts the patient at high risk for cholangitis. Palliation can be achieved by surgery or endoscopic modalities, most commonly stenting. Plastic stents are an effective alternative to open palliative surgical bypass, with lower procedure-related mortality (3% versus 14%; P ¼ .01), major complication rate (11% versus 29%; P ¼ .02), and median total hospital stay (20 days versus 26 days; P ¼ .001) [42]. A study comparing cost of endoscopic stenting, in which most stents placed were polythene (31 of 33), showed stenting to be of significantly lower cost than surgery with the median total lifetime cost for surgical therapy being $60,986 versus $24,251 for endoscopic therapy [43]. Plastic stents frequently become occluded by a bacterial biofilm, often needing replacement every 3 to 4 months. Larger stents of 10F or 11.5F diameter take two to three times longer to occlude than 7F or 8F diameter stents [44,45]. Plastic stents can be made of either a polyethylene polymer or polytef (Teflon). Studies have not found a significant difference in vivo between the two different materials in terms of maintaining patency [46–48]. SEMS also are available for the stenting of biliary strictures. Although SEMS are much more expensive than plastic stents, they expand to a larger diameter and stay patent for a longer time [49,50]. Increased patency duration and decreased stent failure rates mean that the global cost for SEMS is less than that associated with plastic stents [50–52]. In patients with shorter expected survival, a single procedure placing a plastic stent may be more cost-effective [51,53,54]. To protect against tumor ingrowth, covered SEMS have been developed. Initial studies indicate that covered stents may have longer patency compared with uncovered stents [55–58]. A prospective, randomized, controlled trial comprising 112 patients with unresectable distal biliary malignancies showed a significantly higher stent patency for covered stents. In the covered stent group, stent occlusion occurred in 8 patients (14%) after a mean of 304 days, whereas in the uncovered group, 21 patients (38%) had stent occlusion after a mean of 166 days (P ¼ .003) [59]. Covering biliary stents opens the possibility

70

ADLER & MERWAT

of stent removal, and in one study, stent removal was successful in 17 of 18 patients [60]. Indications for stent removal included migration, malpositioning, bleeding, and cholestasis. A disadvantage of covered stents is that they cannot be used in the intrahepatic bile ducts because of the risk of occluding the pancreatic, cystic, or intrahepatic duct orifices, whereas uncovered stents can be placed in any section of the biliary tree [53,61]. Stent migration is another potential complication that may occur with covered stents. Better documented studies are needed to assess this risk further [59,62]. Several different types of SEMS are available. The Wallstent is commonly used and may be covered or uncovered. It has the advantage of being reconstrainable, so if positioning is inaccurate, it may be repositioned. The JoStent SelfX (Abbott Vascular Devices, Redwood City, California) and Zilver stent (Wilson-Cook Medical Inc.), through a design innovation, shorten minimally as they expand (<10%), allowing for more accurate initial positioning. No single type of SEMS has been proven to be more efficacious than others. A more recent development in stents is the advent of a double-layer stent. This stent is constructed in three layers: an inner perfluoro alkoxy (chemically smoothed Teflon) layer, an outer stiff polyamide elastomer, and a middle stainless steel mesh that holds the inner and outer layers together (Figs. 2 and 3). This stent was compared with traditional polyethylene stents in a prospective, controlled, randomized trial of 120 patients [63]. In the double-layer stent group, 26 patients (43%) had clinical symptoms of stent clogging after a mean of 144
11 days, whereas in the polyethylene stent group, 38 patients (63%) had stent clogging after a mean of 99
9 days (P <.05). The cost of the double-layer stent is comparable to the cost of polyethylene stents and is much less than the cost of SEMS. Further studies are needed to confirm these findings and to compare the performance of double-layer stents with SEMS. Intraluminal brachytherapy has been attempted as a palliative measure in biliary obstruction secondary to cholangiocarcinoma. Iridium 192 seeds are

Fig. 2. Double-layer stent for biliary obstruction.

ENDOSCOPIC PALLIATION

71

Fig. 3. Close up end-on view of double-layer stent showing metal struts (arrow) within a plastic outer lining.

placed directly across the biliary stricture in the hopes of locally controlling tumor growth through a radiation source at close proximity. Studies performed have shown mixed results, with some studies showing improved survival and others showing no benefit with a possible increase in complications [64–66]. MALIGNANT GASTRIC OUTLET OBSTRUCTION Malignant GOO refers to the inability of the stomach to empty, either secondary to gastric obstruction or secondary to proximal small bowel obstruction. Patients with malignant GOO typically develop the condition as a consequence of advanced upper gastrointestinal malignancy, although metastatic tumors to the gastroduodenum can cause identical symptoms. The most common tumors that cause malignant GOO are typically pancreatic, ampullary, or gastric in origin, although cholangiocarcinoma also has been reported in this context [67,68]. Clinically, patients present with nausea and vomiting, often with associated dehydration and malnutrition. A simple scoring system has been used to grade the degree of GOO [67]. The Gastric Outlet Obstruction Scoring System (GOOSS) can be used to assess baseline oral intake and is particularly helpful when assessing response to therapy by any modality (endoscopic, surgical, or oncologic) (Table 1).

Table 1 The Gastric Outlet Obstruction Scoring System Level of oral intake

GOOSS score

No oral intake Liquids only Soft solids Low residue or full diet

0 1 2 3

72

ADLER & MERWAT

Malignant GOO is frequently unrecognized and is often undetected until malnutrition is severe. Many patients with advanced malignancy and GOO are undergoing concomitant chemotherapy, radiation therapy, or both. Nausea and vomiting reported in this context can be misinterpreted as a side effect of oncologic therapy when a mechanical problem may be to blame. Two clues that can suggest malignant GOO are a history of vomiting undigested food hours after eating (suggesting that the stomach cannot empty) or the absence of bile from the vomitus (suggesting that the duodenum has been isolated from the stomach). Historically, patients with malignant GOO were treated surgically via the creation of a gastrojejunostomy (with or without a concurrent surgical biliary bypass). This procedure can be performed in an open manner or laparoscopically, although the laparoscopic biliary bypass portion of the operation is still in development [69–71]. Percutaneous endoscopic gastrostomy tubes with jejunal extension or endoscopic or surgical jejunostomy tubes also are options for these patients, although these are not placed with the intent of restoring oral intake [72]. Currently, enteral stents are the mainstay of nonsurgical management of malignant GOO. Although ethanol injection and laser therapy have been used successfully in this context, these modalities are not used frequently at present [73,74]. Enteral stents are similar to esophageal, biliary, and colonic stents in that they are metal mesh stents that can be placed across malignant gastroduodenal strictures or obstructions with the intent of providing lasting luminal patency and the restoration of oral intake. As with all metal stents, enteral stents are permanent devices that are not intended to be removed after placement. Currently, there is only one stent approved by the Food and Drug Administration for palliation of malignant gastroduodenal obstruction, the Enteral Wallstent (Boston Scientific). This device also is approved for deployment in the colon to treat colonic obstruction and is an uncovered lattice stent that can be deployed through the working channel of a therapeutic upper endoscope, a duodenoscope, or a colonoscope. Some esophageal stents have been used in an off-label manner to palliate malignant GOO, although some modification of the delivery system may be required given the increased insertion depth required. The use of fluoroscopy is highly recommended, but not strictly required for proper stent placement. Fluoroscopy allows identification of contrast dye to delineate the stricture, defining its length and geometry, both of which greatly aid in the selection of stent type and size. Use of fluoroscopy during deployment helps to ensure proper position and may minimize misdeployments or other complications. Enteral stents have been studied extensively with regard to their ability to palliate patients with malignant GOO. More than 30 published studies, mostly retrospective in nature, describe the use of enteral stents in more than 600 patients, the results of which have been summarized [75,76]. Technical success (placement of the stent itself) occurred in 97% of patients, and clinical success (improvement in oral intake) was achieved in 89% of patients after stent placement.

ENDOSCOPIC PALLIATION

73

Complications of enteral stents include malpositioning or misdeployment, tumor ingrowth or overgrowth (17%), migration (5%), bleeding (<1%), and perforation (<1%) [76]. The Enteral Wallstent can be reconstrained if positioning is suboptimal, minimizing malpositioning or misdeployments. Tumor ingrowth and overgrowth can be treated via the placement of further stents, a technique that works well and still avoids surgical intervention in most patients. Most patients can have luminal patency restored with a single stent, but for long strictures multiple stents can be deployed in a stent-within-stent fashion. Oral intake after enteral stent placement also has been studied extensively, often with the GOOSS score [67,76,77]. In an analysis of 526 patients treated with enteral stents, the mean preprocedure GOOSS score was 0.4 and improved to 2.4 after stent placements [76]. Most patients can maintain oral intake until death from underlying cancer, which typically occurs within 4 months [67]. Although no prospective trials comparing enteral stents with gastrojejunostomy exist, a retrospective comparison of these two modalities found superior results for enteral stents with regard to rapidity of return to oral feedings and overall clinical performance [78]. Concomitant biliary obstruction is seen frequently in patients with malignant GOO. Enteral stents, when placed, can make endoscopic access to the major papilla biliary tree difficult, if not impossible. The authors recommend that before placing an enteral stent strong consideration should be given to the prophylactic placement of a metal biliary stent. In this manner, the endoscopic equivalent of a gastrojejunostomy with a biliary bypass can be achieved (Fig. 4). If biliary obstruction develops after enteral stent placement, and the papilla cannot be accessed endoscopically, percutaneous approaches still can allow biliary stenting to be completed. MALIGNANT LARGE BOWEL OBSTRUCTION In the United States, colon and rectal cancer are predicted to develop in 106,000 and 40,000 individuals, with malignant bowel obstruction being a common presentation, often requiring emergency surgery [79–83]. As would be expected, patients with malignant large bowel obstruction from colorectal cancer almost always are found to have advanced cancers [84]. Patients with acute malignant large bowel obstruction tend to be quite ill, with a painful, distended abdomen. Patients with symptoms of chronic obstruction may have a more debilitated appearance. Dehydration is seen in both groups. The bowel above the level of the obstruction typically is dilated, and the bowel wall itself is friable. These patients are at high risk for spontaneous perforation or for perforation incurred during any endoscopic procedures. Likewise, the unprepared, dilated, friable colons of these patients make surgical interventions in these patients high risk [85,86]. Given these difficulties, much interest has focused on the use of endoscopic modalities to temporize or treat definitively malignant large bowel obstruction. In patients with advanced cancer who require palliation, these techniques can allow decompression and clinical stabilization and avoidance or temporizing before surgery. Standard surgical approach to patients with advanced disease

74

ADLER & MERWAT

Fig. 4. Fluoroscopic image after combined endoscopic biliary and duodenal stenting for nonsurgical treatment of malignant biliary obstruction and malignant GOO in a patient with pancreatic cancer.

involves a diverting ostomy with or without resection [87]. Laparoscopic colectomy seems to be possible in patients with malignant large bowel obstruction who have undergone prior endoscopic decompression [88–90]. In patients who are found to have advanced or unresectable disease, endoscopic therapies often can obviate the need for any surgical intervention. Colonic Decompression Tubes Colon decompression tubes are an inexpensive and widely available treatment modality for patients with malignant large bowel obstruction. These devices can be implanted with or without fluoroscopic assistance. In patients with a residual colonic lumen, decompression tubes can be placed under direct vision at endoscopy. In patients with complete obstruction, guidewires and fluoroscopy can be of great value in allowing access to the proximal colon and subsequent placement of the decompression tube over the guidewire. As with any endoscopy in patients with malignant large bowel obstruction, the use of air insufflation should be minimized, and there are risks of bleeding and perforation with the placement of the tubes. Colon decompression tubes are often successful in allowing clinical decompression, and patients often can proceed directly to surgery. Commonly available colon decompression tubes typically allow only for the passage of air and not stool [91–93]. A large-bore colorectal decompression tube is now available that can allow preoperative bowel preparation after clinical decompression has occurred, creating the opportunity for one-stage surgical procedures [94].

ENDOSCOPIC PALLIATION

75

Self-Expanding Metal Stents Colonic stents have been available for use in patients with malignant large bowel obstruction for several years, and their use is becoming more commonplace (Fig. 5). Three manufacturers currently produce SEMS designed for use in the large bowel. The Enteral Wallstent is available in widths of 18 to 22 mm and lengths of 60 to 90 mm. The Enteral Wallstent is approved by the Food and Drug Administration for use in malignant small bowel obstruction. In contrast to other colonic stents, the Enteral Wallstent is an example of through-thescope technology; the stent can be placed and deployed through the working channel of an endoscope under direct endoscopic guidance, allowing placement in the proximal colon. The Colonic Z-Stent (Wilson-Cook Medical Inc.) is available in different lengths (4–12 cm) and has a shaft diameter of 25 mm and an end diameter of 35 mm. Boston Scientific also manufactures a nitinol wire colonic stent (Ultraflex Precision Colonic), which has a proximal end of 35 mm diameter. The latter stents are deployed from catheters similar to that used in esophageal stents and cannot pass through the working channel of an endoscope because of size limitations. If endoscopic visualization is needed, an endoscope can be advanced alongside the delivery catheters of these devices. Although in some cases endoscopic imaging alone can be sufficient to place a colonic stent, the authors strongly advocate the use of fluoroscopy during the deployment of any colonic stent. As with all SEMS, fluoroscopy can provide the physician with information regarding stricture length and geometry, endoscope and stent positioning, and a global assessment of the device after stent deployment that is not available with endoscopic imaging alone. In patients with even a small retained lumen, the advancement of the stent delivery system across the stricture is straightforward. In patients with 100% obstruction, careful evaluation of the blockage with a guidewire can allow

Fig. 5. Fluoroscopic image of a 22  90 mm Enteral Wallstent after deployment in the right colon.

76

ADLER & MERWAT

access to the proximal colon and advancement of the constrained stent across the stricture and over the wire. This maneuver is similar to that used to gain access to the proximal biliary tree above a stricture during endoscopic retrograde cholangiopancreatography. Guidewire access to the proximal colon is essential to the ability to deploy stents safely and accurately in patients with complete colonic obstruction. In studies of colonic SEMS, two key parameters always are addressed. Technical success refers to satisfactory placement of the stent in what is believed to be a good position across the stricture. Clinical success is defined as decompression of the obstructed large bowel, but does not always include the ability to purge the colon fully. Technical and clinical success rates with colonic SEMS are greater than 90% [81,95–98]. Few randomized studies comparing colonic SEMS with surgery have been performed, but benefits of SEMS include shorter hospitalizations and decreased costs [99,100]. One case-control series that showed a significant cost reduction in patients receiving SEMS compared with surgery also found that the use of SEMS is associated with shorter stays in ICUs and shorter hospitalization [95]. Colonic stents have not been shown to prolong life [100,101]. It has been shown that the placement of a colonic stent does not preclude subsequent chemoradiation therapy [83]. Complications of colonic stents include improper placement, migration, perforation, and bleeding. These complications are more likely to occur during or immediately after stent deployment. Tumor ingrowth and tumor overgrowth are late complications that can be seen days to months after stent placement. Laser Therapy Laser therapy represents an ablative technique aimed at physically destroying malignant tissue via laser light at endoscopy in an attempt to recanalize an obstructed lumen. The most commonly used devices are Nd:YAG lasers. The technique cannot be used in patients with acute obstruction or extrinsically compressing lesions (eg, genitourinary or gynecologic cancers) because the tumor itself must be endoscopically visible to be ablated. Laser therapy has been evaluated extensively. The technique can lead to relief of obstructive symptoms in 75% to 80% of patients. As is seen in patients with esophageal cancer who undergo laser therapy, patients with obstructing colorectal cancer require multiple laser treatments to maintain a patent lumen. On average, patients require three to four sessions during the course of the disease. Complications occur in 10% to 15% of patients and include perforation, bleeding, postprocedure pain, and formation of fistulas or abscesses [102–106]. There is a definite learning curve to laser therapy, and experience is correlated directly to better outcomes and fewer complications [103,107]. Argon Plasma Coagulation Argon plasma coagulation causes superficial tissue necrosis via the use of ionized plasma gas delivered through endoscopic catheters. The technique cannot cause as deep a degree of tissue ablation as laser therapy. Despite this

ENDOSCOPIC PALLIATION

77

significant limitation, limited data in the form of case reports have shown that argon plasma coagulation can be useful in patients with malignant large bowel obstruction or, perhaps more importantly, as a tool to control bleeding from colorectal cancers [108,109]. References [1] Pera M, Cameron AJ, Trastek VF, et al. Increasing incidence of adenocarcinoma of the esophagus and esophagogastric junction. Gastroenterology 1993;104:510–3. [2] Mellow MH, Pinkas H. Endoscopic laser therapy for malignancies affecting the esophagus and gastroesophageal junction: analysis of technical and functional efficacy. Arch Intern Med 1985;145:1443–6. [3] Guitron A, Adalid R, Huerta F, et al. [Palliative treatment of esophageal cancer with transendoscopic injection of alcohol]. Rev Gastroenterol Mex 1996;61:208–11. [4] Nwokolo CU, Payne-James JJ, Silk DB, et al. Palliation of malignant dysphagia by ethanol induced tumour necrosis. Gut 1994;35:299–303. [5] Carazzone A, Bonavina L, Segalin A, et al. Endoscopic palliation of oesophageal cancer: results of a prospective comparison of Nd:YAG laser and ethanol injection. Eur J Surg 1999;165:351–6. [6] Freitas D, Gouveia H, Sofia C, et al. Endoscopic Nd-YAG laser therapy as palliative treatment for esophageal and cardial cancer. Hepatogastroenterology 1995;42:633–7. [7] Ahlquist DA, Gostout CJ, Viggiano TR, et al. Endoscopic laser palliation of malignant dysphagia: a prospective study. Mayo Clin Proc 1987;62:867–74. [8] Brennan FN, McCarthy JH, Laurence BH. Endoscopic Nd-YAG laser therapy for palliation of upper gastrointestinal malignancy. Med J Aust 1990;153:27–31. [9] Gevers AM, Macken E, Hiele M, Rutgeerts P. A comparison of laser therapy, plastic stents, and expandable metal stents for palliation of malignant dysphagia in patients without a fistula. Gastrointest Endosc 1998;48:383–8. [10] Xinopoulos D, Dimitroulopoulos D, Moschandrea I, et al. Natural course of inoperable esophageal cancer treated with metallic expandable stents: quality of life and cost-effectiveness analysis. J Gastroenterol Hepatol 2004;19:1397–402. [11] Dallal HJ, Smith GD, Grieve DC, et al. A randomized trial of thermal ablative therapy versus expandable metal stents in the palliative treatment of patients with esophageal carcinoma. Gastrointest Endosc 2001;54:549–57. [12] Eriksen JR. Palliation of non-resectable carcinoma of the cardia and oesophagus by argon beam coagulation. Dan Med Bull 2002;49:346–9. [13] Heindorff H, Wojdemann M, Bisgaard T, Svendsen LB. Endoscopic palliation of inoperable cancer of the oesophagus or cardia by argon electrocoagulation. Scand J Gastroenterol 1998;33:21–3. [14] Jin M, Yang B, Zhang W, Wang Y. Photodynamic therapy for upper gastrointesitinal tumours over the past 10 years. Semin Surg Oncol 1994;10:111–3. [15] Luketich JD, Christie NA, Buenaventura PO, et al. Endoscopic photodynamic therapy for obstructing esophageal cancer: 77 cases over a 2-year period. Surg Endosc 2000;14: 653–7. [16] Moghissi K, Dixon K, Thorpe JA, et al. The role of photodynamic therapy (PDT) in inoperable oesophageal cancer. Eur J Cardiothorac Surg 2000;17:95–100. [17] Litle VR, Luketich JD, Christie NA, et al. Photodynamic therapy as palliation for esophageal cancer: experience in 215 patients. Ann Thorac Surg 2003;76:1687–92. [18] Lightdale CJ, Heier SK, Marcon NE, et al. Photodynamic therapy with porfimer sodium versus thermal ablation therapy with Nd:YAG laser for palliation of esophageal cancer: a multicenter randomized trial. Gastrointest Endosc 1995;42:507–12. [19] Marcon NE. Photodynamic therapy and cancer of the esophagus. Semin Oncol 1994;21(Suppl 15):20–3.

78

ADLER & MERWAT

[20] Saidi RF, Marcon NE. Nonthermal ablation of malignant esophageal strictures: photodynamic therapy, endoscopic intratumoral injections, and novel modalities. Gastrointest Endosc Clin N Am 1998;8:465–91. [21] De Palma GD, di Matteo E, Romano G, et al. Plastic prosthesis versus expandable metal stents for palliation of inoperable esophageal thoracic carcinoma: a controlled prospective study. Gastrointest Endosc 1996;43:478–82. [22] Mosca F, Consoli A, Stracqualursi A, et al. [Our experience with the use of a plastic prosthesis and self-expanding stents in the palliative treatment of malignant neoplastic stenoses of the esophagus and cardia: comparative analysis of results]. Chir Ital 2002;54:341–50. [Italian]. [23] Knyrim K, Wagner HJ, Bethge N, et al. A controlled trial of an expansile metal stent for palliation of esophageal obstruction due to inoperable cancer. N Engl J Med 1993;329: 1302–7. [24] Mosca F, Consoli A, Stracqualursi A, et al. Comparative retrospective study on the use of plastic prostheses and self-expanding metal stents in the palliative treatment of malignant strictures of the esophagus and cardia. Dis Esophagus 2003;16:119–25. [25] Shimi SM. Self-expanding metallic stents in the management of advanced esophageal cancer: a review. Semin Laparosc Surg 2000;7:9–21. [26] Kozarek RA, Raltz S, Brugge WR, et al. Prospective multicenter trial of esophageal Z-stent placement for malignant dysphagia and tracheoesophageal fistula. Gastrointest Endosc 1996;44:562–7. [27] Vakil N, Gross U, Bethge N. Human tissue responses to metal stents. Gastrointest Endosc Clin N Am 1999;9:359–65. [28] Vakil N, Morris AI, Marcon N, et al. A prospective, randomized, controlled trial of covered expandable metal stents in the palliation of malignant esophageal obstruction at the gastroesophageal junction. Am J Gastroenterol 2001;96:1791–6. [29] Hills KS, Chopra KB, Pal A, Westaby D. Self-expanding metal oesophageal endoprostheses, covered and uncovered: a review of 30 cases. Eur J Gastroenterol Hepatol 1998;10: 371–4. [30] Siersema PD, Hop WC, van Blankenstein M, et al. A comparison of 3 types of covered metal stents for the palliation of patients with dysphagia caused by esophagogastric carcinoma: a prospective, randomized study. Gastrointest Endosc 2001;54:145–53. [31] May A, Hahn EG, Ell C. Self-expanding metal stents for palliation of malignant obstruction in the upper gastrointestinal tract: comparative assessment of three stent types implemented in 96 implantations. J Clin Gastroenterol 1996;22:261–6. [32] Riccioni ME, Shah SK, Tringali A, et al. Endoscopic palliation of unresectable malignant oesophageal strictures with self-expanding metal stents: comparing Ultraflex and Esophacoil stents. Dig Liver Dis 2002;34:356–63. [33] Sabharwal T, Hamady MS, Chui S, et al. A randomised prospective comparison of the Flamingo Wallstent and Ultraflex stent for palliation of dysphagia associated with lower third oesophageal carcinoma. Gut 2003;52:922–6. [34] Dua KS. Antireflux stents in tumors of the cardia. Am J Med 2001;111(Suppl 8A): 190S–6S. [35] Dua KS, Kozarek R, Kim J, et al. Self-expanding metal esophageal stent with anti-reflux mechanism. Gastrointest Endosc 2001;53:603–13. [36] Evrard S, Le Moine O, Lazaraki G, et al. Self-expanding plastic stents for benign esophageal lesions. Gastrointest Endosc 2004;60:894–900. [37] Adler DG, Pleskow DK. Closure of a benign tracheoesophageal fistula by using a coated, self-expanding plastic stent in a patient with a history of esophageal atresia. Gastrointest Endosc 2005;61:765–8. [38] Costamagna G, Shah SK, Tringali A, et al. Prospective evaluation of a new self-expanding plastic stent for inoperable esophageal strictures. Surg Endosc 2003;17:891–5.

ENDOSCOPIC PALLIATION

79

[39] Dormann AJ, Eisendrath P, Wigginghaus B, et al. Palliation of esophageal carcinoma with a new self-expanding plastic stent. Endoscopy 2003;35:207–11. [40] Bethge N, Vakil N. A prospective trial of a new self-expanding plastic stent for malignant esophageal obstruction. Am J Gastroenterol 2001;96:1350–4. [41] Decker P, Lippler J, Decker D, Hirner A. Use of the Polyflex stent in the palliative therapy of esophageal carcinoma: results in 14 cases and review of the literature. Surg Endosc 2001;15:1444–7. [42] Smith AC, Dowsett JF, Russell RC, et al. Randomised trial of endoscopic stenting versus surgical bypass in malignant low bile duct obstruction. Lancet 1994;344:1655–60. [43] Martin RC 2nd, Vitale GC, Reed DN, et al. Cost comparison of endoscopic stenting vs surgical treatment for unresectable cholangiocarcinoma. Surg Endosc 2002;16: 667–70. [44] Speer AG, Cotton PB, MacRae KD. Endoscopic management of malignant biliary obstruction: stents of 10 French gauge are preferable to stents of 8 French gauge. Gastrointest Endosc 1988;34:412–7. [45] Pedersen FM. Endoscopic management of malignant biliary obstruction: is stent size of 10 French gauge better than 7 French gauge? Scand J Gastroenterol 1993;28:185–9. [46] van Berkel AM, Huibregtse IL, Bergman JJ, et al. A prospective randomized trial of Tannenbaum-type Teflon-coated stents versus polyethylene stents for distal malignant biliary obstruction. Eur J Gastroenterol Hepatol 2004;16:213–7. [47] van Berkel AM, Boland C, Redekop WK, et al. A prospective randomized trial of Teflon versus polyethylene stents for distal malignant biliary obstruction. Endoscopy 1998;30: 681–6. [48] Terruzzi V, Comin U, De Grazia F, et al. Prospective randomized trial comparing Tannenbaum Teflon and standard polyethylene stents in distal malignant biliary stenosis. Gastrointest Endosc 2000;51:23–7. [49] Davids PH, Groen AK, Rauws EA, et al. Randomised trial of self-expanding metal stents versus polyethylene stents for distal malignant biliary obstruction. Lancet 1992;340: 1488–92. [50] Knyrim K, Wagner HJ, Pausch J, Vakil N. A prospective, randomized, controlled trial of metal stents for malignant obstruction of the common bile duct. Endoscopy 1993;25: 207–12. [51] Kaassis M, Boyer J, Dumas R, et al. Plastic or metal stents for malignant stricture of the common bile duct? Results of a randomized prospective study. Gastrointest Endosc 2003;57: 178–82. [52] Lammer J, Hausegger KA, Fluckiger F, et al. Common bile duct obstruction due to malignancy: treatment with plastic versus metal stents. Radiology 1996;201:167–72. [53] Prat F, Chapat O, Ducot B, et al. Predictive factors for survival of patients with inoperable malignant distal biliary strictures: a practical management guideline. Gut 1998;42: 76–80. [54] Yeoh KG, Zimmerman MJ, Cunningham JT, Cotton PB. Comparative costs of metal versus plastic biliary stent strategies for malignant obstructive jaundice by decision analysis. Gastrointest Endosc 1999;49(4 Pt 1):466–71. [55] Kahaleh M, Tokar J, Conaway MR, et al. Efficacy and complications of covered Wallstents in malignant distal biliary obstruction. Gastrointest Endosc 2005;61:528–33. [56] Bezzi M, Zolovkins A, Cantisani V, et al. New ePTFE/FEP-covered stent in the palliative treatment of malignant biliary obstruction. J Vasc Interv Radiol 2002;13:581–9. [57] Schoder M, Rossi P, Uflacker R, et al. Malignant biliary obstruction: treatment with ePTFEFEP-covered endoprostheses initial technical and clinical experiences in a multicenter trial. Radiology 2002;225:35–42. [58] Han YM, Jin GY, Lee SO, et al. Flared polyurethane-covered self-expandable nitinol stent for malignant biliary obstruction. J Vasc Interv Radiol 2003;14:1291–301.

80

ADLER & MERWAT

[59] Isayama H, Komatsu Y, Tsujino T, et al. A prospective randomised study of ‘‘covered’’ versus ‘‘uncovered’’ diamond stents for the management of distal malignant biliary obstruction. Gut 2004;53:729–34. [60] Kahaleh M, Tokar J, Le T, Yeaton P. Removal of self-expandable metallic Wallstents. Gastrointest Endosc 2004;60:640–4. [61] Isayama H, Komatsu Y, Tsujino T, et al. Polyurethane-covered metal stent for management of distal malignant biliary obstruction. Gastrointest Endosc 2002;55:366–70. [62] Wamsteker EJ, Elta GH. Migration of covered biliary self-expanding metallic stents in two patients with malignant biliary obstruction. Gastrointest Endosc 2003;58:792–3. [63] Tringali A, Mutignani M, Perri V, et al. A prospective, randomized multicenter trial comparing DoubleLayer and polyethylene stents for malignant distal common bile duct strictures. Endoscopy 2003;35:992–7. [64] Bruha R, Petrtyl J, Kubecova M, et al. Intraluminal brachytherapy and selfexpandable stents in nonresectable biliary malignancies—the question of long-term palliation. Hepatogastroenterology 2001;48:631–7. [65] Montemaggi P, Costamagna G, Dobelbower RR, et al. Intraluminal brachytherapy in the treatment of pancreas and bile duct carcinoma. Int J Radiat Oncol Biol Phys 1995;32: 437–43. [66] Gerhards MF, van Gulik TM, Gonzalez Gonzalez D, et al. Results of postoperative radiotherapy for resectable hilar cholangiocarcinoma. World J Surg 2003;27: 173–9. [67] Adler DG, Baron TH. Endoscopic palliation of malignant gastric outlet obstruction using self-expanding metal stents: experience in 36 patients. Am J Gastroenterol 2002;97: 72–8. [68] Yim HB, Jacobson BC, Saltzman JR, et al. Clinical outcome of the use of enteral stents for palliation of malignant upper GI tract obstruction. Gastrointest Endosc 2001;53: 329–32. [69] Choi YB. Laparoscopic gastrojejunostomy for palliation of gastric outlet obstruction in unresectable gastric cancer. Surg Endosc 2002;16:1620–6. [70] Bergamaschi R, Arnaud JP, Marvik R, Myrvold HE. Laparoscopic antiperistaltic versus isoperistaltic gastrojejunostomy for palliation of gastric outlet obstruction in advanced cancer. Surg Laparosc Endosc Percutan Tech 2002;12:393–7. [71] Ali AS, Ammori BJ. Concomitant laparoscopic gastric and biliary bypass and bilateral thoracoscopic splanchnotomy: the full package of minimally invasive palliation for pancreatic cancer. Surg Endosc 2003;17:2028–31. [72] Scheidbach H, Horbach T, Groitl H, Hohenberger W. Percutaneous endoscopic gastrostomy/jejunostomy (PEG/PEJ) for decompression in the upper gastrointestinal tract: initial experience with palliative treatment of gastrointestinal obstruction in terminally ill patients with advanced carcinomas. Surg Endosc 1999;13:1103–5. [73] Payne-James JJ, Spiller R, Misiewicz JJ, Silk DB. Use of ethanol-induced tumor necrosis to palliate dysphagia in patients with esophagogastric cancer. Gastrointest Endosc 1990;36:43–6. [74] Spinelli P, Cerrai FG, Dal Fante M. Endoscopic treatment of upper gastrointestinal tract malignancies. Endoscopy 1993;25:675–8. [75] Baron TH, Harewood G. Enteral self-expandable stents. Gastrointest Endosc 2003;58: 421–33. [76] Dormann A, Meisner S, Verin N, Wenk Lang A. Self-expanding metal stents for gastroduodenal malignancies: systematic review of their clinical effectiveness. Endoscopy 2004;36:543–50. [77] Holt AP, Patel M, Ahmed MM. Palliation of patients with malignant gastroduodenal obstruction with self-expanding metallic stents: the treatment of choice? Gastrointest Endosc 2004;60:1010–7.

ENDOSCOPIC PALLIATION

81

[78] Maetani I, Tada T, Ukita T, et al. Comparison of duodenal stent placement with surgical gastrojejunostomy for palliation in patients with duodenal obstructions caused by pancreaticobiliary malignancies. Endoscopy 2004;36:73–8. [79] Jemal A, Tiwari RC, Murray T, et al. Cancer statistics, 2004. CA Cancer J Clin 2004;54: 8–29. [80] Deans GT, Krukowski ZH, Irwin ST. Malignant obstruction of the left colon. Br J Surg 1994;81:1270–6. [81] Baron TH, Dean PA, Yates MR 3rd, et al. Expandable metal stents for the treatment of colonic obstruction: techniques and outcomes. Gastrointest Endosc 1998;47: 277–86. [82] Buechter KJ, Boustany C, Caillouette R, Cohn I Jr. Surgical management of the acutely obstructed colon: a review of 127 cases. Am J Surg 1988;156(3 Pt 1):163–8. [83] Adler DG, Young-Fadok TM, Smyrk T, et al. Preoperative chemoradiation therapy after placement of a self-expanding metal stent in a patient with an obstructing rectal cancer: clinical and pathologic findings. Gastrointest Endosc 2002;55:435–7. [84] Gandrup P, Lund L, Balslev I. Surgical treatment of acute malignant large bowel obstruction. Eur J Surg 1992;158:427–30. [85] Gukovsky-Reicher S, Lin RM, Sial S, et al. Self-expandable metal stents in palliation of malignant gastrointestinal obstruction: review of the current literature data and 5-year experience at Harbor-UCLA Medical Center. Med Gen Med 2003;5:16. [86] Arnell T, Stamos MJ, Takahashi P, et al. Colonic stents in colorectal obstruction. Am Surg 1998;64:986–8. [87] Hartmann H. Nouveau procede d’ablation des cancers de la partie terminale du colo pelvien. Congress Fr Chir 1923;29:662–8. [88] Morino M, Bertello A, Garbarini A, et al. Malignant colonic obstruction managed by endoscopic stent decompression followed by laparoscopic resections. Surg Endosc 2002;16: 1483–7. [89] Balague C, Targarona EM, Sainz S, et al. Minimally invasive treatment for obstructive tumors of the left colon: endoluminal self-expanding metal stent and laparoscopic colectomy: preliminary results. Dig Surg 2004;21:282–6. [90] Law WL, Choi HK, Lee YM, Chu KW. Laparoscopic colectomy for obstructing sigmoid cancer with prior insertion of an expandable metallic stent. Surg Laparosc Endosc Percutan Tech 2004;14:29–32. [91] Lelcuk S, Merhav A, Klausner JM, et al. Rectoscopic decompression of acute recto-sigmoid obstruction. Endoscopy 1987;19:209–10. [92] Rattan J, Klausner JM, Rozen P, et al. Acute left colonic obstruction: a new nonsurgical treatment. J Clin Gastroenterol 1989;11:331–4. [93] Lelcuk S, Ratan J, Klausner JM, et al. Endoscopic decompression of acute colonic obstruction: avoiding staged surgery. Ann Surg 1986;203:292–4. [94] Horiuchi A, Maeyama H, Ochi Y, et al. Usefulness of Dennis Colorectal Tube in endoscopic decompression of acute, malignant colonic obstruction. Gastrointest Endosc 2001;54: 229–32. [95] Binkert CA, Ledermann H, Jost R, et al. Acute colonic obstruction: clinical aspects and costeffectiveness of preoperative and palliative treatment with self-expanding metallic stents—a preliminary report. Radiology 1998;206:199–204. [96] Wholey MH, Levine EA, Ferral H, Castaneda-Zuniga W. Initial clinical experience with colonic stent placement. Am J Surg 1998;175:194–7. [97] Meisner S, Hensler M, Knop FK, et al. Self-expanding metal stents for colonic obstruction: experiences from 104 procedures in a single center. Dis Colon Rectum 2004;47:444. [98] Mauro MA, Koehler RE, Baron TH. Advances in gastrointestinal intervention: the treatment of gastroduodenal and colorectal obstructions with metallic stents. Radiology 2000;215: 659–69.

82

ADLER & MERWAT

[99] Xinopoulos D, Dimitroulopoulos D, Theodosopoulos T, et al. Stenting or stoma creation for patients with inoperable malignant colonic obstructions? Results of a study and costeffectiveness analysis. Surg Endosc 2004;18:421–6. [100] 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;47:1455–61. [101] Johnson R, Marsh R, Corson J, Seymour K. A comparison of two methods of palliation of large bowel obstruction due to irremovable colon cancer. Ann R Coll Surg Engl 2004;86: 99–103. [102] Loizou LA, Grigg D, Boulos PB, Bown SG. Endoscopic Nd:YAG laser treatment of rectosigmoid cancer. Gut 1990;31:812–6. [103] Gevers AM, Macken E, Hiele M, Rutgeerts P. Endoscopic laser therapy for palliation of patients with distal colorectal carcinoma: analysis of factors influencing long-term outcome. Gastrointest Endosc 2000;51:580–5. [104] Daneker GW Jr, Carlson GW, Hohn DC, et al. Endoscopic laser recanalization is effective for prevention and treatment of obstruction in sigmoid and rectal cancer. Arch Surg 1991;126:1348–52. [105] Mandava N, Petrelli N, Herrera L, Nava H. Laser palliation for colorectal carcinoma. Am J Surg 1991;162:212–4. [106] Tan CC, Iftikhar SY, Allan A, Freeman JG. Local effects of colorectal cancer are well palliated by endoscopic laser therapy. Eur J Surg Oncol 1995;21:648–52. [107] Brunetaud JM, Maunoury V, Cochelard D. Lasers in rectosigmoid tumors. Semin Surg Oncol 1995;11:319–27. [108] Nozoe Y, Araki Y, Fukushima H, Shirouzu K. A case of argon plasma coagulation therapy for hemorrhagic rectal tumor in a highly aged patient. Kurume Med J 2004;51:159–61. [109] Solecki R, Zajac A, Richter P, Szura M. Bifocal esophageal and rectal cancer palliatively treated with argon plasma coagulation. Surg Endosc 2004;18:346.

Gastroenterol Clin N Am 35 (2006) 83–100

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

The Role of Endoscopically Placed Feeding or Decompression Tubes Stephen A. McClave, MDa,*, Christine S. Ritchie, MD, MSPHb a

Division of Gastroenterology/Hepatology, University of Louisville School of Medicine, 550 South Jackson Street, Louisville, KY 40202, USA b Department of Medicine, University of Alabama at Birmingham, 1530 Third Avenue S., Birmingham, AL 35294, USA

I

n the area of palliative care, the application of percutaneous endoscopically placed polyurethane or silicone tubes into the stomach, proximal small bowel, or colon has expanded exponentially over the past 25 years since the original introduction of the percutaneous endoscopic gastrostomy (PEG) [1]. The patient populations who have undergone the greatest number of applications of such tubes include those patients who have progressive neurologic disorders (such as amyotrophic lateral sclerosis [ALS], Parkinson’s disease, and dementia), stroke, and advanced malignancy. The percutaneously placed tubes are used most commonly to establish enteral access through which to provide specialized nutritional support and to enable decompression of the gastrointestinal (GI) tract above an area of malignant obstruction or altered motility. The tubes are also used as an adjunctive measure to facilitate other procedures (such as placement of an esophageal stent) or to create an externalized extraintestinal bypass around an obstructed segment of the GI tract. The use of such tubes in palliative care, however, is controversial. The literature guiding the clinician is problematic. Few prospective randomized trials exist. Selection bias occurs, and results of studies are difficult to interpret because of confounding variables and underlying comorbidities. In palliative care, end-of-life decision making must take into account basic ethical principles as well as the cultural, personal, and religious beliefs that the patient brings to the situation. As a result of these factors, decisions are reduced to an honest if imperfect discussion of the potential harm and benefit of PEGs in a particular setting, taking into account the values of each patient and his or her family.

*Corresponding author. E-mail address: [email protected] (S.A. McClave). 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.003

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

84

MCCLAVE & RITCHIE

USE OF PERCUTANEOUS ENDOSCOPIC GASTROSTOMY FOR NUTRITIONAL PURPOSES Ethics and Medicolegal Principles The basic principles of medical ethics create the framework by which PEG tubes are used in palliative care. Most importantly, the decision to undergo PEG placement should represent the desires, beliefs, and wishes of the patient, reflecting the fact that a competent adult is the final arbiter of his or her own health care (the principal of autonomy). The tube should not be placed without full, informed, and educated consent. The current quality of informed consent for the placement of gastrostomy tubes is poor. In a review of 154 consecutive hospitalized adults undergoing placement of gastrostomy tubes, only one medical record documented a procedure-specific discussion of benefits and burdens of and alternatives to tube feeding [2]. The placement of the PEG tube should benefit the patient overall in some way (the principal of beneficence). The placement of the PEG should not result in net harm to the patient, and the benefit should clearly outweigh the risks (the principle of nonmalificence). Legal case precedents over the past 30 years have forged the medicolegal concepts that now guide placement of feeding tubes in end-of-life situations [3]. Artificial, specialized nutritional support is no different from any other lifesustaining medical therapy that supports bodily function, such as antibiotics, oxygen therapy, or dialysis. Artificial nutrition is not considered ‘‘basic care,’’ and health care practitioners are no more obligated to provide artificial nutrition than any other medical therapy. Patients have just as much right to refuse as they do to accept medical therapy. Likewise, legally, the decision to withhold or withdraw therapy is no different from the decision to start therapy. Thus, clinicians are not ‘‘stuck’’ continuing to provide therapy once enteral feeding has been initiated. The underlying disease process results in the patient death, not the withdrawal of nutritional therapy [3]. Nevertheless, as highlighted by the recent Schiavo case, public acceptance of withdrawal of nutritional support varies greatly. To foster patient autonomy, physicians need to promote advanced directives to help families make end-of-life decisions. Living wills provide treatment directives that become activated only when the patient is terminally ill; they are also influenced and regulated by the state in which they are being used. In contrast, establishing a health care proxy or power of attorney, which guides treatment preferences through the patient’s own designee, has greater flexibility and may be activated or used whenever the patient is unable to voice his or her wishes, even at times other than end-stage terminal illness. If it is addressed through such a directive, the final decision to place a feeding tube more likely reflects the patient’s own desires. Surprisingly, in one case series, 14% of patients who underwent PEG placement already had advanced directives specifically indicating the desire to not have feeding tubes placed [4]. Physicians must carefully provide a balanced education to patients and their families, based on the limited scientific evidence available, without injecting their own system of beliefs into the situation.

ENDOSCOPICALLY PLACED PEG TUBES

85

Indication The most common reasons for placing PEG tubes in palliative care are weight loss, decreasing body mass index, and deteriorating nutritional status as a result of anorexia, nausea and vomiting, or dysphagia from the underlying disease process. With progressive neurologic disorders, nutritional deficiency caused by dysphagia occurs because of disturbed motility of the tongue, pharynx, and esophagus, which in turn lead to choking and the risk of aspiration. In patients who have head and neck cancers, impaired deglutition occurs from altered pharyngeal anatomy caused by the tumor itself, from adynamic reconstructive flaps, chemotherapy-induced xerostomia, or decreased oral competence [5]. PEG tube placement in a patient who has advanced malignancy theoretically helps facilitate completion of the chemoradiation therapeutic regimen and may allow the patient to be managed in a setting other than the hospital [6]. Overall, 12% to 27% of patients who have ALS undergo placement of a PEG tube [7,8]. Usually, clinically subjective parameters (aspiration and choking) are more commonly cited rather than objective measures (weight loss, decreasing body mass index, or abnormal swallowing study) in the final decision to undergo PEG placement [7]. PEG placement in ALS patients usually follows failed efforts at teaching swallowing compensation techniques and providing nutritional supplements [7]. Up to 34% of patients who have dementia or impaired cognitive function undergo PEG placement [9]. In addition to reasons stated for the other patient populations, specific reasons often given for placement in patients who have dementia include prevention of aspiration, maintenance of skin integrity and prevention of pressure sores, and prolongation of life; however, the scientific literature has not provided meaningful evidence to date that PEG placement achieves these outcomes. The use of PEG tubes is popular in stroke patients because, generally, they tend to have an intact mental status despite dysphagia and have a higher likelihood for recovering swallowing function [10]. At 4 months after stroke, the incidence of dysphagia drops nearly twofold (from 29%–45% down to 20%) [10]. The largest controlled trial of oral feeding versus PEG feeding was in the trial by Dennis and colleagues [11], in which patients enrolled within 7 days of admission were randomly allocated to early enteral tube feeding or no tube feeding for more than 7 days. Early tube feeding was associated with a nonsignificant reduction in absolute risk of death of 5.8% (95% CI, 0.8%–12.5%; P ¼ .09) [11]. There was no increase in pneumonia associated with early tube feeding; however, the improved survival was offset by a 4.7% excess of survivors who had a poor outcome with a worse quality of life. Thus, early feeding might have kept patients alive in a severely disabled state when they would otherwise have died [11]. The use of PEG tubes in palliative care has expanded tremendously in part because the alternative therapies presented are often less acceptable to the patients. The placement of nasogastric tubes may adversely affect recovery of swallowing function, promote sinusitis and gastroesophageal reflux, irritate suture lines at anastomotic sites, and are socially unacceptable [5,12]. The

86

MCCLAVE & RITCHIE

placement of surgical gastrostomy tubes requires laparotomy, a surgical incision, and a need for general anesthesia. Compared with PEG tubes, surgically placed gastrostomy tubes tend to be associated with higher cost, longer recovery from surgery, longer procedure time, and a higher complication rate (6%– 56% for surgical gastrostomy tubes versus 2%–15% for PEGs) [12,13]. Hand feeding (which is often not offered as an alternative) requires more time and some training to optimize intake and minimize aspiration. The ultimate decision to place a PEG tube often reflects the clinical characteristics of the specific patient and the fiscal organization and demographic features of the nursing home or medical facility housing that patient. In a recent report [9] based on the 1999 National Repository Resident Assessment Instrument Minimum Data Set, which evaluated certified Medicare and Medicaid facilities across the United States, a 10-fold variation in the use of PEGs was demonstrated across the United States. A specific patient was more likely to get a feeding tube if they were younger, nonwhite, male, divorced, lacked advance directives, had a recent decline in functional status, or if they did not have a diagnosis of Alzheimer’s dementia [9]. From the standpoint of the facilities, a patient was more likely to get a PEG tube if he was housed in a facility that was for profit, in an urban setting, had more than 100 beds, lacked a dementia care unit, had a smaller percentage of patients who had a ‘‘do not resuscitate’’ status, had a higher percentage of nonwhite patients or lacked a nurse practitioner or physician’s assistant on the staff [9]. Several factors may underlie these findings. The use of a PEG tube may be seen as a cost-savings device by nursing homes [9]. The staff time that is required to feed patients is expensive, and some evidence suggests that Medicaid reimbursement may be higher for tube-fed patients than nontube-fed patients in a nursing home [9]. Cultural issues may explain differences because nonwhites tend to choose more aggressive end-of-life care [9]. These tendencies may reflect different cultural attitudes toward death, the apprehension of nonwhites toward a predominantly white medical system, or other, less well-understood differences [9]. Efficacy Survival benefit Determining whether PEG placement provides a survival benefit in palliative care is difficult in the absence of prospective randomized controlled trials. In patients who have esophageal cancer, one study showed that the use of PEG placement minimized weight loss through chemotherapy (consecutive nonrandomized patients who underwent PEG versus those managed on oral intake alone) but saw no difference in 4-year survival between the two groups [14]. In a second study, however, patients who underwent PEG placement had a better survival at 12 months and better attainment of the full chemotherapeutic regimen compared with patients who opted not to have PEG placement (70% of the PEG patients achieved full chemotherapeutic regimen versus 37.5% of the non-PEG patients; P ¼ .03) [15]. In patients who had head and neck cancers, one study showed a 61% shorter hospital length of stay in patients who

ENDOSCOPICALLY PLACED PEG TUBES

87

underwent PEG placement, compared with those who were maintained on nasogastric (NG) tubes [16]. In the absence of randomized controls, however, these results may be influenced by selection bias or confounding variables because some patients’ decision to accept PEG placement reflected differences in unmeasured underlying characteristics that were more likely to lead to positive outcomes [5]. In addition, in many instances, the comparison group received nasogastric feeding but not oral feeding. PEG placement in dementia probably neither improves nor worsens survival [17]. However, because eating difficulty is a marker of advanced disease in dementia, the benefit of placing a tube must be questioned in a patient who has advanced disease, in which no survival benefit has been demonstrated. One retrospective study [18] in nursing home patients (51% of whom had dementia) showed increased 1-year mortality with PEG placement compared with patients who refused PEG placement. A different retrospective study [19] of patients in whom swallowing study findings were abnormal (20% to 30% of whom had dementia) showed a decreased mortality for patients who underwent PEG placement versus those who opted not to have PEGs placed. Three other studies, in which the incidence of dementia in the patient population ranged from 25% to 100%, showed no difference in mortality between patients who accepted PEGs and those who refused [20–22]. Aspiration The use of PEG tube placement to reduce aspiration has been particularly criticized in patients who have dementia, with some reports stating ‘‘no study shows that tube feeding reduces risk of regurgitation of gastric contents’’ [23]. Studies of patients who have ALS similarly have failed to show a benefit of PEG placement in reducing aspiration pneumonia [24–28]. Again, the interpretation of these results are skewed by the retrospective nonrandomized nature of most of these studies [29,30], in addition to the fact that varying definitions of pneumonia and clinical endpoints exist. No studies have been conducted randomizing patients to oral feeding versus PEG tube placement in patients who have dementia that address the risk for aspiration pneumonia. These studies are needed critically. Studies in the critical care setting have shown that aspiration of contaminated oropharyngeal contents, a factor that would be affected very little by manipulation of tube feeds, is a greater factor in determining whether the patient develops aspiration pneumonia than aspiration of bacteria-laden gastric contents [31–33]. Nonetheless, recent studies would suggest that diverting the feeding stream lower in the GI tract may decrease the risk of reflux, regurgitation, and aspiration. Converting from an NG tube to a PEG tube reduces the risk for regurgitation and aspiration [34]. The incidence of aspiration with an NG tube was shown to be nearly twice that of a PEG tube (40% versus 20.3%, respectively; P <.01) in one study [35]. Moving the level of infusion of enteral feeding from the stomach downward to the third portion of the duodenum in a separate study using a radioisotope marker was shown to significantly decrease reflux and aspiration

88

MCCLAVE & RITCHIE

[36]. In a third study, comparing gastric with jejunal infusion, reflux was decreased twofold (24.0% versus 12.9%, respectively, comparing gastric with jejunal infusion; P <.01) [37]. As a result of these factors, the incidence of aspiration pneumonia may be less with post-pyloric feeding compared with gastric feeding. In one meta-analysis of nine studies, aspiration pneumonia was reduced by 24% in a comparison of post-pyloric and gastric feeding [38]. In a second meta-analysis, which excluded two studies that were part of the previous study, a trend toward reduced aspiration pneumonia with post-pyloric feeds compared with gastric feeds was observed, but the difference was not statistically significant [39]. The results of these studies in other patient populations would suggest that for dementia patients, diverting the feeding stream lower in the GI tract, either by switching from oral feeds to PEG tube feeds or from gastric to jejunal feeds, reflux, regurgitation, and aspiration should be reduced. Whether this difference would result in a reduced risk of actual aspiration pneumonia in a particular patient is probably a weak clinical factor and certainly should not be the overlying reason that the PEG tube should be placed. Quality of life Studies evaluating changes in quality of life after PEG tube placement are surprisingly disparate. In patients who have chronic dementia, quality of life may actually be diminished by PEG placement [4]. Often, most demented patients (>70% in one study) [10] require wrist restraints once PEGs are placed. The immobility may increase psychic stress, the need for sedation, and the likelihood for developing pressure sores. The placement of a PEG may deprive the patient of the social interaction and pleasure related to the provision of meals. Complications directly related to the PEG itself, such as cellulitis, infection, and pain, may develop and diminish the patient’s immediate well being [4,10]. The failure to provide artificial nutrition may not change quality of life at all. In many instances, oral feeding continues to provide sensory stimulation and social interaction, even if weight is not maintained. In patients who are unable to eat orally, studies suggest that many do not experience significant amounts of discomfort related to hunger. In a study of 32 terminally ill cancer patients, up to 63% never experienced hunger, and 62% never experienced thirst [40]. For those patients who did complain of thirst and hunger, symptoms were often transient and easily alleviated by small amounts of food, mouth care, or provision of ice chips or lip lubrication. Only minimal intake was required to provide comfort [40]. It is difficult to show that the quality of life of a specific patient improves directly as a result of PEG placement, particularly in patients who have dementia. In one large study [41], 70% of patients (a third of whom had dementia) showed no improvement in functional status or overall subjective health status after the placement of a PEG tube. In a separate study [42], objective quality of life measures showed essentially no change in follow-up from weeks to months after PEG placement because 35% of patients improved, but 28% of patients worsened. The likelihood of a positive response varies depending on the

ENDOSCOPICALLY PLACED PEG TUBES

89

patient population. Quality of life improvements may be more likely to be observed when PEGs are placed in patients who have stroke, head trauma, or abnormalities in the GI tract than when tubes are placed in patients who have dementia. The severity of the disability or underlying disease may preclude improvements in quality of life, and thus, it is important that physicians be careful not to promote this issue as a main reason for PEG placement. In contrast to the minimal change in quality of life for the patient who undergoes PEG placement, the quality of life for the family and caregivers may be enhanced. Often, the care of the patient becomes easier and more manageable after PEG placement. Manual feeding takes more time. Taking an active role in the care and maintenance of the feeding tube may provide the family or caregiver a greater sense of accomplishment and less sense of failure, and it allows them to better cope with the terminal illness [43,44]. In one study, 55% of patients said their quality of life improved after PEG placement, but 80% of their caregivers claimed an improvement in quality of life [44]. In a second study that showed no change in overall quality of life for the patient, 68% of relatives said the patient’s quality of life improved [42]. In a large study of patients who had ALS, the PEG was seen as beneficial, according to 91% of caregivers and relatives, and as reduced or as no benefit by only 9% of relatives [8]. Overall, this would suggest a positive effect on quality of life, at least for those involved in the care and support of patients who have end-stage disease [8]. Pressure sores The use of PEG tube placement to promote healing of pressure sores, particularly in patients who have dementia, has been criticized for the lack of evidence of beneficial effect. Surprisingly, only a weak correlation exists between pressure sores and nutritional status [23,45]. Unfortunately, no prospective randomized controlled trials exist in patients who have dementia comparing the effect of oral feeding versus the use of PEG tubes for the treatment of pressure sores. In prospective nonrandomized longitudinal studies, PEG tube placement did not appear to heal pre-existing ulcers [46,47]. The value of the nutritional support provided by PEG tube placement is offset by the need for the patient to be restrained, and this combination is disastrous in a setting of diarrhea and incontinence [10]. The placement of a PEG tube does not guarantee necessarily that nutritional requirements are met. A negative caloric balance regardless of PEG tube placement correlates with the development of pressure sores [48]. In one study [48], patients in negative caloric balance were twice as likely to get pressure sores as patients in positive caloric balance (36.4% versus 15.4%, respectively), but the difference did not reach statistical significance. Nursing homes, however, can be held liable if pressure sores progress in the absence of a feeding tube or with a volume of feeding insufficient to meet caloric requirements and no clear advance directive refusing feeding tubes [49]. To improve the risk-benefit ratio of PEG placement and the provision of specialized enteral nutrition to these patients, certain measures should be taken. To avoid the use of restraints and to minimize immobility in the patient at high risk

90

MCCLAVE & RITCHIE

for dislodging a PEG tube, T-fasteners can be used at the time of PEG placement to secure the stomach to the anterior abdominal wall. The use of skin- level ‘‘button’’ PEG tubes can be used in this setting such that when the patient pulls on tubing, he is actually putting traction on the connector feeding tube and not the actual PEG device. Securing devices such as a horizontal Hollister clamping device may be used to secure the PEG tube against dislodgement. To optimize nutritional therapy, measuring caloric requirements through indirect calorimetry and tracking provision of enteral nutrients help to ensure a positive caloric balance and adequate nutritional support (however, no studies have been performed in the palliative care setting to confirm a change in outcome from such efforts). Timing Determining when to place a PEG tube is difficult in palliative care. The appropriate timing for placement appears to be a balance between perceived potential benefit and increased risk from the procedure itself. For patients who have ALS, dietary changes usually precede consideration for PEG placement [50]. Patients are counseled about foods that are easy to chew and rich in calories. Speech pathologists teach swallowing techniques such as supraglottic swallowing [50]. PEGs are considered at the point at which caloric intake becomes insufficient, weight loss is less than 10% of normal weight before the diagnosis occurs, the point at which choking makes oral intake intolerable, or the period just before the development of any significant respiratory distress [50]. As forced vital capacity (FVC) falls with respiratory deterioration from disease progression, there comes a point for patients who have ALS when the morbidity and even mortality related to the procedure itself increases dramatically. In a large ALS database [51], the average FVC for all ALS patients was 53%. ALS patients with a mean FVC of 36% who were recommended for PEG placement suffered a 30-day mortality rate of 9.6% after PEG placement. In contrast, ALS patients with an FVC greater than 50% who underwent PEG placement had a 30-day mortality rate of 0% [51]. Therefore, the timing of PEG placement in ALS patients may affect the long-term survival benefit from the procedure. In one report of 55 ALS patients being considered for PEG placement, all patients had an FVC greater than 1 L [52]. Those patients who underwent PEG at that point in their disease process had better long-term survival than those who delayed PEG placement [52]. In a second study [53] of 69 ALS patients deemed suitable for PEG because of mild to severe dysphagia and weight loss greater than 5%, 31 accepted PEG placement and 35 refused. Mortality was not significantly different between the two groups at 6 months, but there was reduced mortality with those accepting PEG placement at 12 and 24 months compared with those who refused the procedure [53]. In this study [53], there was a strong correlation among survival, duration of bulbar symptoms, and FVC value at the time of PEG placement. Unfortunately, none of these studies was performed in a randomized controlled fashion, so the findings are subject to confounding factors.

ENDOSCOPICALLY PLACED PEG TUBES

91

For patients who have head and neck cancers, the timing of PEG placement with regard to surgery may affect the rate of complications related to the device. Patients who had PEGs placed before surgery for their head and neck cancer had the highest complication rate of 57% [5]. Those who underwent PEG placement either intraoperatively or postoperatively had a much lower complication rate, between 11% and 17% [5]. In comparison, patients who had unresectable tumors who underwent PEG placement had a complication rate of 31% [5]. Certainly, placing the PEG tube after resection of the tumor avoids the complication of tumor seeding of the PEG tract [5]. For patients who have chronic dementia, determining the best timing for PEG placement is much more difficult. The point at which patients experience a loss of speech, smile, decreased ambulation, increasing dysphagia, and refusal to eat marks the end stage of the disease process and a corresponding lifespan of less than 12 to 18 months [9,23]. At the point at which the patient who has dementia is actually considered for the placement of a PEG tube, the downward spiral of disease progression may limit the opportunity for a therapeutic response. Risks and Complications Certainly, even minimally invasive procedures carry the risks and complications related directly to the procedure, risks that would not be present for a similar patient who refuses to undergo the procedure. Overall, the incidence of minor complications associated with PEG placement (such as wound infection, clogging, leakage, hemorrhage, ileus, and tube malfunction) is 5% to 13% [54,55]. Major complications associated with PEG placement (which are usually defined by the requirement for repeat endoscopic procedures and include complications such as peritonitis, inadvertent removal of the tube, buried bumper syndrome, fasciitis, and tumor implantation of the PEG site) occur in only 1.3% to 3.0% of cases [54,55]. With regard to alternative methods for tube placement, radiologic gastrostomy tube placement has a comparable efficacy and profile of complications [56], whereas the surgical placement of a gastrostomy tube has a significantly higher morbidity than either radiologic or endoscopic techniques [24]. The mortality rate for PEG placement is very low, at 0.2% to 1.0% [54,55,57]. The overall morbidity and mortality rates for the patient undergoing PEG placement relate to the underlying disease process. PEG placement does not appear to hasten the patient’s demise. In patients who have ALS, the morbidity and mortality rate secondary to PEG placement rises as the FVC falls [24]. Retrospective nonrandomized studies that suggest an increased rate of complications with PEG placement (such as one study that reported increased incidence of pressure sores with PEG placement) [58] reflect selection bias (PEGs are more likely to be requested in patients who have pressure sores than in those without). Recent studies would suggest that the morbidity rate related directly to the PEG procedure itself is no worse for patients who have dementia than for those patients of similar age and comorbidities who do not have dementia [21].

92

MCCLAVE & RITCHIE

Technical Issues and Concerns There are several technical issues related to the procedure of PEG placement that may add to patient morbidity. In patients who have malignant esophageal carcinoma, surgeons are concerned that the placement of a PEG preoperatively jeopardizes the use of the stomach later as a conduit after esophagectomy. Injury to the gastroepiploic artery may increase the chance for ischemic injury, leakage, or delays in gastric emptying, postoperatively. The PEG placement may altogether prevent the use of the stomach or at least prolong surgery at the time of esophagectomy because of the need to take the stomach down from the anterior wall. More recent studies would suggest that these concerns are of little clinical significance. In a study [59] of 229 patients who underwent PEG placement for esophageal carcinoma, 67 eventually underwent esophagectomy. There was injury to the gastroepiploic artery in only one patient, which was deemed to be clinically insignificant. The stomach was able to be used as a conduit after PEG placement in 96% (64 of 67 patients) of patients who underwent esophagectomy. Leaks occurred postoperatively in only 6% of patients, comparable with the rest of the literature for these operations [59]. In a retrospective study [15], 61 patients who had esophageal carcinoma who underwent PEG placement were compared with 19 patients who did not get PEG placement, all of whom eventually underwent esophagectomy. Overall, there was no difference in the incidence of complications between PEG patients and non-PEG patients in regard to leakage (8.0% versus 10.5%, respectively), stricture (37.0% versus 32.5%, respectively), or delay of gastric emptying (9.8% versus 10.5%, respectively) [15]. Placing a PEG tube in patients who have a bulky exophytic oropharyngeal or esophageal carcinoma (and actually pulling the tube down through the tumor) increases the risk for seeding the PEG tract with cancer cells and subsequent tumor recurrence at the PEG site [60,61]. Although it was previously thought to be a rare complication, tumor seeding of the PEG tract has been estimated to occur in up to 3% of patients who have such tumors who undergo PEG placement [60,61]. In these reports, the average time from PEG placement to the development of the complication was 7.3 to 8.0 months [60,61]. Tumor recurrence usually presented as increased drainage, bleeding, superinfection, or the presence of a malodorous fungating mass at the PEG site [60,61]. Such complications can have a huge negative psychologic impact on the patient. Unfortunately, many of these patients lived long enough to require specific management of the complication, such as radiation to the site or surgical excision. This complication may be avoided entirely by using the Russell introducer technique at the time of PEG placement. With this technique, no tube is drawn through the proximal oropharynx or esophagus but, instead, is placed directly through the anterior abdominal wall into the stomach [62]. A third technical issue arises in patients who have esophageal carcinoma and a previously placed esophageal stent, who then choose to undergo PEG placement. Pulling the PEG down through an esophageal stent can result in the displacement or migration of the stent into the stomach. This complication may be avoided in three ways [63]. The dome of the internal bolster may be wrapped

ENDOSCOPICALLY PLACED PEG TUBES

93

by surgical suture to decrease the diameter, allowing it to be pulled through the stent. Once in the stomach, a laser or endoscopic scissors can be used to release the suture and allow for expansion of the dome. A second method involves holding the stent in place with rat-tooth biopsy forceps as the PEG is pulled through the stent. A third and more reliable technique involves the use of the Russell technique for initial PEG placement, in which the PEG is pushed directly through the anterior gastric wall and only the endoscope passes through the esophageal stent [63]. PERCUTANEOUS ENDOSCOPIC GASTROSTOMY PLACEMENT FOR DECOMPRESSION OF THE GASTROINTESTINAL TRACT Percutaneous Endoscopic Gastrostomy Placement in the Proximal Gastrointestinal Tract: Stomach and Small Bowel Approximately 3% of all advanced malignancies are complicated by malignant obstruction and paralytic ileus [64]. The two cancers most often associated with this complication are ovarian cancer (occurring in 5%–41% of cases) and colorectal carcinoma (occurring in 10%–28% of cases) [64]. Other gynecologic cancers such as endometrial, uterine, and cervical carcinoma are also associated frequently with malignant bowel obstruction. Approximately 25% of patients who have ovarian carcinoma must ultimately undergo exploratory laparotomy to correct small bowel obstruction [65]. The use of a percutaneously placed gastrostomy tube for the decompression of malignant bowel obstruction was first described in 1986 by Malone, using radiographic techniques [66], and then 1 year later by Gauderer, in 1987, using endoscopic techniques [67]. For patients who have ovarian cancer, the duration from the time of diagnosis to the development of malignant obstruction is, on average, 3.1 years [64]. The tumor may involve encasement of the stomach in 28% of patients or peritoneal metastasis with malignant ascites in 63% of patients [64]. Although surgery is clearly the first option, the use of PEGs for decompression is a viable alternative for patients who are deemed to be poor surgical candidates or for patients who refuse a surgical procedure. The PEG procedure itself, when used for purposes of decompression and malignant obstruction, differs little from the technique used for placement for nutritional purposes. Large 28F PEGs may provide better drainage than smaller, 14- to 20F PEGs [64,65]. Efforts should be made to avoid placing the PEG directly through areas of tumor in the gastric wall because this may be technically more difficult and may increase the risk of bleeding, infection, leakage, or tube displacement [65]. Nonetheless, placement of the PEG directly through tumor is still feasible, especially in patients in whom cancer encases the entire stomach [64]. Adding a gastropexy, in which the stomach is actually sutured to the anterior abdominal wall, to the procedure may help to promote maturation of the tract formation and prevent leakage after PEG placement in patients who have ascites [64]. The efficacy of this procedure is good. PEG tubes for decompression are placed successfully in 86% to 98% of these patients [68–70]. Symptom relief (with the absence of nausea and vomiting and the ability to tolerate oral liquids

94

MCCLAVE & RITCHIE

and soft foods) within 7 days of the procedure is observed in 84% to 91% of patients [68–70]. Frequently, patients are continued on parenteral nutrition for the maintenance of nutritional status. PEG placement for decompression often allows continuation of palliative chemotherapy, observed in 25% to 31% of patients [64,69]. For patients who continue to have nausea and vomiting after PEG placement for decompression, relief of symptoms may be achieved with an intravenous infusion of octreotide, 0.6 mg every 24 hours (to decrease secretions from the GI tract) [64,69]. The failure to place decompressive PEG tubes in these patients usually occurs because of malignant ascites, multiple previous surgeries, or tumor involvement of the abdominal wall, which precludes transillumination [65,69]. For these patients, transcutaneous ultrasonography may help localize the best site. Ascites should be drained as much as possible both before and for several days after the procedure [65,69,71]. Up to 29% of patients may require a direct percutaneous endoscopic jejunostomy because PEG placement is not feasible [71]. Overall, the morbidity rate related to the PEG procedure used for decompression is only slightly higher than when the same technique is used for nutritional purposes. Minor complications associated with PEG placement for decompression range from 19% to 25% and include infection, leakage, tube malfunction, and bleeding [64,68]. Major complications requiring repeat endoscopic procedures, such as dislodgement of the tube, occur in 4% to 9% of cases [64,68]. The alternative technique of decompression through a chronic NG tube is less acceptable and tolerated poorly by patients. Over time, such a tube becomes increasingly uncomfortable, interferes with coughing, and is associated with aspiration, alar necrosis, hemorrhage, and erosion with stricture at the esophagogastric junction. NG tubes are cosmetically unacceptable and confining to the patient and may increase the incidence of infection (such as otitis media and sinusitis) [69]. Percutaneous Endoscopic Gastrostomy Placement in the Distal Gastrointestinal Tract: Colon Similar percutaneous tubes may be placed endoscopically in the colon for palliative care of patients who have a chronic neurogenic colon. Cancer patients who have malignant large bowel obstruction or constipation because of highdose narcotic analgesia can be managed by such tubes, as can patients who have acute colonic pseudo-obstruction (Ogilvie’s syndrome) [72–75]. Percutaneous cecostomy tubes may be placed radiologically [76] or endoscopically [77]. These tubes can be used for drainage or decompression of the proximal colon for obstructed patients or to deliver laxatives for antegrade irrigation for patients who have chronic constipation caused by a neurogenic colon. Only slight modifications of the technique used for PEG placement for nutritional purposes are required to successfully place these tubes into the cecum. Antibiotic prophylaxis for percutaneous endoscopic cecostomy (PEC) must cover not only gram-positive organisms from the skin (piperacillin or amoxicillin) but also gram-negative organisms from fecal bacteria (tazobactam or clavulanic acid) [74]. Usually, the antibiotic prophylaxis is continued for

ENDOSCOPICALLY PLACED PEG TUBES

95

24 hours after the PEC procedure. For chronic neurogenic colon, a full preparation must be performed with polyethylene glycol, but such preparations cannot be performed for acute colonic pseudo-obstruction. In the latter patients, the cecum can be lavaged once it has been reached by the colonoscope. The preparation is difficult in patients who have chronic constipation and, as such, colonoscopy may be performed 1 day before the PEC procedure to deliver polyethylene glycol directly into the cecum (endoscopists have prefilled the water bottle of the endoscopy system with the polyethylene glycol) [75]. Usually, 20F PEG tubes are used for the PEC procedure. A longer polypectomy snare is required to grab the wire once it has been placed through into the cecum [74]. A 21- to 23-gauge spinal needle may be used as a sounding device to gain access to the cecum or to chart the initial site for the trocar puncture. Some reports encourage placement of T-fasteners to secure the cecum to the anterior abdominal wall to decrease the chance for leakage afterward [78]. Care must be taken in a patient with Ogilvie’s syndrome to avoid excess insufflation of air in an already distended cecum [74]. Once placed, the cecostomy should be flushed every 6 hours with 30 to 50 mL of water to prevent clogging. Surprisingly, in some patients it has been shown that the cecostomy tube may be removed endoscopically at a later time without complications. The tube probably should not be removed at the bedside by manual traction (to avoid tearing of the cecum) but instead should be cut at the skin and pulled back into the colon using an endoscope [74]. Complications reported with the PEC procedure include formulation of profuse granulation tissue, cellulitis down the abdominal wall, sepsis, and pressure necrosis from the external bolster [74]. The PEC procedure does provide some advantage over surgical cecostomy in that there is less likelihood for stenosis of the stoma, and theoretically there may be reduced postoperative complications because of less risk for the formation of adhesions [74]. PERCUTANEOUS ENDOSCOPIC GASTROSTOMY TUBE AS AN ADJUNCTIVE MEASURE FOR OTHER PROCEDURES Although some truly innovative applications of the PEG procedure have been used in isolated, unique situations, PEGs are used most commonly to facilitate the passage of a jejunostomy tube farther down in the GI tract (conversion of a PEG to a percutaneous endoscopic gastrojejunostomy or PEGJ) [79,80]. Such a conversion is used most often in cancer patients who experience nausea and vomiting from partial gastric outlet obstruction, tumor involving the autonomic nervous system, linitis plastica, abdominal carcinomatosis, superior mesenteric artery syndrome from weight loss, intolerance to chemotherapy, or delayed gastric emptying from the chronic use of narcotic analgesia. PEG placement has been used to help position an esophageal stent through a tight angulated malignant stricture [81]. Usually, a pulsion technique is used in which the stent is delivered by a pusher tube. With acute angulation of a stricture, however, this may be difficult. In one report, a PEG was placed initially and cut down to a 4-cm length from the external bolster [81]. The guide wire was left in place to facilitate passage of a Savory polyvinyl dilator. Nylon strands

96

MCCLAVE & RITCHIE

(270 cm long) were attached to the distal end of the esophageal prosthesis, tied to the guide wire, and then pulled out through the PEG tube. The stent was then placed over the Savory dilator, which was passed down through the stricture. The nylon strands were used to apply traction, pulling the stent into position. Once the stent was in position across the malignant stricture, the nylon strands were simply pulled off the stent and out through the PEG. This procedure was successful in three patients (two of whom had failed the pulsion technique). Success was attributed to the thought that the pulling technique tended to stretch the stent and thin or decrease its diameter, whereas the usual technique tended to push and compress the stent, increasing the diameter of the device [81]. In another unique situation, a PEG tube was used to reposition and stabilize the stomach before stent placement. In an unusual case, a patient who was deemed unfit for surgery was found to have an adenocarcinoma involving the distal esophagus and an irregular ‘‘upside-down’’ stomach [82]. The surgeon was able to reposition the stomach by endoscopy and then anchor it to the abdominal wall by subsequent PEG placement. Restoring the normal esophagogastric axis allowed for stent placement in a routine fashion [82]. More commonly, PEG tube placement is used to create an external extraintestinal bypass around an obstructed segment of the GI tract. Endoscopists at Sloan-Kettering Memorial Cancer Institute (NY) collaborated with the bioengineers to develop battery-powered pumps to promote the passage of luminal nutrients out through a PEG tube, through the pump, and back into the GI tract through a direct percutaneous endoscopic jejunostomy (PEJ) tube [83]. The presence of the PEG tube above and the PEJ tube below the level of obstruction allowed for decompression and control of nausea and vomiting as well as the reinstillation of biliary secretions back into the GI tract (to help control diarrhea). For patients who have malignant biliary obstruction, percutaneous transhepatic biliary drains may be connected back to a PEG or PEJ, again to allow the reinstillation of biliary secretions back into the GI tract [84]. SUMMARY The minimally invasive nature of endoscopically placed gastrostomy tubes makes them a viable consideration in palliative care. Complications related to the procedure appear to correlate with age and underlying comorbidities. However, in many instances, the scientific basis for establishing benefit or harm from tube placement is methodologically inadequate. Decisions must be preceded by a discussion of the value and potential risk of artificial nutrition in a particular setting, respecting the wishes and beliefs of each patient and his or her family. The decision to use PEG placement for any reason should be consistent with legal and ethical principles, reflect patient autonomy over any other consideration (including beneficence), and arise from a clear determination of the goals of care (and whether the PEG placement will truly help meet those goals). Whenever possible, further studies with better design are needed to evaluate whether the use of PEG truly affects quality of life and patient outcome in palliative care.

ENDOSCOPICALLY PLACED PEG TUBES

97

PEG tubes for decompression are placed successfully most of the time. Symptom relief occurs usually within 7 days of the procedure. Overall, the morbidity related to the PEG procedure used for decompression is only slightly higher than when the same technique is used for nutritional purposes. The appropriate timing for PEG tube placement for nutritional support and for decompression throughout the course of disease progression may be difficult to determine and yet may be a factor in its overall efficacy. Only minor modifications of the basic technique used for PEG placement for nutritional purposes are required to adapt the technique to a variety of other applications in palliative care. References [1] Gauderer MWL, Ponsky JL, Izant RJ. Gastrostomy without laparotomy: a percutaneous endoscopic technique. J Pediatr Surg 1980;15:872–5. [2] Brett AS, Rosenberg JC. The adequacy of informed consent for placement of gastrostomy tubes. Arch Intern Med 2001;161:745–8. [3] Mayo TW. Forgoing artificial nutrition and hydration: legal and ethical considerations. Nutrit Clin Pract 1996;11(6):254–64. [4] Monteleoni C, Clark E. Using rapid-cycle quality improvement methodology to reduce feeding tubes in patients with advanced dementia: before and after study. BMJ 2004; 329(7464):491–4. [5] Raynor EM, Williams MF, Martindale RG, et al. Timing of percutaneous endoscopic gastrostomy tube placement in head and neck cancer patients. Otolaryngol Head Neck Surg 1999;120(4):479–82. [6] Dresner SM, Wayman J, Lovel T, et al. Percutaneous endoscopic duodenostomy: the relief of obstruction in advanced gastric carcinoma. Palliat Med 1999;13(2):165–7. [7] Chio A, Silani V for the Italian ALS Study Group:. Amyotrophic lateral sclerosis care in Italy: a nationwide study in neurological centers. J Neurol Sci 2001;191(1–2):145–50. [8] Neudert C, Oliver D, Wasner M, et al. The course of the terminal phase in patients with amyotrophic lateral sclerosis. J Neurol 2001;248(7):612–6. [9] Mitchell SL, Teno JM, Roy J, et al. Clinical and organizational factors associated with feeding tube use among nursing home residents with advanced cognitive impairment. JAMA 2003;290(1):73–80. [10] Gillick MR. Rethinking the role of tube feeding in patients with advanced dementia. N Engl J Med 2000;342(3):206–10. [11] Dennis MS, Lewis SC, Warlow C for the FOOD Trial Collaboration. Effect of timing and method of enteral tube feeding for dysphagic stroke patients (FOOD): a multicentre randomized controlled trial. Lancet 2005;26(365):764–72. [12] Gutt CN, Held S, Paolucci V, et al. Experiences with percutaneous endoscopic gastrostomy. World J Surg 1996;20(8):1006–8. [13] Ponsky JL, Gauderer MW, Stellato TA. Percutaneous endoscopic gastrostomy: review of 150 cases. Arch Surg 1983;118(8):913–4. [14] Bozzetti F, Cozzaglio L, Gavazzi C, et al. Nutritional support in patients with cancer of the esophagus: impact on nutritional status, patient compliance to therapy, and survival. Tumori 1998;84(6):681–6. [15] Margolis M, Alexander P, Trachiotis GD, et al. Percutaneous endoscopic gastrostomy before multimodality therapy in patients with esophageal cancer. Ann Thorac Surg 2003;76(5):1694–7. [16] Gibson S, Wenig BL. Percutaneous endoscopic gastrostomy in the management of head and neck carcinoma. Laryngoscope 1992;102(9):977–80. [17] Mitchell SL, Kiely DK, Lipsitz LA. The risk factors and impact on survival of feeding tube placement in nursing home residents with severe cognitive impairment. Arch Intern Med 1997;157(3):327–32.

98

MCCLAVE & RITCHIE

[18] Mitchell SL, Kiely DK, Lipsitz LA. Does artificial enteral nutrition prolong the survival of institutionalized elders with chewing and swallowing problems? J Gerontol A Biol Sci Med Sci 1998;53(3):M207–13. [19] Cowen ME, Simpson SL, Vettese TE. Survival estimates for patients with abnormal swallowing studies. J Gen Intern Med 1997;12(2):88–94. [20] Croghan JE, Burke EM, Caplan S, et al. Pilot study of 12-month outcomes of nursing home patients with aspiration on videofluoroscopy. Dysphagia 1994;9(3):141–6. [21] Bourdel-Marchasson I, Dumas F, Pinganaud G, et al. Audit of percutaneous endoscopic gastrostomy in long-term enteral feeding in a nursing home. Int J Qual Health Care 1997;9(4): 297–302. [22] Murphy LM, Lipman TO. Percutaneous endoscopic gastrostomy does not prolong survival in patients with dementia. Arch Intern Med 2003;163(11):1351–3. [23] Finucane TE, Christmas C, Travis K. Tube feeding in patients with advanced dementia: a review of the evidence. JAMA 1999;282(14):1365–70. [24] Hardiman O. Symptomatic treatment of respiratory and nutritional failure in amyotrophic lateral sclerosis. J Neurol 2000;247(4):245–51. [25] Jarnagin WR, Duh QY, Mulvihill SJ, et al. The efficacy and limitations of percutaneous endoscopic gastrostomy. Arch Surg 1992;127(3):261–4. [26] Kadakia SC, Sullivan HO, Starnes E. Percutaneous endoscopic gastrostomy or jejunostomy and the incidence of aspiration in 79 patients. Am J Surg 1992;164(2):114–8. [27] Patel PH, Thomas E. Risk factors for pneumonia after percutaneous endoscopic gastrostomy. J Clin Gastroenterol 1990;12(4):389–92. [28] Cogen R, Weinryb J. Aspiration pneumonia in nursing home patients fed via gastrostomy tubes. Am J Gastroenterol 1989;84(12):1509–12. [29] DiSario JA, Foutch PG, Sanowski RA. Poor results with percutaneous endoscopic jejunostomy. Gastrointest Endosc 1990;36(3):257–60. [30] Wolfsen HC, Kozarek RA, Ball TJ, et al. Tube dysfunction following percutaneous endoscopic gastrostomy and jejunostomy. Gastrointest Endosc 1990;36(3):261–3. [31] Pingleton SK, Hinthorn DR, Liu C. Enteral nutrition in patients receiving mechanical ventilation: multiple sources of tracheal colonization include the stomach. Am J Med 1986;80(5): 827–32. [32] Torres A, el-Ebiary M, Gonzalez J, et al. Gastric and pharyngeal flora in nosocomial pneumonia acquired during mechanical ventilation. Am Rev Respir Dis 1993;148(2):352–7. [33] Bonten MJ, Gaillard CA, van Tiel FH, et al. The stomach is not a source for colonization of the upper respiratory tract and pneumonia in ICU patients. Chest 1994;105(3):878–84. [34] Johnson DA, Hacker JF III, Benjamin SB, et al. Percutaneous endoscopic gastrostomy effects on gastroesophageal reflux and the lower esophageal sphincter. Am J Gastroenterol 1987;82(7):622–4. [35] McClave SA, Lukan JK, Stefater AJ, et al. Poor validity of residual volume as a marker for risk of aspiration in critically ill patients. Crit Care Med 2005;33(2):324–30. [36] Heyland DK, Drover JW, MacDonald S, et al. Effect of postpyloric feeding on gastroesophageal regurgitation and pulmonary microaspiration: results of a randomized controlled trial. Crit Care Med 2001;29(8):1495–501. [37] Lien HC, Chang CS, Chen GH. Can percutaneous endoscopic jejunostomy prevent gastroesophageal reflux in patients with preexisting esophagitis? Am J Gastroenterol 2000;95(12):3439–43. [38] Heyland DK, Drover JW, Dhaliwal R, et al. Optimizing the benefits and minimizing the risks of enteral nutrition in the critically ill: role of small bowel feeding. JPEN J Parenter Enteral Nutr 2002;26(Suppl 6):S51–7 [discussion: S56–7]. [39] Marik PE, Zaloga GP. Gastric versus post-pyloric feeding: a systematic review. Crit Care 2003;7(3):R46–51. [40] McCann RM, Hall WJ, Groth-Junker A. Gastrostomy placement and mortality among hospitalized Medicare beneficiaries. JAMA 1999;282:1365–70.

ENDOSCOPICALLY PLACED PEG TUBES

99

[41] Callahan CM, Haag KM, Weinberger M, et al. Outcomes of percutaneous endoscopic gastrostomy among older adults in a community setting. J Am Geriatr Soc 2000;48(9):1048–54. [42] Weaver JP, Odell P, Nelson C. Evaluation of the benefits of gastric tube feeding in an elderly population. Arch Fam Med 1993;2(9):953–6. [43] Boyd KJ, Beeken L. Tube feeding in palliative care: benefits and problems. Palliat Med 1994;8(2):156–8. [44] Bannerman E, Pendlebury J, Phillips F, et al. A cross-sectional and longitudinal study of health-related quality of life after percutaneous gastrostomy. Eur J Gastroenterol Hepatol 2000;12(10):1101–9. [45] Finucane TE. Malnutrition, tube feeding and pressure sores: data are incomplete. J Am Geriatr Soc 1995;43(4):447–51. [46] Berlowitz DR, Ash AS, Brandeis GH, et al. Rating long-term care facilities on pressure ulcer development: importance of case-mix adjustment. Ann Intern Med 1996;124(6):557–63. [47] Berlowitz DR, Brandeis GH, Anderson J, et al. Predictors of pressure ulcer healing among long-term care residents. J Am Geriatr Soc 1997;45(1):30–4. [48] Kleber MJ, Lowen CC, McClave SA, et al. Is there a role for indirect calorimetry in maximizing patient outcome from nutritional alimentation in the long-term nursing care setting? Nutr Clin Pract 2000;15:227–33. [49] Hoffman DR. The False Claims Act as a remedy to the inadequate provision of nutrition and wound care to nursing home residents. Adv Wound Care 1996;9(5):25–9. [50] Borasio GD, Miller RG. Clinical characteristics and management of ALS. Semin Neurol 2001;21(2):155–66. [51] Kasarskis EJ, Scarlata D, Hill R, et al. A retrospective study of percutaneous endoscopic gastrostomy in ALS patients during the BDNF and CNTF trials. J Neurol Sci 1999;169(1–2):118–25. [52] Mathus-Vliegen LM, Louwerse LS, Merkus MP, et al. Percutaneous endoscopic gastrostomy in patients with amyotrophic lateral sclerosis and impaired pulmonary function. Gastrointest Endosc 1994;40(4):463–9. [53] Mazzini L, Corra T, Zaccala M, et al. Percutaneous endoscopic gastrostomy and enteral nutrition in amyotrophic lateral sclerosis. J Neurol 1995;242(10):695–8. [54] Larson DE, Burton DD, Schroeder KW, et al. Percutaneous endoscopic gastrostomy: indications, success, complications, and mortality in 314 consecutive patients. Gastroenterology 1987;93(1):48–52. [55] Grant JP. Percutaneous endoscopic gastrostomy: initial placement by single endoscopic technique and long-term follow-up. Ann Surg 1993;217(2):168–74. [56] Righi PD, Reddy DK, Weisberger EC, et al. Radiologic percutaneous gastrostomy: results in 56 patients with head and neck cancer. Laryngoscope 1998;108(7):1020–4. [57] Kohli H, Bloch R. Percutaneous endoscopic gastrostomy: a community hospital experience. Am Surg 1995;61(3):191–4. [58] Michocki RJ, Lamy PP. The problem of pressure sores in a nursing home population: statistical data. J Am Geriatr Soc 1976;24(7):323–8. [59] Stockeld D, Fagerberg J, Granstrom L, et al. Percutaneous endoscopic gastrostomy for nutrition in patients with oesophageal cancer. Eur J Surg 2001;167(11):839–44. [60] Pickhardt PJ, Rohrmann CA Jr, Cossentino MJ. Stomal metastases complicating percutaneous endoscopic gastrostomy: CT findings and the argument for radiologic tube placement. AJR Am J Roentgenol 2002;179(3):735–9. [61] Brown MC. Cancer metastasis at percutaneous endoscopic gastrostomy stomata is related to the hematogenous or lymphatic spread of circulating tumor cells. Am J Gastroenterol 2000;95(11):3288–91. [62] Adler DJ, Nunez DC. Large bore percutaneous endoscopic gastrostomy using the Russell technique. Gastrointest Endosc 1989;35(3):239–41. [63] Seibert DG, Matulis SR, Gaskins RD. Utilization of an esophageal stent for subsequent placement of a percutaneous endoscopic gastrostomy tube. Am J Gastroenterol 1997;92(6):1048–50.

100

MCCLAVE & RITCHIE

[64] Pothuri B, Montemarano M, Gerardi M, et al. Percutaneous endoscopic gastrostomy tube placement in patients with malignant bowel obstruction due to ovarian carcinoma. Gynecol Oncol 2005;96(2):330–4. [65] Adelson MD, Kasowitz MH. Percutaneous endoscopic drainage gastrostomy in the treatment of gastrointestinal obstruction from intraperitoneal malignancy. Obstet Gynecol 1993;81(3):467–71. [66] Malone JM Jr, Koonce T, Larson DM, et al. Palliation of small bowel obstruction by percutaneous gastrostomy in patients with progressive ovarian carcinoma. Obstet Gynecol 1986;68(3):431–3. [67] Stellato TA, Gauderer MW. Percutaneous endoscopic gastrostomy for gastrointestinal decompression. Ann Surg 1987;205(2):119–22. [68] Scheidbach H, Horbach T, Groitl H, et al. Percutaneous endoscopic gastrostomy/jejunostomy (PEG/PEJ) for decompression in the upper gastrointestinal tract: initial experience with palliative treatment of gastrointestinal obstruction in terminally ill patients with advanced carcinomas. Surg Endosc 1999;13(11):1103–5. [69] Campagnutta E, Cannizzaro R, Gallo A, et al. Palliative treatment of upper intestinal obstruction by gynecological malignancy: the usefulness of percutaneous endoscopic gastrostomy. Gynecol Oncol 1996;62(1):103–5. [70] Ryan JM, Hahn PF, Mueller PR. Performing radiologic gastrostomy or gastrojejunostomy in patients with malignant ascites. AJR Am J Roentgenol 1998;171(4):1003–6. [71] Campagnutta E, Cannizzaro R. Percutaneous endoscopic gastrostomy (PEG) in palliative treatment of non-operable intestinal obstruction due to gynecologic cancer: a review. Eur J Gynaecol Oncol 2000;21(4):397–402. [72] Morrison MC, Lee MJ, Stafford SA, et al. Percutaneous cecostomy: controlled transperitoneal approach. Radiology 1990;176(2):574–6. [73] Salim AS. Percutaneous decompression and irrigation for large bowel obstruction: a new approach. Dis Colon Rectum 1991;34(11):973–80. [74] Ramage JI Jr, Baron TH. Percutaneous endoscopic cecostomy: a case series. Gastrointest Endosc 2003;57(6):752–5. [75] Wills JC, Trowbridge B, Disario JA, et al. Percutaneous endoscopic cecostomy for management of refractory constipation in an adult patient. Gastrointest Endosc 2003;57(3):423–6. [76] Casola G, Withers C, vanSonnenberg E, et al. Percutaneous cecostomy for decompression of the massively distended cecum. Radiology 1986;158(3):793–4. [77] Ponsky JL, Aszodi A, Perse D. Percutaneous endoscopic cecostomy: a new approach to nonobstructive colonic dilation. Gastrointest Endosc 1986;32(2):108–11. [78] Chevallier P, Marcy PY, Francois E, et al. Controlled transperitoneal percutaneous cecostomy as a therapeutic alternative to the endoscopic decompression for Ogilvie’s syndrome. Am J Gastroenterol 2002;97(2):471–4. [79] Chaurasia OP, Chang KJ. A novel technique for percutaneous endoscopic gastrojejunostomy tube placement. Gastrointest Endosc 1995;42(2):165–8. [80] Shike M, Wallach C, Bloch A, et al. Combined gastric drainage and jejunal feeding through a percutaneous endoscopic stoma. Gastrointest Endosc 1990;36(3):290–2. [81] Foutch PG, Talbert G, Sanowski RA. Nonoperative traction method for placement of esophageal stents: a new use for the percutaneous endoscopic gastrostomy. Gastrointest Endosc 1988;34(3):259–62. [82] Criblez DH. Percutaneous endoscopic gastrostomy to treat upside-down stomach before stent insertion in a patient with distal esophageal carcinoma. Am J Gastroenterol 1998;93(10):1938–41. [83] Shike M. Percutaneous endoscopic stomas for enteral feeding and drainage. Oncology (Williston Park, NY) 1995;9(1):39–44. [84] Shike M, Gerdes H, Botet J, et al. External biliary duodenal drainage through a percutaneous endoscopic duodenostomy. Gastrointest Endosc 1989;35(2):104–5.

Gastroenterol Clin N Am 35 (2006) 101–112

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Palliation of Malignant Obstructive Jaundice Todd H. Baron, MD Division of Gastroenterology & Hepatology, Mayo Clinic College of Medicine, 200 First Street Southwest, Charlton 8A, Rochester, MN 55905, USA

M

alignant obstructive jaundice in the patient with incurable malignancy may cause symptoms that reduce quality of life. Palliation of malignant obstructive jaundice leads to improvement in quality of life. Nonsurgical methods using percutaneous and endoscopic methods have supplanted surgical bypass as the primary method for the palliation of malignant obstructive jaundice. This article reviews management strategies for palliation of malignant obstructive jaundice. CAUSES OF MALIGNANT OBSTRUCTIVE JAUNDICE Malignant obstructive jaundice may be caused by various malignancies, most commonly pancreatic carcinoma. Ampullary cancer, primary bile duct cancer (cholangiocarcinoma), and metastatic lesions to the head of the pancreas or porta hepatis also may cause biliary obstruction (Box 1). The location of the obstruction within the biliary tree is important with regards to the palliative approach within each discipline, and it generally is divided into nonhilar and hilar biliary obstruction (see Box 1). In general, it is more difficult both technically and clinically to successfully relieve jaundice from hilar biliary obstruction regardless of the method used to alleviate obstruction. Nonetheless, even patients with metastatic disease as the cause of hilar biliary obstruction usually can be palliated effectively using a nonsurgical approach [1]. SYMPTOMS AND QUALITY OF LIFE Obstructive jaundice can produce various symptoms. Anorexia, usually attributed to cytokines, may be caused by biliary obstruction. Weight loss, another symptom attributed to the tumor itself, may be caused by malabsorption of fat, even in the absence of overt steatorrhea [2]. Additionally, because of the lack of secretion of bile salts into the small intestine, vitamin K malabsorption occurs. This is important for two reasons. First, in patients who are undergoing invasive procedures during the evaluation and treatment of obstructive jaundice, E-mail address: [email protected] 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2006.01.001

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

102

BARON

Box 1: Common causes of malignant obstructive jaundice Nonhilar biliary obstruction 

Pancreatic cancer



Metastatic disease to the head of pancreas (renal cell carcinoma, breast cancer)



Ampullary cancer



Nodal compression



Gallbladder cancer



Extrinsic masses

Hilar obstruction 

Cholangiocarcinoma



Metastatic disease



Central intrahepatic masses, primary or secondary



Gallbladder cancer

the coagulopathy should be reversed to minimize and prevent bleeding complications. Second, in patients who are already on warfarin sodium, excessive anticoagulation may occur and result in severe bleeding complications. Pruritus usually occurs with higher levels of serum bilirubin and may be intractable and disabling. Finally, social embarrassment may occur as a result of the overt physical presence of jaundice. Cholangitis rarely occurs in patients with malignant obstructive jaundice, unless there has been previous instrumentation of the biliary tree. Relief of obstructive jaundice has been shown to improve quality if life [3,4]. METHODS OF PALLIATION There are three methods for palliation of obstructive jaundice: surgical bypass, percutaneous insertion of stents, and endoscopic insertion of stents. Each of these approaches has advantages and disadvantages (Table 1). There are very few comparative studies of outcomes for each of the disciplines, although nonsurgical palliation of obstructive jaundice is preferred in those with poor performance status, intra-abdominal ascites, and/or expected survival of less than 4 to 6 months. Surgical Palliation Surgical palliation of malignant obstructive jaundice is achieved by bypassing the obstructed system through the creation of a choledocho– or hepatico–enterostomy, usually a jejunostomy [5]. Traditionally, biliary bypass was performed using open techniques, but more recently, it has performed laparoscopically [6].

MALIGNANT OBSTRUCTIVE JAUNDICE

103

Table 1 Approaches to palliation of obstructive jaundice

Surgical

Percutaneous Endoscopic

Advantages

Disadvantages

Life-long palliation usual Simultaneous palliation of gastric outlet obstruction Simultaneous palliation of pain (intraoperative nerve block) Nonsurgical

Most invasive Morbidity and mortality

Nonsurgical Relatively painless Outpatient treatment possible

Pain, bleeding External drains Stents may occlude Expertise variable Complications (perforation, pancreatitis, bleeding)

Most of the data for surgical bypass are derived from the treatment of distal biliary obstruction caused by pancreatic cancer. Palliative surgical bypass may be performed at the time of laparotomy for attempted curative resection. If the patient is found to be unresectable, a surgical palliative approach may be undertaken. Alternatively, patients deemed inoperable based upon extent of disease, comorbid medical illnesses, or advanced age are referred for nonsurgical palliation. Since the evolution of better preoperative staging modalities (endoscopic ultrasound, thin section CT, and MRI) and the improvement in nonoperative palliative strategies, it appears that fewer patients are undergoing operative palliation. Nonetheless, surgical palliation has the advantage of allowing simultaneous palliation of biliary and gastric outlet obstruction and pain control in one setting [5]. Lifelong palliation of biliary obstruction without the need for further procedures usually is obtained. The average postoperative length of hospital stay for patients who undergo surgical palliation is less than 15 days. The average survival of patients who receive surgical palliation alone for nonmetastatic, unresectable pancreatic cancer is approximately 8 months [5]. The disadvantage of surgical palliation is the morbidity, especially in those with advanced disease. Surgical palliation of distal biliary obstruction Surgical palliation is performed using a biliary–enteric bypass, most commonly the bile duct to the jejunum (choledocho–jejunostomy). The use of the gallbladder as a conduit (cholecysto–jejunostomy) is utilized rarely, because premature occlusion of the cystic duct by tumor growth leads to subsequent loss of palliation and need for additional palliative procedures [7]. Successful palliation of jaundice is achieved in nearly all patients. Complications include wound infection, pneumonia, and bleeding. Surgical palliation of hilar biliary obstruction Surgical palliation is performed infrequently for palliation of hilar biliary obstruction. The bypass is created by sewing intrahepatic ducts to the jejunum

104

BARON

(hepatico–jejunostomy) [8,9]. The success rate for relief of jaundice is lower; the complication rate is higher, and the recurrence of jaundice is more likely with hilar obstruction than with distal biliary obstruction [5,10]. Biliary enteric anastomosis specifically for hilar cholangiocarcinoma carries a morbidity and mortality as high as 50% and 12%, respectively [10]. One serious complication is the development of an anastomotic bile leak. Nonsurgical Palliation Nonsurgical palliation of malignant obstructive jaundice is achieved with stent placement by means of the percutaneous, endoscopic, or rarely combined percutaneous–endoscopic approaches. For nonsurgical techniques, biliary stents used are composed of plastic (Fig. 1) or metal materials. Plastic stents occlude, because the formation of bacterial biofilm and plant materials [11] results in recurrent jaundice, frequently with cholangitis, and necessitates repeat procedures with stent replacement. Self-expandable metal stents (SEMS) have combined the advantage of a small predeployment delivery system with a large postdeployment stent diameter (Fig. 2). Additionally, they are less likely to occlude than plastic stents. Metal stents still may occlude because of tumor ingrowth through the mesh, tumor overgrowth (tumor growing beyond the ends of the stent), or tissue hyperplasia (excessive normal tissue growth in response to the stent). SEMS may be uncovered or covered. Covered SEMS resist occlusion from tumor ingrowth or tissue hyperplasia, although they are

Fig. 1. Examples of 10Fr plastic biliary stents.

MALIGNANT OBSTRUCTIVE JAUNDICE

105

Fig. 2. Examples of deployed self-expanding metal biliary stents. Note the third stent from the left is covered.

more likely to migrate. They cannot be used across the bifurcation, because the covering prevents drainage of intrahepatic bile ducts. Percutaneous Palliation Percutaneous therapy is performed by interventional radiologists by means of a transhepatic approach (percutaneous transhepatic cholangiography, PTC). The intrahepatic bile ducts initially are accessed with a skinny needle under fluoroscopic guidance and an anterograde approach. Three possible outcomes may occur following attempted percutaneous drainage. The least desirable is termed external drainage. This occurs when the bile ducts are entered, but there is a failure to traverse the stricture to place catheters and guide wires into the duodenum. Thus, the only route for bile drainage is externally, through the percutaneous tube (Fig. 3). The second possible scenario is called internal–external drainage. This occurs when the stricture is traversed, and a catheter is passed into the duodenum; the external portion exits the skin (Fig. 4). Bile can drain internally and externally, and eventually the external tube can be capped or completely internalized. Usually, the most desirable outcome is internal drainage, whereby the entire drainage is within the patient by means of an internal stent. Some patients may have internalization on the first attempt, or rarely, stages of advancement may occur from external to internal. For example, one procedure may be required for initial external tube placement and another for internalization to a stent. Uncommonly, internalization is not technically possible, and the patient is left with an external drain. Prior to the advent of SEMS, the percutaneous tract through the liver required dilation to accommodate large-bore plastic stents. This was associated with a higher incidence of postprocedural pain and bleeding. Because SEMS have a small predeployment delivery system, the tract through the liver requires less dilation. The main disadvantage of percutaneous therapy is the pain felt as a result of the catheter placement procedure through body wall and the liver

106

BARON

Fig. 3. Radiograph of external biliary drain (arrowheads). The stricture below the catheter was not traversable at the time of percutaneous transhepatic cholangiography.

capsule, and the subsequent unpleasantness of having one’s biliary secretions visible as a reminder of their illness. Percutaneous palliation of distal biliary obstruction Percutaneous insertion of plastic or metal stents is achievable in nearly all patients with distal bile duct obstruction. With the advent of endoscopic

Fig. 4. Radiograph of patient with internal–external biliary drain. Note the external portion (arrowheads) leads to the skin. Bile flows through the drain into the duodenum, where the distal end is coiled (arrow).

MALIGNANT OBSTRUCTIVE JAUNDICE

107

palliation, percutaneous therapy has become less commonly performed and is used when endoscopic techniques fail or when the papilla is inaccessible because of duodenal invasion by tumor or because of postsurgical anatomy. There is only one randomized, prospective trial comparing percutaneous and endoscopic therapy for the palliation of distal malignant obstructive jaundice. This landmark study showed that the endoscopic method had a significantly higher success rate for relief of jaundice (81% versus 61%) and significantly lower 30-day mortality (15% versus 33%). The higher mortality in the percutaneous group was caused by complications associated with liver puncture (bleeding and bile leaks) [12]. This study is dated, because it was performed using plastic stents prior to the development of expandable metal stents, the insertion of which is less likely to require aggressive dilation of the percutaneous tract through the liver, which in turn decreases bleeding and bile leakage [13,14]. Nonetheless, more recent studies have shown that major complications following percutaneous SEMS placement are still more common than major complications following endoscopic stent placement [15]. Percutaneous stent placement, however, is effective in most patients who have failed attempts at endoscopic stent placement [16]. Additionally, in centers without expertise in endoscopic retrograde cholangiopancreatography (ERCP), percutaneous therapy is used as the primary method of nonsurgical palliation of malignant obstructive jaundice. Overall, the technical success rate of percutaneous expandable metal stent placement is high, with a high rate of clinical resolution of jaundice [17]. In addition to primary tumors of the biliary tree, metastatic disease to the biliary system may occur. Successful palliation is also achievable in this group using the percutaneous approach [18]. Percutaneous palliation of hilar biliary obstruction Palliation of hilar obstruction is more difficult because of the inability to place one stent to drain the entire liver. Indeed, the early reports of percutaneous drainage showed that only 50% of patients achieved either fair or good palliation [19]. More recently, the use of expandable stents has produced better results, although when compared with nonhilar tumors, the outcome is inferior. For example, one study comparing the outcome of 58 patients who underwent percutaneous intervention for palliation of hilar and nonhilar tumors showed that relief of jaundice was significantly less in hilar tumors and that the long-term patency of expandable stents was significantly lower in the presence of hilar involvement [20]. It appears from the literature that only one side of the liver (ie, one stent or tube) is required to palliate jaundice in most patients [9,10,21]. Endoscopic Endoscopic palliation of distal biliary obstruction ERCP and biliary stent placement (Fig. 5) has been shown in randomized trials to be an acceptable alternative to palliative surgical bypass [22]. Biliary stents

108

BARON

Fig. 5. Illustration of endoscopic placement of plastic biliary stent for relief of jaundice from malignant biliary obstruction. (Courtesy of Wilson-Cook Medical, Winston-Salem, NC; with permission.)

can be placed safely in an outpatient setting [23]. The comparative studies of endoscopy and surgery for palliation of distal biliary obstruction were performed using plastic stents prior to the advent SEMS. As previously mentioned, the main limitation to plastic stent placement is stent occlusion as a result of bacterial biofilm. Thus, in the comparative trials of surgery and endoscopy, the lower initial hospital stay in the endoscopy group was offset by the need for subsequent hospitalization and need for subsequent ERCP to manage stent occlusion. The median time for stent occlusion for standard large-bore stents is approximately 3 months. Stent occlusion results in recurrent jaundice, usually with cholangitis. Expandable metal stents have overcome the problem of bacterial biofilm, and randomized controlled trials have shown superior patency rates to plastic stents [24]. Because the cost of expandable metal stents is much greater than that of plastic stents, they are cost-effective only if the patient lives greater than 3 to 6 months. Therefore, projected life expectancy should be considered when choosing between plastic and metal stents [25]. Other factors to be considered include patient compliance, ability to return for care, and early stent occlusion of plastic stents. Metal stent occlusion generally is managed easily with placement of a plastic stent or another metal stent within the existing metal stent (Fig. 6) [26]. More recently, covered metal stents have been developed to overcome occlusion caused by tumor overgrowth and tissue hyperplasia. Early comparative studies have shown prolonged patency with covered metal stents compared with uncovered stents [27].

MALIGNANT OBSTRUCTIVE JAUNDICE

109

Fig. 6. (A) Endoscopic photo of occluded metal stent caused by tissue hyperplasia and/or tumor. (B) Same patient. Plastic stent (blue) has been inserted through the metal stent.

Endoscopic palliation of hilar biliary obstruction Hilar strictures may be caused by cholangiocarcinoma or metastatic diseases. The clinical success rates for achieving adequate palliation for hilar tumors is less than that for distal tumors [28]. Furthermore, technical success rates for bilateral endoscopic stent placement (right and left hepatic ducts) are also lower. Most patients with hilar obstruction will be palliated adequately when only one side of the liver (unilateral drainage) is drained, assuming only one side has been accessed [9]. Patients who have had contrast instilled in both systems require stenting of both systems to prevent progressive cholangitis [29]. It is not as firmly established that metal stents offer superior prolongation of palliation compared with plastic stents, as is seen in distal strictures. In a prospective, single-arm pilot study of metal stent placement in 17 patients with Bismuth type II and III obstruction, however, median stent patency was 12 months [30]. A more recent noncomparative, single-arm study showed that Wallstent insertion is safe and feasible, and achieves successful palliation without the need for further biliary reintervention in most (69%) patients with nonresectable hilar cholangiocarcinoma [31]. This suggests that metal stents may offer the same benefits in hilar biliary obstruction as in distal biliary obstruction. Recently, photodynamic therapy (PDT) has been used for palliation of patients with unresectable hilar cholangiocarcinoma who did not resolve their jaundice after adequate endoscopic stent placement [32]. Significant improvement in cholestasis, quality of life, and survival (as compared with historical controls) has been demonstrated in one study [33]. In another study, improvement in cholestasis, performance, and quality of life was seen and maintained for an extended period [34]. These studies were performed outside of the United States, where smaller, more flexible laser fibers are available. Currently, passage of laser fibers available for the treatment of esophageal cancer into the biliary system is difficult, but feasible.

110

BARON

Overall, in patients with unresectable cholangiocarcinoma who are undergoing ERCP for palliation, unilateral stenting should be performed with contrast injection confined to one lobe. A pre-ERCP abdominal CT may reveal atrophy of one lobe, and this lobe specifically is avoided, because contamination will require drainage to prevent cholangitis but likely will not add to palliation. Alternatively, MRI can define the anatomy prior to ERCP. One stent, plastic or metal is usually adequate to achieve palliation [35]. PDT for cholangiocarcinoma should be performed in select centers until randomized trials (currently underway in Europe) clearly demonstrate superiority of PDT over stents alone. Brachytherapy is yet another approach to palliation of hilar cholangiocarcinoma, although it is used infrequently and in very few centers. Brachytherapy is performed by placing Iridium92 catheters into the biliary tree so as to deliver high-dose intraductal radiation therapy. Combined Percutaneous and Endoscopic Approach to Biliary Obstruction In some situations in which endoscopic therapy fails to access the bile and/or traverse the stricture, a percutaneous approach is performed to facilitate the endoscopic procedure [10]. The advantage of this approach versus the completion of the entire procedure percutaneously may be realized in selected situations [36]. SUMMARY Palliation of obstructive jaundice can be achieved in most patients using various approaches. The method chosen should be individualized to the patient and based upon performance status, patient preferences, and available expertise. The best approach ideally should be determined by a multi-discipline approach with endoscopists, interventional radiologists, oncologists, and surgeons. References [1] Van Laethem JL, De Broux S, Eisendrath P, et al. Clinical impact of biliary drainage and jaundice resolution in patients with obstructive metastases at the hilum. Am J Gastroenterol 2003;98(6):1271–7. [2] Wakasugi H, Hara Y, Abe M. A study of malabsorption in pancreatic cancer. J Gastroenterol 1996;31(1):81–5. [3] Ballinger AB, McHugh M, Catnach SM, et al. Symptom relief and quality of life after stenting for malignant bile duct obstruction. Gut 1994;35:467–70. [4] Abraham NS, Barkun JS, Barkun AN. Palliation of malignant biliary obstruction: a prospective trial examining impact on quality of life. Gastrointest Endosc 2002;56(6):835–41. [5] House MG, Choti MA. Palliative therapy for pancreatic/biliary cancer. Surg Clin North Am 2005;85(2):359–71. [6] Date RS, Siriwardena AK. Laparoscopic biliary bypass and current management algorithms for the palliation of malignant obstructive jaundice. Ann Surg Oncol 2004;11(9):815–7. [7] Urbach DR, Bell CM, Swanstrom LL, et al. Cohort study of surgical bypass to the gallbladder or bile duct for the palliation of jaundice due to pancreatic cancer. Ann Surg 2003;237(1): 86–93.

MALIGNANT OBSTRUCTIVE JAUNDICE

111

[8] Jarnagin WR, Burke E, Powers C, et al. Intrahepatic biliary enteric bypass provides effective palliation in selected patients with malignant obstruction at the hepatic duct confluence. Am J Surg 1998;175(6):453–60. [9] Freeman ML, Sielaff TD. A modern approach to malignant hilar biliary obstruction. Rev Gastroenterol Disord 2003;3(4):187–201. [10] Singhal D, van Gulik TM, Gouma DJ. Palliative management of hilar cholangiocarcinoma. Surg Oncol 2005;14(2):59–74. [11] van Berkel AM, van Marle J, Groen AK, et al. Mechanisms of biliary stent clogging: confocal laser scanning and scanning electron microscopy. Endoscopy 2005;37(8):729–34. [12] Speer AG, Cotton PB, Russell RC, et al. Randomised trial of endoscopic versus percutaneous stent insertion in malignant obstructive jaundice. Lancet 1987;2(8550):57–62. [13] Inal M, Aksungur E, Akgul E, et al. Percutaneous placement of metallic stents in malignant biliary obstruction: one-stage or two-stage procedure? Predilate or not? Cardiovasc Intervent Radiol 2003;26(1):40–5. [14] Beissert M, Wittenberg G, Sandstede J, et al. Metallic stents and plastic endoprostheses in percutaneous treatment of biliary obstruction. Z Gastroenterol 2002;40(7):503–10. [15] Pinol V, Castells A, Bordas JM, et al. Percutaneous self-expanding metal stents versus endoscopic polyethylene endoprostheses for treating malignant biliary obstruction: randomized clinical trial. Radiology 2002;225(1):27–34. [16] Doctor N, Dick R, Rai R, et al. Results of percutaneous plastic stents for malignant distal biliary obstruction following failed endoscopic stent insertion and comparison with current literature on expandable metallic stents. Eur J Gastroenterol Hepatol 1999;11(7):775–80. [17] Kaskarelis IS, Papadaki MG, Papageorgiou GN, et al. Long-term follow-up in patients with malignant biliary obstruction after percutaneous placement of uncovered wall stent endoprostheses. Acta Radiol 1999;40(5):528–33. [18] Kaskarelis IS, Minardos IA, Abatzis PP, et al. Percutaneous metallic self-expandable endoprostheses in biliary obstruction caused by metastatic cancer. Hepatogastroenterology 1996;43(10):785–91. [19] Gibson RN, Yeung E, Hadjis N, et al. Percutaneous transhepatic endoprostheses for hilar cholangiocarcinoma. Am J Surg 1988;156(5):363–7. [20] Becker CD, Glattli A, Maibach R, et al. Percutaneous palliation of malignant obstructive jaundice with the wall stent endoprosthesis: follow-up and reintervention in patients with hilar and non-hilar obstruction. J Vasc Interv Radiol 1993;4(5):597–604. [21] Inal M, Akgul E, Aksungur E, et al. Percutaneous placement of biliary metallic stents in patients with malignant hilar obstruction: unilobar versus bilobar drainage. J Vasc Interv Radiol 2003;14(11):1409–16. [22] Taylor MC, McLeod RS, Langer B. Biliary stenting versus bypass surgery for the palliation of malignant distal bile duct obstruction: a meta-analysis. Liver Transpl 2000;6(3):302–8. [23] Cvetkovski B, Gerdes H, Kurtz RC. Outpatient therapeutic ERCP with endobiliary stent placement for malignant common bile duct obstruction. Gastrointest Endosc 1999;50(1):63–6. [24] Levy MJ, Baron TH, Gostout CJ, et al. Palliation of malignant extrahepatic biliary obstruction with plastic versus expandable metal stents: an evidence-based approach. Clin Gastroenterol Hepatol 2004;2(4):273–85. [25] Arguedas MR, Heudebert GH, Stinnett AA, et al. Biliary stents in malignant obstructive jaundice due to pancreatic carcinoma: a cost-effectiveness analysis. Am J Gastroenterol 2002;97(4):898–904. [26] Bueno JT, Gerdes H, Kurtz RC. Endoscopic management of occluded biliary wall stents: a cancer center experience. Gastrointest Endosc 2003;58(6):879–84. [27] Isayama H, Komatsu Y, Tsujino T, et al. A prospective randomised study of covered versus uncovered diamond stents for the management of distal malignant biliary obstruction. Gut 2004;53(5):729–34. [28] Liu CL, Lo CM, Lai EC, et al. Endoscopic retrograde cholangiopancreatography and endoscopic endoprosthesis insertion in patients with Klatskin tumors. Arch Surg 1998;133:293–6.

112

BARON

[29] Chang WH, Kortan P, Haber GB. Outcome in patients with bifurcation tumors who undergo unilateral versus bilateral hepatic duct drainage. Gastrointest Endosc 1998;47:354–62. [30] Peters RA, Williams SG, Lombard M, et al. The management of high-grade hilar strictures by endoscopic insertion of self-expanding metal endoprostheses. Endoscopy 1997;29:10–6. [31] Cheng JLS, Bruno MJ, Bergman JJ, et al. Endoscopic palliation of patients with biliary obstruction caused by nonresectable hilar cholangiocarcinoma: efficacy of self-expandable metallic wall stents. Gastrointest Endosc 2002;56:33–9. [32] Ortner MA, Liebetruth J, Schreiber S, et al. Photodynamic therapy of nonresectable cholangiocarcinoma. Gastroenterology 1998;114:536–42. [33] Ortner M. Photodynamic therapy for cholangiocarcinoma. J Hepatobiliary Pancreat Surg 2001;8:137–9. [34] Wiedmann M, Berr F, Schiefke I, et al. Photodynamic therapy in patients with nonresectable hilar cholangiocarcinoma: 5-year follow-up of a prospective phase II study. Gastrointest Endosc 2004;60(1):68–75. [35] Costamagna G, Tringali A, Petruzziello L, et al. Hilar tumours. Can J Gastroenterol 2004;18(7):451–4. [36] Wayman J, Mansfield JC, Matthewson K, et al. Combined percutaneous and endoscopic procedures for bile duct obstruction: simultaneous and delayed techniques compared. Hepatogastroenterology 2003;50(52):915–8.

Gastroenterol Clin N Am 35 (2006) 113–123

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Medical Palliation of the Jaundiced Patient with Pruritus Nora V. Bergasa, MD Department of Medicine, State University of New York at Downstate, 451 Clarkson Avenue, Brooklyn, NY 11203, USA

C

holestasis is defined as the impaired secretion of bile [1], at any point in its trajectory. Pruritus can be a complication of cholestasis regardless of the cause [2]. In the context of malignancy, two situations need to be considered: one relates to intrahepatic cholestasis secondary to infiltration of the liver by malignant tissue; and the other relates to cholestasis secondary to biliary obstruction [3] from tumors (eg, biliary or pancreatic) or nodes (eg, lymphoma) [4]. Successful relief of biliary obstruction is associated with complete relief of pruritus [3]. Unfortunately, biliary obstruction cannot be relieved completely in many cases. In addition, when cholestasis results from the malignant infiltration of the liver, pruritus persists if effective surgical resection (eg, single colon metastasis to the liver) cannot be performed or there is a lack of response to chemotherapy. Accordingly, the provision of alternative measures to relieve pruritus in these patients is necessary because the quality of their lives is markedly affected by this symptom. This article reviews the pharmacologic treatment of the pruritus of cholestasis, with an emphasis on that secondary to malignancy. There are few data from controlled clinical trials in this group of patients; thus, much of the discussion will center on the treatments used for the pruritus of cholestasis that is not secondary to malignancy but, nevertheless, may be useful in that difficult situation.

CAUSE OF PRURITUS OF CHOLESTASIS In cholestasis, substances that normally are excreted in bile [1] accumulate in tissues. The nature of the pruritogens is unknown, but it is inferred that pruritogens accumulate in plasma as a result of cholestasis. The idea that these pruritogens are made in the liver is supported by the disappearance of chronic pruritus in patients who have liver disease after liver transplantation [5], the resolution of pruritus after complete relief of biliary obstruction [3], and the disappearance of pruritus as patients who have chronic liver disease progress to liver failure [6]. E-mail address: [email protected] 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.008

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

114

BERGASA

It is proposed that the pruritus of cholestasis is mediated centrally by endogenous opioids [7]. Specific data on the processing of pruritus and the scratch reflex at the central level are not available; however, there is evidence to suggest that in cholestasis there is increased opioidergic neurotransmission [7]. These observations were not made specifically in patients who had cholestasis secondary to malignancy; however, it can be inferred that when malignancy results in cholestasis, the consequences are similar in this context. The supporting evidence of the idea that increased central opioidergic neurotransmission contributes to the pathophysiology of cholestasis can be summarized as follows: (1) patients who have cholestasis may experience an unpleasant constellation of symptoms and signs with some similarities to an opiate withdrawal reaction when opiate antagonists are administered [8,9]; (2) a stereo-specific naloxone reversible state of antinociception (analgesia) can be displayed by rats with cholestasis secondary to bile duct resection [10]; and (3) there is down-regulation of mu, delta, and kappa opioid receptors [11] in rats with cholestasis secondary to bile duct resection. There are experimental data to suggest that opiateinduced pruritus is mediated by the mu opioid receptor [12]. Because the pharmacologic increase in opioidergic neurotransmission (eg, intrathecal administration of morphine) can be associated with pruritus [13–15], it has been proposed that the increased opioidergic tone of cholestasis can also result in pruritus [7]. The reason there is increased opioidergic tone in cholestasis is unknown; however, increased concentrations of methionine and leucine enkephalins, two endogenous opioid peptides, have been reported in cholestatic patients [8,16]. This observation, nonetheless, should not be interpreted as evidence for a specific role of enkephalins in pruritus but as a suggestion that related or independent alterations exist in the endogenous opioid system in cholestasis. The source of peripheral endogenous opioids in cholestasis is unknown, but the liver may contribute to the increased availability of opioids in liver disease [17–19]. Whether endogenous opioid ligands derived from the liver in cholestasis mediate what has been interpreted as centrally increased opioidergic neurotransmission is unknown; however, transport proteins found in the liver [20] are also found in the brain and can transport opiates in vitro [21] and may potentially transport periphery-derived opioids into the central nervous system (CNS) [22,23]. Furthermore, the increased availability of opioid peptides in the periphery may facilitate their entrance into the CNS [24,25]. In the context of pruritus secondary to malignancy, even in the absence of cholestasis, endogenous opioids may play a role because their expression has been detected in neural and non-neural human tumors [26]. PRURITUS AS A PARANEOPLASTIC SYMPTOM Pruritus is a recognized manifestation of malignancy including breast and gastric cancers and lymphoma (eg, lymphoma originating in the small intestine) [27] and has been reported recently as a variant of Stauffer’s syndrome, in which cholestasis, jaundice, and pruritus resolved after resection of a renal cell carcinoma [28]. Tumors of the central nervous system can also be associated with

MEDICAL PALLIATION OF THE JAUNDICED PATIENT WITH PRURITUS

115

pruritus [29]. The cause of this paraneoplastic manifestation is unknown; the treatment is directed to the primary tumor, but the provision of treatment of the pruritus specifically may be necessary if the primary cause is not cured. METHODOLOGY IN THE SCIENTIFIC STUDY OF PRURITUS Pruritus is a sensation; accordingly, it cannot be measured directly. Scratching as a result of pruritus can be recorded by instruments that have been developed for that purpose. The prototype of these instruments is a scratching activity monitoring system [30] based on piezoelectric technology, which permits the recording of vibrations produced by the fingernail in the act of scratching. Other systems have been developed that allow for outpatient recording of scratching [31–33]. These tools can be applied in clinical trials of pruritus from which objective data can be analyzed, in contrast to the subjective data provided by visual analog scales and questionnaires. It is unlikely that behavioral studies in patients who have pruritus from cholestasis secondary to malignancy will be conducted because patients may not be in an emotional or medical state to participate in clinical trials; however, objective data gathered from patients who have cholestasis from nonmalignant disease may still apply to patients who have cholestasis from malignancy. MANAGEMENT OF THE PRURITUS OF CHOLESTASIS The quality of life of patients who have pruritus is compromised beyond their primary disease. The palliation of pruritus in patients who have cholestasis secondary to malignancy can be limited by the need to administer medications that may interfere with opiate-induced analgesia and absorption of medications. There are no specific recommendations to treat the pruritus secondary to cholestasis from malignancy; however, the need to develop guidelines to treat this maddening complication has been recognized [34]. Information from selected publications on the use of medications to treat this type of pruritus in patients who have malignancy is provided in Table 1 [34–37]. Antihistamines Some patients who receive antihistamines (eg, hydroxyzine) report improvement of the pruritus of cholestasis, but it is unlikely that reported relief of pruritus is caused by a specific antipruritic effect because histamine does not appear to play a role in this type of pruritus. A sedative effect [38] is likely to be related to the reported relief of the pruritus. Sedatives The use of sedatives can be helpful in the management of patients who have chronic pruritus secondary to malignancy, particularly in terminal situations. Benzodiazepines (eg, lorazepam) may be considered. Antidepressants Pruritus is considered the second order of nociception, the first one being pain. The serotonin system mediates nociception; accordingly, changes in

BERGASA

116

Table 1 Selected pharmacologic interventions for the treatment of the pruritus of cholestasis secondary to malignancy Class of drugs Serotonin type-3 receptor antagonists

Serotonin reuptake inhibitors

Opioid activity

Antidepressant (noradrenaline and specific serotonin antidepressant)

Causes of cholestasis, (N), age group

Drug, dose, mode of administration, frequency

Study [ref]

Metastasis to liver from renal (1), colon (1), and cholangiocarcinoma (1), adults Hepatoblastoma, (1), child Metastasis to liver from breast cancer (1), rectal cancer (1), and pancreatic carcinoma (1), adults Metastasis to liver from colon cancer

Ondansetron, 8 mg IV followed by 8 mg PO twice/d

[35]

Paroxetine, 5 mg PO once/d Paroxetine, 20 mg PO at breakfast time

[36]

Buprenorphine, 0.2 mg SL 3 times/d, increased to every 3–4 hours; and naloxone infusions, 0.2 mg/24 h (ultra-low dose) Mirtazapine, 15 mg PO at bedtime

[34]

Metastasis to liver from adenocarcinoma of unknown origin (1), adult

[36]

[37]

Abbreviations: PO, orally; SL, sublingually.

serotoninergic neurotransmission may have some impact on the perception of pruritus. In this context, the selective serotonin reuptake inhibitor (SSRI) paroxetine has been reported to provide relief (often dramatically) of pruritus secondary to malignancy and specifically secondary to cholestasis from malignancy. In a placebo-controlled trial [39] that applied subjective methodology and that included 26 patients, three of whom were classified as having pruritus of cholestasis, the administration of paroxetine, 20 mg per day orally, was reported to be associated with relief in pruritus, although it was associated with nausea and vomiting. The inhibition of the hepatic enzyme CYP2D6, which is involved in the metabolism of codeine to morphine, was postulated as a potential mechanism to explain the observed antipruritic effect [39]. Based on this interpretation, paroxetine may inhibit a step in the conversion of a putative endogenous substance (eg, an opioid) to a pruritus-generating form of that substance. Low doses of paroxetine (5–10 mg) also have been reported to be useful in this type of pruritus, with minimal side effects. Another SSRI, sertraline (75 mg per day orally), was reported in a retrospective review [40] and in a prospective, dose-finding study [41] to relieve the pruritus of cholestasis in patients who had primary biliary cirrhosis. Mirtazapine [42], a noradrenalin and specific serotonin antidepressant with H1 antihistamine properties,

MEDICAL PALLIATION OF THE JAUNDICED PATIENT WITH PRURITUS

117

was reported to relieve pruritus in patients who had cholestasis from malignancy at doses ranging from 7.5 to 30 mg per day [37]. Serotonin Antagonists The serotoninergic system participates in the neurotransmission of nociceptive stimuli. There are encouraging reports from the early 1990s on the use of ondansetron, a serotonin type-3 receptor (5-HT3) antagonist, in patients who had cholestasis and pruritus, some of whom had liver metastasis from malignant tumors. Intravenous ondansetron (4 and 8 mg) but not placebo was reported to be associated with a decrease in pruritus in a small number of patients [35,43]. These results were not confirmed in a study that applied behavioral methodology [44]. There is no explanation as to how serotonin antagonists and SSRIs improve the pruritus of cholestasis. Drugs with Affinity for Opioid Receptors The ameliorating effect of pruritus of cholestasis by the administration of opiate antagonists supports the hypothesis that endogenous opioids contribute to the pruritus. The antagonists that have been investigated include nalmefene [8,9,45], naloxone [46,47], and naltrexone [48–50]. The potential development of an opiate withdrawal-like reaction makes it necessary to introduce these medications gradually and, in certain cases, intravenously. By starting at ultra-low doses of naloxone (0.002 lg/kg) [51,52] and gradually increasing to 0.8 lg/kg, the withdrawal reaction may be avoided or its intensity decreased. After the patient receives 1 or 2 days of a naloxone drip regimen, the opiate antagonist naltrexone may be started at a dose of 6.25 or 12.5 mg per day and increased as necessary to a maximum of 100 mg per day. Doses higher than 100 mg per day have been used to treat the pruritus of cholestasis (ie, 400 mg per day); however, the requirement for high doses of naltrexone tends to predict a suboptimal response to opiate antagonists in some patients. Naltrexone metabolites can accumulate in patients who have decompensated liver disease, which requires dose reduction. However, it is not common for patients who have decompensated liver disease to report pruritus because this symptom tends to cease as liver failure ensues [6]. Tolerance may develop with the use of opiate antagonist with the recurrence of pruritus. Other interventions that alter opioidergic neurotransmission have been used in uncontrolled studies, including buprenorphine, a partial opioid receptor agonist [53], and codeine. In this context, it is hypothesized that codeine, like morphine, binds to the mu opioid receptor [54–56] and competes with the endogenous ‘‘pruritogenic’’ ligand for receptor occupancy. Indeed, receptor selectivity is relevant to the opioid-mediated pruritus of cholestasis. For example, it has been shown that the activation of kappa receptors by the use of a specific kappa receptor ligand [57,58] prevented morphine-induced, centrally mediated scratching behavior [57,59]. By analogy, because the pruritus of cholestasis appears to be mediated, at least in part, by increased opioidergic neurotransmission, activating kappa receptors in patients who have cholestasis may decrease

118

BERGASA

the pruritus. Evidence in support of this idea was provided by the relief of pruritus associated with butorphanol in a patient with cholestasis. Butorphanol is an antagonist at the mu opioid receptor and an agonist at the kappa receptor. The spray form of butorphanol (one application delivers close to 1 mg of drug) was used in a patient with severe pruritus secondary to liver disease from chronic hepatitis C with marked reduction of the symptom and healing of cutaneous excoriations (N.V. Bergasa, unpublished data, 2003). Butorphanol was considered initially to have a low potential for abuse, but dependence has been a concern. In patients who have pruritus secondary to cholestasis from malignancy, the risk of dependence should not be a major concern, and relief should be the goal of therapy. Nonabsorbable Resins The nonabsorbable anion exchange resins used to lower serum cholesterol include cholestyramine [60], colestipol, and colesevelam [61]. Cholestyramine and other agents are used to increase fecal excretion of the presumed pruritogens. This usage assumes that the resins reach the intestine. Therefore, in cases of marked biliary obstruction from malignancy, the use of these agents seems to lack rationale. There are no published reports on the use of these resins to treat the pruritus from cholestasis caused by malignancy; furthermore, the side effects of resins, which include bloating and interference with the absorption of medications, may not be ideal for a terminally ill patient. The dose of cholestyramine is 4 g administered immediately before and after breakfast, because it is presumed that the pruritogens are stored in the gallbladder during the overnight fast. Additional doses may be added at lunch and dinnertime, not to exceed 16 g per day. The taste of these resins is unpleasant. The addition of a flavored, cooked liquid gelatin to each 4-g dose can be solidified with refrigeration for ingestion to improve the palatability. If cholestyramine is not associated with relief in the pruritus, the drug should be discontinued. Removal of Circulating Substances Invasive procedures have been used to remove pruritogens, including plasmapheresis [62,63] and extracorporeal albumin dialysis [64–66]. These modalities have not been submitted to clinical trials, and their use in patients who have a terminal illness may not be appropriate. Partial external diversion of bile [67,68] and ileal diversion [69] in children with chronic cholestasis and pruritus are reported to decrease pruritus and improve quality of life. The nature of any relevant substances removed by these interventions has not been defined. Hepatic Enzyme Inducers Phenobarbital [70–72] and rifampicin [73] are hepatic enzyme inducers used to manage the pruritus of cholestasis. The reported decrease in pruritus associated with phenobarbital may be caused by sedation. Rifampicin, at doses of 300 to 450 mg per day [74] or 10 mg/kg [73], has been reported to improve the pruritus of cholestasis secondary to primary biliary cirrhosis. The mechanism by which rifampicin exerts the reported antipruritic effect is unknown. Rifampicin

MEDICAL PALLIATION OF THE JAUNDICED PATIENT WITH PRURITUS

119

induces drug-metabolizing enzymes and transporters through activation of the pregnane X receptor [75]. Rifampicin may have anti-opiate activity in conditions associated with increased opioidergic tone (eg, cholestasis), based on a reported experience in patients maintained on methadone treatment [76]. Rifampicin can be hepatotoxic at doses used to manage pruritus [77], and monitoring of liver tests is recommended. Metronidazole, another antibiotic, at doses of 250 mg 3 times per day orally for 1 week has been reported to control pruritus in patients who have primary biliary cirrhosis [78]. Experience with the use of Anesthetics Taking into account the neural trajectory of pruritus, it can be considered that some form of anesthesia may decrease this sensation. Propofol, administered at subhypnotic doses, has been reported to decrease the pruritus of cholestasis [79,80] and also has been reported to decrease the pruritus induced by intrathecal morphine [80]. The use of intravenous anesthetics, including lidocaine (British approved name, lignocaine) [81], have been reported to relieve pruritus that was not relieved by other medications [82–84]. Recently, a controlled trial that applied subjective methodology reported a decrease in pruritus associated with the administration of 100 mg (in 5 mL of normal saline) of lidocaine and not by the administration of placebo [82]. Novel Therapies for the Pruritus of Cholestasis Dronabinol Cannabinoids have analgesic properties mediated by CNS receptors [85,86]. Dronabinol is a sesame oil preparation of D9-tetrahydrocannabinol, the psychoactive compound extracted from Cannabis sativa (marijuana). In a report that included three patients who had cholestasis and intractable pruritus, dronabinol, 5 mg at bedtime, was associated with symptom relief [87]. One proposed mechanism for the relief of pruritus is an increased threshold to the perception of pruritus [88]. Gabapentin Gabapentin is used in the treatment of neuropathy [89], and it has been reported to abolish brachioradial pruritus [90]. Gabapentin (up to 2400 g/d in divided doses) was not found to be better than placebo in reducing scratching activity in a double-blind, randomized, placebo-controlled trial but may be effective in individual cases. Sedation commonly occurs. In patients who had cholestasis, a starting dose of 100 mg three times per day orally was chosen to prevent side effects. General Measures Consultation with a dermatologist to identify potential primary skin lesions and to select the most effective physical measures (eg, emollients) is necessary. Baths with sodium bicarbonate (0.5 kg/bath) in the water have been helpful. It is useful to identify factors that worsen the pruritus, including changes in environmental temperature and the use of certain fabrics for clothing and bedding and avoid them.

120

BERGASA

SUMMARY Cholestasis secondary to infiltration of the liver by malignant tumors or by obstruction of the biliary tree can be complicated by pruritus. The clinician and ancillary personal must recognize how debilitating pruritus is and identify the treatment of this symptom as a priority. Because robust clinical trials have not been conducted in patients who have pruritus with cholestasis, a network connecting the services that provide care for these patients (eg, hospices) may be useful for disseminating information. References [1] Reichen J, Simon F. Cholestasis. In: Arias IM, Jakoby WB, Popper H, et al, editors. The liver: biology and pathobiology. 2nd edition. New York: Raven Press; 1988. p. 1105–24. [2] Sherlock S, Dooley J. Cholestasis. In: Diseases of the liver and biliary system. 9th edition. London: Blackwell; 1991. p. 214–35. [3] Hall RI, Denyer ME, Chapman AH. Percutaneous-endoscopic placement of endoprostheses for relief of jaundice caused by inoperable bile duct strictures. Surgery 1990;107(2):224–7. [4] Feller E, Schiffman FJ. Extrahepatic biliary obstruction by lymphoma. Arch Surg 1990;125(11):1507–9. [5] Elias E. Liver transplantation. Clin Med 1993;27:224–32. [6] Lloyd-Thomas HG, Sherlock S. Testosterone therapy for the pruritus of obstructive jaundice. BMJ 1952;2(4797):1289–91. [7] Jones EA, Bergasa NV. The pruritus of cholestasis: from bile acids to opiate agonists. Hepatology 1990;11(5):884–7. [8] Thornton JR, Losowsky MS. Opioid peptides and primary biliary cirrhosis. BMJ 1988;297(6662):1501–4. [9] Bergasa NV, Talbot TL, Schmitt JP, et al. Open label trial of oral nalmefene therapy for the pruritus of cholestasis. Hepatology 1998;27:679–84. [10] Bergasa NV, Alling DW, Vergalla J, et al. Cholestasis in the male rat is associated with naloxone-reversible antinociception. J Hepatol 1994;20:85–90. [11] Bergasa NV, Rothman RB, Vergalla J, et al. Down-regulation of delta opioid receptors in bile duct resected rats: further evidence for alteration in the opioid system in cholestasis. Gastroenterology 1992;102:A946. [12] Inan S, Cowan A. Reduced kappa-opioid activity in a rat model of cholestasis. Eur J Pharmacol 2005;518(2–3):182–6. [13] Ko MC, Song MS, Edwards T, et al. The role of central mu opioid receptors in opioid-induced itch in primates. J Pharmacol Exp Ther 2004;310(1):169–76. [14] Ballantyne JC, Loach AB, Carr DB. Itching after epidural and spinal opiates. Pain 1988;33(2):149–60. [15] Abbound TK, Lee K, Zhu J, et al. Prophylactic oral naltrexone with intrathecal morphine for cesarean section: effects on adverse reactions and analgesia. Anesth Analg 1990;71(4): 367–70. [16] Thornton JR, Losowsky MS. Plasma methionine enkephalin concentration and prognosis in primary biliary cirrhosis. BMJ 1988;297:1241–2. [17] Bergasa NV, Sabol SL, Young WSD, et al. Cholestasis is associated with preproenkephalin mRNA expression in the adult rat liver. Am J Physiol 1995;268(2 Pt 1):G346–54. [18] Bergasa NV, Vergalla J, Swain MG, et al. Hepatic concentration of proenkephalin-derived opioids are increased in a rat model of cholestasis. Liver 1996;16:298–302. [19] Bergasa N. Hepatic met-enkephalin immunoreactivity is enhanced in primary biliary cirrhosis. Liver 2002;22(2):107–13. [20] Dietrich CG, Geier A, Oude Elferink RP. ABC of oral bioavailability: transporters as gatekeepers in the gut. Gut 2003;52(12):1788–95.

MEDICAL PALLIATION OF THE JAUNDICED PATIENT WITH PRURITUS

121

[21] Dombrowski SM, Desai SY, Marroni M, et al. Overexpression of multiple drug resistance genes in endothelial cells from patients with refractory epilepsy. Epilepsia 2001;42(12): 1501–6. [22] Hoffmaster KA, Zamek-Gliszczynski MJ, Pollack GM, et al. Hepatobiliary disposition of the metabolically stable opioid peptide [D-Pen2, D-Pen5]-enkephalin (DPDPE): pharmacokinetic consequences of the interplay between multiple transport systems. J Pharmacol Exp Ther 2004;311(3):1203–10. [23] Hoffmaster KA, Zamek-Gliszczynski MJ, Pollack GM, et al. Multiple transport systems mediate the hepatic uptake and biliary excretion of the metabolically stable opioid peptide [D-penicillamine2, 5]enkephalin. Drug Metab Dispos 2005;33(2):287–93. [24] Banks WA, Kastin AJ. Peptide transport systems for opiates across the blood–brain barrier. Am J Physiol 1990;259(1 Pt 1):E1–10. [25] Banks WA, Kastin AJ. Saturable transport of peptides across the blood-brain barrier. Life Sci 1987;41(11):1319–38. [26] Zagon IS, McLaughlin PJ, Goodman SR, et al. Opioid receptors and endogenous opioids in diverse human and animal cancers. J Natl Cancer Inst 1987;79(5):1059–65. [27] Caretta E. Three cases of paraneoplastic syndrome caused by lymphoma of the small intestine. Minerva Med 1981;72(43):2905–12. [28] Karakolios A, Kasapis C, Kallinikidis T, et al. Cholestatic jaundice as a paraneoplastic manifestation of prostate adenocarcinoma. Clin Gastroenterol Hepatol 2003;1(6): 480–3. [29] Adreev VC, Petkov I. Skin manifestations associated with tumours of the brain. Br J Dermatol 1975;92(6):675–8. [30] Talbot TL, Schmitt JM, Bergasa NV, et al. Application of piezo film technology for the quantitative assessment of pruritus. Biomed Instrum Technol 1991;25(5):400–3. [31] Stein H, Bijak M, Heerd E, et al. [Pruritometer 1: portable measuring system for quantifying scratching as an objective measure of cholestatic pruritus]. Biomed Tech (Berl) 1996;41(9): 248–52 [in German]. [32] Molenaar HA, Oosting J, Jones EA. Improved device for measuring scratching activity in patients with pruritus. Med Biol Eng Comput 1998;36(2):220–4. [33] Bijak M, Mayr W, Rafolt D, et al. [Pruritometer 2: portable recording system for the quantification of scratching as objective criterion for the pruritus]. Biomed Tech (Berl) 2001;46(5): 137–41 [in German]. [34] Zylicz Z, Stork N, Krajnik M. Severe pruritus of cholestasis in disseminated cancer: developing a rational treatment strategy: a case report. J Pain Symptom Manage 2005;29(1):100–3. [35] Raderer M, Muller C, Scheithauer W. Ondansetron for pruritus due to cholestasis. N Engl J Med 1994;21:1540. [36] Zylicz Z, Smits C, Krajnik M. Paroxetine for pruritus in advanced cancer. J Pain Symptom Manage 1998;16(2):121–4. [37] Davis MP, Frandsen JL, Walsh D, et al. Mirtazapine for pruritus. J Pain Symptom Manage 2003;25(3):288–91. [38] Douglas W, editor. The pharmacological basis of therapeutics. 7th edition. New York: Macmillan Publishing Company; 1975. [39] Zylicz Z, Krajnik M, Sorge AA, et al. Paroxetine in the treatment of severe non-dermatological pruritus: a randomized, controlled trial. J Pain Symptom Manage 2003;26(6): 1105–12. [40] Browning JD, Combes B, Mayo M. Long-term efficacy of sertraline as a treatment for cholestatic pruritus in patients with primary biliary cirrhois. Am J Gastroenterol 2003;98(12): 2736–41. [41] Mayo M. Effect of sertraline on pruritus in cholestatic liver disease: a randomized double blind placebo controlled crossover study. Hepatology 2005;4(Suppl 1):S60. [42] Kasper S. Clinical efficacy of mirtazapine: a review of meta-analyses of pooled data. Int Clin Psychopharmacol 1995(Suppl 4):S25–35.

122

BERGASA

[43] Schworer H, Hartmann H, Ramadori G. Relief of cholestatic pruritus by a novel class of drugs: 5-hydroxytryptamine type 3 (5–HT3) receptor antagonists: effectiveness of ondansetron. Pain 1995;61(1):33–7. [44] O’Donohue JW, Pereira SP, Ashdown AC, et al. A controlled trial of ondansetron in the pruritus of cholestasis. Aliment Pharmacol Ther 2005;21(8):1041–5. [45] Bergasa NV, Alling DW, Talbot TL, et al. Oral nalmefene therapy reduces scratching activity due to the pruritus of cholestasis: a controlled study. J Am Acad Dermatol 1999;41(3):431–4. [46] Bergasa NV, Talbot TL, Alling DW, et al. A controlled trial of naloxone infusions for the pruritus of chronic cholestasis. Gastroenterology 1992;102(2):544–9. [47] Bergasa NV, Alling DW, Talbot TL, et al. Naloxone ameliorates the pruritus of cholestasis: results of a double-blind randomized placebo-controlled trial. Ann Intern Med 1995; 123(3):161–7. [48] Wolfhagen FHJ, Sternieri E, Hop WCJ, et al. Oral naltrexone treatment for cholestatic pruritus: a double-blind, placebo-controlled study. Gastroenterology 1997;113(4):1264–9. [49] Carson KL, Tran TT, Cotton P, et al. Pilot study of the use of naltrexone to treat the severe pruritus of cholestatic liver disease. Am J Gastroenterol 1996;91:1022–3. [50] Terg R, Coronel E, Sorda J, et al. Efficacy and safety of oral naltrexone treatment for pruritus of cholestasis: a crossover, double blind, placebo-controlled study. J Hepatol 2002;37(6):717–22. [51] Jones EA, Bergasa NV. The pathogenesis and treatment of pruritus and fatigue in patients with PBC. Eur J Gastroenterol Hepatol 1999;11(6):623–31. [52] Jones EA, Neuberger J, Bergasa NV. Opiate antagonist therapy for the pruritus of cholestasis: the avoidance of opioid withdrawal-like reactions. QJM 2002;95(8):547–52. [53] Juby LD, Wong VS, Losowsky MS. Buprenorphine and hepatic pruritus. British Journal of Clinical Practice 1994;48(6):331. [54] Fukuda K, Kato S, Mori K. Location of regions of the opioid receptor involved in selective agonist binding. J Biol Chem 1995;270(12):6702–9. [55] Mignat C, Wille U, Ziegler A. Affinity profiles of morphine, codeine, dihydrocodeine and their glucuronides at opioid receptor subtypes. Life Sci 1995;56(10):793–9. [56] Yu L. The mu opioid receptor: from molecular cloning to functional studies. Addict Biol 1996;1(1):19–30. [57] Umeuchi H, Togashi Y, Honda T, et al. Involvement of central mu-opioid system in the scratching behavior in mice, and the suppression of it by the activation of kappa-opioid system. Eur J Pharmacol 2003;477(1):29–35. [58] Ko MC, Lee H, Song MS, et al. Activation of kappa-opioid receptors inhibits pruritus evoked by subcutaneous or intrathecal administration of morphine in monkeys. J Pharmacol Exp Ther 2003;305(1):173–9. [59] Ko MC, Naughton NN. An experimental itch model in monkeys: characterization of intrathecal morphine-induced scratching and antinociception. Anesthesiology 2000;92(3):795–805. [60] Datta DV, Sherlock S. Cholestyramine for long term relief of the pruritus complicating intrahepatic cholestasis. Gastroenterology 1966;50:323–32. [61] Berg C. Use of colesevelam hydrochloride (Welchol) as a novel therapeutic agent for the management of refractory pruritus in chronic liver disease. Hepatology 2001;34:541. [62] Ambinder EP, Cohen LB, Wolke AM, et al. The clinical effectiveness and safety of chronic plasmapheresis in patients with primary biliary cirrhosis. J Clin Apheresis 1985;2(3):219–23. [63] Cohen LB, Ambinder EP, Wolke AM, et al. Role of plasmapheresis in primary biliary cirrhosis. Gut 1985;26(3):291–4. [64] Joannidis M, Bellmann R, Graziadei I, et al. Current indications for the albumin dialysis: intractable pruritus. In: Proceedings of the 4th International Symposium on Albumin Dialysis in Liver Disease. Rostock-Warnemu¨nde; 2002. p. 27. [65] Strange J, Hassanein TL, Mehta R, et al. The molecular adsorbents recycling system as a liver support system based on albumin dialysis: a summary of preclinical investigations, prospective, randomized, controlled clinical trial, and clinical experience from 19 centers. Artif Organs 2002;26:103–10.

MEDICAL PALLIATION OF THE JAUNDICED PATIENT WITH PRURITUS

123

[66] Sturm E, Franssen CF, Gouw A, et al. Extracorporeal albumin dialysis (MARS) improves cholestasis and normalizes low apo A-I levels in a patient with benign recurrent intrahepatic cholestasis (BRIC). Liver 2002;22(Suppl 2):S72–5. [67] Whitington P, Whitington G. Partial external diversion of bile for the treatment of intractable pruritus associated with intrahepatic cholestasis. Gastroenterology 1988;95:130–6. [68] Emerick KM, Whitington PF. Partial external biliary diversion for intractable pruritus and xanthomas in Alagille syndrome. Hepatology 2002;35(6):1501–6. [69] Hollands CM, Rivera-Pedrogo FJ, Gonzalez-Vallina R, et al. Ileal exclusion for Byler’s disease: an alternative surgical approach with promising early results for pruritus. J Pediatr Surg 1998;33(2):220–4. [70] Bloomer JR, Boyer JL. Phenobarbital effects in cholestatic liver diseases. Ann Intern Med 1975;82(3):310–7. [71] Ghent CN, Bloomer JR, Hsia YE. Efficacy and safety of long-term phenobarbital therapy of familial cholestasis. J Pediatr 1978;93(1):127–32. [72] Bachs L, Pare´s A, Elena M, et al. Comparison of rifampicin with phenobarbitone for treatment of pruritus in biliary cirrhosis. Lancet 1989;1(8638):574–6. [73] Bachs L, Pare´s A, Elena M, et al. Effects of long-term rifampicin administration in primary biliary cirrhosis. Gastroenterology 1992;102(6):2077–80. [74] Ghent CN, Carruthers SG. Treatment of pruritus in primary biliary cirrhosis with rifampin: results of a double-blind, crossover, randomized trial. Gastroenterology 1988;94(2): 488–93. [75] Tirona RG, Leake BF, Wolkoff AW, et al. Human organic anion transporting polypeptide-C (SLC21A6) is a major determinant of rifampin-mediated pregnane X receptor activation. J Pharmacol Exp Ther 2003;304(1):223–8. [76] Kreek MJ, Garfield JW, Gutjahr CL, et al. Rifampin-induced methadone withdrawal. N Engl J Med 1976;294(20):1104–6. [77] Prince MI, Burt AD, Jones DE. Hepatitis and liver dysfunction with rifampicin therapy for pruritus in primary biliary cirrhosis. Gut 2002;50(3):436–9. [78] Berg C, Gollan J. Primary biliary cirrhosis. Scand J Gastroenterol 1992;192(Suppl):S43–9. [79] Borgeat A, Wilder-Smith OHG, Mentha G. Subhypnotic doses of propofol relieve pruritus associated with liver disease. Gastroenterology 1993;104:244–7. [80] Borgeat A, Wilder-Smith O, Saiah M, et al. Subhypnotic doses of propofol relieve pruritus induced by epidural and intrathecal morphine. Anesthesiology 1992;76:510–2. [81] Watson WC. Intravenous lignocaine for relief of intractable itch. Lancet 1973;1(7796):211. [82] Villamil AG, Bandi JC, Galdame OA, et al. Efficacy of lidocaine in the treatment of pruritus in patients with chronic cholestatic liver diseases. Am J Med 2005;118(10):1160–3. [83] Joffe P, Andersen LW, Molvig J, et al. Intravenous lidocaine in the treatment of pruritus in hemodialysis patients. Clin Nephrol 1985;24(4):214. [84] Fishman SM, Caneris OA, Stojanovic MP, et al. Intravenous lidocaine for treatment-resistant pruritus. Am J Med 1997;102(6):584–5. [85] Meng ID, Manning BH, Marting WJ, et al. An analgesia circuit activated by cannabinoids. Nature 1998;395:381–3. [86] Fuentes JA, Ruiz-Gayo M, Manzanares J, et al. Cannabinoids as potential new analgesics. Life Sci 1999;65(6–7):675–85. [87] Neff GW, O’Brien CB, Regev A, et al. The remedy for intractable cholestatic related pruritus: delta-9-tetrahydrocannabinol (Marinol) [abstract]. Hepatology 1999;30:328. [88] Neff GW, O’Brien CB, Reddy KR, et al. Preliminary observation with dronabinol in patients with intractable pruritus secondary to cholestatic liver disease. Am J Gastroenterol 2002;97(8):2117–9. [89] Backonja M, Glanzman RL. Gabapentin dosing for neuropathic pain: evidence from randomized, placebo-controlled clinical trials. Clin Ther 2003;25(1):81–104. [90] Bueller HA, Bernhard JD, Dubroff LM. Gabapentin treatment for brachioradial pruritus. J Eur Acad Dermatol Venereol 1999;13(3):227–8.

Gastroenterol Clin N Am 35 (2006) 125–130

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

The Role of Radiation Therapy in the Palliation of Gastrointestinal Malignancies David D. Howell, MD Department of Radiation Oncology, University of Michigan School of Medicine, UH B2C 490 UH 0010, Ann Arbor, MI 48109, USA

T

here is a wide array of options for the palliation of various symptoms of gastrointestinal and intra-abdominal malignancies. In addition, there is a role for radiation therapy in certain clinical situations for the patient with significant symptoms. ESOPHAGUS Primary and secondary malignancies of the esophagus can cause many debilitating symptoms, including pain, bleeding, odynophagia, and dysphagia. Advanced malignancies of the esophagus, by virtue of direct extension to the trachea, also can cause tracheoesophageal fistulas with symptoms that include aspiration, cough, and the significant potential for aspiration pneumonia. Treatment using external beam radiotherapy or intraluminal radiation therapy (brachytherapy) has been shown to provide palliative benefit in some patients who have locally advanced and unresectable esophageal cancers [1]. A course of external radiation may vary from 20 Gy in 5 fractions to 50.4 Gy in 28 fractions. This radiation treatment may be delivered with cytotoxic systemic chemotherapy, generally using multiple field techniques, in an attempt to enhance the response with an acceptable rate of morbidity. Intraluminal brachytherapy using radiation sources such as iridium 192 may be helpful in providing high-dose treatment to the lumen and submucosal muscle tissue while limiting higher doses of radiation to surrounding organs [2,3]. Generally, brachytherapy is delivered once or twice per week. Various dose-fractionation schemas have been used. The use of radiation has been shown to have an impact on symptoms. The degree of symptom relief depends on the initial severity of symptoms, tumor burden, and other patient factors, including performance status. Patients may notice palliation as early as 1 week into a course of therapy but may require

E-mail address: [email protected] 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.011

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

126

HOWELL

an entire course of treatment to see a response. Generally, focal radiation is tolerated reasonably well, with acceptable side effects. Acute esophagitis can be caused by the effects of radiation. Patients may note odynophagia or burning symptoms, which generally begin during the third or fourth week of treatment. In patients who receive concurrent chemotherapy, symptoms occur generally on an average of 1 week earlier. The management of radiation esophagitis will vary depending on the patient’s symptoms. Early in the treatment, a mixture of diphenhydramine (Benadryl), magnesium (Maalox), and lidocaine (Xylocaine) may be used. Combination analgesics or narcotics may be used in patients who do not respond to this regimen. Acid reflux irritation also may occur. The use of a proton pump inhibitor, either prophylactically or in response to symptoms, may help to reduce acid reflux or associated symptoms. Late effects of external beam radiation therapy, or brachytherapy, to the esophagus can include a small but finite risk of clinical stricture or perforation. Although the incidence of these complications is low, the risks are affected by the total dose of radiation used, the dose of radiation per fraction delivered, tumor burden, and the volume of the esophagus that is irradiated. STOMACH Locally advanced and technically unresectable malignancies of the stomach can be particularly debilitating. Symptoms can include melena, pain, nausea, vomiting, and hematemesis. In patients who have unresectable, locally advanced, or metastatic disease with these symptoms, radiation can be used for palliation. Generally, external beam radiation therapy, with or without cytotoxic chemotherapy, can be used [4,5]. Care must be taken in planning radiation fields to try to limit radiation dose to other intra-abdominal organs, trying to keep the doses of radiation to the kidneys and small intestine to as low a dose as reasonably possible. A course of radiation may vary from 30 Gy in 10 fractions to 54 Gy in 30 fractions, in an attempt to provide palliation of symptoms. Generally, bleeding can be suppressed, often within the first week of treatment. Factors that affect symptom control include the histology and extent of disease, the depth of disease penetration in the gastric wall, and the volume of stomach that is involved with the malignancy. Controlling symptoms of nausea and vomiting is more difficult with radiation, and the radiation itself may temporarily aggravate these symptoms by virtue of its local effects. The morbidity of gastric radiation can indeed mimic or aggravate some of the symptoms associated with gastric malignancies. Side effects of radiation can include abdominal pain, nausea, and vomiting. Gastritis and gastric ulcers also may be caused by radiation. Morbidity of damage to gastric parietal cells can result in reduced acid secretion, resulting in dyspepsia. Generally, these symptoms are seen in patients treated with high doses of radiation to the entire organ. Modern techniques using partial gastric radiation, when possible, with progressively coned-down fields is associated with lower morbidity. Antacids, H2 blockers, and various antiemetic regimens may be used for symptom control. Narcotic and combination analgesics may used for pain.

RADIATION THERAPY IN THE PALLIATION OF GI MALIGNANCY

127

LIVER Today, there is a wide array of choices for treating hepatic metastasis, which historically has been treated with radiation therapy. Symptoms from liver metastasis include upper abdominal fullness, bloating, and local and referred pain, although the disease may be quite advanced before the onset of symptoms. When using radiation, one must be mindful of the volume of the organ that is treated as well as the total dose used, to gauge the potential morbidity of the treatment [6]. Historically, radiation delivery parameters ranged from 20 Gy in 8 to 12 fractions up to 40 Gy in 4 weeks, depending on the volume of liver to be treated (among other factors). Radiation therapy has been used in some centers for focal treatment of hepatic metastases or treatment of as much as two thirds of the liver. Special planning techniques, including multiple fields, intensity modulated radiation therapy, and inverse treatment planning, can be used in these circumstances to allow for the protection of at least a portion of the liver. The side effects of hepatic radiation can include malaise, nausea, and vomiting, and in some circumstances, may lead to radiation-induced liver disease in situations in which significant doses of radiation are affecting substantial volumes of the liver. SMALL INTESTINE Although primary malignancies of the small intestine are rare, the small intestine is a major source of morbidity for patients receiving radiation to the abdomen. Small bowel toxicity is generally the dose-limiting toxicity for patients receiving radiation to the abdomen and pelvis. The incidence and severity of morbidity caused by small bowel toxicity depends on the total dose of radiation, dose per fraction of radiation, and volume of small intestine included in radiation treatment fields [7]. Toxicity increases with the use of concurrent chemotherapy, previous abdominal or pelvic surgery, and vascular disease. Manifestations of small bowel toxicity can range acutely from diarrhea and pain to chronic changes of ischemia, obstruction, ulceration, and fibrosis. Acute side effects are treated symptomatically with antidiarrheals and antiemetics. Diet controls, including a low-residue, low-fat, and lactose-free diet, may improve symptoms. If obstruction occurs, it is generally managed with bowel rest; if obstruction is severe, surgical resection or bypass of the obstructed segment or lysis of adhesions may be needed. PANCREAS The majority of pancreatic neoplasms are incurable at diagnosis. Presenting symptoms can include pain, anorexia, jaundice, and weight loss. Radiation therapy is frequently used in an attempt to suppress local disease and to palliate symptoms associated with this malignancy. Radiation therapy is generally delivered using multiple field techniques, using doses between 45 and 50.4 Gy in 5 to 5.5 weeks to the pancreas and surrounding tissue. The chance that pain will be reduced is significant, although the prognosis remains poor [8–10].

128

HOWELL

Side effects of pancreatic radiation can be similar to those associated with radiation of the stomach, including nausea and vomiting. Diarrhea is less common because the volume of small intestine that is the radiation field is usually small. Antiemetics or H2 blockers can be used either prophylactically or for symptoms in patients receiving radiation to this area. BILIARY TREE Most patients who have cancers to the biliary tract require palliative surgical or endoscopic procedures to alleviate obstructive symptoms. Significant palliation of symptoms of pain and occasional long-term survival can be seen with the use of focal external beam radiation with or without cytotoxic systemic chemotherapy [11]. The use of intraluminal brachytherapy has been described as well. Permanent control of disease and symptoms is uncommon. LARGE INTESTINE Local radiation can be used in an attempt to palliate recurrent colon cancer, symptoms of which may include local pain or bleeding. A physician needs to be mindful of the location and size of recurrence, whether radiation has been used previously in the region, and other organs in the treatment fields, factors that may affect the total dose of radiation that can be delivered. Systemic therapy may also be used in an attempt to augment response [12]. Because the large intestine courses throughout the pelvis and abdomen, the potential morbidity of therapy hinges on the location of recurrent or unresectable disease, the field arrangement planned, and dose of radiation used. RECTUM Pelvic recurrences of rectal neoplasms may be among the most difficult clinical situations faced by radiation oncologists. Depending on the size and location of these pelvic recurrences, patients may experience significant problems with pain that can be very challenging to manage. Locally advanced and technically unresectable rectal neoplasms may cause symptoms that include bleeding, local pain, obstruction, and in more advanced cases, may be associated with fistula formation. Local radiation may be helpful in an attempt to alleviate symptoms. Classically, a course of radiation may range from 40 Gy in 20 fractions to 54 Gy in 30 fractions. Frequently, cytotoxic systemic chemotherapy will be delivered concurrently with the radiation in the hope of augmenting response [13]. Relief of bleeding may occur as soon as the second to third week of treatment. The dose of radiation required for pain control is variable. Frequently, bypass surgery is necessary for patients who have advanced malignancies of the rectum that are causing obstructive symptoms. The morbidity of radiation to the rectum can sometimes be seen early in the course of pelvic radiation. Symptoms may include bleeding and tenesmus. Later changes involving damage to rectal mucosa may lead to bleeding, ulcers, strictures, and fistulas. Symptoms of urgency and tenesmus can be treated with diphenoxylate hydrochloride and atropine sulfate (Lomotil) or loperamide.

RADIATION THERAPY IN THE PALLIATION OF GI MALIGNANCY

129

Local pain can be controlled with narcotics or combination analgesics. Occasionally, suppositories containing opiates and belladonna are used. Bleeding and chronic proctitis may lead to stricture formation or fistulas in a small number of cases. The dose of radiation delivered, volume treated, and tumor burden are all factors that may have an impact on the risk of late effects. Pharmacologic interventions for chronic proctitis can include steroid enemas or other locally applied formulations containing steroids, with surgical and endoscopic procedures used for more advanced toxicities. SUMMARY Radiation therapy can provide significant palliation in many patients who have unresectable, metastatic, or incurable malignancies of the gastrointestinal tract and is a modality that can be used in many clinical situations. In considering radiation therapy, the radiation oncologist should be mindful of the actual expectation of palliation, the potential degree of relief of symptoms, and the time involved for the patient and those involved in his or her care and transport. When a course of radiation is planned, the dose per fraction, total number of fractions, and total dose delivered should all be taken into account. The use of concurrent systemic chemotherapy needs to be reviewed with respect to chances for augmentation of palliation as well as morbidity. Although long-term disease control is often not achieved, there are reasonable expectations for control of symptoms of bleeding and pain. The side effects associated with radiation are generally well tolerated and, if they occur, usually can be controlled with conservative measures. Significant late effects, although uncommon, are generally a function of the volume of the organ treated, the total dose used, and whether systemic therapy is used in conjunction with radiation. When considering the options for palliation in these situations, multidisciplinary collaboration among all those involved in a patient’s care, surgeon, radiation oncologist, medical oncologist, primary care provider, and palliative care specialist, is ideal to address each situation in each patient, for whom there may be a variety of options for palliative treatment. Generally, combinations of interventions are needed to optimize the palliation of a patient’s various problems. The goal of providing relief to the patient in the least amount of time with the least amount of morbidity and the greatest expectation for durability of response is paramount. Acknowledgments The author thanks Cyndi Wunderlich for stenographic assistance. References [1] Albertsson M, Ewers SB, Widmark H, et al. Evaluation of the palliative effect of radiotherapy for esophageal carcinoma. Acta Oncol 1989;28(2):267–70. [2] Sharma V, Mahantshetty U, Dinshaw KA, et al. Palliation of advanced/recurrent esophageal carcinoma with high-dose-rate brachytherapy. Int J Radiat Oncol Biol Phys 2002;52(2):310–5.

130

HOWELL

[3] Sur RK, Levin CV, Donde B, et al. Prospective randomized trial of HDR brachytherapy as a sole modality in palliation of advanced esophageal carcinoma: an International Atomic Energy Agency study. Int J Radiat Oncol Biol Phys 2002;53(1):127–33. [4] Safran H, Wanebo HJ, Hesketh PJ, et al. Paclitaxel and concurrent radiation for gastric cancer. Int J Radiat Oncol Biol Phys 2000;46(4):889–94. [5] Gunderson LL, Hoskins RB, Cohen AC, et al. Combined modality treatment of gastric cancer. Int J Radiat Oncol Biol Phys 1983;9(7):965–75. [6] Ben-Josef E, Lawrence TS. Radiotherapy for unresectable hepatic malignancies. Semin Radiat Oncol 2005;15(4):273–8. [7] Coia LR, Myerson RJ, Tepper JE. Late effects of radiation therapy on the gastrointestinal tract. Int J Radiat Oncol Biol Phys 1995;31(5):1213–36. [8] Rich T, Harris J, Abrams R, et al. Phase II study of external irradiation and weekly paclitaxel for nonmetastatic, unresectable pancreatic cancer: RTOG-98–12. Am J Clin Oncol 2004;27(1):51–6. [9] Willett CG, Clark JW. Update on combined-modality treatment options for pancreatic cancer. Oncology (Huntingt) 2003;17(Suppl 13):S29–36. [10] Ceha HM, van Tienhoven G, Gouma DJ, et al. Feasibility and efficacy of high dose conformal radiotherapy for patients with locally advanced pancreatic carcinoma. Cancer 2000;89(11):2222–9. [11] Crane CH, Macdonald KO, Vauthey JN, et al. Limitations of conventional doses of chemoradiation for unresectable biliary cancer. Int J Radiat Oncol Biol Phys 2002;53(4):969–74. [12] Mohiuddin M, Kudrimoti M, Regine WF, et al. Concurrent infusional gemcitabine and radiation in the treatment of advanced unresectable GI malignancy: a phase I study. Cancer J 2002;8(3):255–62. [13] Ratto C, Valentini V, Morganti AG, et al. Combined-modality therapy in locally advanced primary rectal cancer. Dis Colon Rectum 2003;46(1):59–67.

Gastroenterol Clin N Am 35 (2006) 131–142

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Managing Nonmalignant Chronic Abdominal Pain and Malignant Bowel Obstruction Lesley K. Bicanovsky, DO, Ruth L. Lagman, MD, MPH*, Mellar P. Davis, MD, Declan Walsh, MSc, FRCP (Edin) The Harry R. Horvitz Center for Palliative Medicine, Taussig Cancer Center, Cleveland Clinic, 9500 Euclid Avenue, M76, Cleveland, OH 44195, USA

A

bdominal pain is challenging for any physician. It is present at some time in 50% of all adults [1]. Prevalence is high across geographical regions, regardless of age and ethnicity [1–3]. The most frequent cause of abdominal pain is irritable bowel syndrome (IBS) [1–3]. Difficulty exists in separating IBS from abdominal pain associated with cancer or mechanical bowel obstruction. The differential diagnosis is often difficult because of the nonspecific nature and overlapping characteristics of abdominal pain, irrespective of the underlying cause. Abdominal pain clusters with other symptoms (back pain, bloating, nausea, vomiting), complicating the diagnosis [4]. A thorough history should include pain location, radiation, duration, severity, and associated symptoms (eg, chills, constipation, diarrhea, fevers nausea, vomiting, weight loss), and family history of bowel disorders, alcohol consumption, current medications, and menstrual history [5]. Physical findings vary depending upon chronicity and etiology. Rebound tenderness occurs with acute abdominal pain caused by perforation or distention. A distended but nontender abdomen and no bowel sounds occur with ileus or pseudo-obstruction. Obstruction and opioid bowel syndrome may be inseparable based upon history and physical examination. Multiple illnesses may have similar nonspecific abdominal complaints. Pain worsened by rectus abdominus contraction is usually somatic and nonvisceral [6]. Initial laboratory tests include complete blood count with differential, complete metabolic panel, amylase, lipase, ferritin, thyroid-stimulating hormone, and urine analysis. Additional invasive testing should be directed at ruling out specific diseases and not as a general screening tool. This article reviews common causes of

A World Health Organization Demonstration Project. Supported in part by a grant from the Project on Death in America of the Open Society Institute.

*Corresponding author. E-mail addresses: [email protected]; www.clevelandclinic.org/ palliative (R.L. Lagman). 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2006.01.002

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

132

BICANOVSKY, LAGMAN, DAVIS, ET AL

chronic abdominal pain and malignant bowel obstruction from a palliative medicine perspective. NEUROLOGIC BASIS OF ABDOMINAL PAIN Abdominal pain severity ranges from slight discomfort to excruciating pain. It often is preceded by nonpainful distention or bloating at low intensity. Visceral pain is often dull, poorly defined, and difficult to pinpoint. Sometimes, pain is referred to distant somatic areas. Referred pain occurs from convergence of somatic and visceral afferents upon the same spinal dorsal horn sensory neuron segments [7]. The intensity of referred pain reflects the severity of visceral pain. It often is misdiagnosed as musculoskeletal, or the diagnosis is delayed because of somatic referral [8,9]. These characteristics indicate that the representation of internal organs within the central nervous system are imprecise. Visceral nociceptors are fewer, are wide receptor field afferents that differ from somatic nociceptors, and are represented poorly in the primary somatosensory cortex (S1), but represented well in the cingulate, insular cortex, and the secondary somatosensory cortex, (S2). Therefore, perception and psychological processing of visceral pain are intense, but localization is poor and differs from somatic pain [10]. Visceral receptors are located on serosal, muscularis, and mucosal surfaces of hollow organs, and within mesenterium, but not in parenchyma [11]. Visceral nociceptors are activated by mechanical (stretch, distention, contraction, traction, compression, or torsion) and chemical (substance P, bradykinin, serotonin, histamine, and prostaglandins) stimuli [12,13]. Certain stimuli such as extensive cutting or burning (visceral injury) will not be perceived as pain [14,16]. Visceral pain becomes the summation of input from recruitment of high threshold and silent nociceptors not specific low threshold nociceptors. These nonspecialized high threshold nociceptors fire continuously once sensory thresholds are surpassed and produce a summation signal interpreted as pain [14]. Inflammation lowers nociceptor firing thresholds, lowering pain thresholds, and pain is experienced at lower distention pressures [14]. The paucity of afferent nerves relative to somatic tissue and poor representation within the primary S1 make localization of visceral pain imprecise. Visceral afferents travel with abdominal sympathetics to the dorsal horn and are designated as splanchnic or pelvic sensory afferents [14,15]. Splanchnic afferents originate from the lower thoracic and upper lumbar dorsal root ganglia and innervate upper abdominal viscera (including liver, spleen, pancreas, and gallbladder), small bowel, and the transverse colon. Pelvic afferents terminate in the S2–S5 sacral cord and innervate the distal transverse, descending, and sigmoid colon, and rectum [14,16]. Therefore, if visceral pain is relieved by blocking the visceral afferents that enter the spinal cord at certain levels, then the neuroanatomical distribution of afferents predicts the potential benefit to bilateral splanchnicectomy, celiac plexus block, or superior hypogastric plexus block [14].

MALIGNANT AND NONMALIGNANT GI COMPLAINTS

133

NONMALIGNANT ETIOLOGIES OF CHRONIC ABDOMINAL PAIN Chronic Pancreatitis Chronic pancreatitis is an inflammatory condition that permanently changes pancreatic structure, impairing of exocrine and endocrine function [17]. Most cases are caused by alcohol abuse; other causes include heredity, duct obstruction, certain systemic diseases (eg, systemic lupus erythematosus, cystic fibrosis), or idiopathic causes. Alcohol use is not an independent risk factor [18]. Smoking and genetic predisposition contribute [18]. The incidence is three to eight cases per 100,000 patients per year [18]. The cardinal feature is abdominal pain [17,19]. This is usually epigastric. It often radiates to the back, occasionally is associated with nausea and vomiting, and may be relieved by sitting upright or leaning forward. Patients writhe and rock back and forth rather than lie quietly, unlike patients who have back pain caused by vertebral disease. Pain worsens 15 to 30 minutes after eating. Early in the disease course, abdominal pain occurs in discrete attacks with pain-free periods. As pancreatitis becomes chronic, abdominal pain becomes continuous. Cullen’s sign, periumbilical ecchymosis, is usually a sign of hemorrhagic pancreatitis, but it also is reported with splenic rupture [20]. Chronic pancreatitis is associated with complications, such as bile duct or duodenal obstruction, pancreatic ascites, pancreatic pseudocysts, pleural effusions, pseudoaneurysms, and splenic vein thrombosis. Symptoms of chronic pancreatitis overlap significantly with pancreatic carcinoma. Pain pathogenesis is largely multifactorial and varies at different stages of the disease [24]. Colic occurs with biliary obstruction. Continuous pain is caused by pancreatic inflammation and edema. The release of excessive oxygen-free radicals, tissue hypoxia, acidosis, and inflammatory infiltration releases substances into damaged afferent nerve endings [24]. Pancreatic ductal and tissue fluid hypertension contributes to morphological changes of the pancreas [24]. Treatment of chronic pancreatitis depends on the stage of the disease and complications. Most patients are managed with low-fat diets, pancreatic enzyme supplements for exocrine insufficiency, insulin for endocrine insufficiency, and opioids for pain. The pancreas is innervated by the sympathetic, parasympathetic, and enteric nervous systems [21]. The celiac plexus is part of the sympathetic fibers of the splanchnic nerves arising from T5–T12, and it serves as the center for pain sensations from upper abdominal viscera [22]. Celiac plexus block should be limited to those with chronic pancreatitis whose pain has not responded to other modalities [23]. An alternative would be a splanchnic block. The American Gastroenterological Association guidelines from 1998 recommend pain management should be stepwise, beginning with high-dose pancreatic enzyme supplementation and H2-blockers before analgesics or invasive treatment. The use of an opioid implies a more severe or chronic process. Morphine sulfate can be used and if the patient is opiate naı¨ve, the starting dose should be the lowest dose possible and adjusted based on response and breakthrough dose.

134

BICANOVSKY, LAGMAN, DAVIS, ET AL

A patient controlled analgesic pump may be used initially to control pain and then changed to an oral dose through appropriate equianalgesic dose conversion. Morphine is the opioid of choice [25]. There is no evidence that other potent opioids are superior for managing pancreatic pain. It is believed that meperidine is safer in pancreatitis, because it causes less biliary spasm [26]. Meperidine, however, should be avoided because of potential toxicity, including seizures. Oral morphine in cancer pain has an established ratio of 1:3 [25] or 1:2 [27] when converting from parenteral to oral dosing with wide individual differences (not the 1:6 in single-dose studies of morphine) [28]. In clinical practice, there is a difference in relative analgesic potency when the route of administration is changed, and adjustment is essential to achieve an equivalent effect and avoid underdosing or overmedicating. The goal of appropriate pain management is for the patient to be alert, awake, and functional on a given opioid dose. Because of the complex etiology of chronic abdominal pain from pancreatitis, pharmacotherapy with analgesics alone is frequently insufficient for managing pain. For example, nonenteric coated enzyme preparations are also useful for treating pain, while enteric-coated enzyme preparations are preferred for managing steatorrhea. Octreotide is used for abdominal pain that is unresponsive to pancreatic enzyme therapy [29]. Enteral feeding, once contraindicated in chronic pancreatitis because it was believed to worsen the course of the disease, has been found to be beneficial in reducing postprandial pain [30] and improving overall outcome [31]. When medical therapy for pain has failed, endoscopic ultrasound-guided celiac plexus block or thoracoscopic splanchnicectomy can be considered. In very carefully screened cases, pancreatic resection can be offered for debilitating abdominal or back pain. The surgical approach to chronic pancreatitis is individualized based on pancreatic and ductal anatomy, pain characteristics, baseline exocrine and endocrine function, and medical comorbidity. No single approach is ideal [32–35]. Complications, such as bile duct or pancreatic duct strictures, require interventional endoscopy as first line treatment followed by surgery [36]. Irritable Bowel Syndrome IBS is characterized by chronic abdominal pain and altered bowel habit without organic etiology. It is the most commonly diagnosed gastrointestinal (GI) condition. Population studies in North America estimate a prevalence of 10% to 15% [37–39]. IBS can affect men or women, young or elderly patients, but it is more likely to be found in younger females (2:1 female:male predominance) [38]. The onset of abdominal pain after the age of 50 is unusual for IBS. Pathogenesis involves altered intestinal motility, visceral hypersensitivity, altered gut wall compliance, psychological disturbances, or any combination of these symptoms [40]. Patients usually describe their abdominal pain as crampiness that varies in intensity from day to day and during the day. The pain is generally in the lower abdomen, often on the left. Location and character, however, vary greatly (mildly annoying to debilitating) between individuals [41,42].

MALIGNANT AND NONMALIGNANT GI COMPLAINTS

135

IBS has no known cure. This is a chronic disorder of pain processing, and in this sense, it has the same pathogenesis (though different clinical manifestation) as fibromyalgia. The focus of treatment is symptom management, to maintain patient function, and secondarily to address patient concerns. An important component to treatment is establishing a therapeutic physician–patient relationship [43]. This allows for agreement on treatment objectives and a plan that targets troublesome symptoms [43]. The choice of therapies varies with the severity and type of symptoms. Management includes medications (antispasmodics, antidepressants, antidiarrheals, benzodiazepines, 5-hydroxytryptamine [serotonin] 3 receptor antagonists, 5- hydroxytryptamine [serotonin] 4 receptor agonists [tegaserod], antibiotics, and complementary therapies), patient education, dietary modification, and psychosocial supportive therapies. No single agent has been shown to be consistently effective in treating IBS symptoms. Antispasmodic drugs are used most frequently. They selectively inhibit GI smooth muscle, reducing colonic motor activity. They may relieve postprandial abdominal pain, bloating, and fecal urgency [44]. These medications should be used on an as-needed basis or in anticipation of symptoms caused by a known stressor. Antispasmodic drugs include dicyclomine and hyoscyamine. It is not clear whether one antispasmodic agent is more effective than another. Physicians will be limited to those which are locally available [44]. Bulk agents (dietary fiber) improve stool consistency and frequency but have no benefit in abdominal pain [44]. Antidepressants are analgesic and improve mood, which may reduce visceral pain [45–49]. Therapeutic effects often are obtained at doses lower than those for depression [45–49]. Tricyclic antidepressants also have anticholinergic properties that slow transit and improve diarrhea-predominant symptoms [45,49]. Antidiarrheal drugs should be used on an as-needed basis instead of a regular schedule. Agents like loperamide improve diarrhea but not abdominal pain [50]. 5HT3-receptor antagonists (eg, alosetron) improve diarrhea-predominant symptoms and pain; however there is a 1 in 350 risk for ischemic colitis, so it is not available [50,51]. Another 5-hydroxytryptamine (serotonin) 3 receptor antagonist (ondansetron) modulates visceral afferent activity and improves abdominal pain in diarrhea-predominant IBS [52,53]. Tegaserod is a partial 5HT4 receptor agonist which decreases constipation, abdominal discomfort, and bloating in some women with constipation-predominant IBS [54,55]. There is no published evidence that it offers any advantage over standard treatment (fiber supplementation or antispasmodics), because direct comparison has not been performed [56]. Tegaserod is unproven in men and those with alternating bowel habits [57]. Complementary therapies are being used increasingly to treat IBS symptoms. The number of randomized trials has doubled every 5 years [58,59]. Many people turn to this when conventional medicine fails or because they believe in the effectiveness of complementary medicine. The active ingredients of herbal medicines are largely unknown. There are increasing reports of liver and renal toxicity, and association with cancer [60–62].

136

BICANOVSKY, LAGMAN, DAVIS, ET AL

Potential targets for drug treatment of IBS are enteric P2X receptors and kappa receptor agonists. Enteric P2X receptors are expressed on GI sensory nerve endings [63]. They are multimeric ATP-gated cation channels expressed by neuronal and non-neuronal cells [63]. Agonists may enhance propulsion and secretion, where antagonists decrease propulsion and secretion, which could be useful in treating constipation-predominant and diarrhea-predominant IBS, respectively [63]. The kappa receptor agonist, fedotozine, blocks hypersensitivity by altering the processing of visceral sensations along gut afferent neuronal pathways [51]. There are limited published data to support its use in IBS. Clinical development in the United States seems to have been suspended because of inconsistent clinical benefits [51]. Malignant Bowel Obstruction A well-recognized complication to advanced abdominal and pelvic malignancy is malignant bowel obstruction (MBO). Obstruction occurs when normal GI peristalsis is interrupted mechanically or neurologically. This causes gastroparesis, small intestine or colon dilatation proximal to the blockage, and collapse of the small bowel or colon distally. As dilation occurs, blood flow is compromised, leading to ischemic bowel, necrosis, and rarely strangulation and sepsis. Loss of myenteric neuronal function will cause obstructive symptoms without a mechanically occluded lumen. Mechanical obstruction can be extrinsic, intraluminal, or mural. Intussusception is a rare presentation in adults most commonly caused by cancer; it rarely occurs in children. Symptoms include abdominal colic, abdominal pain, nausea, constipation, and vomiting. MBO may develop any time in the course of cancer; it usually occurs in advanced stages with progression of disease. The incidence ranges from 5% to 51% in ovarian cancer and 10% to 28% in primary intestinal malignancies [64,78]. Cancer is the second most common cause of bowel obstruction in adults, and the third most common cause in children. Nonmalignant causes are numerous and must be included in the differential diagnosis. Hernias were the most common cause during the first third of the last century, but presently postoperative adhesions cause approximately 75% of cases because of longer survival and the large number of patients undergoing laparotomy [65]. Postoperative adhesions cause extrinsic compression of the intestine. The diagnosis of obstruction is suspected based on clinical findings confirmed by radiographs. Mechanical obstruction must be distinguished from nonmechanical obstruction. A detailed history should be obtained (including symptoms; prior surgeries; history of Crohn’s disease, ulcerative colitis, or cancer; and review of medications). Physical findings suggesting malignant obstruction include Sister Mary Joseph nodes [66–68], ascites, supraclavicular adenopathy, palpable mass, hepatosplenomegaly, Blumer’s shelf [69], and succussion splash [70]. Laboratory studies are generally unhelpful to determine the cause or presence of an obstruction, but they can aid assessment of dehydration. Iron deficiency is a sign of GI bleeding and indicative of an intraluminal etiology. Radiographic studies should include an upright chest film to rule out free air

MALIGNANT AND NONMALIGNANT GI COMPLAINTS

137

and supine and upright abdominal films. Multiple air fluid levels with distended loops of bowel are seen unless there is a gastric outlet obstruction with frequent vomiting. Occasionally, a similar pattern is seen with a paralytic ileus or pseudo-obstruction. The level of the obstruction determines the symptoms and the severity. The higher the obstruction, the more severe the symptoms, the fewer the signs. Most patients require no further radiographic testing for adequate confirmation of an obstruction. Plain films can be equivocal in 20% to 30% of cases. They can be normal or nonspecific or misleading in 10% to 20% of cases secondary to obesity, bowel loops filled with fluid (gasless abdomen), adynamic ileus, IBD, or prior abdominal malignant tumor [71,72]. In the past, a small bowel series confirmed the diagnosis. CT has replaced the barium study as the adjuvant study of choice for those who are surgical candidates. The CT scan is not only more sensitive and specific compared with plain abdominal films, but it also provides objective evaluation of the global extent of the disease, important in subsequent therapeutic decision-making. MRI should be considered in patients with allergies to intravenous contrast or those unable to tolerate oral barium secondary to nausea and vomiting. Virtual colonoscopy may be helpful for those who present with a high-grade occlusive colon cancer unable to complete a total colon evaluation (standard colonoscopy and barium contrast enema) preoperatively. CT scanning, gentle air insufflation, and glucagons preparation can accurately visualize the proximal colon preoperatively [73]. The bowel preparation and air insufflation necessary for the virtual colonoscopy are as difficult to tolerate as the preparation for standard colonoscopy. Radiographic procedures should be reserved for those who are surgical candidates. Management is challenging. Surgery is the treatment of choice. Bypass procedures or debulking with reanastomosis procedures to re-establish bowel continuity usually are attempted. Some with advanced disease, however, are inoperable and require medical management to palliate symptoms. The success of medical management depends on the level and degree of obstruction. The higher the level, the less abdominal distention, the greater the vomiting, and the more difficult to medically manage. Complete obstructions are less responsive to medical management than partial obstructions. Most obstructions in advanced cancer are partial and rarely emergencies [74]. Continuous pain from mesenteric implants or peritoneal carcinomatosis is palliated by continuous opioids. Corticosteroids or parenteral nonsteroidal anti-inflammatory drugs such as ketorolac [75,76] may be added as adjuvants to improve pain control. Corticosteroids improve nausea, vomiting, and oral intake. Abdominal cramps are relatively opioid-resistant, but they are relieved with anticholinergics. Glycopyrrolate is used with a starting dose of 0.1 to 0.2 mg every 6 hours as needed, adjusted to renal or hepatic function [77]. Glycopyrrolate is a quaternary amine and has a low risk for causing delirium, because it does not cross the blood–brain barrier. If abdominal cramps are absent in partial bowel obstruction, a trial of metoclopramide (30 to 60 mg subcutaneous or intravenous over 24 hours, with 10 mg every 6 hours as needed, and can be escalated to 120 mg over 24 hours) may relieve gastroparesis,

138

BICANOVSKY, LAGMAN, DAVIS, ET AL

nausea, and partial obstruction [78,79]. Metoclopramide should not be used with anticholinergics. Anticholinergics block the peristaltic action, and doses need to be halved for hepatic or renal failure. Nausea and vomiting initially should be treated with haloperidol if the obstruction is complete or if there is colic. Starting doses are 5 mg intravenously or subcutaneously over 24 hours, with 1 mg every 6 hours as needed and increased if indicated. Vomiting is the most difficult symptom to control. Complete relief of vomiting can be achieved in only 30% of cases. Successful management limits emesis to once daily [78]. Delirium can occur with prerenal azotemia because of accumulation of opioid metabolites. Gentle hydration by either intravenous or hypodermoclysis hydration can be done safely in the hospital or at home [80,81]. The choice between methods depends upon venous access and location of care. Hydration also can be used to replace vomitus and insensible loss. Over-rehydration can lead to ascites, pleural effusions, and pulmonary edema, particularly if renin levels are elevated [82]. Hydration is based on prognosis and performance status, not routine. Refractory bowel obstruction by definition is the failure to respond to the combination of opioid, anticholinergic, and antiemetic medications at optimal doses. These individuals may respond by expanding the initial drug regimen. Octreotide, a somatostatin analog, can be added to anticholinergics for relief of colic, vomiting; they also may reduce pain [83]. Octreotide is given at 100 to 600 lg per 24 hours intermittent or as continuous infusion [84–86]. Corticosteroids may also be added, with reported responses in two-thirds of patients. Palliation of these symptoms for this inoperable group also may be aided by stenting obstructing lesions or by placing a percutaneous gastrostomy tube or a percutaneous transesophageal gastrostomy tube [87,88]. SUMMARY Evaluation of abdominal pain requires an understanding of the possible causes (benign or malignant) and recognition of typical patterns and clinical presentation. Abdominal pain has multiple causes; associated signs and symptoms may aid in the diagnosis. Remember that some patients will not have a textbook presentation, and unusual causes for pain must be considered. Those with chronic pancreatitis with structural complications should be operated on early, whereas those with other types of chronic pancreatitis should receive medical therapy focusing on alleviating symptoms. Control of the most troublesome symptoms will provide the best management for IBS. Pharmacologic success in bowel obstruction depends on the level and degree of obstruction. Decision making is based on reasonable expectations of survival, treatment-related success, performance status, and goals of care. Quality of life will be enhanced by appropriate symptom management. References [1] Heading RC. Prevalence of upper gastrointestinal symptoms in the general population: a systematic review. Scand J Gastroenterol Suppl 1999;231:3.

MALIGNANT AND NONMALIGNANT GI COMPLAINTS

139

[2] Jones R, Lydeard S. Irritable bowel syndrome in the general population. BMJ 1992;304: 87. [3] Sandler R. Epidemiology of irritable bowel syndrome in the United States. Gastroenterology 1993;99:409. [4] Fleischer AB Jr, Gardner EF, Feldman SR. Are patients. Chief complaints generally specific to one organ system? Am J Manag Care 2001;7:299. [5] Kamin RA, Nowicki TA, Courtney DS, et al. Pearls and pitfalls in the emergency department evaluation of abdominal pain. Emerg Med Clin North Am 2003;21(1):10. [6] Choi YK, Chou S. Rectus syndrome. Another cause of upper abdominal pain. Reg Anesth 1995;20(4):347–51. [7] Cervero F. Visceral pain—central sensitization. Gut 2000;47(Suppl 4):56–7. [8] Giamberardino MA. Recent and forgotten aspects of visceral pain. Eur J Pain 1999;3: 77–92. [9] Hobson AR, Aziz Q. Central nervous system processing of human visceral pain in health and disease. News Physiol Sci 2003;18:109–14. [10] Cervero F. Visceral hyperalgesia revisited. Lancet 2000;356(9236):1127–8. [11] Gebhart GF. Pathobiology of visceral pain: molecular mechanisms and therapeutic implications v. central nervous system processing of somatic and visceral sensory signals. Am J Physiol Gastrointest Liver Physiol 2000;278(6):G834–8. [12] Cervero F. Neurophysiology of gastrointestinal pain. Baillieres Clin Gastroenterol 1988;2: 183. [13] Haupt P, Janig W, Kohler W. Response pattern of visceral afferent fibers, supplying the colon upon chemical and mechanical stimuli. Pflugers Arch 1983;398:41. [14] Cervero F. Sensory innervation of the viscera: peripheral basis of visceral pain. Physiol Rev 1994;7(1):95–138. [15] Gebhart GF. Visceral nociception: consequences, modulation, and the future. Eur J Anaesthesiol Suppl 1995;10:24–7. [16] McMahon SB. Are there fundamental differences in the peripheral mechanisms of visceral and somatic pain? Behav Brain Sci 1997;20:381–91. [17] Steer ML, Waxman I, Freedman SD. Chronic pancreatitis. N Engl J Med 1995;332:1482. [18] Ihse I, Andersson R, Albiin N, et al. Guidelines for management of patients with chronic pancreatitis. Report from a consensus conference. Lakartidningen 2003;100:2518–25. [19] Mergener K, Baillie J. Chronic pancreatitis. Lancet 1997;350:1379. [20] Chung MA, Oung C, Szilagyi A. Cullen’s sign: it doesn’t always mean hemorrhagic pancreatitis. Am J Gastroenterol 1992;87(8):1026–8. [21] Toskes PP. Gastroenterology and hepatology: the comprehensive visual reference. Volume 8, pancreas. In: Feldman M, editor. Gastroenterology and hepatology: the comprehensive visual reference series. Current Medicine, Incorporated; 1998. p. 1.4–1.5. [22] Lee MJ, Mueller PR, van Sonnenberg E, et al. CT-guided celiac ganglion block with alcohol. Am J Radiology 161(9): 633–6. [23] Draganov P, Toskes PP. Chronic pancreatitis: controversies in etiology, diagnosis and treatment. Rev Esp Enferm Dig 2004;96(9):649–54. [24] Bornman PC, Marks IN, Girdwood AW, et al. Pathogenesis of pain in chronic pancreatitis: ongoing enigma. World J Surg 2003;27(11):1175–82. [25] Hanks GW, Conno F, Cherng N, et al. Morphine and alternative opioids in cancer pain: the EAPC recommendations. Br J Cancer 2001;84:587–93. [26] Latta KS, Ginsberg B, Barkin RL. Meperidine: a critical review. Am J Ther 2002;9:53–68. [27] Twycross RG. Therapeutic equivalence of oral and subcutaneous intramuscular morphine sulfate in cancer patients. J Palliat Care 1988;2:67–8. [28] Houde RW, Wallenstein S, Beaver WT. Clinical measurement of pain. In: Stevens G, editor. Analgesics. New York: Academic Press; 1965. p. 75–122. [29] Singh VV, Toskes PP. Medical therapy for chronic pancreatitis pain. Curr Gastroenterol Rep 2003;5(2):110–6.

140

BICANOVSKY, LAGMAN, DAVIS, ET AL

[30] Shea JC, Bishop MD, Parker EM, et al. An enteral therapy containing medium-chain triglycerides and hydrolyzed peptides reduces postprandial pain associated with chronic pancreatitis. Pancreatology 2003;3(1):36–40. [31] Mayerle J, Simon P, Lerch MM. Medical treatment of acute pancreatitis. Gastroenterol Clin North Am 2004;33(4):855–69. [32] Duffy JP, Reber HA. Surgical treatment of chronic pancreatitis. J Hepatobiliary Pancreat Surg 2002;9(6):659–68. [33] Morrison CP, Wemyss-Holden SA, Partensky C, et al. Surgical management of intractable pain in chronic pancreatitis: past and present. J Hepatobiliary Pancreat Surg 2002;9(6): 675–82. [34] Hartel M, Tempia-Caliera AA, Wente MN, et al. Evidence-based surgery in chronic pancreatitis. Langenbecks Arch Surg 2003;388(2):132–9. [35] Dite P, Ruzicka M, Zboril V, et al. A prospective, randomized trial comparing endoscopic and surgical therapy for chronic pancreatitis. Curr Gastroenterol Rep 2004;35(7):553–8. [36] Mossner J, Keim V. Therapy of chronic pancreatitis. Internist (Berl) 2003;44(12):1515–23. [37] Talley NJ, Zinsmeister AR, Van Dyke C, et al. Epidemiology of colonic symptoms and the irritable bowel syndrome. Gastroenterology 1991;101:927. [38] Brandt LJ, Bjorkman D, Fennerty MB, et al. Systematic review on the management of irritable bowel syndrome in North America. Am J Gastroenterol 2002;97:S7. [39] Drossman DA, Zhiming L, Andruzzi E, et al. US householders survey of functional gastrointestinal disorders: prevalence, sociodemography, and health impact. Dig Dis Sci 1993;93: 1569. [40] DePonti F, Malagegelada J. Functional gut disorders from motility to sensitivity disorders. Pharmacol Ther 1998;89:49–88. [41] Lynn RB, Friedman LS. Irritable bowel syndrome. N Engl J Med 1993;329:1940. [42] Swarbrick ET, Bat L, Hegarty JE, et al. Site of pain from irritable bowel. Lancet 1980;2:443. [43] Drossman DA, Thompson WG. The Irritable bowel syndrome: Review and a graduated multicomponent treatment approach. Ann Intern Med 1992;116:1009. [44] Quartero AO, Meineche-Schmidt V, Muris J, et al. Bulking agents, antispasmodic and antidepressant medication for the treatment of irritable bowel syndrome. Cochrane Collaboration 2005;18:CD003460. [45] Eisendrath SJ, Kodama KT. Fluoxetine management of chronic abdominal pain. Psychosomatics 1992;33:227. [46] Hameroff SR, Weiss JL, Lerman JC, et al. Doxepin’s effects on chronic pain and depression: A controlled study. J Clin Psychiatry 1984;45:47. [47] Pilowsky K, Barrow CG. A controlled study of psychotherapy and amitriptyline used individually and in combination in the treatment of chronic intractable psychogenic pain. Pain 1990;40:3. [48] Gorard DA, Libby GW, Farthing MJ. Effect of tricyclic antidepressant on small intestinal motility in health and diarrhea predominant irritable bowel syndrome. Dig Dis Sci 1995;40:86. [49] Clouse RE, Lustman PJ, Geisman RA, et al. Antidepressant therapy in 138 patients with irritable bowel syndrome: a five-year clinical experience. Aliment Pharmacol Ther 1994;8:409. [50] Talley NJ. Evaluation of drug treatment in irritable bowel syndrome. Br J Clin Pharmacol 2003;56(4):362–9. [51] Callahan MJ. Irritable bowel syndrome neuropharmacology. A review of approved and investigational compounds. J Clin Gastroenterol 2002;35(Suppl 1):S58–67. [52] Zighelboim J, Talley NJ, Phillips SF, et al. Visceral perception in irritable bowel syndrome: Rectal and gastric responses to distention and serotonin type 3 antagonism. Dig Dis Sci 1995;40:819. [53] Prior A, Read NW. Reduction of rectal sensitivity and post-prandial motility by granisetron, a 5–HT3-receptor antagonist, in patients with irritable bowel syndrome. Aliment Pharmacol Ther 1993;7:175.

MALIGNANT AND NONMALIGNANT GI COMPLAINTS

141

[54] Prather CM, Camilleri M, Zinsmeister AR, et al. Tegaserod accelerates orocecal transit in patients with constipation-predominant irritable bowel syndrome. Gastroenterology 2000;118:463. [55] Muller-Lissner SA, Fumagalli I, Bardhan KD, et al. Tegaserod, a 5-HT(4) receptor partial agonist, relieves symptoms in irritable bowel syndrome patients with abdominal pain, bloating, and constipation. Aliment Pharmacol Ther 2001;15:1655. [56] Camilleri M. Tegaserod in the treatment of irritable bowel syndrome. UpToDate; 2004. [57] Talley NJ. Pharmacologic therapy for the irritable bowel syndrome. Am J Gastroenterol 2003;98(4):750–8. [58] Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. JAMA 1998;280(18):1569–75. [59] Vickers A. Recent advances: complementary medicine. BMJ 2000;321:683–6. [60] Bensoussan A, Myers SP, Carlton Al. Risks associated with the practice of traditional Chinese medicine: an Australian study. Arch Fam Med 2000;9(10):1071–8. [61] Koh HL, Woo SO. Chinese proprietary medicine in Singapore: regulatory control of toxic heavy metals and undeclared drugs. Drug Saf 2000;23(5):351–62. [62] Melchart D, Linde K, Weidenhammer W, et al. Liver enzyme elevations in patients treated with traditional Chinese medicine. JAMA 1999;282(1):28–9. [63] Galligan JJ. Enteric P2X receptors as potential targets for drug treatment of irritable bowel syndrome. Br J Pharmacol 2004;141(8):1294–302. [64] Feuer DJ, Broadley KE, Shepard JH, et al. Systematic review of surgery in malignant bowel obstruction in advanced gynecological and gastrointestinal cancer. Gynecol Oncol 1999;75:313–22. [65] Parker MC, Ellis H, Morgan BJ, et al. Postoperative adhesions: ten-year follow-up of 12,584 patients undergoing lower abdominal surgery. Dis Colon Rectum 2001;44:822. [66] Schneider V, Smyczek B. Sister Mary Joseph’s nodule. Diagnosis of umbilical metastases by fine needle aspiration. Acta Cytol 1990;34(4):555–8. [67] Khan AJ, Cook B. Metastatic carcinoma of umbilicus: Sister Mary Joseph’s nodule. Cutis 1997;60(6):297–8. [68] Coll DM, Meyer JM, Mader M, et al. Imaging appearances of Sister Mary Joseph nodule. Br J Radiol 1999;72(864):1230–3. [69] Blumer G. Classic articles in colonic and rectal surgery: the rectal shelf. Dis Colon Rectum 1980;23(6):445–8. [70] Davis NC. Intestinal succussion splash—a valuable clinical sign insufficiently appreciated. Med J Aust 1964;10:360–1. [71] Balthazar EJ. CT of small-bowel obstruction. AJR Am J Roentgenol 1994;162:255–61. [72] Dunn JT, Halls JM, Berne TV. Roentgenographic contrast studies in acute small-bowel obstruction. Arch Surg 1984;119:1305–8. [73] Felon HM, McAneny DB, Nunes DP, et al. Occlusive colon carcinoma: virtual colonoscopy in the preoperative evaluation of the proximal colon. Radiology 1999;210:423–8. [74] Rubin S. Intestinal obstruction in advanced ovarian cancer: what does the patient want? Gynecol Oncol 1999;75:311–2. [75] Joishy SK, Walsh D. The opioid-sparing effects of intravenous ketorolac as an adjuvant analgesic in cancer pain: application in bone metastases and the opioid bowel syndrome. J Pain Symptom Manage 1998;16(5):334–9. [76] Chen JY, Wu GJ, Mok MS, et al. Effect of adding ketorolac to intravenous morphine patientcontrolled analgesia on bowel function in colorectal surgery patients—a prospective, randomized, double-blind study. Acta Anaesthesiol Scand 2005;49(4):546–51. [77] Davis MP, Furste A. Glycopyrrolate: a useful drug in the palliation of mechanical bowel obstruction. J Pain Symptom Manage 1999;18(3):153–4. [78] Davis MP, Nouneh C. Modern management of cancer-related intestinal obstruction. Curr Oncol Rep 2000;2:343–50.

142

BICANOVSKY, LAGMAN, DAVIS, ET AL

[79] Ripamonti C, De Conno F, Ventafridda BV, et al. Management of bowel obstruction in advanced and terminal cancer patients. Ann Oncol 1993;4:15–21. [80] Fainsinger RL, MacEachern T, Miller MJ, et al. The use of hypodermoclysis for rehydration in terminally ill cancer patients. J Pain Symptom Manage 1994;9(5):298–302. [81] Steiner N, Bruera E. Methods of hydration in palliative care patients. J Palliat Care 1998;14(2):6–13. [82] Morita T, Tei Y, Inoue S, et al. Fluid status of terminally ill cancer patients with intestinal obstruction: an exploratory observational study. Support Care Cancer 2002;10:474–9. [83] Pandha HS, Waxman J. Octreotide in malignant intestinal obstruction. Anticancer Drugs 1996;7:5–10. [84] Mystakidou K, Tsilika E, Kalaidopoulou O, et al. Comparison of octreotide administration vs conservative treatment in the management of inoperable bowel obstruction in patients with far advanced cancer: a randomized, double-blind, controlled clinical trial. Anticancer Res 2002;22:1187–92. [85] Mercadante S, Maddaloni S. Octreotide in the management of inoperable gastrointestinal obstruction in terminal cancer patients. J Pain Symptom Manage 1992;7(8):496–8. [86] Mercadante S, Spoldi E, Caraceni A, et al. Octreotide in relieving gastrointestinal symptoms due to bowel obstruction. Palliat Med 1993;7:295–9. [87] Oishi H, Shindo H, Shirotani N, et al. The experience of improved quality of life at home for the long term, using percutaneous trans-esophageal gastro-tubing drainage for a case with terminal stage cancer. J Tokyo Wom Med Univ 2001;71(3):188–92. [88] Oishi H, Shindo H, Shirotani N, et al. A nonsurgical technique to create an esophagostomy for difficult cases of percutaneous endoscopic gastrostomy. Surg Endosc 2003;17:1224–7.

Gastroenterol Clin N Am 35 (2006) 143–151

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Surgical Palliation of Bowel Obstruction Robert S. Krouse, MD* University of Arizona, Tucson, AZ, USA Arizona Cancer Center, Tucson, AZ, USA

T

he treatment of malignant bowel obstruction (MBO) is a common dilemma encountered by all general surgeons treating cancer patients. MBO may be the presenting picture of a previously undiagnosed intraluminal tumor, or it can herald recurrent disease. Because of the advanced, incurable disease that bowel obstruction from a malignancy usually portends, treatment is considered palliative. Because bowel obstruction can occur in as many as 5% to 43% of patients with a diagnosis of advanced primary or metastatic intra-abdominal malignancy [1], it is necessary for surgeons, gastroenterologists, and oncologists to have a thorough understanding of this condition. The most common primary malignancies causing MBO are ovarian (5.5% to 51%) and colorectal (10% to 28%) cancers [2]. Nonabdominal cancers, including lung cancer, breast cancer, and melanoma, also are known to occasionally cause intestinal obstruction. The long-term known survival for these patients is poor (4 to 9 months) [3]. Although newer chemotherapeutic options may improve long-term survival [4], patients ultimately succumb to their cancers. Lower-grade tumors, such as pseudomyxoma peritonii, may have a better outlook, further warranting consideration of more aggressive treatment [5]. CAUSES OF MALIGNANT BOWEL OBSTRUCTION AND RELEVANCE TO DECISION-MAKING There are multiple causes of bowel obstruction in the cancer patient. Clearly, the cause of the obstruction, if it can be discerned, will play a role in the treatment decision-making process. Benign causes include adhesions, radiationinduced strictures, and internal hernias. The rate of benign intestinal obstruction in patients with previously diagnosed malignancy is variably reported but significant (3% to 48%) [6], and a benign cause in the setting of an MBO must be considered even for a patient with known metastatic or unresectable

*Southern Arizona Veterans Affairs Health Care System Surgical Care Line, 2-112 Tucson, AZ 85723, USA. E-mail address: [email protected] 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2006.01.003

Published by Elsevier Inc. gastro.theclinics.com

144

KROUSE

cancer. In addition, radiation enteritis should be considered in patients with a history of abdominal exposure to radiation therapy. Bowel obstruction resulting from tumors includes mechanical and functional mechanisms. Carcinomatosis, rather than local tumor recurrence, is likely the cause of small bowel obstruction from neoplasm [3–7], in contrast with large bowel obstructions, which are usually at a single site [7]. The effects of mechanical obstruction by tumor may be accentuated by inflammatory edema, constipation, cancer- or treatment-induced fibrosis, abnormalities in intestinal mobility, decreased production of intestinal enzymes and secretions, a change in fecal flora, or the adverse effects of medications [1]. SELECTION FOR SURGERY Although the overall patient status, including physical, social, psychologic, and spiritual domains, would seem mandatory considerations for palliative intervention, this big picture view is not reflected in previously used selection criteria for surgery. The Krebs and Gopland prognostic index may be used to determine optimal patients for surgery [8]. This index uses age, nutritional status, tumor status, ascites, previous chemotherapy, and previous radiation therapy as the prognostic parameters. The authors define successful palliation or benefit from surgery as survival for at least 2 months. Although this index based on physical and objective parameters is used commonly, it assumes that optimal palliation is reflected in survival time and not quality-of-life outcomes. Therefore, the usefulness of this index is unclear in assessing surgical palliation. Surgical intervention for MBO can have serious complications, including death. If the patient can resume an acceptable quality of life, based on the patient’s, caretaker’s, and surgeon’s assessments, an operation can be considered. It is also vital that the patient have a reasonable expected survival time. If a patient can be expected to live for several months after an operation, barring major complications, surgical options should be contemplated. If it is estimated that a patient has only days to a few weeks of life remaining, then a major operation seems unreasonable. Because even experienced oncologic surgeons have been shown to be frequently incorrect in their predictions of survival time following a palliative procedure [9], prognoses should be based on performance status, literature-based procedure-specific operative mortality rates, and on preoperative extent of disease. Treatment may have a secondary gain of improved survival, and this possibility should be taken into account. If the patient is deemed too sick for a surgical procedure, if an operation is unlikely to help achieve goals, or the patient refuses an operative intervention, other modalities must be recommended. It is unclear if specific contraindications to surgery can be mandated. A literature review of surgical procedures for MBO [10] found nearly all studies were retrospective, whereby no defined criteria could be discerned. Reports are variable as to the percentage of patients deemed inoperable (6.2% to 50%), with the most frequent reasons noted being extensive tumor, multiple partial

SURGICAL PALLIATION

145

obstructions, and inability to correct obstructions surgically [11]. Most authors agree that the criteria that suggest surgery would unlikely benefit a patient include ascites, carcinomatosis, palpable intra-abdominal masses, multiple bowel obstructions, and very advanced disease with poor overall clinical status [11]. Recommendations are not uniform, as in one series of MBO caused by colorectal cancer that demonstrated that the number of sites of obstruction did not correlate with resumption of bowel function or postoperative death [12]. In elderly patients with nonovarian cancers and known large-volume ascites (at least 3 L) who undergo abdominal procedures, the perioperative mortality of 41% [13] seems prohibitive. Recurrence of MBO has been noted to be significantly increased and symptom-free survival to be significantly decreased for patients with as little as 100 mL of ascites [3]. Blair and colleagues found that ascites accompanying a small bowel obstruction predicted a worse outcome following surgery [14]. In one study, quality of life seemed to be significantly better for patients with local recurrence compared with those with carcinomatosis [5]. Reoperation following a previous operation for MBO usually should be discouraged, as it has been shown that these patients have a low likelihood of successful palliation (30%) and limited survival rates [15]. Occasionally, a surgeon is faced with the dilemma of a patient with an acute or surgical abdomen and an MBO. Although a true emergency of this nature is rare [16], and decision-making is more difficult in this setting, the same operative criteria should be applied. If reasonable survival time cannot be anticipated, it is warranted not to pursue surgical options. Although it has been shown in the case of colorectal cancer obstructions that survival is quite good for emergent procedures [17,18], it is unclear from the literature how pressing were the indications to operate emergently. For large bowel obstruction causing threat of proximal perforation in a patient with limited survival expectations, intestinal stenting should be considered. If intestinal stenting is not an option, a loop colostomy, ileostomy, or blow-hole colostomy are reasonable alternatives. Surgery should be performed only after thorough discussion and clear understanding by the patient, family, and surgical team about the expected goals and outcomes of a surgical intervention. Informed consent for this necessitates explanation of possible operative complications, mortality rates from surgery and the underlying disease process, recovery time, and symptom recurrence rate. If an operation is considered, parameters should be set for future operative intervention. It is imperative that there be no miscommunication resulting in an unrealistic expectation for cure. Improvement or preservation of quality of life should be understood as the primary goal of treatment. The patient and family should be reassured they will not be abandoned regardless of the outcome of the operation. Because pain after surgery is a predictable concern, preoperative planning should include discussion of postoperative analgesia. An epidural catheter and other postoperative pain management techniques should be considered, not only for procedure-related pain, but also for ongoing cancerinduced pain. Although opioids may contribute to bowel dysfunction, they are usually necessary in the perioperative period.

146

KROUSE

SURGICAL OPTIONS When attempting an operation for intestinal obstruction with a palliative intent, the surgeon’s repertoire should include a wide range of options because of the wide spectrum of operative findings. Although the quickest and safest procedure that can alleviate the obstruction or favorably impact on symptoms should be considered, there is limited information related to the optimal procedure in the setting of a MBO. This reflects the heterogeneity of the diseases that cause obstruction and the lack of patient-defined considerations in earlier published series. In localized disease, resection of the obstructed segment may afford the best outcome. Resection may entail restoration of bowel continuity. In the setting of carcinomatosis, or if a prohibitive amount of dissection is needed, alternative approaches should be used. This may include a bypass procedure, either by an enterostomy or an intestinal stoma, or simply a drainage procedure using a large gastrostomy tube. Although a gastrostomy tube alone for obstruction may not improve the ability to eat, it usually will alleviate the problems of nausea, vomiting, and pain. Patients may be able to take some form of oral diet in time. Although it has been noted that patients with resection may live somewhat longer [2] and have a slightly better quality of life [5] than those with bypass procedures, no study has addressed this issue in a randomized, prospective fashion. Based on the many different clinical presentations related to MBO, it is unlikely that such a study will take place. If a benign cause of bowel obstruction is encountered, usual surgical techniques are often applicable, but bypass or intestinal stoma may be recommended, if they will shorten the procedure and lead to less morbidity. For patients undergoing an urgent procedure for large bowel obstruction, it may be unwise to reanastamose bowel, and exteriorization of colon or ileum may be a more prudent option. In addition, a blow-hole colostomy is a potentially simpler procedure that has been used in this setting [19]. Intestinal stomas can benefit the quality of life for patients with MBO greatly. Unfortunately, they frequently are avoided, as they are unsightly to patients and regarded as a failure by surgeons. Although the threat of an intestinal stoma is a major concern for patients, colostomies and ileostomies may provide a greater ease of care and minimize patient symptoms. For example, in a patient who has a complete low rectal obstruction caused by tumor along with a substantial metastatic tumor burden, placement of a loop colostomy or ileostomy is a procedure that can be performed quickly with minimal morbidity. A blowhole colostomy may be a reasonable procedure for patients who have a malignant toxic megacolon [19]. As always, ostomy placement should be considered preoperatively so that proper stoma placement can be attained. Involving a stoma nurse to help mark the possible site on the abdomen may facilitate stoma planning. In addition, involving a stoma nurse in the preoperative setting allows the patient to discuss issues and, one hopes, to allay fears [20,21]. It is imperative the patient and family members receive adequate postoperative training, because caring for a stoma is one more challenge in an already overwhelming situation. The ability for a follow-up visit with the surgical team, and

SURGICAL PALLIATION

147

with the ostomy nurse, is mandatory, as difficulties and anxieties are sure to arise. Radiation enteritis can mimic the presentation of bowel obstruction caused by tumor even in the absence of tumor. Because of the impaired perfusion of the bowel wall resulting from radiation arteritis, trivial serosal injuries can result in major complications. The incidence of radiation enteritis is related directly to the amount of small bowel in the radiation field and the dose of radiation received. For patients with radiation-induced obstructions, intestinal bypass rather than resection may be the preferred surgical treatment [22]. Intestinal bypass has been shown to lead to long-term alleviation of gastrointestinal obstructions and successful palliation in most patients [22]. Severe adhesions may make surgery impossible [23], and other treatments must be considered in this situation. Recently, laparoscopy has been used to bypass obstruction caused by radiation injury [24]. Laparoscopic approaches to MBO are increasingly an option, as surgeons expand the limits of these procedures. Potential advantages include reduced pain and a shorter hospital stay. Laparoscopy also may allow the diagnosis of unforeseen disease in the peritoneal cavity, which may limit surgical intervention. Laparoscopic techniques are highly dependent on surgeon experience and ability, and are usually not appropriate for patients with dense adhesions and carcinomatosis. In properly selected patients, especially those who present initially with incurable cancer and a bowel obstruction or who had only laparoscopic procedures previously, this may be a less invasive surgical option. Although indications for surgical intervention remain the same, it is likely that treatment-related morbidity and mortality can be lessened through laparoscopic approaches. The need to convert to an open procedure may be increased for obstruction related to malignant disease [25,26]. Reports for laparoscopic palliative procedures are small [24–27], and definitive conclusions are not available. Given the quality-of-life benefits identified in the treatment of primary colorectal malignancies, minimally invasive approaches may allow more latitude in selection of surgical procedures. Averbach and Sugarbaker [4] advocate an aggressive approach to surgical palliation that entails maximal debulking and possible intraperitoneal chemotherapy for patients with symptomatic recurrent intra-abdominal cancer. Success is much more likely in patients who have low-grade tumors such as pseudomyxoma peritonii, complete cytoreduction, and a relatively long interval between primary surgery and the obstruction. It should be noted that their group reports that aggressive surgery is time-consuming (mean, greater than 10 hours), involves large blood losses (approximately 1500 cm3), and carries a high morbidity (55%), and mortality (7.14%). They advocate this treatment only in the most experienced hands. Because the authors focused on survival outcomes, it is not clear that these techniques improve overall quality of life or reduce symptoms. Therefore, it is difficult to interpret the role of these interventions for patients who have a disease with an aggressive biology and terminal status. Clearly, this treatment can be advocated only in a minority of the

148

KROUSE

patients who present with intestinal obstruction in the palliative setting, but aggressive cytoreductive surgery can be considered in limited settings. During surgery for MBO, in addition to addressing the immediate problem of bowel obstruction, it is also important to anticipate other potential complications. Survival time, recurrent obstruction, treatment-related morbidity, nonobstructive symptoms, and quality-of-life considerations are all part of this calculation. In some cases, a combination of staged invasive approaches may be warranted. For example, initial stent placement may allow for later one-stage [28], laparoscopic [29], or open [30] resection in a more controlled and safe manner. At the time of operation, a venting gastrostomy may be used in addition to a resection or bypass procedure for a patient with carcinomatosis who is likely to become reobstructed but would not tolerate an additional invasive procedure.

SURGICAL OUTCOMES Resumption of oral intake following operation for MBO is variable and not well documented in the literature [10]. Patients who are chronically obstructed or have had a complicated procedure may not be ready to eat for a longer time. Others may be anorexic from cancer cachexia syndrome, which will not reverse unless the underlying cancer is eradicated. The inability to eat can prompt initiation of forced nutritional supplementation in patients with limited life expectancy. There are no simple solutions to this. There are few data to support forced nutrition in the setting of advanced, progressive cancer, although feeding and hydration are deeply emotional issues that have the potential for becoming ethical dilemmas. Although the total parenteral nutrition (TPN) in these settings may be considered and initiated, limitations should be set based on preoperative objectives (eg, ability to go home, eating with family, independence). TPN has not been shown to improve survival or nutritional parameters in the patient with advanced cancer, and malnourished patients may sustain increased complications from the central venous access that is necessary [31]. If TPN is instituted, enhancement of quality of life should be the measure of efficacy of this treatment. This standard is reached more readily if discussions about nutrition and hydration are part of the continuous process of shared decision making. Extended survival time often can be an expected secondary benefit of the primary intention to surgically palliate symptoms of MBO. Unfortunately, survival time is frequently used as the only measure of success [8]. Although aggressive surgical approaches have been advocated, most outcomes have been reported as survival time and procedure-related morbidity, frequently omitting quality-of-life concerns. Operative mortality is frequent (5% to 32%), most often related to progression of neoplasm [5,10,32]. Morbidity is also common (42%) [32], and reobstruction after surgery may be quite high (10% to 50%) [10,33]. Although it is recognized that improvement in quality of life after surgery is variable

SURGICAL PALLIATION

149

(42% to 85%) [10,34], there is no consistent parameter used to determine this clinical outcome. Measures include the ability to tolerate solid food [12], the ability to tolerate oral feeding at discharge, the ability to resume a normal diet, restoration of bowel function, ability to return home, ability to live without a recurrent obstruction, and survival for more than 60 days [10]. Without uniform and clearly defined outcome measures, it is difficult to measure success. The primary goals of treatments are frequently multiple: the alleviation of pain, nausea, and vomiting. Outpatient care should be an important goal of therapy, no matter what treatment approach is initiated. Outpatient care depends on the availability of appropriate resources, including hospice and caretakers who are willing and able to take on the multiple tasks necessary to care for these patients, and on financial and insurance limitations. Although it is important to be familiar with the retrospective literature related to MBO, the aims for surgical and other approaches must be considered and prioritized based on each individual patient’s and family’s goals and desires. SUMMARY MBO is a common but difficult problem for surgeons caring for cancer patients. Nonsurgical interventions should be considered in all patients who have MBO, especially in those with limited expected survival time or for whom surgery will have little effect on disease control. Surgical options can be helpful in the setting of MBO, as long as reasonable goals and realistic outcomes are clear. There is no defined algorithm for all patients with MBO, and decision-making is based on reasonable estimates of survival and treatmentrelated success. Therefore, better prospective data need to be collected for this population of patients. In addition, a randomized prospective trial comparing treatments based on the clinical scenario could help practitioners who care for patients with this condition. References [1] Baines M. The pathophysiology and management of malignant bowel obstruction. In: Doyle D, Hanks GWC, MacDonald N, editors. Oxford textbook of palliative medicine. Oxford, UK: Oxford University Press; 1998. p. 526–34. [2] Davis MP, Nouneh D. Modern management of cancer-related intestinal obstruction. Curr Pain Headache Rep 2001;5:257–64. [3] Aranha GV, Folk FA, Greenlee HB. Surgical palliation of small bowel obstruction due to metastatic carcinoma. Am Surg 1981;47(3):99–102. [4] Higashi H, Shida H, Ban K, et al. Factors affecting successful palliative surgery for malignant bowel obstruction due to peritoneal dissemination from colorectal cancer. Jpn J Clin Onc 2003;33(7):357–9. [5] Averbach AM, Sugarbaker PH. Recurrent intra-abdominal cancer with intestinal obstruction. International Surgery 1995;80(2):141–6. [6] Legendre H, Vahhuyse F, Caroli-Bosc FX, et al. Survival and quality of life after palliative surgery for neoplastic gastrointestinal obstruction. Eur J Surg Oncol 2001;27:364–7. [7] Aabo K, Pedersen H, Bach F, Knudsen J. Surgical management of intestinal obstruction in the late course of malignant disease. Acta Chir Scand 1984;150(2):173–6. [8] Krebs HB, Goplerud DR. Surgical management of bowel obstruction in advanced ovarian carcinoma. Obstet Gynecol 1983;61:327–30.

150

KROUSE

[9] McCahill LE, Smith DD, Borneman T, et al. A prospective evaluation of palliative outcomes for surgery of advanced malignancies. Ann Surg Onc 2003;10(6):654–63. [10] Feuer DJ, Broadley KE, Shepherd JH, et al. Systematic review of surgery in malignant bowel obstruction in advanced gynecological and gastrointestinal cancer. Gynecol Oncol 1999;75:313–22. [11] Ripamonti C. Management of bowel obstruction in advanced cancer. Curr Opin Oncol 1994;6:351–7. [12] Lau PWK, Lorentz TG. Results of surgery for malignant bowel obstruction in advanced, unresectable, recurrent colorectal cancer. Dis Colon Rectum 1993;36:61–4. [13] Yazdi GP, Miedema BW, Humphrey LJ. High mortality after abdominal operation in patients with large-volume malignant ascites. J Surg Oncol 1996;62:93–6. [14] Blair SL, Chu DZ, Scwarz RE. Outcome of palliative operations for malignant bowel obstruction in patients with peritoneal carcinomatosis from nongynecological cancer. Ann Surg Oncol 2001;8:632–7. [15] Pothuri B, Montemarano M, Gerardi M, et al. Percutaneous endoscopic gastrostomy tube placement in patients with malignant bowel obstruction due to ovarian carcinoma. Gynecol Oncol 2004;96:330–4. [16] Rubin SC. Intestinal obstruction in advanced ovarian cancer: what does the patient want? Gynecol Oncol 1999;75:311–2 [17] Huang TJ, Wang JY, Lee LW, et al. Emergency one-stage surgery for obstructing left-sided colorectal carcinomas. Kaohsiung J Med Sci 2002;18:323–8. [18] Tsugawa K, Koyanagi N, Hashizume M, et al. Therapeutic strategy of emergency surgery for colon cancer in 71 patients over 70 years of age in Japan. Hepato-Gastroenterology 2002;49:393–8. [19] Remzi FH, Oncel M, Hull TL, et al. Current indications for blow-hole colostomy: ileostomy procedure. A single center experience. Int J Colorectal Dis 2003;18(4):361–4. [20] Comb J. Role of the stoma care nurse: patient with cancer and colostomy. Br J Nurs 2003;12(14):852–6. [21] Thomson I. Teaching the skills to cope with a stoma. Nurs Times 1998;94:55–6. [22] Lillemoe KD, Brigham RA, Harmon JW, et al. Surgical management of small bowel radiation enteritis. Arch Surg 1983;118:905–7. [23] Kwitko AO, Peiterse AS, Hecker R, et al. Chronic radiation injury to the intestine: a clinico– pathologic study. Aust N Z J Med 1982;12:272–7. [24] Lauter DM. Laparoscopic enterocolostomy of the palliation of malignant bowel obstruction. J Laparoendosc Adv Surg Tech A 2000;10:275–6. [25] Agresta F, Piazza A, Michelet I, et al. Small bowel obstruction. Laparoscopic approach. Surg Endosc 2000;14(2):154–6. [26] Ibrahim IM, Wolodiger F, Sussman B, et al. Laparoscopic management of acute small-bowel obstruction. Surg Endosc 1996;10(10):1012–4. [27] Milsom JW, Kim SH, Hammerhofer KA, et al. Laparoscopic colorectal cancer surgery for palliation. Dis Colon Rectum 2000;43:1512–6. [28] Khot UP, Lang AW, Murali K, et al. Systematic review of the efficacy and safety of colorectal stents. Br J Surg 2002;89:1096–102. [29] Morino M, Bertello A, Garbarini A, et al. Malignant colonic obstruction managed by endoscopic stent decompression followed by laparoscopic resection. Surg Endosc 2002;16(10): 1483–7. [30] Martinez-Santos C, Lobato RF, Fradejas JM, et al. Self-expandable stent before elective surgery vs. emergency surgery for the treatment of malignant colorectal obstructions: comparison of primary anastomosis and morbidity rates. Dis Colon Rectum 2002;45: 401–6. [31] Easson AM, Hinshaw DB, Johnson DL. The role of tube feeding and total parenteral nutrition in advanced illness. J Am Coll Surg 2002;194(2):225–9.

SURGICAL PALLIATION

151

[32] Makela J, Kiviniemi H, Laitinen S, et al. Surgical management of intestinal obstruction after treatment for cancer. Eur J Surg 1991;157:73–7. [33] Miner TJ, Jaques DP, Paty PB, et al. Symptom control in patients with locally recurrent rectal cancer. Ann Surg Oncol 2003;10(1):72–9. [34] Miner TJ, Jaques DP, Shriver CD. A prospective evaluation of patients undergoing surgery for the palliation of an advanced malignancy. Ann Surg Oncol 2002;9(7):696–703.

Gastroenterol Clin N Am 35 (2006) 153–165

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Endoscopic Ultrasound-Guided Pain Control for Intra-abdominal Cancer Michael J. Levy, MDa,*, Maurits J. Wiersema, MDb a

Developmental Endoscopy Unit, Division of Gastroenterology and Hepatology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA b Indiana Medical Associates, 7900 West Jefferson Boulevard, Suite 201, Fort Wayne, IN 46804, USA

P

atients who have intra-abdominal cancer, particularly pancreatic cancer, commonly experience pain that is often difficult to control, thereby necessitating opioid administration [1,2]. Although opioids effectively relieve pain, opioid use is associated commonly with dry mouth, constipation, nausea, vomiting, drowsiness, delirium, impaired immune function, and, recently, associated with death [3,4]. Therefore, nonpharmacologic therapies, such as celiac plexus neurolysis, are often given with the goal of improving pain control and quality of life (QOL) while reducing the risk of drug-related side effects. The term celiac plexus block (CPB) has been applied to the use of steroids or a local anesthetic injection to temporarily inhibit celiac plexus function in patients who have benign diseases such as chronic pancreatitis. The term celiac plexus neurolysis (CPN) is often applied to techniques in which alcohol is injected to induce permanent neural destruction in patients who have pancreatic cancer and other malignancies. Although many clinicians use these terms interchangeably, they signify different procedures performed for different indications. This article summarizes percutaneous and surgical methods for performing CPN and focuses on the technical aspects of endoscopic ultrasound-guided (EUS) CPN. Published literature concerning EUS CPN is reviewed, indications are proposed, and opinions are offered concerning potential future applications and investigational needs as they apply to this technique. LITERATURE REVIEW METHODOLOGY A comprehensive literature review using OVID (http://www.ovid.com) and the expanded version of each of the following key words: celiac plexus neurolysis, celiac plexus block, celiac ganglion neurolysis, celiac ganglion block, celiac plexus neurolysis, celiac plexus block, celiac ganglion neurolysis, celiac ganglion block, and sympathetic block. These terms were used to search

*Corresponding author. E-mail address: [email protected] (M.J. Levy). 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.005

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

154

LEVY & WIERSEMA

MEDLINE from 1960 through 2005. Only studies published in English were included. Results of the search were augmented by a review of the references from each of these reports, with the inclusion of additional articles considered germane to the topic. CURRENT TECHNOLOGY Although the terms ‘‘celiac plexus’’ and ‘‘splanchnic nerves’’ are often used interchangeably, they are anatomically distinct structures (Fig. 1) [5–7]. The splanchnic nerves are located cephalic to the diaphragm (in a retrocrural position), most often anterior to the 12th thoracic vertebra. The celiac plexus is located caudal to the diaphragm (in an antecrural position); it surrounds the origin of the celiac trunk and is composed of a dense network of ganglia and interconnecting fibers. Ganglia vary in number (1–5), size (diameter 0.5– 4.5 cm), and location (T12–L2) [5]. The celiac plexus transmits the sensation of pain for the pancreas and most of the abdominal viscera, except the left colon, rectum, and pelvic organs (Fig. 2) [8]. The neurons that innervate the pancreas [9] can receive nociceptive stimulation and then transmit this pain information to the celiac plexus [10]. Stimuli reach the thalamus and cortex of the brain, which induce the sensation of pain. Descending inhibitory mechanisms may also modulate the ascending pain information. Kappis described the classic technique of CPN in 1914 [11]. Modifications have been created in the attempt to improve the accuracy of needle placement and pain relief while reducing procedure-related complications. These techniques differ with respect to the route of needle insertion (Figs. 3–7), use of radiologic guidance versus a blind procedure, and chemical composition of the injectate. For CPN in cancer patients, the injectate usually includes a local anesthetic (bupivacaine or lidocaine) and a neurolytic agent (phenol or alcohol). The local

Fig. 1. Anatomy of the splanchnic nerves and celiac plexus. (Copyright 2005 by the Mayo Clinic; with permission.)

EUS-GUIDED PAIN CONTROL FOR INTRA-ABDOMINAL CANCER

155

Fig. 2. The celiac plexus transmits the sensation of pain for the pancreas and most of the abdominal viscera except the left colon, rectum, and pelvic organs. (Copyright 2005 by the Mayo Clinic; with permission.)

anesthetic reduces the discomfort caused by the neurolytic agent. Phenol produces minimal pain because of its local anesthetic effect. Although direct comparisons between alcohol and phenol have not been performed, alcohol is favored because it induces greater neurolysis and presumably greater pain relief [12]. Three meta-analyses have reached conflicting conclusions regarding percutaneous CPN for intra-abdominal malignancy, including therapy for patients who have cancer involving the breast, lungs, esophagus, stomach, colon, rectum, liver, gallbladder, bile ducts, adrenal glands, kidneys, and pancreas [13–15].

Fig. 3. Classic approach, also referred to as the posterolateral, retrocrural, or deep splanchnic approach. (Copyright 2005 by the Mayo Clinic; with permission.)

156

LEVY & WIERSEMA

Fig. 4. Transcrural (or antecrural) approach. (Copyright 2005 by the Mayo Clinic; with permission.)

Lebovits and Lefkowitz [13] concluded that CPN leads to very successful relief of pancreatic cancer pain. Sharfman and Walsh [14] on the other hand, found the data insufficient to judge the efficacy, long-term morbidity, or cost effectiveness. Most recently, Eisenberg and colleagues [15] reviewed the literature from 1966 to 1993, including 24 studies in which two were randomized controlled trials, one was prospective, and 21 were retrospective uncontrolled trials. The cancer type was specified in 1117 patients (63% pancreatic and 37% nonpancreatic). Good to excellent pain relief was reported in 89% of patients during the first 2 weeks after they underwent CPN. Partial to complete pain relief was reported

Fig. 5. Transaortic approach. (Copyright 2005 by the Mayo Clinic; with permission.)

EUS-GUIDED PAIN CONTROL FOR INTRA-ABDOMINAL CANCER

157

Fig. 6. Anterior approach. (Copyright 2005 by the Mayo Clinic; with permission.)

in approximately 90% of patients at 3 months and 70% to 90% at the time of death. Interestingly, pain relief was not influenced by the technical approach or the use of radiologic guidance. Outcomes were not influenced by the primary site of malignancy. The most common side effects, local pain (96%), diarrhea (44%), and hypotension (38%), were generally mild and transient. The authors conclude that CPN has a long-lasting benefit for 70% to 90% of patients who have pancreatic and other intra-abdominal cancers, regardless of the technique used, and that adverse effects are common but generally transient and mild. More recently, a prospective, randomized, double-blind study of 24 patients who had pancreatic cancer who underwent percutaneous CPN was reported [16]. The CPN group experienced a significant reduction in analgesic use and drug-induced side effects versus patients who received drug therapy alone. Kawamata and colleagues [17] showed that CPN results in less deterioration in quality of life for pancreatic cancer patients when added to morphine therapy versus morphine or nonsteroidal anti-inflammatory drug therapy alone. This was believed to occur as a result of an increased duration of the analgesic effect and reduced opioid side effects. Major complications develop in approximately 1% to 2% of patients, including lower extremity weakness and paresthesia, paraplegia, puncture of adjacent

Fig. 7. Transintervertebral approach. (Copyright 2005 by the Mayo Clinic; with permission.)

158

LEVY & WIERSEMA

organs, and chronic gastroparesis and diarrhea [7,15,18]. Neurologic complications result from spinal cord ischemia or direct injury to the spinal cord or somatic nerves. Spinal cord ischemia may result from thrombosis or spasm of the artery of Adamkiewicz, located on the left of the spine between vertebrae T8 and L4, which perfuses the lower two thirds of the spinal cord (Fig. 8) [19,20]. Despite theoretical advantages of the given methods, it is believed that the risk of neural dysfunction is not influenced by the technical approach. Paraplegia has been reported with each percutaneous method regardless of the use of radiologic guidance. There are even several reports of paraplegia following the most direct approach (surgical neurolysis) [21]. CPN may also be performed surgically (Fig. 9). Lillemoe and colleagues [22] published a prospective randomized trial of 137 patients who had unresectable pancreatic adenocarcinoma. Neurolysis improved pain control, compared with placebo, at 2, 4, and 6 months’ follow-up. Although the study was not designed to assess survival benefit, patients who had preoperative pain had a significant survival advantage compared with controls. The reason is unclear but may relate to reduced opioid-induced side effects, improved nutritional status, and emotional well being. To date, no study has reproduced these results. The potential disadvantages of surgical neurolysis include reduced pain relief reported by some, uncertainty regarding response to therapy because of difficulty differentiating postoperative pain from cancer pain, and limited surgical access that

Fig. 8. Ischemia of the spinal cord may result from thrombosis or spasm of the artery of Adamkiewicz, which is located on the left of the spine between T8 and L4 and perfuses the lower two thirds of the spinal cord. a, artery; Ant, anterior. (Copyright 2005 by the Mayo Clinic; with permission.)

EUS-GUIDED PAIN CONTROL FOR INTRA-ABDOMINAL CANCER

159

Fig. 9. Surgical neurolysis. (Copyright 2005 by the Mayo Clinic; with permission.)

prolongs procedure time [7,13]. Additionally, the frequency of surgery for unresectable pancreatic adenocarcinoma would be expected to decline with more sensitivity cross-sectional imaging, thereby limiting the population of patients who may benefit from this approach. Wong and colleagues [23] evaluated 100 patients who had unresectable pancreatic cancer to determine the affect of percutaneous CPN on pain relief, QOL, and survival. Patients were randomized to either percutaneous CPN or systemic analgesic therapy alone with sham injection. At 1 week after therapy, patients in both groups experienced improved pain control (P <.01) and QOL (P <.001), with an advantage for those who were randomized to percutaneous CPN (P ¼ .005). The analgesic benefit of percutaneous CPN compared with analgesic therapy alone was sustained over the 12-month follow-up period or until death. However, opioid consumption, the frequency of opioid-induced adverse effects, and QOL were similar between the groups. As opposed to the findings of Lillemoe and colleagues [22], CPN failed to result in a survival advantage. TECHNIQUE FOR PERFORMING ENDOSCOPIC ULTRASOUND-GUIDED CELIAC PLEXUS NEUROLYSIS EUS CPN has been developed for the purpose of offering a method of pain control at the same setting of a diagnostic examination for pancreatic malignancy. Furthermore, enhanced image resolution may permit optimal injection of the neurolytic agents [24]. When EUS CPN is offered to patients, allergic conditions and the use of anticoagulant drugs should be addressed. Informed consent is obtained with specific attention to the unique complications associated with CPN. Patients are hydrated initially with 500 to 1000 mL normal saline to minimize the risk of hypotension. Patients are placed in the left lateral decubitus position and sedated using medications such as midazolam, meperidine, and in some cases, promethazine. Throughout the procedure, patients are monitored continuously by an automated noninvasive blood pressure device, pulse oximeter, and electrocardiogram.

160

LEVY & WIERSEMA

Linear array endosonographic imaging from the posterior lesser curve of the gastric fundus allows identification of the aorta, which appears in a longitudinal plane (Fig. 10). The aorta is traced distally to the celiac trunk, which is the first major branch below the diaphragm. Targeting the celiac plexus is based on the position of the celiac plexus relative to the position of the celiac trunk. Color Doppler ultrasonography is used to confirm the vascular nature of the structures. A 22-gauge needle is primed with saline to remove air and then placed through the biopsy channel and affixed to the hub. The needle is inserted under EUS guidance immediately adjacent and anterior to the lateral aspect of the aorta at the level of the celiac trunk (Fig. 11). The needle is flushed with 3 mL of normal saline to remove any tissue acquired during insertion. An aspiration test is performed to rule out vessel penetration before each injection. For performing CPN in cancer patients, 10 mL (0.25%) of bupivacaine is injected followed by 10 mL (98%) of dehydrated alcohol. The alcohol, which produces an echogenic cloud, may lead to discomfort. Before withdrawing the needle, it should be flushed with 3 mL normal saline to prevent seeding of the needle track with alcohol, which may produce transient severe postprocedural pain. The entire process is then repeated on the opposite side of the aorta. Occasionally, altered anatomy resulting from significant lymphadenopathy or bulky tumors may necessitate injection of the entire solution into one preferably anterior site. There are limited data regarding the efficacy of a ‘‘single’’ versus ‘‘bilateral’’ injection. Lemelin and colleagues [25] compared unilateral with bilateral injection in a mixed population of patients who had pancreatic cancer or chronic pancreatitis. The investigators determined that bilateral CPN or CPB was clearly more effective than a single injection, with a mean pain reduction of 70.4% versus 45.9% (P ¼ .0016). However, the report raises several questions regarding the injection technique and study methodology, which cannot be clarified because their findings were been reported only in abstract form. Nevertheless, their study offers preliminary data to suggest that bilateral injection may offer enhanced pain relief.

Fig. 10. Endoscopic ultrasonography imaging from the posterior lesser curve of the gastric fundus allows identification of the aorta, which appears in a longitudinal plane. (Copyright 2005 by the Mayo Clinic; with permission.)

EUS-GUIDED PAIN CONTROL FOR INTRA-ABDOMINAL CANCER

161

Fig. 11. Under EUS guidance, the needle is inserted immediately adjacent and anterior to the lateral aspect of the aorta at the level of the celiac trunk. (Copyright 2005 by the Mayo Clinic; with permission.)

After the procedure, the vital signs are monitored for 2 hours. Before the patient is discharged, blood pressure is checked, with the patient in both a supine and erect position to assess for orthostasis. CPN is performed routinely as an outpatient procedure and rarely necessitates hospitalization. OUTCOMES Although there is extensive literature assessing the efficacy of percutaneous CPN for a broad spectrum of intra-abdominal malignancies, reports evaluating EUS CPN are limited to those with pancreatic cancer. Wiersema and Wiersema [24] published the initial study evaluating EUS CPN in 30 patients who had malignancy, 25 of whom had pancreatic cancer. The same group published a follow-up study that included all 25 patients who had pancreatic cancer from their initial report [26]. This latter prospective study involved 58 patients who underwent EUS CPN for pain secondary to inoperable pancreatic cancer. The neurolysis procedure, which averaged 15 minutes, was performed by injecting 3 to 6 mL (0.25%) of bupivacaine and 10 mL (98%) of alcohol into both sides of the celiac region. Pain scores were assessed using a standardized 11-point visual analog scale. Forty-five patients (78%) experienced a drop in pain score after EUS CPN. Overall pain scores were significantly lower (P <.0001) 2 weeks after the procedure. On multivariate analysis, sustained pain relief was seen for 24 weeks and was independent of morphine use or adjuvant therapy. However, patients who received chemotherapy alone or chemotherapy plus radiation experienced pain relief in addition to that offered by EUS CPN. Pain relief resulting from adjuvant therapy increased over time and at 24 weeks was statistically significant (P ¼ .002). Although opioid administration increased throughout the study, the increase was not statistically significant. There were no major complications. Minor complications were mild and transient and included hypotension (20%), diarrhea (17%), and pain exacerbation (9%).

162

LEVY & WIERSEMA

Although this study offers preliminary data suggesting the efficacy and safety of EUS CPN, the small sample size, absence of a placebo control group, and no physician or patient blinding limit the strength of the conclusions. Despite the 45 patients (78%) who experienced a drop in pain score, only 31 (54%) experienced a decline of greater than two points, which is a measure of improvement that some consider necessary to signify efficacy. The efficacy of EUS CPN diminished at 8 to 12 weeks, after which pain scores in patients not receiving adjuvant therapy trended upward. Although the results of this study are promising, these data considered in isolation do not allow definitive conclusions to be made regarding the safety and efficacy of EUS CPN in pancreatic cancer. INDICATIONS There are several potential advantages of EUS-guided CPN compared with percutaneous CPN. The proximity of the posterior lesser curve of the stomach to the celiac plexus, the use of continuous real-time visualization of the target area, and the availability of Doppler color to assess the vasculature facilitate accurate needle placement. By doing so, EUS theoretically may improve pain relief and reduce complications. Unfortunately, the lack of well-designed studies comparing studies between EUS and percutaneous approaches prohibits the verification of these assumptions. Although the EUS study results are encouraging, methodological shortcomings limit the strength of the conclusions. In addition, these studies were conducted by an expert endosonographer, and one must assume that as the use of this method broadens so too will the reported rate of complications, and the potential for diminished efficacy may vary. Those who perform this procedure should have a thorough knowledge of the anatomy, mechanism of action, and technical aspects concerning linear endosonography, color Doppler, and aspiration needles. This level of expertise is necessary to optimize the efficacy and minimize complications. Despite these shortcomings, the present authors believe the following statements are true: (1) The efficacy of CPN is similar regardless of the technique (EUS or percutaneous). This view is supported by the finding of a meta-analysis that has concluded that the efficacy of CPN is independent of the percutaneous approach or the use of radiologic guidance [15]. The reported efficacy rates of EUS CPN have been similar to those reported for percutaneous methods. Although comparative studies have not been performed, we suspect that the efficacy is similar. (2) The risk of EUS CPN is similar or slightly lower than percutaneous methods. Serious complications (such as paraplegia) have never been reported with EUS. This is likely because percutaneous methods are used far more often than EUS as much because of any difference in the inherent risk for a particular procedure. However, EUS is an ‘‘anterior’’ approach, which thereby avoids the retrocrural space and may reduce the risk of neurologic dysfunction and pulmonary complications. Furthermore, as opposed to the percutaneous anterior approach, with EUS, the needle traverses only the gastric wall, presumably eliminating complications resulting from

EUS-GUIDED PAIN CONTROL FOR INTRA-ABDOMINAL CANCER

163

Box 1: Classification system Class I: Conditions for which there is evidence or general agreement that a given procedure or treatment is useful and effective Class II: Conditions for which there is conflicting evidence or a divergence of opinion about the usefulness or efficacy of a procedure or treatment Class IIa: Weight of evidence or opinion is in favor of usefulness or efficacy Class IIb: Usefulness or efficacy is less well established by evidence/opinion Class III: Conditions for which there is evidence or general agreement that the procedure or treatment is not useful or effective and in some cases may be harmful

inadvertent penetration of surrounding organs. The authors assume that the risk of pain, hypotension, diarrhea, and abscess formation would be similar for EUS and percutaneous approaches. If EUS guidance offers no advantage in terms of pain relief and no minimal risk advantage, then what is its role versus percutaneous techniques for patients who have intra-abdominal cancers? The major disadvantage of EUS CPN is the inherent cost associated with the endoscopy. Also, when a diagnostic block is desired, the necessary conscious sedation will confound the assessment of any benefit derived from injection of the local anesthetic. Given the inaccuracies of a diagnostic block, few investigators consider this later point an issue [6,16,27]. Furthermore, the use of conscious sedation alleviates acute discomfort associated with CPN. The present authors reserve EUS CPN for patients in whom disease spread precludes a satisfactory percutaneous approach. Although most studies have found that CPN reduces cancer pain, it rarely eliminates pain, and nearly all patients require continued opioid use, albeit often at a lower dose. When counseling patients, it is important to emphasize a realistic goal, which is not to eliminate pain but to optimize oral pharmacologic therapy and to allow a dose reduction to minimize the side effects. The timing of the neurolysis relative to the onset of pain may predict response. In one study, CPN was more effective when performed early after pain onset rather than late in its course [6]. This may be explained by the fact that early pancreatic cancer pain appears to derive mainly from the celiac plexus. Using the classification system (Box 1) and definitions for the level of evidence (Box 2) adopted by the American Heart Association, data are summarized as follows (Box 3). The present authors consider pancreatic cancer pain

Box 2: Level of evidence Level of evidence A: Data derived from multiple randomized clinical trials Level of evidence B: Data derived from a single randomized trial or nonrandomized studies Level of evidence C: Consensus opinions of experts

164

LEVY & WIERSEMA

Box 3: Recommendations for EUS CPN Class IIa: 1. Pancreatic cancer pain (level of evidence B)

a class IIa indication because the weight of evidence favors the efficacy of EUS CPN. For this indication, the level of evidence is considered category B. By default, one may consider that EUS CPN is equally effective for other intraabdominal cancers, as is the case for percutaneous CPN. However, in the absence of any data, such a claim is only conjectural. The only absolute contraindications for EUS CPN for the palliation of pain caused by malignancy are (1) coagulopathy (international normalized ratio
1.5); (2) thrombocytopenia (platelets 50,000); or (3) an uncooperative patient. A relative contraindication is altered anatomy that prohibits access. In summary, despite the paucity of data, EUS CPN appears to be as effective as and potentially safer than other methods of CPN for providing pain relief from unresectable pancreatic cancer. EUS may be the most cost-effective approach because CPN can be performed at the time of biopsy and staging. Larger, long-term, prospective randomized controlled trials that evaluate cost, clinical outcomes, and quality of life are needed. Only then can we verify the efficacy and safety of EUS CPN and identify advantages and disadvantages of this approach over conventional techniques. Future trials should focus on remaining issues including (1) the optimal timing and route for CPN; (2) the ideal composition of the injectate; (3) relative cost; (4) patient preference; (5) influence on quality of life; (6) the effect on hospitalization duration (in those who have chronic pancreatitis); and (7) potential survival advantage. References [1] Ventafridda GV, Caraceni AT, Sbanotto AM, et al. Pain treatment in cancer of the pancreas. Eur J Surg Oncol 1990;16(1):1–6. [2] Lankisch PG. Natural course of chronic pancreatitis. Pancreatology 2001;1(1):3–14. [3] Ventafridda V, Tamburini M, Caraceni A, et al. A validation study of the WHO method for cancer pain relief. Cancer 1987;59(4):850–6. [4] Yeager MP, Colacchio TA, Yu CT, et al. Morphine inhibits spontaneous and cytokine-enhanced natural killer cell cytotoxicity in volunteers. Anesthesiology 1995;83(3):500–8. [5] Ward EM, Rorie DK, Nauss LA, et al. The celiac ganglia in man: normal anatomic variations. Anesth Analg 1979;58(6):461–5. [6] Ischia S, Ischia A, Polati E, et al. Three posterior percutaneous celiac plexus block techniques: a prospective, randomized study in 61 patients with pancreatic cancer pain. Anesthesiology 1992;76(4):534–40. [7] Brown DL, Moore DC. The use of neurolytic celiac plexus block for pancreatic cancer: anatomy and technique. J Pain Symptom Manage 1988;3(4):206–9. [8] Plancarte R, Velasquez R, Patt R. Neurolytic blocks of the sympathetic axis. In: Pratt RB, editor. Pain. Philadelphia: Lippincott Williams & Wilkins; 1993. p. 377. [9] Nagakawa T, Mori K, Nakano T, et al. Perineural invasion of carcinoma of the pancreas and biliary tract. Br J Surg 1993;80(5):619–21. [10] Gebhardt GF. Visceral pain mechanisms. In: Chapman CR, Foley KM, editors. Current and emerging issues in cancer pain. New York: Raven Press; New York; 1993. p. 99–111.

EUS-GUIDED PAIN CONTROL FOR INTRA-ABDOMINAL CANCER

165

[11] Kappis M. Erfahrungen mit local anasthesie bie bauchoperationen. Vehr Dtsch Gesellsch Chir 1914;43:87–9. [12] Mercadante S, Nicosia F. Celiac plexus block: a reappraisal [commentary]. Reg Anesth Pain Med 1998;23(1):37–48. [13] Lebovits AH, Lefkowitz M. Pain management of pancreatic carcinoma: a review. Pain 1989;36(1):1–11. [14] Sharfman WH, Walsh TD. Has the analgesic efficacy of neurolytic celiac plexus block been demonstrated in pancreatic cancer pain? Pain 1990;41(3):267–71. [15] Eisenberg E, Carr DB, Chalmers TC. Neurolytic celiac plexus block for treatment of cancer pain: a meta-analysis. Anesth Analg 1995;80(2):290–5 [erratum appears in Anesth Analg 1995;80(2):213]. [16] Polati E, Finco G, Gottin L, et al. Prospective randomized double-blind trial of neurolytic coeliac plexus block in patients with pancreatic cancer. Br J Surg 1998;85(2):199–201. [17] Kawamata M, Ishitani K, Ishikawa K, et al. Comparison between celiac plexus block and morphine treatment on quality of life in patients with pancreatic cancer pain. Pain 1996;64(3):597–602. [18] Davies DD. Incidence of major complications of neurolytic coeliac plexus block. J R Soc Med 1993;86(5):264–6. [19] De Conno F, Caraceni A, Aldrighetti L, et al. Paraplegia following coeliac plexus block [commentary]. Pain 1993;55(3):383–5. [20] van Dongen RT, Crul BJ. Paraplegia following coeliac plexus block. Anaesthesia 1991;46(10):862–3. [21] Hayakawa J, Kobayashi O, Murayama H. Paraplegia after intraoperative celiac plexus block. Anesth Analg 1997;84(2):447–8. [22] Lillemoe KD, Cameron JL, Kaufman HS, et al. Chemical splanchnicectomy in patients with unresectable pancreatic cancer: a prospective randomized trial. Ann Surg 1993;217(5): 447–55 [discussion: 456–7]. [23] Wong GY, Schroeder DR, Carns PE, et al. Effect of neurolytic celiac plexus block on pain relief, quality of life, and survival in patients with unresectable pancreatic cancer: a randomized controlled trial. JAMA 2004;291(9):1092–9. [24] Wiersema MJ, Wiersema LM. Endosonography-guided celiac plexus neurolysis. Gastrointest Endosc 1996;44(6):656–62. [25] Lemelin V, et al. A Prospective Trial of Central versus Bilateral Celiac Plexus Bloc/Neurolysis in 160 Patients: Bilateral Injection Is Safe and Is More Effective. Gastrointestinal Endoscopy 2005;61(5):AB77. [26] Gunaratnam NT, Sarma AV, Norton ID, et al. A prospective study of EUS-guided celiac plexus neurolysis for pancreatic cancer pain. Gastrointest Endosc 2001;54(3):316–24. [27] Fujita Y. CT-guided neurolytic splanchnic nerve block with alcohol [commentary]. Pain 1993;55(3):363–6.

Gastroenterol Clin N Am 35 (2006) 167–188

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Percutaneous-Guided Pain Control: Exploiting the Neural Basis of Pain Sensation Michael D. Adolph, MD, Costantino Benedetti, MD* Arthur G. James Cancer Hospital, Richard J. Solove Research Institute, Ohio State University College of Medicine and Public Health, 400 West 10th Avenue, Suite 511, Columbus, OH 43210, USA

P

atients who are suffering from chronic gastrointestinal illnesses, particularly cancer, are human beings in the throes of physical, psychosocial, and spiritual crises that comprise their total pain [1–3]. The patient and primary care physician may seek the gastroenterologist to manage symptoms of advanced illnesses [4,5]. When cure is not feasible, palliative care tools are available to ease a patient’s suffering. When pain is a prominent symptom, the gastroenterologist may include percutaneous-guided techniques in the therapeutic toolbox. Help in this area may be as close as the nearest pain medicine specialist, anesthesiologist, or interventional radiologist. To date, no percutaneous procedure has replaced systemic analgesic therapy (SAT) with opioids. However, percutaneous-guided pain control interventions have earned a role in the multidisciplinary approach to pain therapy. Given that afferent sensory fibers of abdominal and pelvic organs travel in sympathetic ganglia, a chemical or surgical sympathectomy for visceral pain is analogous to a peripheral nerve block or dorsal rhizotomy for somatic pain. Most percutaneous-guided procedures in the abdomen are performed with imaging localization and local anesthetic or contrast medium injections to confirm effective spread [6]. Percutaneous techniques may be complex and highly operator-dependent, but time-tested procedures have demonstrated efficacy and safety in the hands of experienced clinicians. Although percutaneous neurolysis is usually performed to treat cancer pain, some investigators have approached noncancer pain with similar techniques.

*Corresponding author. E-mail address: [email protected] (C. Benedetti). 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.009

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

168

ADOLPH & BENEDETTI

PAIN THERAPY: CONTEXT AND PRACTICE IN GASTROENTEROLOGY Cancer patients experience pain in over 80% of cases [7]. A hierarchy of pain therapy was developed by the World Health Organization (WHO), and these principles apply to the majority of patients experiencing cancer pain [8]. Oral opioids are effective in 75% of such patients. However, fear and ignorance in prescribing opioids may serve as a barrier to adequate pain therapy [9]. Additional parenteral opioids enable more patients to achieve relief, but a subset of patients may require procedural interventions to control pain. In addition, unwanted opioid side effects may limit the ability to safely titrate opioids to achieve pain control. For this reason, Bonica and Benedetti [10] proposed a five-step model for pain control, with initial therapy based on the severity of pain on presentation (Fig. 1). Given these complexities, pain therapy is best delivered in a multidisciplinary approach [11]. Acute Versus Chronic Pain The gastroenterologist may benefit from a conceptual model of acute (physiologic) pain versus chronic (pathologic) pain, given the prevalence of both entities in the clinical practice of gastroenterology [5,12,13]. In addition, coexisting

Fig. 1. A stepwise approach to escalating cancer pain therapy, which emphasizes that severe pain at onset may warrant omitting prior steps and proceeding directly to the treatment that will alleviate the patient’s pain. CNS, central nervous system. (From Bonica JJ, Benedetti C. Management of cancer pain. In: Moosa AR, Robson MC, Schimpff SC, editors. Comprehensive textbook of oncology. Baltimore (MD): Williams & Wilkins; 1986. p. 462; with permission.)

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

169

acute and chronic pain create a challenging clinical problem; for example, when considering an opioid-dependent patient who has failed back syndrome and is undergoing a laparotomy for colon cancer, what underlying mechanisms make pain therapy difficult? Sensory nerve fibers have specific terminal peripheral branches that, under normal conditions, are activated by high intensity, potentially harmful stimuli. When the stimulus ceases, normal nociceptors return to a steady state, with a normal activation threshold. These nociceptors are thinly myelinated (A delta) and unmyelinated C afferent nerve fibers that respond to chemicals released from damaged tissue (eg, hydrogen ions, potassium ions, or bradykinins) [14], mechanical compression or traction [15], or thermal stimuli [16]. Although ascending and descending spinal cord pathways modulate neural inputs, the sensation of pain is ultimately a cortical experience. Impulses transmitted in the nociceptive pathways, before reaching or activating the sensory cortex, do not elicit the sensation of pain but generate nociceptive reflexes such as tachycardia, hypertension, and hyperventilation. These responses are mediated mostly by the release of catecholamines. The brain also permits the behavioral components to pain sensation. The major descending pathway for modulating pain sensation starts in the periaqueductal gray area and projects to the spinal cord laminae. The targets of opioid analgesics are opioid receptors in the brain, spinal cord, and possibly at peripheral sites. The most potent analgesic effect in humans from the smallest amount of administered opioids has been infused through an intracerebroventricular catheter providing pain relief from advanced cancers [17]. In contrast, clinical chronic pain represents a nervous system gone awry. The abnormal nociceptive system develops in progressive stages. When it is subjected to repetitive injury, to a series of strong stimuli, or to an inflammatory process, the activation threshold of nociceptive fibers decreases significantly [18]. Subsequently, the nociceptor may generate an action potential when it is exposed to stimuli of lower intensities; therefore, innocuous stimuli may generate a sensation of pain. Neural adaptation, however, is not limited to the periphery. The spinal cord and brain respond to chronically painful impulses by altering their chemical, electrical, or microanatomic [19] structure and function; a phenomenon known as ‘‘neuroplasticity’’ [20]. On a long-term basis, when a barrage of impulses from nociceptive fibers reaches the spinal cord, ascending amplification of central nervous system nociceptive pathways occurs in the spinal cord. In animals, the brain’s somatosensory cortex representing the forepaw expands on exposure to avoidance behavior imposed by forepaw shock for 8 minutes per day [21]. Recent studies also demonstrate a neuroplastic response of the central nervous system to visceral stimuli [22]. Silverman and associates [23] documented patterns of brain activation in patients who had irritable bowel syndrome and showed that these patterns are different from those of healthy volunteers [23–25].

170

ADOLPH & BENEDETTI

Another pathologic entity in chronic pain syndromes is an abnormal neural connection that creates ‘‘cross talk’’ between damaged efferent and afferent nerves [26]. Cross talk between somatic and visceral nerves has been documented in the rat model [27] and may occur in humans, for example, when fibromyalgia syndrome is associated with visceral abdominal pain [28]. With inflammatory insults or infiltrating tumor, a ‘‘short circuit’’ may occur in a damaged segment of efferent myelinated nerve, where the integrity of the myelin sheath is breached. Impulses from damaged efferent nerves may ‘‘cross over’’ to normal unmyelinated afferent nociceptor fibers, returning to the central nervous system as a painful sensation. Therefore, isolated nerve injuries, such as those caused by tumor invasion or radiation, can serve as a substrate for intense pain. Also, somatic and autonomic symptoms and signs may overlap, creating a confusing clinical picture. An extreme example of this phenomenon is causalgia, or reflex sympathetic dystrophy. Are there useful physical signs to substantiate the diagnosis of chronic pain? Painful impulses may be referred to healthy tissue at the periphery of injured, invaded, or irradiated tissue. As afferent nociceptive impulses repeatedly stimulate the spinal cord, segmental interneurons, through neurotransmitters such as substance P [29], amplify areas of hypersensitivity beyond the injured tissue itself (Fig. 2). Impulses are converted so that even stimuli that are not usually

Fig. 2. Impulses transmitted by spinothalamic tract cells and the neural communication circuits between first- and second-order neurons in different segments of the spinal cord. When central neurons become sensitized, the patient may experience pain during stimulation of normal tissue near the original area of injury, cancer, or inflammation. (From Willis WD. Hyperalgesia and allodynia: summary and overview. In: Willis WD, editor. Hyperalgesia and allodynia: the Bristol-Myers Squibb Symposium on Pain Research. New York: Raven Press; 1992. p. 5; with permission.)

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

171

painful now are experienced as painful. In this situation, a normally nonpainful stimulus such a light touch may cause intense pain and grimacing (allodynia). A minimally painful stimulus such as a gentle pinch of the skin in entirely normal tissue may yield an exaggerated painful response (hyperalgesia) (Fig. 3) [30]. Despite the patient’s report of intense pain, there may be a complete absence of systemic signs that are manifested typically during acutely painful conditions such as tachycardia, hypertension, diaphoresis, or avoidance behaviors. In fact, at the bedside of a patient suffering from chronic pain, it may appear as though the patient is faking, malingering, or drug seeking [31]. Recent studies of visceral pain have documented similar phenomena, although the perception and psychologic processing of visceral pain may be different from the perception of somatic pain [32,33]. In chronic pain patients, until pain is relieved with adequate therapy, drugseeking behavior may reflect ‘‘pseudoaddiction’’ [34] and not true addiction. The term pseudoaddiction refers to the iatrogenic syndrome of abnormal behavior that develops as a direct consequence of inadequate pain management. The natural history of pseudoaddiction is a progression through three characteristic stages: (1) inadequate prescription of analgesics to counteract the primary pain stimulus; (2) escalation of analgesic demands by the patient, associated with behavioral changes to convince others of the pain’s severity; and (3) a crisis of mistrust between the patient and caregivers. Treatment strategies include empathy and improving communication, establishing trust, and providing appropriate and timely analgesics to control the patient’s level of pain. The sin qua non of pseudoaddiction is a therapeutic endpoint: if pseudoaddiction is present, adverse behaviors cease when adequate pain therapy

Fig. 3. How injury alters pain sensation. The curve on the right represents a normal relationship between stimulus intensity and pain sensation in uninjured tissue. Pain is sensed with noxious stimuli. The vertical line signifies the pain threshold. The curve on the left shows an abnormal relationship in injured tissue, where innocuous stimuli evoke pain sensation (allodynia), and minimally noxious stimuli evoke an excessive pain sensation (hyperalgesia). (Adapted from Cervero F, Laird JM. Mechanisms of touch-evoked pain (allodynia): a new model. Pain 1996;68:14; with permission.)

172

ADOLPH & BENEDETTI

is established. Otherwise, pseudoaddiction may be indistinguishable from true addiction behaviors. Although the evidence for these phenomena is best described as a consequence of injury in somatic nerves, recent studies suggest that similar mechanisms in visceral nerves underlie painful pathology of the gastrointestinal tract [35]. In fact, some investigators propose that defective neural regulation of inflammation itself may contribute to gastrointestinal inflammatory disease [36]. Assessment and Clinical Progress Notes In clinical practice, the gastroenterologist and patient will benefit mutually from establishing a baseline pain assessment and follow-up pain assessments to track the patient’s course. Pain is entirely subjective and inherently dependent on the patient’s report, particularly in the case of chronic pain [37], in which physical examination findings may be limited or absent. A 0-to-10 scale is a validated useful measure to document the level of pain at each clinical visit in which pain is a complaint (Fig. 4) [38]. A simple question such as, ‘‘How would you rate your pain now if 0 means ‘no pain’ and 10 means the ‘worst pain imaginable’?,’’ is a reasonable query. Some patients find a mild-moderatesevere scale more understandable (see Fig. 4). With this simple measure, the clinician will be able to follow the patient’s progress and the patient’s response to therapeutic measures. A pain rating is also useful when communicating to the referring physician. Studies demonstrate that, in the patient’s perception, the 0-to-10 progression in pain is not linear but exponential [39]. Also, keep in mind that it is common for patients to complain of pain as the first sign of malignancy that may precede findings on physical examination or radiologic images by weeks or months [40,41].

Fig. 4. Example of a pain assessment tool used by the present authors in each day’s clinical progress note. A tolerable level of pain is the goal, by subjective report of the patient. The grid also includes a tool to monitor side effects of opioids.

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

173

MECHANISM OF PAIN, SYMPTOMS, AND SIGNS Ascribing a mechanism to pain is helpful in formulating a pain treatment plan. The mechanism of pain can be confirmed by history and physical examination findings. Somatic pain is well localized. For example, chronic abdominal wall pain may be caused by a musculoskeletal source, such as repetitive trauma and muscle spasm related to an underlying ventral hernia defect [42]. Pain in somatic tissues is related to the activation of nociceptor fibers by mechanical or chemical tissue damage and inflammatory cascades. The pain may be exacerbated by activity and musculoskeletal function. With visceral pain, for example, localized Crohn’s disease without parietal peritonitis, the chief quality is a vague distribution. There are two types of pain receptors in visceral organs [32]. Hollow visceral organ distension reproducibly produces pain. The location of a pain by general visceral afferent nerves will broadly parallel the embryonic foregut (epigastrium), midgut (periumbilical region), and hindgut (lower abdominal quadrants). Visceral pain may also manifest as referred pain to a distant site, for instance hepatic metastases may cause right shoulder pain. Some solid organs, such as the lungs, kidneys, and liver, are not sensitive to tumor until capsular stretch, invasion, or spread to nearby organs occurs. Visceral pain may be associated with an autonomic response such as bradycardia, nausea, or vomiting. An example of neuropathic pain occurs with retroperitoneal lymph node metastasis from testicular cancer or pancreatic cancer. When the retroperitoneum is involved directly by tumor or indirectly by nodal metastasis, back pain will be a prominent feature, and patients may prefer sleeping in chair or laying in a fetal position. Back pain may be caused by retroperitoneal nodal nerve invasion or nerve traction and is found in 21% of patients who have periarotic lymph node metastasis related to testicular carcinoma [43]. Advanced cancers may present with mixed pain mechanisms, for example, in a patient with pancreatic carcinoma. The invasion of surrounding visceral organs occurs late in tumors originating in the body or tail of the pancreas, and up to 87% of these patients have pain. Cancer in the head of the pancreas will be associated with pain in 70% of cases. At the time of diagnosis, 40% of pancreatic cancer patients have peripancreatic spread and retroperitoneal lymph node involvement [44]. In addition, over 50% of cases have metastasis to visceral organs such as liver and peritoneum. Malignant ascites is another source of visceral pain, somatic pain, or referred shoulder pain because the peritoneum, abdominal wall, and diaphragm are stretched by tense fluid accumulation. The location of the pain in left, central, right epigastrium or mid back sensed by the patient is attributed to the complex array of visceral nerves supplying the head, body, and tail of the pancreas [45]. Pain may also be related to surgery, bowel obstruction, or complications of surgical diversion performed on the biliary or gastrointestinal tract. Carcinomatosis creates pain by inflammatory peritonitis related to diffuse tumor nodules. Hepatic metastases bulky enough to stretch Glisson’s capsule of the liver produces a dull and aching pain located in the right subcostal, right mid back, or flank regions.

174

ADOLPH & BENEDETTI

Treatments themselves may cause acute or chronic pain. Chemotherapy with vinca-derived alkaloids causes dysesthesia, hyperesthesia, or paresthesia in up to 100% of patients [46]. Following abdominopelvic radiotherapy, up to 50% of patients may experience acute or chronic radiation proctocolitis with rectal pain, bleeding, tenesmus, and diarrhea [47]. Pain syndromes are often associated with other significant symptoms such as anorexia and dyspnea, which complicate medical management [48]. Patients eventually dying from their illnesses do so because of starvation and dehydration, particularly when the gastrointestinal tract is the source of neoplasia. Before these events, however, patients may live weeks, months, or years with symptoms and the sequelae of incurable disease or its treatments. PAIN THERAPY PLAN At the first visit, establishing realistic goals with the patient will help to meet the patient’s needs and expectations. Reducing pain to a zero level may not be feasible for most patients. However, designing a therapy plan to make the patient’s pain tolerable is a reasonable goal. Frequent reevaluation and modification of therapeutic measures are important features to alleviate a patient’s suffering. Each patient has unique pharmacokinetic, disease-specific, psychologic, and situational characteristics that create a dynamic therapeutic relationship. When nerves conveying painful stimuli to the brain serve a region of involved anatomy, the anesthesiologist turns to ‘‘regional anesthesia’’ techniques. Regional anesthesia refers to interventions that limit nerve transmission from a specific anatomic region. Nerve transmission can be altered temporarily with an injection of local anesthetics, opioids, or other agents or by applying radiofrequency pulsation, thereby leaving intact the neurilemma around individual nerves and assuring that painful transmissions will eventually return. Nerves can also be stimulated electrically to alter central nervous system sensation of painful stimuli. Alternatively, nerve cells can be destroyed by neurolytic techniques such as phenol or ethanol injections, thermal application, radiofrequency ablation, or surgical division of nerves, thereby permanently preventing the affected nerves from further pain transmission. Although there are data to demonstrate the efficacy and safety of percutaneous-guided pain therapy techniques, controlled trials are limited in describing the best routes, injectable agents, and localization techniques. In addition to safety and efficacy, important endpoints in any pain relief study should also include the intensity and duration of pain relief, quality of life, and survival. Basic precautions in performing nerve blocks help to avoid complications, such as identifying anatomic landmarks, performing serial aspiration to prevent intravascular infusion, administering a test dose of local anesthetic, and observing for paresthesia [49]. The targets of percutaneous-guided pain control are nerves, specifically, nerves conveying painful sensations (nociception) to the brain from damaged visceral, somatic, or neural tissue (or a mixed combination of these substrates). Normal or pathologically involved nerves (neuropathic mechanism) may

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

175

convey painful sensations to the central nervous system. The neuropathic component of pain syndromes may be caused by compressed (eg, retroperitoneal lymphadenopathy), invaded (eg, cancer), tractioned (eg, perineural invasion), irradiated, or chemotherapeutically damaged (eg, cisplatin) nerves. One would anticipate that pain syndromes with a component of neuropathic pain are most likely to benefit from a percutaneous approach. An epidural or intrathecal catheter is also a percutaneous technique, targeting the alteration or blockade of impulses through an important nerve: the spinal cord. Opioids delivered by the spinal route may be as low as 1% of the dose delivered systemically in intravenous administration, thereby limiting the likelihood of side effects promulgated by opioid receptors in the gastrointestinal tract, brain (sedation), and skin (pruritus). Combination agents such as an opioid plus a local anesthetic can further limit opioid side effects and exploit the benefits of a secondary agent [50]. CANCER PAIN Intractable cancer pain is complicated by a host of other conditions that complicate pain therapy. Cachexia and advanced illness tend to cause a narrower therapeutic window, leading to a higher likelihood of side effects from opioid use. The gastrointestinal route of oral analgesic delivery and absorption is affected by previous surgery, radiation, chemotherapy, xerostomia, mucositis, tumor bulk, bowel obstruction, gastrointestinal dysmotility, or ileus. Bulky or disseminated tumors tend to involve mixed pain mechanisms. Any neuropathic component to a pain complex is less likely to be responsive to opioids alone. Transdermal delivery of opioids is less predictable, given the lack of adipose tissue, as in a cachectic patient, to serve as a depot for fentanyl, or dehydration that serves to limit drug diffusion from the site of dermal absorption. On the other hand, a limited lifespan opens the door for neurolytic destruction of plexi that deliver painful stimuli to the central nervous system. Most cancer pain relief may be achieved with SAT, according to the World Health Organization analgesic ladder. The evaluation of clinical outcomes from managing cancer pain has yielded favorable evidence for opioid use. Opioid therapy improves the quality of life and relieves pain in up to 70% to 90% of patients who have cancer pain [51]. However, advanced cancers or complex cancer pain may require additional interventional techniques. In this clinical scenario, an attempt to escalate systemic opioids alone often provides inadequate relief or excessive opioid side effects such as sedation or myoclonus. In practice, there is a fourth and fifth analgesic step above the WHO three-step analgesic ladder that represents invasive techniques such as percutaneous regional analgesia (see Fig. 1). NONCANCER PAIN It may be difficult to view patients who experience nonmalignant pain through the same lens as cancer patients are viewed. For example, patients who have chronic pancreatitis, inflammatory bowel disease, and other painful

176

ADOLPH & BENEDETTI

gastrointestinal conditions do not fit into a ‘‘pain medicine’’ template designed for cancer pain. Portenoy [52] has described some significant differences in managing pain with opioids between cancer and noncancer patient populations. There are few controlled trials of opioid therapy for nonmalignant pain [53]. Neither analgesic tolerance nor physical dependence is a frequent clinical problem in the cancer pain population. Drug-seeking behaviors are rare in patients who have cancer who have no history of substance abuse. Although pain is considered ‘‘permanent,’’ it may recur after neurolytic block, weeks, months, or years after injection neurolysis. Nerve regeneration can occur as long as the nerve cell body is intact. Unfortunately, recurrent pain after neurolysis may be more severe than the original pain syndrome. Therefore, neurolysis is limited typically to patients who have a limited lifespan and patients who fail all other therapies yet still suffer from intractable pain. Investigators are applying targeted injection therapy using newer localization techniques in nonmalignant chronic pain syndromes such as those seen in chronic pancreatitis [54,55]. PROCEDURES Neurolytic Celiac Plexus Block The celiac plexus transmits nociceptive signals from the solid and hollow viscera of the upper abdomen. The distribution of organs includes stomach, small and large bowel to the splenic flexure, liver, pancreas, biliary tract, kidneys, adrenals, spleen, and omentum. The celiac plexus is anterolateral to the aorta, just inferior to the origin of the celiac artery (Fig. 5). A reticular nerve formation measuring approximately 3 by 4 cm in the retroperitoneum, the celiac plexus is located behind the stomach and pancreas, enveloping the anterolateral surface of the aortic wall. It extends from the T12 to L1 intervertebral disk, anterior to the diaphragmatic crura. There is significant anatomic variation in the size and number of celiac ganglia [56]. Given the density of important anatomic structures in the region, its location makes for a potentially hazardous needle stick. The prototypical percutaneous pain therapy procedure for pancreatic cancer pain is neurolytic celiac plexus block (NCPB). The blockade of afferent visceral and neuropathic pain impulses from the celiac plexus improves pain sensation and quality of life [57]. The technique was first described in 1914 for intraoperative anesthesia [58]. Pancreatic cancer may cause relentless upper abdominal and mid-back pain on the basis of visceral (foregut organ destruction) and neuropathic (involvement of retroperitoneal nerves) mechanisms. Autopsy studies of pancreatic cancer patients reveal invasion of peripancreatic nerves in up to 84% of cases [59]. Such cancers are typically advanced and unresectable. Although controlled trials of techniques and agents are limited and subject to methodological constraints, historical reports demonstrate safety and efficacy of NCPB. Cousins and Bridenbaugh [49] reviewed 31 studies of 1599 patients receiving 2750 celiac plexus blocks. The authors found that as many as 85% to 90% of patients achieved good to excellent relief. With an injection of 20

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

177

Fig. 5. Anatomic relationships of the celiac plexus and autonomic ganglia of the abdominopelvic cavity. The sympathetic chain terminates in the ganglion impar. The celiac plexus is a highly variable network of nerves in an area 4 cm wide by 3 cm long, anterior to the aorta and between the two adrenal glands. (From Bonica JJ, Buckley FP, Moricca G, et al. Neurolytic blockade and hypophysectomy. In: Bonica JJ, editor. The Management of pain. 2nd edition. Media (PA): Lea & Febiger; 1990. p. 2017; with permission.)

to 80 mL of either dilute phenol or alcohol, this approach involves the chemical destruction of the nerve or its ability to convey electrical impulses, thereby achieving a sustained analgesic effect until the time of death in 70% to 90% of patients. The suffering patient will typically note an improvement immediately or within the first 24 hours after injection. The efficacy and safety of

178

ADOLPH & BENEDETTI

NCPB was also supported in a meta-analysis of 24 studies [60]. However, only two studies were randomized, controlled trials (RCTs). The authors noted long-term benefit in up to 79% to 90% of patients who had upper abdominal pain most frequently related to pancreatic cancer. Approximately 6% to 8% of patients may require a second block procedure to establish pain control. However, some authors suggest the efficacy of NCPB has not been proven, given that pre- and post-NCPB pain assessment data are lacking in many studies [61]. Complications of injection are limited, despite the injection of a relatively large volume of neurotoxic agent; preventative measures tend to minimize their morbidity. Eisenberg and colleagues [60] found that back pain (96%), diarrhea (44%), and hypotension (38%) were the most common adverse effects. Hypotension and diarrhea are related to the decrease in sympathetic outflow relative to vagal tone. Most skilled clinicians anticipate these problems and administer fluids to counteract splanchnic vasodilatation, monitoring blood pressure and providing low-dose vasoconstrictors on a transient basis. Diarrhea typically is transient and is often welcomed in the face of chronic constipation caused by opioid use. Severe, life-threatening diarrhea is rare but has been reported and must be considered after NCPB. Although the risk of serious complications is low, the potential for major problems has relegated this procedure to cancer patients who have a limited lifespan. Serious complications occur in less than 2% of injections [60]. If symptoms arise after the procedure, the gastroenterologist may be the first physician contacted and should be aware of more serious uncommon pitfalls that may occur. To reduce the risk of cardiovascular complications and bleeding, coagulation studies should be performed and normalized before the procedure. Phenol has caused cardiotoxicity with arrhythmia and arrest after injection in at least one patient. Significant bleeding has not been a problem, despite puncture of the aorta by accident or by intention [62,63]. Major neurologic complications are uncommon after NCPB. Somatic back pain is related to the procedure itself and typically is self-limited. The patient should be advised that back pain may occur after the procedure. Large-volume chemical neurolytic agents also may diffuse inferiorly in the retroperitoneum, toward the lumbosacral plexus. Imaging localization and injection of contrast with a neurolytic agent may limit this complication (Fig. 6). Less common are reports of persistent lower extremity weakness and abdominal muscle weakness caused by spillage of neurolytic agent on somatic motor nerves. Localization of the celiac trunk was performed blindly until angiography was used in 1969 to identify the celiac artery [64]. Angiographic localization for NCPB, however, proved to be too cumbersome. The advent of CT permitted Moore and colleagues [65] to build on the transcural approach first proposed by Boas [66] in 1978. Retrocural, as opposed to transcural, chemical injection decreases the likelihood that the agent will diffuse toward the lumbar plexus and cause somatic nerve damage and probably requires a lesser volume of injectate to achieve the same effect caused by proximity. Ischia and associates [67]

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

179

Fig. 6. Neurolytic celiac plexus block with CT imaging permits direct injection (injection needle is the white line) anterior to the aorta, with a minimal volume of neurolytic agent (ethanol or phenol) mixed with contrast, thus avoiding injury to the kidney and surrounding retroperitoneal organs. The patient is in the prone position.

took this concept a step further and sought to achieve solely preaortic injection by completely traversing the aorta itself with a needle from the posterior approach. Hemorrhagic complications are uncommon (0.1%–0.5%), despite the use of 14- to 18-gauge needles. Currently, most NCPB procedures are performed with 20- to 22-gauge needles. The evolution of NCPB to include imaging localization may have served to reduce the risk of major complications [68]; however, there are no clear data that localization techniques have served to make the procedure safer and more effective. To minimize the likelihood of idiosyncratic serious complications during NCPB, radiologic localization has become the standard of care (Fig. 7). In the 1995 meta-analysis conducted by Eisenberg and colleagues [60], 68% of the patients underwent radiologic localization by plain radiography (34%), CT (28%), fluoroscopy (5%), and ultrasonography (1%). Which imaging technique is superior? The answer is unknown; no randomized controlled trials have directly compared radiologic imaging localization techniques and outcomes. Endoscopic ultrasonographically guided localization for celiac plexus neurolysis (EUS-CPN) in cancer patients was first reported by Wiersema and Wiersema in 1995 [69]. There is preliminary evidence that the EUS technique is superior to CT for celiac plexus localization and neurolysis in nonmalignant pancreatic pain [54] and is helpful in managing malignant cancer pain (see article by Levy and Wiersema elsewhere in this issue). An anterior approach warrants CT-guided imaging and may improve efficacy by injection concentrated anterior to the aorta. To minimize the risk of retroperitoneal abscess, preprocedural antibiotics are recommended. The placement of an indwelling catheter by imaging localization permits serial injections

180

ADOLPH & BENEDETTI

Fig. 7. Moore’s original CT localization technique demonstrating the distribution of largevolume neurolytic agent mixed with contrast in the retroperitoneal space. (A) Needle on the left is cephalad to the diaphragm. (B) Needle on the right is caudal to the diaphragm. (C) Spread of solution after injection of 30 mL of 50% alcohol with contrast through the left needle and 50 mL through the right needle. a, aorta; C, crus of diaphragm; K, kidneys; L-1, first lumbar vertebra; N, needle’s point; R, spread of radiopaque solution; v, vena cava. (From Moore DC. Intercostal nerve block and celiac plexus block for pain therapy. In: Benedetti C, Chapman CR, Moricca G, et al, editors. Advances in pain research and therapy. volume 7. New York: Raven Press; 1984. p. 309–29; with permission.)

to achieve efficacy. Case reports have not described clearly the benefits of an indwelling catheter for symptom control. What is the most appropriate timing for celiac neurolysis? If the rationale for NCPB is to maximize the patient’s quality-of-life, earlier may be better. Ischia and colleagues [70] propose that NCPB should be an early pain therapy intervention to improve the patient’s little remaining time. Earlier percutaneous

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

181

intervention is supported by the finding that patients who have unresectable pancreatic cancer have pain that is difficult to control by systemic pharmacotherapy. Four investigational groups claim that early NCPB improves a patient’s quality of life until death [71–74]. Given its neuropathic component, pancreatic cancer pain may be less responsive to opioid therapy alone. Problems with nonsteroidal anti-inflammatory drug use in systemically ill patients such as renal insufficiency, diabetes, coagulopathy, and bone marrow suppression may limit their use. Oral delivery of medications may not be feasible in patients who have vomiting, diarrhea, xerostomia, dysphagia, and ileus, particularly near the end of life. So far, no investigator has shown by RCT that NCPB increases survival or prevents pain until death. Mercandante and Arcuri [6] have noted an increase in pain during the final 4 weeks of life in patients dying of advanced pancreatic cancer. The authors postulate that cancer growth proceeds to areas not innervated by the celiac plexus, leading to an escalation in pain. The authors also followed nonpain symptoms in the natural history of their patient cohorts and found that multiple symptoms are observed frequently. These findings underscore the necessity of a total patient approach, not focusing solely on pain, to achieve maximum symptom control and quality of life for this patient population. Is percutaneous injection more effective than systemic opioid therapy? Three published trials have attempted to demonstrate differences in outcomes in patients who have pancreatic cancer treated with NCPB versus SAT. Smaller studies (10 patients in each treatment arm) have compared NCPB with systemic analgesic therapy [73,75,76]. These studies have found that the greatest benefit in pain control for pancreatic cancer occurs in the first 4 weeks after injection. Pain control may be sustained for up to 50 days after injection, thereby improving quality of life until the time of death. A reduction in total opioid consumption in injected patients also contributes to an improved quality of life by reducing the occurrence of opioid-related side effects. Mercadante [75] followed 22 patients who had pancreatic cancer and whose pain was treated with either systemic analgesic therapy alone or by NCPB. Patients survived 2 to 38 weeks after the intervention. The therapeutic effect of NCPB resolved after 4 to 5 weeks. Other investigators agree that the more advanced the tumor is, the more difficulty in achieving pain control [77]. These beneficial findings of percutaneous neurolysis (reduced opioid requirements, effective pain control, and sustained pain control) echoed the results demonstrated in open surgical neurolysis at the time of laparotomy, when unresectable pancreatic cancer was identified [72]. In Lillemoe and colleagues’ [72] 1993 surgical series, intraoperative chemical splanchnicectomy was superior to placebo for pain control in a double-blind study of 137 patients. If cancer pain was present at the time of the operation, patients who received chemical splanchnicectomy survived longer than the placebo-treated cohort did. This improvement in survival has not been duplicated to date in surgical or percutaneous neurolysis. However, in many studies, patients are not followed uniformly until they die.

182

ADOLPH & BENEDETTI

No study has demonstrated that NCPB will replace opioids. Available data suggest that opioid use may diminish with successful NCPB. In the present authors’ experience, NCPB is an adjunct in pain therapy for pancreatic cancer patients. Reducing overall opioid use, however, has beneficial effects that cannot be discounted. In keeping with the findings of other investigators, the present authors believe that the final 4 weeks of life for these patients is characterized by an escalation of pain requiring strong opioids, including morphine, hydromorphone, or methadone, delivered by an oral, intravenous, or a subcutaneous route. The benefits of early administration of NCPB are perhaps more theoretical than real but deserve the focus of future controlled trials. Lumbar Plexus Block for Lumbosacral Plexopathy A lumbosacral sympathetic block should be distinguished from a lumbosacral plexus block; the former affects the transmission of visceral pain impulses, whereas the latter affects somatic impulses. The plexus lies in the psoas muscle. It is associated with a lymphatic plexus of lymph nodes and connecting vessels adjacent to the distal aorta and the common iliac vessels. Pelvic tumors, for example, advanced colorectal carcinoma, sarcoma, cervical cancer, ovarian cancer, or prostate cancer, can precipitate a rapid decline in function caused by disabling pain and progressive lower extremity weakness. Initial hip and leg pain are typically radicular in nature, progressing eventually within weeks to months to numbness and weakness in the lower extremity. The lack of mobility precedes infections such as pneumonia as well as deep venous thrombosis of the lower extremity, often associated with painful, acquired lymphedema. Flank pain and lumbar back pain may occur if tumor extends superior to the pelvic rim. The combination of leg pain, edema, weakness, a rectal mass, and hydronephrosis implies an underlying lumbosacral plexopathy by cancer [78]. Radiation fibrosis causes neuropathy in the lumbosacral plexus after intracavitary implants or external beam radiation and is accompanied by pain in up to 10% of affected patients. Pain may occur between 1 and 30 years after radiation therapy [79]. Weakness, paresthesia, and numbness begin in the distal lower extremities, with a slow rate of progression. Plain films, MRI, or bone scanning may confirm the presence of radiation necrosis, which frequently coexists with symptoms and signs of pelvic radiation fibrosis. To perform a lumbar plexus block, one- and two-needle techniques have been described. Fluoroscopic guidance and contrast injection are used to localize and confirm good position, typically at the L2 to L3 level. The position of the needle before injection should be anterior to the psoas muscle fascia. Neurolysis is considered after local anesthetic blocks of the sympathetic chain are effective as a temporary test measure. A characteristic contrast spread during injection is reproducible during both local anesthetic injection and neurolysis. Pelvic and Perineal Pain: Superior Hypogastric Block and Ganglion Impar Block The superior hypogastric plexus, an autonomic ganglion, is the continuation of the aortic plexus inferior to the aortic bifurcation across the fifth lumbar

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

183

vertebra. These nerves continue inferiorly into the pelvis where they divide into two hypogastric nerves on both sides of the rectum. These nerves join the pelvic splanchnic nerves to form the inferior hypogastric plexi, supplying the pelvic organs. The plexus itself lies anterior to the L5 vertebral body and contains postganglionic sympathetic fibers and afferent pain fibers. The plexus innervates the vagina, rectum, bladder, perineum, vulva, prostate, and uterus. Management options for severe pelvic pain include (1) superior hypogastric plexus (also referred to as the presacral nerve) block; (2) ganglion impar block; and (3) bilateral lumbar sympathetic block. Rectal and gynecologic cancers are prone to local recurrence and advanced local growth. Bladder spasms are involved frequently, and positional changes tend to exacerbate the pain. Tenesmus may be a prominent feature. Pain that radiates to the groin is characteristic of ureteral obstruction. The nature of local visceral pain is often complicated by impaired bowel or bladder function. Hip and lower extremity pain can escalate when somatic nerve involvement occurs. In the present authors’ experience, pelvic malignancies or those that lead to advanced lymphatic spread such as melanoma and lymphoma also cause pain related to severe acquired lymphedema or postphlebitic syndrome after deep venous thrombosis. The gastroenterologist may be consulted in these patients to evaluate pain, lower gastrointestinal symptoms, or gastrointestinal bleeding. Superior hypogastric plexus block is effective at reducing pain scores in 70% of patients experiencing malignant pelvic pain [80]. In addition, average daily opioid use was shown to drop by 67% in patients who underwent a successful block. It is a safe procedure, and significant neurologic complications are rare. The ganglion impar innervates the perineum. This ganglion is located anterior to the sacrococcygeal junction and represents the endpoint of the left and right sympathetic chains (Fig. 8). Cancers of the cervix, colon, bladder, rectum, and endometrium are the most common types of malignancy causing pain in this area. A diagnosis of intractable sympathetic pain will predict a response to this intervention. This type of pain is associated with urgency and burning and is vague and poorly localized, but it presents in the perineum. Pain may also be referred to the rectum, perineum, or vagina. In one study [80] of ganglion impar neurolytic block, 8 of 16 patients achieved 100% relief of pain, whereas the remaining patients achieved 60% or better relief, according to visual analog scores. A safe approach to the ganglion impar traverses the sacrococcygeal junction to avoid the underlying rectum. The injection of contrast agent under fluoroscopy confirms proper needle placement. The technique can be performed with local anesthetic and steroid for nonmalignant painful conditions such as coccygodynia or perirectal pain from tumor involvement. Neurolytic blockade may be performed for malignant pain conditions and only after a diagnostic block is performed with local anesthetic. The advantage of the ganglion impar block over other neurolytic procedures for rectal pain is that bowel and bladder function generally is unaffected.

184

ADOLPH & BENEDETTI

Fig. 8. Lateral view of sacrococcygeal anatomy and neurolytic injection for ganglion impar block. (From Rauck RL, de Leon-Casasola OA. Interventional techniques. In: Raj PP, editor. Pain medicine: a comprehensive review. 2nd edition. St. Louis (MO): Mosby; 2003. p. 269; with permission.)

Spinal (Epidural and Intrathecal) Analgesia The spine and brain are the primary targets of opioid analgesia. The delivery of analgesic or anesthetic medication directly into or near the spinal cord requires low dosages of drug and minimizes systemic side effects. Intraspinal pain therapy is accomplished by a catheter placed in the epidural space or into the intrathecal (subarachnoid) space [50]. Intrathecal catheters may be tunneled subcutaneously, exiting the skin or connected to a subcutaneous port. The catheter may also be attached to an external pump or to a subcutaneous implantable pump. Indications include malignant pain refractory to strong systemic opioids, pain in conjunction with excess opioid side effects, or failed percutaneous sympathetic plexus blocks. Recent advances in implantable pump devices and techniques have expanded intraspinal analgesics to patients who have chronic nonmalignant pain. Surveys of experienced pain medicine practitioners reveal that morphine is the most popular intraspinal agent, with 62% of patients receiving morphine alone. The remaining 38% of patients are switched to other agents or combination agents because of a lack of efficacy in 80% and untoward side effects in the other 20% [81]. Patients who have neuropathic pain only or mixed pain syndromes tend to use higher doses of morphine than other patients [82]. There are significant variations in how clinicians use intraspinal catheters. Guidelines established by a consensus panel have focused largely on agents and not indications [81]. The most common complications of intraspinal analgesia are skin site or catheter infections.

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

185

An intrathecal catheter does not require an implantable subcutaneous pump. For severe pain related to malignancy, the present authors usually place an intrathecal catheter, tunnel it beneath the skin to the lateral abdomen, and exit the skin. The catheter is attached to a CAD pump placed in a waist belt. The patient may use the device as a continuous intrathecal infusion, a patient-controlled analgesia apparatus, or a combination of both approaches [50]. Intraspinal catheters for pain therapy appear to improve the quality of life for the patients who require them and may permit a patient who was previously incapacitated by excruciating pain to go home with a tolerable pain level. The presence of infection, epidural or paraspinal metastasis, immunosuppression, anticoagulation, and availability of home care resources may affect the decision whether to place an intrathecal catheter. SUMMARY The gastroenterologist deals frequently with painful conditions and suffering patients. Performing regular pain assessments and applying basic pain medicine principles will augment the care of patients in pain. Percutaneous-guided pain therapy techniques play a role in the multidisciplinary approach to pain medicine. Systemic opioid analgesia is the primary means of controlling cancer pain. However, 10% to 15% of cancer patients may need additional interventions to control pain. Sympathetic ganglion nerve blocks with neurolytic agents such as alcohol or phenol are reserved mostly for cancer pain. The efficacy and safety of these tools are validated by several decades of clinical application and published studies. Although the procedures are operator-dependent, in the hands of experienced clinicians, patients achieve sustained relief in the majority of cases. Although these techniques have been attempted in some benign conditions, such as chronic pancreatitis, with limited success, studies of newer imaging localization techniques such as endoscopic ultrasonography may expand future indications. Patients of the gastroenterologist who experience malignant abdominal pain may benefit from referral for percutaneous-guided pain control techniques. References [1] Portenoy RK, Lesage P. Management of cancer pain. Lancet 1999;353:1695–700. [2] Reidenberg MM, Portenoy RK. The need for an open mind about the treatment of chronic nonmalignant pain. Clin Pharmacol Ther 1994;55:367–9. [3] Saunders C. Spiritual pain. J Palliat Care 1988;4:29–32. [4] Clearfield HR. Consultant strategies for the gastroenterologist. Am J Gastroenterol 1999;94:1453–6. [5] Switz DM. What the gastroenterologist does all day: a survey of a state society’s practice. Gastroenterology 1976;70:1048–50. [6] Mercadante S, Arcuri E. Sympathetic blocks and disease progression modifying pain mechanisms. Reg Anesth Pain Med 2003;28:586–7. [7] Melzack R. The tragedy of needless pain. Sci Am 1990;262:27–33. [8] Stjernsward J, Colleau SM, Ventafridda V. The World Health Organization Cancer Pain and Palliative Care Program: past, present, and future. J Pain Symptom Manage 1996;12: 65–72.

186

ADOLPH & BENEDETTI

[9] Scanlon MN, Chugh U. Exploring physicians’ comfort level with opioids for chronic noncancer pain. Pain Res Manag 2004;9:195–201. [10] Bonica JJ, Benedetti C. Management of cancer pain. In: Moosa AR, Robson MC, Schimpff SC, editors. Comprehensive textbook of oncology. Baltimore (MD): Williams & Wilkins; 1986. [11] Flor H, Fydrich T, Turk DC. Efficacy of multidisciplinary pain treatment centers: a meta-analytic review. Pain 1992;49:221–30. [12] Devine R, Merskey H. The description of pain in psychiatric and general medical patients. J Psychosom Res 1965;9:311–6. [13] Mekhjian HS, Switz DM, Melnyk CS, et al. Clinical features and natural history of Crohn’s disease. Gastroenterology 1979;77:898–906. [14] McHugh JM, McHugh WB. Pain: neuroanatomy, chemical mediators, and clinical implications. AACN Clin Issues 2000;11:168–78. [15] Hallin RG, Torebjork HE. Electrically induced A and C fibre responses in intact human skin nerves. Exp Brain Res 1973;16:309–20. [16] Fitzgerald M, Lynn B. The sensitization of high threshold mechanoreceptors with myelinated axons by repeated heating. J Physiol 1977;265:549–63. [17] Lobato RD, Madrid JL, Fatela LV, et al. Intraventricular morphine for control of pain in terminal cancer patients. J Neurosurg 1983;59:627–33. [18] Swamydas M, Skoff AM, Adler JE. Partial sciatic nerve transection causes redistribution of pain-related peptides and lowers withdrawal threshold. Exp Neurol 2004;188:444–51. [19] Li Z, Okamoto K, Hayashi Y, et al. The importance of dendritic mitochondria in the morphogenesis and plasticity of spines and synapses. Cell 2004;119:873–87. [20] Zhuo M. Central plasticity in pathological pain. Novartis Found Symp 2004;261:132–45 [discussion: 145–54]. [21] Spinelli DN, Jensen FE. Plasticity: the mirror of experience. Science 1979;203:75–8. [22] Sato J, Perl ER. Adrenergic excitation of cutaneous pain receptors induced by peripheral nerve injury. Science 1991;251:1608–10. [23] Silverman DH, Munakata JA, Ennes H, et al. Regional cerebral activity in normal and pathological perception of visceral pain. Gastroenterology 1997;112:64–72. [24] Han JS, Neugebauer V. Synaptic plasticity in the amygdala in a visceral pain model in rats. Neurosci Lett 2004;361:254–7. [25] Sarkar S, Hobson AR, Furlong PL, et al. Central neural mechanisms mediating human visceral hypersensitivity. Am J Physiol Gastrointest Liver Physiol 2001;281:G1196–202. [26] Pezzone MA, Liang R, Fraser MO. A model of neural cross-talk and irritation in the pelvis: implications for the overlap of chronic pelvic pain disorders. Gastroenterology 2005;128: 1953–64. [27] Miranda A, Peles S, Rudolph C, et al. Altered visceral sensation in response to somatic pain in the rat. Gastroenterology 2004;126:1082–9. [28] Chang L, Berman S, Mayer EA, et al. Brain responses to visceral and somatic stimuli in patients with irritable bowel syndrome with and without fibromyalgia. Am J Gastroenterol 2003;98:1354–61. [29] Li P, Zhuo M. Substance P and neurokinin A mediate sensory synaptic transmission in young rat dorsal horn neurons. Brain Res Bull 2001;55:521–31. [30] Cervero F, Laird JM. Mechanisms of touch-evoked pain (allodynia): a new model. Pain 1996;68:13–23. [31] Porter-Williamson K, Heffernan E, Von Gunten CF. Pseudoaddiction. J Palliat Med 2003;6: 937–9. [32] Cervero F, Laird JM. Understanding the signaling and transmission of visceral nociceptive events. J Neurobiol 2004;61:45–54. [33] Cervero F. Visceral hyperalgesia revisited. Lancet 2000;356:1127–8. [34] Weissman DE. Understanding pseudoaddiction. J Pain Symptom Manage 1994;9:74.

PERCUTANEOUS-GUIDED PAIN CONTROL: NEURAL BASIS OF PAIN

187

[35] Bernstein CN, Niazi N, Robert M, et al. Rectal afferent function in patients with inflammatory and functional intestinal disorders. Pain 1996;66:151–61. [36] Anisman H, Baines MG, Berczi I, et al. Neuroimmune mechanisms in health and disease: 2. disease. CMAJ 1996;155:1075–82. [37] Grossman SA, Sheidler VR, Swedeen K, et al. Correlation of patient and caregiver ratings of cancer pain. J Pain Symptom Manage 1991;6:53–7. [38] Daut RL, Cleeland CS, Flanery RC. Development of the Wisconsin Brief Pain Questionnaire to assess pain in cancer and other diseases. Pain 1983;17:197–210. [39] Myles PS, Troedel S, Boquest M, et al. The pain visual analog scale: is it linear or nonlinear? Anesth Analg 1999;89:1517–20. [40] Kori SH, Foley KM, Posner JB. Brachial plexus lesions in patients with cancer: 100 cases. Neurology 1981;31:45–50. [41] Portenoy RK, Kornblith AB, Wong G, et al. Pain in ovarian cancer patients. Prevalence, characteristics, and associated symptoms. Cancer 1994;74:907–15. [42] Srinivasan R, Greenbaum DS. Chronic abdominal wall pain: a frequently overlooked problem. Practical approach to diagnosis and management. Am J Gastroenterol 2002;97: 824–30. [43] Cantwell BM, Mannix KA, Harris AL. Back pain: a presentation of metastatic testicular germ cell tumours. Lancet 1987;1:262–4. [44] Kelsen DP, Portenoy RK, Thaler HT, et al. Pain and depression in patients with newly diagnosed pancreas cancer. J Clin Oncol 1995;13:748–55. [45] Bliss WR, Burch B, Martin MM, et al. Localization of referred pancreatic pain induced by electric stimulation. Gastroenterology 1950;16:317–23. [46] Mollman JE. Neuromuscular toxicity of therapy. Curr Opin Oncol 1992;4:540–6. [47] Denham JW, O’Brien PC, Dunstan RH, et al. Is there more than one late radiation proctitis syndrome? Radiother Oncol 1999;51:43–53. [48] Donnelly S, Walsh D. The symptoms of advanced cancer. Semin Oncol 1995;22: 67–72. [49] Cousins MJ, Bridenbaugh PO, editors. Neural blockade in clinical anesthesia and management of pain. 2nd edition. Philadelphia: Lippincott; 1988. [50] Benedetti C. Intraspinal analgesia: an historical overview. Acta Anaesthesiol Scand Suppl 1987;85:17–24. [51] Portenoy RK. Cancer pain: epidemiology and syndromes. Cancer 1989;63:2298–307. [52] Portenoy RK. Opioid therapy for chronic nonmalignant pain: a review of the critical issues. J Pain Symptom Manage 1996;11:203–17. [53] Moulin DE, Iezzi A, Amireh R, et al. Randomised trial of oral morphine for chronic noncancer pain. Lancet 1996;347:143–7. [54] Gress F, Schmitt C, Sherman S, et al. Endoscopic ultrasound-guided celiac plexus block for managing abdominal pain associated with chronic pancreatitis: a prospective single center experience. Am J Gastroenterol 2001;96:409–16. [55] Wong GY, Sakorafas GH, Tsiotos GG, et al. Palliation of pain in chronic pancreatitis: use of neural blocks and neurotomy. Surg Clin North Am 1999;79:873–93. [56] Ward EM, Rorie DK, Nauss LA, et al. The celiac ganglia in man: normal anatomic variations. Anesth Analg 1979;58:461–5. [57] Wong GY, Schroeder DR, Carns PE, et al. Effect of neurolytic celiac plexus block on pain relief, quality of life, and survival in patients with unresectable pancreatic cancer: a randomized controlled trial. JAMA 2004;291:1092–9. [58] Patt RB, Cousins MJ. Techniques for neurolytic neural blockade. In: Cousins MJ, Bridenbaugh PO, editors. Neural blockade in clinical anesthesia and management of pain. 3rd edition. Philadelphia (PA): Lippencott; 1998. [59] Nagakawa T, Mori K, Nakano T, et al. Perineural invasion of carcinoma of the pancreas and biliary tract. Br J Surg 1993;80:619–21.

188

ADOLPH & BENEDETTI

[60] Eisenberg E, Carr DB, Chalmers TC. Neurolytic celiac plexus block for treatment of cancer pain: a meta-analysis. Anesth Analg 1995;80:290–5. [61] Sharfman WH, Walsh TD. Has the analgesic efficacy of neurolytic celiac plexus block been demonstrated in pancreatic cancer pain? Pain 1990;41:267–71. [62] Gaudy JH, Tricot C, Sezeur A. [Serious heart rate disorders following perioperative splanchnic nerve phenol nerve block]. Can J Anaesth 1993;40:357–9 [in French]. [63] Hilgier M, Rykowski JJ. One needle transcrural celiac plexus block: single shot or continuous technique, or both. Reg Anesth 1994;19:277–83. [64] Jackson SH, Jacobs JB, Epstein RA. A radiographic approach to celiac plexus block. Anesthesiology 1969;31:373–5. [65] Moore DC, Bush WH, Burnett LL. Celiac plexus block: a roentgenographic, anatomic study of technique and spread of solution in patients and corpses. Anesth Analg 1981;60: 369–79. [66] Boas RA. Sympathetic blocks in clinical practice. Int Anesthesiol Clin 1978;16:149–82. [67] Ischia S, Luzzani A, Ischia A, et al. A new approach to the neurolytic block of the coeliac plexus: the transaortic technique. Pain 1983;16:333–41. [68] Moore DC. The dreaded complications from neurolytic celiac plexus blocks are preventable! Reg Anesth Pain Med 2004;29:377–8. [69] Wiersema MJ, Wiersema LM. Endosonography-guided celiac plexus neurolysis. Gastrointest Endosc 1996;44:656–62. [70] Ischia S, Polati E, Finco G, et al. 1998 Labat lecture: the role of the neurolytic celiac plexus block in pancreatic cancer pain management: do we have the answers? Reg Anesth Pain Med 1998;23:611–4. [71] Ischia S, Ischia A, Polati E, et al. Three posterior percutaneous celiac plexus block techniques: a prospective, randomized study in 61 patients with pancreatic cancer pain. Anesthesiology 1992;76:534–40. [72] Lillemoe KD, Cameron JL, Kaufman HS, et al. Chemical splanchnicectomy in patients with unresectable pancreatic cancer: a prospective randomized trial. Ann Surg 1993;217: 447–55. [discussion: 456–7]. [73] Polati E, Finco G, Gottin L, et al. Prospective randomized double-blind trial of neurolytic coeliac plexus block in patients with pancreatic cancer. Br J Surg 1998;85:199–201. [74] Patt RB, Reddy SK, Black RG. Neural blockade for abdominopelvic pain of oncologic origin. Int Anesthesiol Clin 1998;36:87–104. [75] Mercadante S. Celiac plexus block versus analgesics in pancreatic cancer pain. Pain 1993;52:187–92. [76] Kawamata M, Ishitani K, Ishikawa K, et al. Comparison between celiac plexus block and morphine treatment on quality of life in patients with pancreatic cancer pain. Pain 1996;64:597–602. [77] Vranken JH, Zuurmond WW, Van Kemenade FJ, et al. Neurohistopathologic findings after a neurolytic celiac plexus block with alcohol in patients with pancreatic cancer pain. Acta Anaesthesiol Scand 2002;46:827–30. [78] Jaeckle KA, Young DF, Foley KM. The natural history of lumbosacral plexopathy in cancer. Neurology 1985;35:8–15. [79] Elliott K, Foley KM. Neurologic pain syndromes in patients with cancer. Neurol Clin 1989;7:333–60. [80] Plancarte R, Amescua C, Patt RB, et al. Superior hypogastric plexus block for pelvic cancer pain. Anesthesiology 1990;73:236–9. [81] Hassenbusch SJ, Portenoy RK. Current practices in intraspinal therapy: a survey of clinical trends and decision making. J Pain Symptom Manage 2000;20:S4–11. [82] Paice JA, Penn RD, Shott S. Intraspinal morphine for chronic pain: a retrospective, multicenter study. J Pain Symptom Manage 1996;11:71–80.

Gastroenterol Clin N Am 35 (2006) 189–199

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Palliation of Malignant Ascites Stefanie M. Rosenberg, PA-C Department of Radiology, Lutheran General Hospital, 1775 Dempster Avenue, Park Ridge, IL 60068, USA

L

arge-volume ascites accumulation is a common, problematic occurrence in patients with advanced malignancies and in patients with intraperitoneal spread of tumor. The mass effect of ascites accumulation can cause symptoms of painful abdominal distention, early satiety, and nausea. In extreme cases, vomiting secondary to the fluid causing external gastric or bowel compression and obstruction may occur. Additionally, patients often complain of shortness of breath, limited mobility, and lower extremity edema. These symptoms not only are distressing, but also adversely affect quality of life. PATHOPHYSIOLOGY The most common cause of ascites is cirrhosis (75% of patients). Ten percent of cases are due to malignancy, 3% are due to cardiac failure, 2% are due to tuberculosis, and the remainder are due to other causes [1]. The cause of ascites in cirrhotic patients has been well described [2,3]. Portal hypertension develops secondary to resistance at the splanchnic and arteriolar level, and the sinusoidal resistance is not relieved by the subsequent development of portosystemic collaterals. The ensuing development of portosystemic shunting and an increase in cardiac output leads to potential lethal complications, including renal failure, gastroesophageal varices, ascites, and hepatic encephalopathy. The cause of ascites development in oncology patients differs from patients with cirrhosis. In 50% of patients with malignancies, ascites development is secondary to invasion of the parietal or visceral peritoneum; 15% are due to liver invasion and portal venous compression, 15% are a combination of the first two, and the remaining 20% are attributed to chylous ascites secondary to lymphatic invasion [4]. Most patients with malignant ascites have epithelial carcinomas as the source. Only 20% of patients have malignancies of unknown origin; 80% are from the breast, ovaries, colon, endometrium, gastrointestinal tract, or pancreas [5].

The author is a consultant for Denver Biomedical, Denver, Colorado.

E-mail address: [email protected] 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.006

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

190

ROSENBERG

Differentiating between malignant and nonmalignant causes of ascites production traditionally has been done by sending fluid for cytology and laboratory chemistries to determine if it is exudative or transudative. The difference between the albumin concentration in serum and ascitic fluid is termed the serum–ascitic fluid albumin gradient (SAAG). If the SAAG is greater than 1.1 g/dL, it is termed a ‘‘high’’ SAAG, and if it is less than 1.1 g/dL, it is termed a ‘‘low’’ SAAG. A high SAAG occurs in portal hypertension, which may be related to either liver disease or cardiac disease. When the SAAG is greater than 1.1 g/dL, the concentration of total proteins in ascitic fluid helps to distinguish cardiac causes from hepatic causes. In cardiac disease, the total protein concentration in ascitic fluid is usually greater than 2.5 g/dL, whereas in liver disease, it is less than 2.5 g/dL. A low SAAG typically is associated with carcinomatosis, tuberculosis, pancreatic ascites, and nephritic syndrome. Fifteen percent of cirrhotic patients have low SAAG ascites, however, and 20% of oncology patients have high SAAG ascites [6–11]. Several other tests have been described to help differentiate between malignant and nonmalignant ascites. Ascitic fluid can be sent for lactate dehydrogenase and for cholesterol. An ascites lactate dehydrogenase level greater than 250 lg/mL and a cholesterol level greater than 70 mg/dL are associated with malignancy [6,12–14]. Oncology patients who have comorbidities of cirrhosis, liver invasion by tumor, congestive heart failure, or cardiac compromise secondary to chemotherapy need further evaluation of ascites to help differentiate the cause because medical management may be appropriate. Cirrhotic patients with symptomatic ascites who undergo frequent paracentesis for symptom relief typically are monitored closely for hypotension. Most of these patients undergo paracentesis at a slow rate with or without concomitant administration of plasma expanders, such as albumin, to maintain homeostasis. Lethal hypotensive episodes after paracentesis typically do not occur in oncology patients because of the different physiologic mechanism of ascites accumulation. Large-volume paracentesis of 5 L/d is recommended as safe [5], but this is a conservative recommendation, and larger volumes can be removed safely if necessary for patient comfort. MANAGEMENT OPTIONS Treatment options for malignant ascites should be developed with a goal of palliation of symptoms that is best suited for the patient. Most patients with advanced disease have a life expectancy of only a few weeks to a few months. Although breast and ovarian cancer patients often have longer survival times [15], palliation of symptoms still should be the goal. Medical therapies, such as diuretics, and sodium and fluid restriction are not effective in most oncology patients. The small percentage of patients with cirrhotic/portal compression as a cause of refractory ascites may find the restrictiveness of medical management unpalatable at the end of life. Available procedure-based therapies should be discussed openly, and the patient should be given the choice of medical versus nonmedical options. The following therapies are discussed in

PALLIATION OF MALIGNANT ASCITES

191

more detail: paracentesis, pigtail catheter (all-purpose drain) placement, tunneled catheter placement, port placement, and shunt creation/placement. Paracentesis Paracentesis is the most common and effective therapy used by physicians for effective palliation of ascites symptoms [16]. Drainage of large-volume ascites can be accomplished without ultrasound guidance in an outpatient or office setting. Many hospitals offer paracentesis in various departments, including the gastrointestinal laboratory, ultrasound unit, and interventional radiology suite. Using ultrasound guidance is important in patients with loculated pockets of fluid or in patients with postsurgical abdomens in which fluid is walled off. Paracentesis is a safe procedure. Rare complications that may occur include infection, bowel perforation, and hemorrhage. Paracentesis offers the advantages of a quick, simple, low-risk procedure with immediate symptom relief. Symptom relief is temporary, and as the patient’s disease progresses, requisite trips to the hospital for the procedure increase as well. Patients are left with the fatigue of frequent hospital visits or waiting as long as possible between procedures until the ascites symptoms are no longer tolerable and the needle-stick pain that occurs with each procedure. Pigtail Catheter Placement Pigtail drainage catheters are used for a wide variety of indications, including percutaneous abscess drainage [17], evacuation of pleural effusions [18], and percutaneous biliary and renal drainage procedures. For management of ascites, catheters typically are placed under ultrasound or fluoroscopic guidance. This can be done as an outpatient procedure or during hospital admission. After placement, the catheters can be capped and drained intermittently via gravity drainage bag or vacuum bottles. Alternatively, they can be attached immediately to gravity drainage bags for continuous drainage. Complications of pigtail catheter management for ascites occur in 35% of patients [19] and include peritonitis, accidental removal, leakage around the drain, and catheter occlusion. Good results have been obtained when pigtail catheters are used to treat symptomatic ascites associated with ovarian hyperstimulation syndrome [20]. Drains were placed in healthy young women for an average duration of less than 2 weeks. There has been speculation that infection rate may vary with continuous drainage as opposed to intermittent drainage. O’Neill and colleagues [21] attributed their four cases of peritonitis to the intermittent technique, whereas Richard and colleagues [22], Barnett and colleagues [23], and Rosenberg and colleagues [24] had no cases of peritonitis secondary to intermittent drainage technique. Infection rates in pigtail catheters are more likely secondary to the duration the catheter is in place, and these catheters should be used in patients with a life expectancy in weeks rather than months. Tunneled Catheter Placement Tunneled catheters have been used for vascular dialysis, apheresis, and peritoneal dialysis for many years. The catheters have Dacron cuffs that reside in the

192

ROSENBERG

subcuticular tunnel. The tissues heal and scar around the cuff, preventing a conduit for bacterial growth along the catheter. The catheters are designed to have one, two, or three cuffs. The third cuff is located near the tunnel exit site and impregnated with an antimicrobial agent. Dialysis Catheters Silastic peritoneal dialysis catheters now are being used effectively in management of malignant ascites [21,23,25,26]. The catheters are inserted into the largest area of ascites in an outpatient setting, usually by radiologists [27] under ultrasound or fluoroscopic guidance (Fig. 1). The catheters can be managed at home by the patient or primary caregiver via intermittent gravity drainage, intermittent vacuum bottle drainage, or continuous gravity drainage. Complications are similar to the complications seen with peritoneal dialysis catheters, including cellulitis, peritonitis, and catheter occlusion. Another

Fig. 1. (A and B) Dialysis catheters with two or three Dacron subcuticular cuffs. The superficial cuff closest to the exit site can be impregnated with antimicrobial chemicals.

PALLIATION OF MALIGNANT ASCITES

193

common problem is leakage of ascites around the catheter tunnel until the cuff heals in properly. This leakage can be minimized by keeping the ascites well drained and pressure off of the tunnel until the cuff is well healed in the tunnel. Pleurx Catheter The Pleurx (Denver Biomedical, Denver, Colorado) catheter is a single-cuff, tunneled, 16Fr Silastic catheter that is approved by the Food and Drug Administration (FDA) for the drainage of malignant pleural effusions. FDA approval for malignant ascites was obtained in November of 2005. It has a one-way rubber valve rather than a clamp device to close the system and a line attached to vacuum bottles to access the catheter (Fig. 2). The valve remains sealed even if the cap becomes dislodged, and the catheter is open only when the valve is accessed. Patients and caregivers can be taught to perform the drainage at home without nursing assistance. Placement of the Pleurx catheter is similar to that of the tunneled dialysis catheters. Problems with tunnel leakage have been resolved by puncturing the peritoneum superior and lateral to the umbilicus, then creating the subcuticular tunnel medially and superiorly (Fig. 3). In patients in whom fluid is located in the lower pelvis, a ‘‘C’’ curved tunnel can be created to help prevent fluid leakage (Fig. 4). Patients are instructed to perform drainage every day for the first 2 weeks to keep pressure off of the tunnel tract while it is healing. Patients are instructed to perform drainage subsequently as needed for comfort. Complications are similar to the complications associated with other tunneled catheters [22,24,28] and include infection, occlusion, and malpositioning.

Fig. 2. Pleurx system with access line to go through the one-way safety valve.

194

ROSENBERG

Fig. 3. Access to peritoneal cavity is obtained superior to umbilicus, and a tunnel is created further superior and medial to avoid leakage through tunnel.

Abdominal Port Placement Venous arm and chest ports for chemotherapy delivery have been used by oncologists for years. When venous access is obtained, a tunnel is made for the catheter portion of the port, and a subcuticular pocket is created for the port well to reside. To help prevent infection, the whole system is subcuticular, and access is achieved through the skin with a noncoring needle. Some centers now place these types of ports in the abdomen to drain malignant ascites. One published series of abdominal port placement in nine patients used either 6F or 8F vascular access ports [29]. A high complication rate was seen

Fig. 4. ‘‘C’’ curved tunnel for when catheter must be placed in the lower pelvis to help prevent tunnel leakage.

PALLIATION OF MALIGNANT ASCITES

195

with three cases of bacterial peritonitis, a case of catheter occlusion, and wound dehiscence. Additionally, ascites formation around the port made it difficult to palpate and prevented access. Other investigators addressed these issues [30,31] by using a port system designed for peritoneal access and by careful anatomic placement of the port. The port base they used was larger so that it could be palpated more easily, and it was attached to a 16F catheter with a Dacron cuff similar to that present on dialysis catheters. The larger, cuffed catheter not only allows for faster drainage time, but also should have fewer tunnel leaks after healing occurs around the cuff. It is recommended that the port hub be placed over the ribs to give undersupport to the port base when accessed (Fig. 5). The complication rate was similar to that seen with the external tunneled catheters previously described. The disadvantages to using a port are the requisite nursing support for home drainage because the port is accessed with a noncoring needle under sterile technique and the associated pain of needle access that occurs with each drainage attempt. Shunt Creation Peritoneovenous shunts The shunting of ascitic fluid back into the main circulatory system can be accomplished through placement of Silastic catheters, such as the Denver or LeVeen [32], through peritoneovenous shunts (PVS), or through surgical manipulation such as transjugular intrahepatic portosystemic shunts (TIPS). Traditionally, PVS placement has been done under general anesthesia by surgeons in the operating room using large venous cutdowns. More recently, interventional radiologists have placed PVS successfully using minimally invasive techniques under conscious sedation [33–36]. Advantages of PVS placement include retention of protein-rich ascitic fluid and avoidance of external drainage devices, making it more acceptable to

Fig. 5. Large peritoneal port placed over ribs to provide stable base for needle access. Note peritoneal port (as opposed to vascular port) with subcuticular Dacron cuff.

196

ROSENBERG

patients and their family members. The disadvantages are not trivial, however. PVS are prone to failure secondary to occlusion [37] and have been associated with pulmonary edema [32,38,39], thrombosis of major veins [38,40], seroma formation and leaks [38,41], and disseminated intravascular coagulation [39,42]. Patients must be evaluated carefully for body habitus, cardiac and renal status, and preexisting coagulopathy. PVS placement may not be available at local medical centers. Success and complication rates are tied directly to the learning curve associated with any complex procedure. A site with little experience in placing PVS may be more likely to experience a higher complication rate. Transjugular intrahepatic portosystemic shunts TIPS creation as a treatment for refractory ascites has been well described [43] and is described in detail in a separate article in this issue. TIPS is an effective treatment for intractable ascites because it reduces portal hypertension and improves sodium excretion and diuretic response. TIPS creation also is associated with the development of hepatic encephalopathy. Many studies have compared TIPS for treating refractory ascites versus repeated paracentesis. Saab and colleagues [44] conducted a Cochrane Database review of four trials that compared TIPS with paracentesis in treating refractory ascites [45–48]. They concluded that TIPS and paracentesis showed similar mortality and major complications except for the increase in hepatic encephalopathy in TIPS patients. TIPS is efficacious in only a small percentage of oncology patients with liver invasion/portal compression as the underlying mechanism of ascites formation. Even in the few patients in whom TIPS may be helpful, careful consideration must be given to the benefits of ascites symptom management versus the risk of TIPS placement and subsequent complications of hepatic encephalopathy. RECOMMENDATIONS Effectively managing the symptoms associated with malignant ascites accumulation is problematic and challenging. No single method has been developed that works satisfactorily for all patients. Effective management has been a frustrating problem for many physicians and their patients. Paracentesis is the most common therapy used to provide immediate symptom relief from large-volume ascites. It is a simple, low-risk procedure and is the therapy of choice when ascites first develops. It also is an excellent choice for symptom relief for ascites development between chemotherapy sessions. Many oncology patients, especially patients with ovarian and breast cancer, have temporary resolution of ascites accumulation after chemotherapy has been initiated or a new regimen begun. If chemotherapy is effective at resolving ascites development, paracentesis is the procedure of choice as a ‘‘bridge’’ between regimens. If the tunneled catheters were to be used immediately, they would need to be removed between each regimen because of the risk of infection and then replaced. PVS also are much more prone to failure if ascites is not present to keep them open and functioning. Paracentesis is the procedure of choice until all available chemotherapy options fail to address ascites accumulation.

PALLIATION OF MALIGNANT ASCITES

197

After first-line therapies are exhausted, patients typically are left with paracentesis as the treatment option for symptom relief. Many patients find the pain with each procedure unpalatable and the frequent trips to the hospital fatiguing. They often wait as long as possible until the symptoms are no longer tolerable before scheduling their next procedure. Often, patients progress to needing drainage several times per week to maintain satisfactory comfort levels. Often at this point patients seek other alternatives to paracentesis. The least invasive methods for palliating ascites symptoms are tunneled catheters or port placement. TIPS and PVS placement require centers that are proficient at these procedures and either a general anesthetic or monitored sedation. PVS placement should be reserved for patients who cannot tolerate external drainage for psychological or physical reasons. The choice of tunneled dialysis catheter, peritoneal port, or Pleurx placement may be dictated by what is available at local medical centers or by patient preference. Tunneled peritoneal catheters are easily placed as an outpatient. They have an open, clamp system with Luer-Lok access for drainage that can be performed at home by the patient or a family member. If vacuum bottles are unavailable, the significant time it would take for gravity drainage into bags may not be desirable for many patients. Peritoneal ports have a similar complication rate to other tunneled devices and can be placed as an outpatient. Drainage requires nursing support, however, for sterile preparation and noncoring needle access (and the associated pain with each access). The Pleurx catheter also is placed as an outpatient. It has a closed, one-way valve access system and comes with vacuum bottles and access lines. Previously, the bottles were 500 mL, but now 1-L bottles are available. A multicenter trial evaluating abdominal placement of the Pleurx catheter in patients with malignant ascites was completed to obtain FDA approval for this indication. Preliminary results show low complication rates similar to that published by Rosenberg et al [24] and Reisfield and Wilson [26], and FDA approval was obtained in November of 2005. Blood protein and chemistry levels were evaluated before placement of the catheter and at 12 weeks after ascites drainage of 2 L/d. There were no statistically significant changes in laboratory values for these patients, despite frequency of drainage [49]. Patients who are near death with less than several weeks to live can be kept comfortable with simple pigtail catheter drainage attached to gravity drainage bags. Most patients would like to travel, participate in family events, work, and go about the remainder of their lives without having to make frequent trips to the hospital for symptomatic drainage or are dependent on family members taking time off from work to get them to their drainage appointments. These patients benefit most from external drainage techniques. For patients who cannot tolerate external drainage, PVS may be an alternative. References [1] Reynolds TB. Ascites. Clin Liver Dis 2000;4:157–68. [2] Groszmann RJ, Abraldes JG. Portal hypertension: from bedside to bench. J Clin Gastroenterol 2005;39:S125–30.

198

ROSENBERG

[3] Gines P, Cardenas A, Arroyo V, Rodes J. Current concepts: management of cirrhosis and ascites. N Engl J Med 2004;350:1646–54. [4] Enck RE. Malignant ascites. Am J Hosp Palliat Care 2002;19:7–8. [5] Muir JC. Ascites. In: Von Roenn J, Smith TJ, Loprinzi CL, von Guten CF, editors. ASCO curriculum: optimizing cancer care.The importance of symptom management, vol. 1. Dubuque (IA): Kendall/Hunt; 2001. p. 1–31. [6] Runyon BA, Montano AA, Akriviadis EA, et al. The serum-ascites albumin gradient is superior to the exudate-transudate concept in the differential diagnosis of ascites. Ann Intern Med 1992;117:215–20. [7] Moore KP, Wong F, Gines P, et al. The management of ascites in cirrhosis: report on the consensus conference of the International Ascites Club. Hepatology 2003;38:258–66. [8] Runyan BA. Cardiac ascites: a characterization. J Clin Gastroenterol 1988;10:410–2. [9] Alexandrakis MG, Moschandrea JA, Koulocheri SA, et al. Discrimination between malignant and nonmalignant ascites using serum and ascitic proteins in a multivariate analysis model. Dig Dis Sci 2000;45:500–8. [10] Pare P, Talbot J, Hoefs TC. Serum-ascites albumin concentration gradient: a physiologic approach to the differential diagnosis of ascites. Gastroenterology 1983;85:240–4. [11] Siddiqui RA, Kochkar R, Singh V. Evaluation of fibronectin as a marker of malignant ascites. J Gastroenterol Hepatol 1992;7:161–4. [12] Mansour-Ghanaei F, Shafaghi A, Bagherzadeh AH, Fallah MS. Low gradient ascites: a seven year course review. World J Gastroenterol 2005;11:2337–9. [13] Runyon BA. Approach to the patient with ascites. In: Yamada T, Alpers DH, Owyang C, et al, editors. Textbook of gastroenterology. 3rd edition, Philadelphia: Lippincott, Williams & Wilkins; 1999. p. 966–91. [14] Rana SV, Babu SGV, Hocchar R. Usefulness of ascitic cholesterol as a marker for malignant ascites. Med Sci Monit 2005;11:CR137–42. [15] Mackey JR, Venner PM. Malignant ascites: demographics, therapeutic efficacy, and predictors of survival. Can J Oncol 2001;6:474–80. [16] Lee CW, Bociek G, Faught W. A survey of practice in the management of malignant ascites. J Pain Symptom Manage 1998;16:96–101. [17] Jansen M, Truong S, Reisener KP, et al. Results of sonographically guided percutaneous catheter drainage of intra-abdominal abscesses in surgery. Chirurg 1999;70:1168–71. [18] Gammie JS, Banks ML, Fuhrman CR, et al. The pigtail catheter for pleural drainage: a less invasive alternative to tube thoracostomy. J Soc Laparoendosc Surg 1999;3:57–61. [19] Lee A, Lau TN. Indwelling catheters for the management of malignant ascites. Support Care Cancer 2000;8:493–9. [20] Abuzeid MI, Nassar Z, Massaad Z, et al. Pigtail catheter for the treatment of ascites associated with ovarian hyperstimulation syndrome. Hum Reprod 2003;18:370–3. [21] O’Neill MJ, Weissleder R, Gervais DA, et al. Tunneled peritoneal catheter placement under sonographic and fluoroscopic guidance in the palliative treatment of malignant ascites. AJR Am J Roentgenol 2001;177:615–8. [22] Richard HM, Coldwell DM, Boyd-Kranis RI, et al. Pleurx tunneled catheter in the management of malignant ascites. J Vasc Interv Radiol 2001;2:373–5. [23] Barnett TD, Rubins J. Placement of a permanent tunneled peritoneal drainage catheter for palliation of malignant ascites: a simplified percutaneous approach. J Vasc Interv Radiol 2002;13:379–83. [24] Rosenberg S, Courtney A, Nemcek AA Jr, Omary RA. Comparison of percutaneous management techniques for recurrent malignant ascites. J Vasc Interv Radiol 2004;15: 1129–31. [25] Lomas DA, Wallis PJ, Stockley RA. Palliation of malignant ascites with a Tenckhoff catheter. Thorax 1989;44:928. [26] Reisfield GM, Wilson GR. Management of intractable cirrhotic ascites with an indwelling drainage catheter. J Palliat Med 2003;6:787–91.

PALLIATION OF MALIGNANT ASCITES

199

[27] Savador SJ, Geschwind JF, Lund GB, Schal PJ. Percutaneous radiologic placement of peritoneal dialysis catheters: long term results. J Vasc Interv Radiol 2000;11:965–70. [28] Iyengar TD, Herzog TJ. Management of symptomatic ascites in recurrent ovarian cancer patients using an intra-abdominal semi-permanent catheter. Am J Hosp Palliat Care 2002;19: 35–8. [29] Rosenbaum DI, Geisinger MA, Newman JS, et al. Use of subcutaneous venous access ports to treat refractory ascites. J Vasc Interv Radiol 2001;12:1343–6. [30] Savin MA, Kirsch MJ, Romano WJ, et al. Peritoneal ports for treatment of intractable ascites. J Vasc Interv Radiol 2005;16:363–8. [31] Sabatelli FW, Glassman ML, Kerns SR, Hawkins IF Jr. Permanent indwelling peritoneal access device for the management of malignant ascites. Cardiovasc Interv Radiol 1994;17: 292–4. [32] Gines P, Arroyo V, Vargas V, et al. Paracentesis with intravenous infusion of albumin as compared with peritoneovenous shunting in cirrhosis with refractory ascites. N Engl J Med 1991;325:829–35. [33] Hussain FF, Meer ZF, Lopez AS. Peritoneovenous shunt insertion for intractable ascites: a district general hospital experience. Cardiovasc Interv Radiol 2004;27:325–8. [34] Zanon C. Palliative treatment of malignant refractory ascites by positioning of Denver peritoneovenous shunt. Tumori 2002;88:123–7. [35] Park JS. Percutaneous peritoneovenous shunt creation for the treatment of benign and malignant refractory ascites. J Vasc Interv Radiol 2001;12:1445–8. [36] Orsi F. Percutaneous peritoneovenous shunt positioning: technique and preliminary results. Eur Radiol 2002;12:1188–92. [37] Deltenre P, Mathurin P, Dharanay S, et al. Transjugular intrahepatic portosystemic shunt in refractory ascites: a meta-analysis. Liver Int 2005;25:349–56. [38] Stanley MM. Treatment of intractable ascites in patients with alcoholic cirrhosis by peritoneovenous shunting (LeVeen). Med Clin North Am 1979;63:523–35. [39] Marimuthu K, Keemar AS, Sabanathan S, et al. Indigenous cost-effective peritoneo-venous shunt for refractory ascites. Int Surg 2004;89:85–9. [40] Lund RH, Moritz MW. Complications of Denver peritoneo-venous shunting. Arch Surg 1982;117:924–8. [41] Berger A, Goldberg MI. Subcutaneous cancer growth complicating the peritoneovenous shunting of malignant ascites. Surgery 1983;93:374–6. [42] Tueche SG, Pector JC. Peritoneovenous shunt in malignant ascites, the Bordet Institute experience from 1975–1998. Hepatogastroenterology 2000;47:1322–4. [43] Ochs A, Rossle M, Haag K, et al. The transjugular intrahepatic stent-shunt procedure for refractory ascites. N Engl J Med 1995;332:1192–7. [44] Saab S, Nieto JM, Ly D, Runyon BA. TIPS versus paracentesis for cirrhotic patients with refractory ascites. The Cochrane Database of Systemic Reviews. Cochrane Library 2005;1. [45] Lebrec D, Giuily N, Hadengue A, et al. Transjugular intrahepatic portosystemic shunts: comparison with paracentesis in patients with cirrhosis and refractory ascites. J Hepatol 1996;25:135–44. [46] Rossle M, Ochs A, Gulberg V, et al. A comparison of paracentesis and transjugular intrahepatic portosystemic shunting in patients with ascites. N Engl J Med 2000;342:1701–7. [47] Gines P, Uriz J, Calahorra B, et al. Transjugular intrahepatic portosystemic shunting versus paracentesis plus albumin for refractory ascites in cirrhosis. Gastroenterology 2002;123: 1839–47. [48] Sanyal A, Genning C, Reddy K, et al. The North American study for the treatment of refractory ascites. Gastroenterology 2003;124:634–41. [49] Nemcek A, Courtney A, Rosenberg S, et al. Efficacy and safety of the Pleurx catheter when used to treat recurrent malignant ascites, in press.

Gastroenterol Clin N Am 35 (2006) 201–219

GASTROENTEROLOGY CLINICS OF NORTH AMERICA

Palliative Care for Patients with End-Stage Liver Disease Ineligible for Liver Transplantation William Sanchez, MDa, Jayant A. Talwalkar, MD, MPHb,* a

Department of Medicine, Division of Gastroenterology & Hepatology, Mayo Clinic College of Medicine, 200 First Street, SW, Rochester, MN 55901, USA b Department of Medicine, Advanced Liver Diseases Study Group, Division of Gastroenterology & Hepatology, Mayo Clinic College of Medicine, 200 First Street, SW, Rochester, MN 55901, USA

A

s the final common end point in patients with chronic liver disease of a variety of causes, end-stage liver disease (ESLD) poses an important clinical challenge. Epidemiologic factors including the increasing numbers of patients with chronic hepatitis C developing cirrhosis, the increasing age of the population, and the obesity epidemic translate into a growing number of patients with ESLD. The number of patients with ESLD is increasing, while the number of donor organs available for transplantation remains relatively stable. These patients with ESLD need to be managed in the community without liver transplantation. These patients face a variety of symptoms and disease-related complications, which affect their survival and health-related quality of life. Understanding the principles of palliative medicine and the management of complications of cirrhosis is important for improving the health status of patients with ESLD.

SCOPE OF THE PROBLEM: CLINICAL EPIDEMIOLOGY OF END-STAGE LIVER DISEASE Patients with cirrhosis who develop complications, such as ascites, portal hypertension, gastrointestinal hemorrhage, or hepatic encephalopathy, are considered to have ESLD. No recent population-based epidemiologic study describing the secular trends of disease burden for cirrhosis is available; however, data support the notion that ESLD is a significant cause of morbidity, resource use, and death in the United States. An estimated 5.5 million people (2% of the US population) are affected by cirrhosis, its consequences from portal hypertension, and the attendant risk for carcinogenesis [1–12]. Approximately 26,000 patients die of chronic liver disease and cirrhosis in the United States each year, making this condition the seventh leading cause of death among *Corresponding author. E-mail address: [email protected] (J.A. Talwalkar). 0889-8553/06/$ – see front matter doi:10.1016/j.gtc.2005.12.007

ª 2006 Elsevier Inc. All rights reserved. gastro.theclinics.com

202

SANCHEZ & TALWALKAR

people between the ages of 25 and 64 [13]. An additional 18,000 patients die of hepatobiliary cancer related most often to cirrhosis. The major causes of death in the absence of liver transplantation are liver failure, hepatocellular carcinoma (HCC), gastrointestinal bleeding, sepsis, and renal failure[1–12,14–19]. With the development of ascites or hepatic encephalopathy, the estimated 2-year survival of patients with ESLD is less than 50% [14–19]. Despite the success of liver transplantation, only 6000 patients undergo this curative procedure annually [20,21]. The shortage of suitable donor organs is expected to continue to worsen as the number of potential transplant recipients increases. Based on current data from the Organ Procurement and Transplantation Network, there are more than 17,000 patients with ESLD awaiting liver transplantation in the United States. With increasing waiting times secondary to the lack of organs, an estimated 1500 patients die awaiting liver transplantation [21]. In addition, an increasing proportion of patients with ESLD ultimately are found to be unsuitable candidates for liver transplantation based on advanced age, obesity, and comorbid illness, such as diabetes mellitus and coronary artery disease. DETERMINING THE PROGNOSIS OF PATIENTS WITH END-STAGE LIVER DISEASE Available data regarding the natural history and prognosis of cirrhosis come largely from studies performed at referral-based centers [12,14–19] and population-based studies from the 1970s [22]. Patients with compensated cirrhosis can expect an estimated median survival of nearly 10 years. With the development of esophageal varices, the median survival of patients with compensated cirrhosis is shortened to 7 to 10 years. With a transition rate to develop ascites at 5% to 10% annually, it is anticipated that only 50% of patients with ESLD survive 2 years [1–12]. Regardless of liver disease severity, patients who experience acute variceal hemorrhage have a risk of bleed-related death of 10% to 20% per episode requiring hospitalization [23]. Patients who develop medically refractory ascites have an even worse prognosis with an estimated 6-month survival rate of 50% [12,24]. To predict long-term outcome more accurately in this patient population, many classification schemes and prognostic models have been developed for clinical use. The two most commonly used indices are the Child-TurcottePugh (CTP) classification and the Model for End-Stage Liver Disease (MELD) (Table 1). The CTP classification, originally introduced in 1964 and subsequently modified in 1973, has been used widely for many years by health care providers [25,26]. Developed as a prognostic tool for patients with cirrhosis undergoing shunt surgery for portal hypertension, the CTP classification incorporates objective parameters (prothrombin time, serum bilirubin, and albumin) and subjective assessments regarding the degree of involvement with ascites and hepatic encephalopathy. Over time, the CTP classification was used in other clinical situations and found to have prognostic utility in patients with cirrhosis. Based on some degree of external validity, the

PALLIATIVE CARE FOR ESLD

203

Table 1 Prognostic models in end-stage liver disease Child-Turcotte-Pugh classification (class A, 5–6 points; class B, 7–9 points; class C, 10–15 points) 1 point 2 points 3 points Bilirubin (mg/dL) Noncholestatic disease <4 4–10 >10 Cholestatic disease <2 2–3 >3 Albumin (g/dL) >3.5 2.8–3.5 <2.8 Prothrombin time INR <1.7 1.7–2.3 >2.3 Hepatic encephalopathy None Stage 1–2 Stage 3–4 Ascites None Mild–moderate Severe MELD—Model for End-Stage Liver Disease (UNOS modification) ÿ MELD ¼ 0:957
Loge ðcreatinine mg=dL Þ þ 0:378
Loge ðbilirubin mg=dLÞ þ 1:120
Loge ðprothrombin time INRÞ þ 0:643Þ
10 Abbreviation: INR, international normalized ratio. *Patients undergoing hemodialysis are considered to have a creatinine of 4 mg/dL.

CTP classification subsequently was used as a disease severity index for organ allocation [27,28]. Numerous shortcomings limit the utility of the CTP classification, including the subjectivity of scores regarding manifestations of portal hypertension. Additionally, the CTP score does not take into account renal dysfunction, which has been shown to have major prognostic importance in patients with advanced liver disease [28–30]. The MELD score initially was developed to assess mortality risk in patients undergoing transjugular intrahepatic portosystemic shunt (TIPS) procedures. The MELD score is calculated using widely available, objective clinical laboratory values, including total bilirubin, creatinine, and international normalized ratio for prothrombin time (see Table 1). Since its conception, the MELD score has been validated in a variety of patient cohorts (ambulatory and hospitalized) to predict accurately 3-month liver-related mortality risk compared with the CTP classification [28,31]. The United Network for Organ Sharing adopted the MELD score as the basis for organ allocation in liver transplantation in February 2002. As with any prognostic model, there are some limitations with the MELD score. Variables that represent complications of portal hypertension, including hyponatremia, ascites and spontaneous bacterial peritonitis, are excluded from the model. Among these patients, there may be individuals with a higher risk of short-term death compared with patients with a similar MELD score without these complications. In addition, the longitudinal ability to predict survival accurately beyond 3 months is uncertain in the absence of liver transplantation. PALLIATIVE AND END-OF-LIFE CARE Patients with ESLD who are ineligible for or do not wish to pursue liver transplantation require increasingly complex medical support. In this context,

204

SANCHEZ & TALWALKAR

a focused discussion on the palliative care aspects of medical management should be initiated. Physician-patient communication addressing palliative care and end-of-life issues is essential and should be initiated early rather than at the time of acute deterioration requiring hospitalization. Definition of Palliative Care The National Consensus Project for Quality Palliative Care defines the goal of palliative care as to ‘‘prevent and relieve suffering and to support the best possible quality of life for patients and their families, regardless of the stage of the disease or the need for other therapies’’ [32]. Palliative care involves a multidisciplinary approach to optimize symptom management and health status without excluding medical therapies to achieve these goals. The concept of palliative care is integrated easily into standard medical management of ESLD. In contrast, hospice care represents the phase at which point lifeprolonging therapies are not pursued by affected individuals [33]. Members of the palliative care team should include physicians, expert nursing staff, and social workers [32,33]. An important component of palliative care is the understanding that death is a normal process with support provided for the bereavement process of the patient and his or her loved ones [34]. Controversy exists as to the role of life-extending therapies in palliative care. The World Health Organization states that palliative care ‘‘intends neither to hasten or postpone death.’’ The World Health Organization goes on to state that palliative care also is ‘‘applicable early in the course of illness, in conjunction with other therapies that are intended to prolong life, such as chemotherapy or radiation therapy’’ [34]. The National Consensus Project for Quality Palliative Care emphasizes that although palliative care is focused on improving the patient’s quality of life, it does not exclude medical therapy aimed at prolonging life [32]. Although frequently thought of in the context of a patient with incurable malignancy, palliative care also is appropriate for patients ‘‘with a persistent or recurring condition that adversely affects their daily functioning or will predictably reduce life expectancy’’ [32]. Patients with advanced chronic disease states including ESLD are candidates for palliative care. As mentioned previously, a palliative focus should be incorporated into the standard care of patients with ESLD. A similar approach has been advocated for patients with congestive heart failure, a common condition that shares several features (chronicity, frequent hospitalizations, volume overload) with ESLD [33]. Hospice Hospice care is directed toward providing medical care and support services at the end of life, frequently in the patient’s home [32]. Medicare provides benefits covering hospice services, as do most private insurance plans and state Medicaid programs. The National Hospice and Palliative Care Organization has published guidelines for the use of hospice care in nonmalignant disease such as ESLD. Hospice criteria for patients with nonmalignant diseases include documented clinical progression of a life-limiting condition, multiple emergency

PALLIATIVE CARE FOR ESLD

205

department visits or inpatient hospitalizations over the past 6 months, and impaired nutritional status related to the disease process [35]. These guidelines, based on expert opinion, have been adopted by the Health Care Finance Administration to determine eligibility for hospice benefits under Medicare. Although the number of patients enrolled in hospice has increased substantially since the 1990s, only 1.6% of 2003 hospice admissions were for ESLD [35,36]. Patients identify dying at home as an important feature of quality health care at the end of life [37]. Most deaths occur in a hospital, however, and only 20% of US patients receive hospice care [36]. Data from the Study to Understand Prognoses and Preferences for Outcomes and Risk of Treatments (SUPPORT) revealed that only 2% (54 of 2607) of eligible patients with advanced organ failure were discharged from the hospital to hospice care. One of the contributing factors to this observation is the gross inaccuracy of clinical assessments used to estimate an individual patient’s life expectancy. The predictive model used in the SUPPORT study was designed to approximate the National Hospice and Palliative Care Organization guidelines used by the Health Care Finance Administration. General (recurrent hospitalizations, home care after discharge, dependency for activities of daily living, weight loss, and hypoalbuminemia) and ESLD-specific (cachexia and renal insufficiency) variables did not effectively identify which patients with ESLD died within 6 months of hospitalization [36]. ESLD also disproportionally affects younger patients relative to other patients with end-stage organ disease. In the absence of liver transplantation, patients with ESLD also typically die at a younger age. Data from the SUPPORT trial showed that patients with cirrhosis died at a mean age of 57 years, significantly younger than patients with other severe chronic medical conditions, such as chronic obstructive lung disease or congestive heart failure [38]. In contrast, the experience from Asia (where seropositivity for viral hepatitis is endemic) shows that a larger proportion of patients are enrolled in hospice care because of HCC compared with patients in the United States. Most of these patients also have coexisting ESLD. Patients admitted to hospice for HCC were typically younger and had a shorter survival time than nonHCC hospice patients. Advanced liver dysfunction predicted shorter survival time, with almost 98% of patients with CTP class C hepatic disease severity dying while in the inpatient hospice ward [39]. Advance Directives An important component of medical care for patients with ESLD is addressing the use of advance directives before the end of life. In patients who are not candidates for organ transplantation, progressive liver disease is most likely to be the cause of death in most cases. Data regarding how frequently health care providers discuss the issue of advance directives with patients affected by ESLD are scarce, however. Available information supports the notion that advance directives are addressed less frequently in these patients compared with patients with malignancy. Among hospitalized patients in the SUPPORT

206

SANCHEZ & TALWALKAR

study, only 20% (618 of 3058) of patients had advance directives [38]. From a study of three United States teaching hospitals, only 16% of patients with ESLD had ‘‘do not resuscitate’’ orders written in the medical record compared with 47% of patients with nonresectable lung cancer. Resident physicians were more likely to discuss ‘‘do not resuscitate’’ orders with patients affected by cancer rather than cirrhosis or congestive heart failure, despite a good understanding of their prognosis [40]. A discussion of patient preferences for end-of-life care also should include wishes for cardiopulmonary resuscitation, intubation with mechanical ventilation, invasive medical procedures (eg, emergency endoscopy), and whether subsequent hospitalization for unstable illness is desired. The physician should validate a patient’s wishes regarding ‘‘do not resuscitate’’ and ‘‘do not hospitalize’’ through written orders and reassure the patient that not proceeding with invasive medical therapy does not equate to discontinuing care or the termination of the physician-patient relationship [33,41]. PALLIATIVE CARE TO MANAGE THE COMPLICATIONS OF END-STAGE LIVER DISEASE Ascites Ascites is the most common complication of ESLD and is typically symptomatic [4,42]. Patients with alcoholic liver disease and ascites may benefit from the cessation of alcohol consumption because reduction in portal pressure and improvement of ascites is a known outcome [42,43]. Most patients with ascites gain control of symptoms with dietary sodium restriction (maximum 2 g/d) and oral diuretic therapy. Initial oral diuretic therapy should consist of the combination of spironolactone with a loop-acting diuretic, such as furosemide. In most cases, the combination of 100 mg of spironolactone with 40 mg of furosemide (administered once daily, usually in the morning to avoid nocturia) is more effective than single-agent diuretic therapy. In addition, the combination of medications can ameliorate potassium derangements that can be seen with monotherapy [42,44,45]. Dose escalation of these medications should occur in the same ratio to control ascites with a suggested maximum of spironolactone 400 mg with furosemide 160 mg daily. The development of azotemia or hyponatremia suggests diuretic-refractory ascites, and dosages of one or both medications should be reduced. As the severity of liver disease progresses, an increasing percentage of patients develop symptomatic ascites refractory to medical therapy. An assessment should be made for the presence of factors that could aggravate sodium retention. Patients with a history of alcoholic liver disease should be questioned regarding abstinence from alcohol. Adherence to sodium restriction should be assessed by measuring urinary sodium excretion. Measured urinary sodium excretion greater than 78 mEq/d suggests nonadherence to sodium restriction. Spot urinary sodium measurements may be as accurate as 24-hour collections and are less cumbersome to collect in the outpatient setting. The use of nonsteroidal anti-inflammatory analgesics also should be discontinued

PALLIATIVE CARE FOR ESLD

207

given their propensity to aggravate ascitic fluid accumulation via prostaglandin inhibition [42]. Large-volume paracentesis (LVP) is a safe and effective therapy in the management of symptomatic ascites for patients with ESLD. Currently, there is no limit to the frequency with which LVP can be performed. LVP should be performed as frequently as necessary to provide symptomatic relief from tense ascites. The use of plasma volume expansion with intravenous albumin has been shown to reduce significantly paracentesis-induced activation of the reninangiotensin system. Activation of the renin-angiotensin system is associated with vasoconstriction, sodium retention, and a more rapid reaccumulation of ascites (ie, postparacentesis circulatory dysfunction) [4,46]. There is some controversy, however, regarding the use of albumin infusion given its high cost, and it has not been uniformly recommended after LVP [42]. The medical management of refractory ascites also may be accomplished by the use of portosystemic shunting. More commonly, TIPS placement is employed. In a prospective, randomized, multicenter study, the North American Study for the Treatment of Refractory Ascites group compared TIPS plus medical therapy with LVP plus medical therapy. Although TIPS was found to be superior to LVP for the control of ascites, it did not reduce the need for hospitalization or improve survival. Although the LVP group and the TIPS group had improvement in health-related quality-of-life scores compared with baseline, the end-of-treatment scores were not different between the two groups. Additionally, TIPS did not eliminate the need for a sodium-restricted diet [24]. New-onset or worsening hepatic encephalopathy is the most common complication of TIPS and is due to increased portosystemic shunting. The incidence of hepatic encephalopathy after TIPS is reported to be as high as 40% in prospective series. In most patients, hepatic encephalopathy after TIPS can be managed using lactulose or nonabsorbable antibiotic preparations. Rarely, occlusion of the shunt to control refractory symptoms is necessary [47–51]. The major disadvantage of TIPS shunts is the development of in-stent stenosis requiring revision. The shunt dysfunction rate is 75% at 1 year and approaches 100% at 2 years, although not all patients with radiographic evidence of shunt dysfunction develop clinically significant portal hypertension [50,52–54]. Relative contraindications for TIPS placement include severe ESLD (CTP class C liver disease), portal vein thrombosis, pulmonary hypertension or right heart failure, intrinsic renal disease, and prior history of hepatic artery chemoembolization. Peritoneovenous shunts (eg, Denver shunt, LeVeen shunt) are surgically placed shunts that have been used for the management of refractory ascites. A high incidence of complications, including shunt malfunction, disseminated intravascular coagulation, and sepsis from bacterial infection, is associated with the use of this technique. Their widespread use has not occurred in recent times [55–57]. The authors do not recommend the use of peritoneovenous shunts for the management of intractable ascites based on the absence of evidence-based data for efficacy and an unfavorable risk/benefit profile. Instead, the use of serial LVP is safer and more effective.

208

SANCHEZ & TALWALKAR

In terminally ill patients with ESLD, the use of an indwelling peritoneal drainage catheter for the management of symptomatic refractory ascites has been described. Indwelling catheters have been used safely for the management of malignant ascites, although it is unknown whether the complication rate would be higher in patients with ESLD, who are typically coagulopathic and prone to spontaneous bacterial peritonitis. The technique of bedside insertion of an indwelling catheter has been described, and an experienced physician can place a catheter without the need for patient hospitalization. An indwelling catheter allows the patient (or care provider) to drain small volumes of ascitic fluid as needed for symptom relief. Given the concerns for infectious complications, indwelling peritoneal drainage catheters seem most appropriate for patients with terminal disease enrolled in hospice care [58–60]. Hepatic Hydrothorax Hepatic hydrothorax is defined by the development of a transudative pleural effusion in patients with cirrhosis and portal hypertension. The pathogenesis is believed to be the preferential movement of ascitic fluid from the positivepressure abdominal cavity into the negative-pressure thoracic cavity in patients with diaphragmatic defects. Although the abdominal cavity can expand, and patients frequently tolerate large volumes of ascitic fluid, relatively small pleural fluid collections are often highly symptomatic. The management of hepatic hydrothorax is important for the relief of dyspnea in patients with ESLD. As with the initial medical approach for management of ascites, the mainstay of therapy is sodium restriction and combination diuretic therapy. Large-volume thoracentesis is frequently necessary for the rapid relief of dyspnea. Repeated thoracentesis can be performed, but if frequently necessary, patients may benefit from more invasive, alternative management strategies. Tube thoracostomy is commonly used to manage recurrent symptomatic pleural effusions. This therapeutic modality is associated with a high rate of complications in patients with ESLD, however, including spontaneous bacterial empyema and hemothorax. Among 56 patients with CTP class B or C cirrhosis who underwent tube thoracostomy for the treatment of hepatic hydrothorax, most patients developed electrolyte disturbances or renal insufficiency. A total of 15 patients died with the chest tube in place. Pleurodesis (most commonly talc) has a low rate of success in ablating the pleural space, likely owing to the rapid reaccumulation of pleural fluid driven by portal hypertension. Incomplete pleurodesis leads to the formation of loculated pleural effusions, which make subsequent management by thoracentesis difficult [61–63]. By decreasing portal venous pressure, the use of TIPS is noted for reductions in the accumulation of pleural fluid among patients with refractory hepatic hydrothorax. TIPS also has been shown to decrease the need for serial therapeutic thoracentesis in these patients. Similarly to patients with ascites, however, TIPS does not obviate the need for sodium restriction. Although TIPS may decrease the need for thoracentesis, there has been no proven survival benefit in

PALLIATIVE CARE FOR ESLD

209

the absence of liver transplantation. To date, there has been no systematic assessment of the impact of TIPS on health-related quality of life in patients with refractory hepatic hydrothorax. Despite the absence of evidence-based data and health status information, it is reasonable to use TIPS for the palliation of recurrent, symptomatic hepatic hydrothorax that is refractory to medical therapy [50,63,64]. Hepatic Encephalopathy Hepatic encephalopathy may be the most disabling complication of portal hypertension experienced by patients with cirrhosis. Hepatic encephalopathy is characterized by several neuropsychiatric disturbances that result in significantly impaired health-related quality of life [65]. Although hepatic encephalopathy occurs commonly, its manifestations range from subclinical encephalopathy to hepatic coma. Neuropsychometric testing can identify subtle disturbances from hepatic encephalopathy in 60% to 70% of patients with ESLD [66–68]. Frequently, patients with overt hepatic encephalopathy present with confusion, ataxia, and somnolence, which can progress rapidly to stupor and coma. Medications, particularly psychoactive agents including narcotic analgesics and sedative-hypnotics, frequently are identified as precipitating factors for the acute exacerbation of hepatic encephalopathy. In terminally ill patients with ESLD, the presence of somnolence from hepatic encephalopathy may not be a distressing symptom for the patient; the pursuit of aggressive therapy to reverse hepatic encephalopathy should be considered carefully on a case-by-case basis. The most widely used first-line medical therapy for hepatic encephalopathy is a cathartic, typically a nonabsorbable disaccharide, such as lactulose or lactitol. The mechanism of action for lactulose is to acidify luminal contents that promote the formation of ammonium chloride from ammonia, which is excreted in stool. Lactulose can be administered orally or, in an obtunded patient, per rectum. The goal of lactulose therapy is to achieve three to four soft, nondiarrheal stools per day. Excessive administration of lactulose results in diarrhea, which may precipitate hepatic encephalopathy further via dehydration and electrolyte disturbance, and negatively affects patient quality of life [67,69]. Of concern is the more recent observation that data supporting the efficacy of lactulose are based on methodologically weak clinical trials, which raises the suspicion that lactulose is no better than placebo. The correction of precipitating factors (eg, dehydration or constipation) may be the most important therapy, but remains obscured by the widespread use of lactulose in patients with moderate-to-severe hepatic encephalopathy. In patients who are refractory or intolerant to lactulose, the use of nonabsorbable antibiotic therapy can be effective for managing relapsing or treatmentdependent hepatic encephalopathy. Oral neomycin (1 to 2 g per day) is poorly absorbed, but long-term administration can lead to nephrotoxicity. Oral metronidazole (250 mg two to three times daily) is effective, but may negatively affect the patient’s sense of taste and can lead to significant peripheral neuropathy [69]. Several other compounds have been evaluated, typically in small trials,

210

SANCHEZ & TALWALKAR

including ornithine aspartate (which is not available in the United States), flumazenil, and sodium benzoate. The use of rifaximin, a nonabsorbable rifampin analogue, has been shown to be equally effective as lactitol in the management of acute exacerbations of hepatic encephalopathy [70]. In the United States, rifaximin has not been approved yet by the Food and Drug Administration for the treatment of hepatic encephalopathy, and its use remains an off-label indication. Pruritus Pruritus is a complication of chronic liver disease and cirrhosis that can cause patient distress. Although most commonly associated with cholestatic liver diseases, the development of pruritus may complicate liver disease of any origin. The mainstay of therapy is treatment of the underlying liver disease whenever possible. In patients with biliary obstruction causing pruritus, obtaining adequate biliary drainage is paramount for symptom relief. If endoscopic biliary drainage is inadequate or not technically feasible, external biliary drainage via percutaneous transhepatic cholangiography is a viable option. If necessary (eg, a difficult central obstructing lesion), the left and right biliary ductal systems can be percutaneously accessed and drained separately by endoscopic or percutaneous methods. A wide range of putative agents or mediators has been proposed as the cause of pruritus in liver disease, although medical therapies used in practice often do not have their basis in well-defined pathophysiology. The most common firstline therapy of choice is an oral antihistamine. Because the pruritus of liver disease is multifactorial, however, antihistamines probably work mainly through their nonspecific sedating effect. They may be of particular benefit administered at bedtime to allow patients to get an adequate night’s sleep [71]. Cholestyramine, a nonabsorbable bile acid exchange resin, is administered orally and prevents uptake of bile acids in the terminal ileum. The drug is most effective in patients with an intact gallbladder. A total of 2 to 16 g daily is administered orally in divided doses. Care must be taken to separate the administration of cholestyramine from other medications because it can interfere with their absorption. Other common side effects of cholestyramine include unpalatability and constipation [71–73]. Endogenous opioids have been implicated in the cause of liver disease– associated pruritus. The beneficial effect of rifampin and its analogues, which upregulate the cytochrome P-450 metabolic pathway, has been postulated to be due to the enhanced metabolism of endogenous opioids. Rifampin is administered orally starting at 150 to 300 mg/d [71,74–76]. The drug may have significant hepatotoxic effects in patients with ESLD, however, and its widespread use in patients with CTP class B and C cirrhosis is not recommended. In patients with compensated cirrhosis, a dose escalation of rifampin to 600 to 900 mg daily is possible. Serum liver tests should be obtained at 2 and 6 weeks after therapy to exclude toxicity. Thereafter, serum liver tests may be obtained every 3 months to monitor for toxicity. Numerous small clinical trials using narcotic receptor

PALLIATIVE CARE FOR ESLD

211

antagonists have shown benefit in patients with pruritus [77–81]. The oral narcotic antagonist naltrexone (starting 12.5 mg twice daily titrated upwards to 50 mg daily) has been reported to provide rapid relief of pruritus in selected patients. Patients receiving opioid antagonists can experience a narcotic withdrawal–like phenomenon, and narcotic antagonist therapy is inappropriate for patients receiving opioid analgesics for the management of pain [71]. Anecdotal experience with the selective serotonin reuptake inhibitor sertraline (25–50 mg daily) indicates improvement of pruritus in patients with noncirrhotic primary biliary cirrhosis [82]. The safety and efficacy of sertraline in patients with ESLD is unknown. In patients with pruritus refractory to these medical therapies, referral to a tertiary center for investigational therapies, such as extracorporeal hemoperfusion, may be appropriate. Hepatocellular Carcinoma HCC is a frequent complication of cirrhosis, occurring at an estimated rate of 1% to 6% per year. The incidence of HCC has been increasing steadily in United Sates patients since the 1980s. In contrast, HCC is the second leading causes of cancer-related mortality in Asia. The prognosis of unresectable HCC is poor with a 5-year survival rate of less than 5% [39,83,84]. Additionally, many patients are not candidates for surgical resection based on ESLD resulting in hyperbilirubinemia and portal hypertension. In patients with CTP class B or C cirrhosis, hepatic resection is fraught with an increased risk of death from progressive liver failure. Although liver transplantation remains the treatment of choice for patients with cirrhosis and HCC, many patients do not meet the eligibility criteria for liver transplantation. Patients with small tumors have an excellent prognosis after liver transplantation. Given concerns for disease recurrence in patients with more extensive tumors, the United Network for Organ Sharing limits allocation of donor organs to patients with HCC according to the previously published Milan criteria (Box 1). Although protocols for extended criteria are being investigated, patients with large tumor burdens or vascular invasion are not typically considered candidates for transplantation [14,85,86]. Aside from surgical interventions, many nonoperative treatment modalities are available for palliative care in HCC. Ablative therapies include percutaneous ethanol injection (PEI), radiofrequency ablation (RFA), and selective hepatic arterial chemoembolization (HACE). PEI is effective in ablating small

Box 1: Milan criteria for liver transplantation for hepatocellular carcinoma Single lesion <5 cm in diameter or Three lesions all <3 cm in diameter and No extrahepatic metastasis and No vascular invasion

212

SANCHEZ & TALWALKAR

tumors and has the advantage of being inexpensive and widely available. Successful therapy with PEI frequently requires multiple injections, and tumors larger than 3 cm have been shown to have a higher local recurrence rate than smaller lesions [87,88]. Data suggest that RFA may provide more effective local tumor ablation with fewer treatment sessions being necessary. The safety of RFA in a large series of 664 patients was found to be similar to PEI [88]. A randomized trial comparing 157 patients with HCC found that RFA was more effective than PEI in controlling HCC smaller than 4 cm [89]. RFA and PEI are limited by the size of tumor amenable for treatment. Tumors close to the liver capsule are not ideal for ablation because of the possible injury to adjacent organs. Percutaneous tumor ablation should be considered carefully in patients with portal vein thrombosis owing to the concern for causing hepatic abscess in the necrosed tissue. Selective HACE has been shown in a meta-analysis of 545 patients to improve survival among patients with unresectable HCC. The greatest and perhaps only real benefit is in patients with compensated or CTP class A cirrhosis. Embolization alone, without instillation of chemotherapeutic agents, has not been shown to be equally effective [90]. HACE is generally well tolerated; the most common side effects are fever, abdominal pain, nausea, and anorexia. Contraindications to HACE include main portal vein thrombosis, CTP class C liver disease, and a history of TIPS shunt placement based on the fact that portal venous inflow is already reduced. Systemic chemotherapy and radiation therapy are limited by increased side effects, dose-limiting hepatotoxicity, and poor efficacy [91]. Several investigations have recognized the palliative effects of thalidomide in patients with advanced HCC. Treatment often is complicated, however, by fatigue, constipation, and peripheral neuropathy even at low doses [92,93]. Ongoing clinical trials are evaluating new treatment modalities for patients with HCC. Symptoms That Reduce Quality of Life Patients with ESLD have significant impairments in health-related quality of life partly as a result of physical symptoms associated with liver disease (ascites, pruritus) and cognitive dysfunction and psychological symptoms. Depression is common in patients with ESLD; prevalence rates are 30% to 40% [94]. A more precise estimate is difficult to assess based on the inability to exclude hepatic encephalopathy as a cause of mood disturbance in terminally ill patients [39]. Depression in patients with ESLD has been shown to be associated with a poor outcome compared with nondepressed patients with ESLD. In a study of 81 patients awaiting liver transplantation, 64% were found to have depression based on the Beck Depression Inventory. Among patients with depression, the presence of diminished health-related quality-of-life scores, lower performance status, and decreased survival among nontransplanted patients was observed [94]. Although an approach similar to managing depression in patients with advanced cancer seems applicable, there is a need for more data regarding the management of depression in patients with ESLD.

PALLIATIVE CARE FOR ESLD

213

Insomnia with sleep disturbance is common and negatively affects healthrelated quality of life in patients with ESLD. Sleep disturbances (hypersomnia and insomnia) are early features of hepatic encephalopathy [69]. Sleep disturbances without hepatic encephalopathy also have been recognized as a significant problem in patients with cirrhosis. From a questionnaire-based study, investigators found nearly 50% of patients with cirrhosis complained of sleep disturbances. Unsatisfactory sleep patterns were associated with higher scores for depression and anxiety [95]. Fatigue is commonly associated with ESLD, particularly primary biliary cirrhosis and hepatitis C virus. Pharmacotherapy with b-adrenergic antagonists and oral diuretics may contribute to fatigue [96,97]. Other contributing factors include muscle wasting, hyperammonemia, depression, and sleep disruption from poorly controlled pruritus and muscle cramps [71,98]. Lower extremity muscle cramps can be a particularly problematic symptom. They frequently occur at night and often disrupt sleep. A large study of factors associated with health-related quality of life in patients with ESLD identified CTP score and muscle cramps as the factors most associated with diminished health-related quality-of-life scores [98]. The cause of muscle cramps in ESLD is unknown; it is unrelated to age, severity of liver disease, electrolyte disturbance, or use of oral diuretics. Currently, there is no identified therapy for leg cramps based on a controlled clinical trial. Frequently, quinine sulfate 260 to 324 mg daily at bedtime is used. Despite widespread use, patients with chronic liver disease have been systematically excluded from randomized controlled trials of quinine for muscle cramps. Adequately controlled studies are needed to define the safety and efficacy of quinine therapy in patients with ESLD. Cachexia and Muscle Wasting Cachexia is a common finding in patients with ESLD. The cause of cachexia in cirrhosis is multifactorial, owing in part to increased metabolic needs and diminished caloric intake. Patients with ESLD often have inadequate caloric intake to meet their metabolic needs based on several factors, including elevated intraabdominal pressure from ascites resulting in early satiety, protein-calorie deficiency from protein loss secondary to frequent paracentesis, or autonomic neuropathy associated with cirrhosis resulting in gastroparesis [99]. It is crucial to avoid protein restriction in these patients, which is frequently done in an effort to avoid hepatic encephalopathy without proof of efficacy [69,100]. Most data regarding cachexia in terminally ill patients are from the cancer literature [101]. The choices of therapeutic options to counteract cachexia and muscle wasting in ESLD are limited. More recent data have shown a potential benefit for branched-chain amino acid supplementation in terms of delaying liver disease progression and improving anorexia and health-related quality of life [102,103]. Improvement in drug delivery (capsule form versus unpalatable powder) and subsequent compliance with therapy is likely responsible for the attainment of beneficial effects. Further studies are needed to verify these findings.

214

SANCHEZ & TALWALKAR

LIMITATIONS AND FUTURE DIRECTIONS The effectiveness of palliative care has been measured in terms of symptom improvement, improved health status, and patient satisfaction with care. The evidence to support these observations is not supported, however, by rigorous investigations based on interventions in a randomized setting. Additionally, the studies to date mainly focus on patients with advanced malignancy rather than advanced organ failure, such as ESLD. Further study in patients with ESLD is needed to define the important aspects of palliative care that are grounded in evidence-based medicine. SUMMARY The proportion of patients with ESLD who will be managed without liver transplantation will increase in the near future, largely as a result of the increasing age of the population. Patients with ESLD are subject to many physical and psychosocial symptoms that negatively affect health-related quality of life. Sleep quality should be maximized by controlling pruritus and leg cramps. Many frequently used therapies are not supported by a strong evidence base. Advance directives should be addressed with all patients with ESLD, preferably in the outpatient setting before an acute deterioration. Medicare provides a hospice benefit for patients with ESLD, and referral to a hospice is appropriate for patients with an expected survival of 6 months or less. References [1] Zoli M, Cordiani MR, Marchesini G, et al. Prognostic indicators in compensated cirrhosis. Am J Gastroenterol 1991;86:1508–13. [2] Fattovich G, Giustina G, Degos F, et al. Morbidity and mortality in compensated cirrhosis type C: a retrospective follow-up study of 384 patients. Gastroenterology 1997;112: 463–72. [3] Gentilini P, Laffi G, La Villa G, et al. Long course and prognostic factors of virus-induced cirrhosis of the liver. Am J Gastroenterol 1997;92:66–72. [4] Gines P, Quintero E, Arroyo V, et al. Compensated cirrhosis: natural history and prognostic factors. Hepatology 1987;7:122–8. [5] Hashizume M, Inokuchi K, Beppu K, et al. The natural history of non-alcoholic cirrhosis. Gastroenterol Jpn 1984;19:430–5. [6] Hu KQ, Tong MJ. The long-term outcomes of patients with compensated hepatitis C virusrelated cirrhosis and history of parenteral exposure in the United States. Hepatology 1999;29:1311–6. [7] Okazaki I, Maruyama K, Funatsu K, et al. Ten year survival rate of 131 patients with liver cirrhosis excluded the association of liver carcinoma at the establishment of diagnosis. Gastroenterol Jpn 1980;15:350–4. [8] Realdi G, Fattovich G, Hadziyannis S, et al. Survival and prognostic factors in 366 patients with compensated cirrhosis type B: a multicenter study. The Investigators of the European Concerted Action on Viral Hepatitis (EUROHEP). J Hepatol 1994;21:656–66. [9] Serfaty L, Aumaitre H, Chazouilleres O, et al. Determinants of outcome of compensated hepatitis C virus-related cirrhosis. Hepatology 1998;27:1435–40. [10] Sugimura T, Tsuji Y, Sakamoto M, et al. Long-term prognosis and prognostic factors of liver cirrhosis in the 1980s. J Gastroenterol Hepatol 1994;9:154–61. [11] Tanaka R, Itoshima T, Nagashima H. Follow-up study of 582 liver cirrhosis patients for 26 years in Japan. Liver 1987;7:316–24.

PALLIATIVE CARE FOR ESLD

215

[12] D’Amico G, Morabito A, Pagliaro L, Marubini E. Survival and prognostic indicators in compensated and decompensated cirrhosis. Dig Dis Sci 1986;31:468–75. [13] Hoyert DL, Kung HC, Smith BL. Deaths: preliminary data for 2003. Natl Vital Stat Rep 2005;53:1–48. [14] Bustamante J, Rimola A, Ventura PJ, et al. Prognostic significance of hepatic encephalopathy in patients with cirrhosis. J Hepatol 1999;30:890–5. [15] Christensen E, Krintel JJ, Hansen SM, et al. Prognosis after the first episode of gastrointestinal bleeding or coma in cirrhosis: survival and prognostic factors. Scand J Gastroenterol 1989;24:999–1006. [16] Fernandez-Esparrach G, Sanchez-Fueyo A, Gines P, et al. A prognostic model for predicting survival in cirrhosis with ascites. J Hepatol 2001;34:46–52. [17] Llach J, Gines P, Arroyo V, et al. Prognostic value of arterial pressure, endogenous vasoactive systems, and renal function in cirrhotic patients admitted to the hospital for the treatment of ascites. Gastroenterology 1988;94:482–7. [18] Magliocchetti N, Torchio P, Corrao G, et al. Prognostic factors for long-term survival in cirrhotic patients after the first episode of liver decompensation. Ital J Gastroenterol Hepatol 1997;29:38–46. [19] Salerno F, Borroni G, Moser P, et al. Survival and prognostic factors of cirrhotic patients with ascites: a study of 134 outpatients. Am J Gastroenterol 1993;88:514–9. [20] Carithers RL Jr. Liver transplantation. American Association for the Study of Liver Diseases. Liver Transpl 2000;6:122–35. [21] 2004 Annual Report of the US Organ Procurement and Transplantation Network and the Scientific Registry of Transplant Recipients: Transplant Data 1994–2003. Rockville (MD): Department of Health and Human Services, Health Resources and Services Administration, Healthcare Systems Bureau, Division of Transplantation; Richmond (VA): United Network for Organ Sharing; Ann Arbor (MI): University Renal Research and Education Association; 2004. [22] Saunders JB, Walters JR, Davies AP, Paton A. A 20-year prospective study of cirrhosis. BMJ (Clin Res Ed) 1981;282:263–6. [23] Carbonell N, Pauwels A, Serfaty L, et al. Improved survival after variceal bleeding in patients with cirrhosis over the past two decades. Hepatology 2004;40:652–9. [24] Sanyal AJ, Genning C, Reddy KR, et al. The North American Study for the Treatment of Refractory Ascites. Gastroenterology 2003;124:634–41. [25] Child CI, Turcotte J. Surgery and portal hypertension. In: Child CI, editor. The liver and portal hypertension. Philadelphia: Saunders; 1964. p. 49–64. [26] Pugh RN, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646–9. [27] Christensen E, Schlichting P, Fauerholdt L, et al. Prognostic value of Child-Turcotte criteria in medically treated cirrhosis. Hepatology 1984;4:430–5. [28] Kamath PS, Wiesner RH, Malinchoc M, et al. A model to predict survival in patients with end-stage liver disease. Hepatology 2001;33:464–70. [29] Freeman RB, Wiesner RH, Edwards E, et al. Results of the first year of the new liver allocation plan. Liver Transpl 2004;10:7–15. [30] Wiesner RH, McDiarmid SV, Kamath PS, et al. MELD and PELD: application of survival models to liver allocation. Liver Transpl 2001;7:567–80. [31] Malinchoc M, Kamath PS, Gordon FD, et al. A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology 2000;31: 864–71. [32] National Consensus Project for Quality Palliative Care. Clinical practice guidelines for quality palliative care. Available at: http://www.aahpm.org/positions. Accessed April 1, 2005. [33] Hauptman PJ, Havranek EP. Integrating palliative care into heart failure care. Arch Intern Med 2005;165:374–8.

216

SANCHEZ & TALWALKAR

[34] WHO definition of palliative care. Available at: http://www.who.int/cancer/palliative/ definition/en. Accessed April 1, 2005. [35] Medical guidelines for determining prognosis in selected non-cancer diagnoses. Arlington (VA): National Hospice and Palliative Care Organization; 1996. [36] Fox E, Landrum-McNiff K, Zhong Z, et al. Evaluation of prognostic criteria for determining hospice eligibility in patients with advanced lung, heart, or liver disease. SUPPORT Investigators. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments. JAMA 1999;282:1638–45. [37] Patrick DL, Curtis JR, Engelberg RA, et al. Measuring and improving the quality of dying and death. Ann Intern Med 2003;139(5 Pt 2):410–5. [38] Freeborne N, Lynn J, Desbiens NA. Insights about dying from the SUPPORT project. The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments. J Am Geriatr Soc 2000;48(5 suppl):S199–205. [39] Lin MH, Wu PY, Tsai ST, et al. Hospice palliative care for patients with hepatocellular carcinoma in Taiwan. Palliat Med 2004;18:93–9. [40] Wachter RM, Luce JM, Hearst N, Lo B. Decisions about resuscitation: inequities among patients with different diseases but similar prognoses. Ann Intern Med 1989;111: 525–32. [41] Pantilat SZ. Care of dying patients: beyond symptom management. West J Med 1999;171:253–6. [42] Runyon BA. Management of adult patients with ascites caused by cirrhosis. Hepatology 1998;27:264–72. [43] Reynolds TB, Geller HM, Kuzma OT, Redeker AG. Spontaneous decrease in portal pressure with clinical improvement in cirrhosis. N Engl J Med 1960;263:734–9. [44] Fogel MR, Sawhney VK, Neal EA, et al. Diuresis in the ascitic patient: a randomized controlled trial of three regimens. J Clin Gastroenterol 1981;3(Suppl 1):73–80. [45] Perez-Ayuso RM, Arroyo V, Planas R, et al. Randomized comparative study of efficacy of furosemide versus spironolactone in nonazotemic cirrhosis with ascites: relationship between the diuretic response and the activity of the renin-aldosterone system. Gastroenterology 1983;84(5 Pt 1):961–8. [46] Sola-Vera J, Minana J, Ricart E, et al. Randomized trial comparing albumin and saline in the prevention of paracentesis-induced circulatory dysfunction in cirrhotic patients with ascites. Hepatology 2003;37:1147–53. [47] Riggio O, Merlli M, Pedretti G, et al. Hepatic encephalopathy after transjugular intrahepatic portosystemic shunt: incidence and risk factors. Dig Dis Sci 1996;41:578–84. [48] Sanyal AJ, Freedman AM, Shiffman ML, et al. Portosystemic encephalopathy after transjugular intrahepatic portosystemic shunt: results of a prospective controlled study. Hepatology 1994;20(1 Pt 1):46–55. [49] Sanyal AJ, Freedman AM, Luketic VA, et al. Transjugular intrahepatic portosystemic shunts compared with endoscopic sclerotherapy for the prevention of recurrent variceal hemorrhage: a randomized, controlled trial. Ann Intern Med 1997;126:849–57. [50] Rosado B, Kamath PS. Transjugular intrahepatic portosystemic shunts: an update. Liver Transpl 2003;9:207–17. [51] Somberg KA, Riegler JL, LaBerge JM, et al. Hepatic encephalopathy after transjugular intrahepatic portosystemic shunts: incidence and risk factors. Am J Gastroenterol 1995;90: 549–55. [52] Sanyal AJ, Freedman AM, Luketic VA, et al. The natural history of portal hypertension after transjugular intrahepatic portosystemic shunts. Gastroenterology 1997;112:889–98. [53] Kamath PS, McKusick MA. Transvenous intrahepatic portosystemic shunts. Gastroenterology 1996;111:1700–5. [54] Casado M, Bosch J, Garcia-Pagan JC, et al. Clinical events after transjugular intrahepatic portosystemic shunt: correlation with hemodynamic findings. Gastroenterology 1998; 114:1296–303.

PALLIATIVE CARE FOR ESLD

217

[55] Stanley MM, Ochi S, Lee KK, et al. Peritoneovenous shunting as compared with medical treatment in patients with alcoholic cirrhosis and massive ascites. Veterans Administration Cooperative Study on Treatment of Alcoholic Cirrhosis with Ascites. N Engl J Med 1989;321:1632–8. [56] Bories P, Garcia Compean D, Michel H, et al. The treatment of refractory ascites by the LeVeen shunt: a multi-centre controlled trial (57 patients). J Hepatol 1986;3:212–8. [57] Zervos EE, McCormick J, Goode SE, Rosemurgy AS. Peritoneovenous shunts in patients with intractable ascites: palliation at what price? Am Surg 1997;63:157–62. [58] Reisfield GM, Wilson GR. Management of intractable, cirrhotic ascites with an indwelling drainage catheter. J Palliat Med 2003;6:787–91. [59] Richard HM 3rd, Coldwell DM, Boyd-Kranis RL, et al. Pleurx tunneled catheter in the management of malignant ascites. J Vasc Interv Radiol 2001;12:373–5. [60] Iyengar TD, Herzog TJ. Management of symptomatic ascites in recurrent ovarian cancer patients using an intra-abdominal semi-permanent catheter. Am J Hosp Palliat Care 2002;19:35–8. [61] Liu LU, Haddadin HA, Bodian CA, et al. Outcome analysis of cirrhotic patients undergoing chest tube placement. Chest 2004;126:142–8. [62] Lazaridis KN, Frank JW, Krowka MJ, Kamath PS. Hepatic hydrothorax: pathogenesis, diagnosis, and management. Am J Med 1999;107:262–7. [63] Siegerstetter V, Deibert P, Ochs A, et al. Treatment of refractory hepatic hydrothorax with transjugular intrahepatic portosystemic shunt: long-term results in 40 patients. Eur J Gastroenterol Hepatol 2001;13:529–34. [64] Gordon FD, Anastopoulos HT, Crenshaw W, et al. The successful treatment of symptomatic, refractory hepatic hydrothorax with transjugular intrahepatic portosystemic shunt. Hepatology 1997;25:1366–9. [65] Arguedas MR, DeLawrence TG, McGuire BM. Influence of hepatic encephalopathy on health-related quality of life in patients with cirrhosis. Dig Dis Sci 2003;48: 1622–6. [66] Das A, Dhiman RK, Saraswat VA, et al. Prevalence and natural history of subclinical hepatic encephalopathy in cirrhosis. J Gastroenterol Hepatol 2001;16:531–5. [67] Riordan SM, Williams R. Treatment of hepatic encephalopathy. N Engl J Med 1997;337: 473–9. [68] Ferenci P, Lockwood A, Mullen K, et al. Hepatic encephalopathy—definition, nomenclature, diagnosis, and quantification: final report of the working party at the 11th World Congresses of Gastroenterology,Vienna,1998. Hepatology 2002;35:716–21. [69] Blei AT, Cordoba J. Practice Parameters Committee of the American College of Gastroenterology Hepatic Encephalopathy. Am J Gastroenterol 2001;96:1968–76. [70] Mas A, Rodes J, Sunyer L, et al. Comparison of rifaximin and lactitol in the treatment of acute hepatic encephalopathy: results of a randomized, double-blind, double-dummy, controlled clinical trial. J Hepatol 2003;38(1):51–8. [71] Mela M, Mancuso A, Burroughs AK. Review article: pruritus in cholestatic and other liver diseases. Aliment Pharmacol Ther 2003;17:857–70. [72] Datta DV, Sherlock S. Cholestyramine for long term relief of the pruritus complicating intrahepatic cholestasis. Gastroenterology 1966;50:323–32. [73] Javitt NB. Timing of cholestyramine doses in cholestatic liver disease. N Engl J Med 1974;290:1328–9 [letter]. [74] Miguet JP, Mavier P, Soussy CJ, Dhumeaux D. Induction of hepatic microsomal enzymes after brief administration of rifampicin in man. Gastroenterology 1977;72(5 Pt 1):924–6. [75] Podesta A, Lopez P, Terg R, et al. Treatment of pruritus of primary biliary cirrhosis with rifampin. Dig Dis Sci 1991;36:216–20. [76] Ghent CN, Carruthers SG. Treatment of pruritus in primary biliary cirrhosis with rifampin: results of a double-blind, crossover, randomized trial. Gastroenterology 1988;94: 488–93.

218

SANCHEZ & TALWALKAR

[77] Bergasa NV, Alling DW, Talbot TL, et al. Effects of naloxone infusions in patients with the pruritus of cholestasis: a double-blind, randomized, controlled trial. Ann Intern Med 1995;123:161–7. [78] Bergasa NV, Schmitt JM, Talbot TL, et al. Open-label trial of oral nalmefene therapy for the pruritus of cholestasis. Hepatology 1998;27:679–84. [79] Terg R, Coronel E, Sorda J, et al. Efficacy and safety of oral naltrexone treatment for pruritus of cholestasis, a crossover, double blind, placebo-controlled study. J Hepatol 2002;37: 717–22. [80] Thornton JR, Losowsky MS. Opioid peptides and primary biliary cirrhosis. BMJ 1988;297: 1501–4. [81] Wolfhagen FH, Sternieri E, Hop WC, et al. Oral naltrexone treatment for cholestatic pruritus: a double-blind, placebo-controlled study. Gastroenterology 1997;113: 1264–9. [82] Browning J, Combes B, Mayo MJ. Long-term efficacy of sertraline as a treatment for cholestatic pruritus in patients with primary biliary cirrhosis. Am J Gastroenterol 2003;98: 2736–41. [83] El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999;340:745–50. [84] El-Serag HB, Davila JA, Petersen NJ, McGlynn KA. The continuing increase in the incidence of hepatocellular carcinoma in the United States: an update. Ann Intern Med 2003;139:817–23. [85] Llovet JM, Fuster J, Bruix J. Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology 1999;30:1434–40. [86] Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693–9. [87] Ishii H, Okada S, Nose H, et al. Local recurrence of hepatocellular carcinoma after percutaneous ethanol injection. Cancer 1996;77:1792–6. [88] Tateishi R, Shiina S, Teratani T, et al. Percutaneous radiofrequency ablation for hepatocellular carcinoma: an analysis of 1000 cases. Cancer 2005;103:1201–9. [89] Lin SM, Lin CJ, Lin CC, et al. Radiofrequency ablation improves prognosis compared with ethanol injection for hepatocellular carcinoma < or ¼ 4 cm. Gastroenterology 2004;127: 1714–23. [90] Llovet JM, Bruix J. Systematic review of randomized trials for unresectable hepatocellular carcinoma: chemoembolization improves survival. Hepatology 2003;37:429–42. [91] Hassoun Z, Gores GJ. Treatment of hepatocellular carcinoma. Clin Gastroenterol Hepatol 2003;1:10–8. [92] Patt YZ, Hassan MM, Lozano RD, et al. Thalidomide in the treatment of patients with hepatocellular carcinoma: a phase II trial. Cancer 2005;103:749–55. [93] Wang TE, Kao CR, Lin SC, et al. Salvage therapy for hepatocellular carcinoma with thalidomide. World J Gastroenterol 2004;10:649–53. [94] Singh N, Gayowski T, Wagener MM, Marino IR. Depression in patients with cirrhosis: impact on outcome. Dig Dis Sci 1997;42:1421–7. [95] Cordoba J, Cabrera J, Lataif L, et al. High prevalence of sleep disturbance in cirrhosis. Hepatology 1998;27:339. [96] Goldblatt J, Taylor PJ, Lipman T, et al. The true impact of fatigue in primary biliary cirrhosis: a population study. Gastroenterology 2002;122:1235. [97] Sene D, Limal N, Cacoub P. Hepatitis C virus-associated extrahepatic manifestations: a review. Metab Brain Dis 2004;19:357. [98] Marchesini G, Bianchi G, Amodio P, et al. Factors associated with poor health-related quality of life of patients with cirrhosis. Gastroenterology 2001;120:170–8. [99] Verne GN, Soldevia-Pico C, Robinson ME, et al. Autonomic dysfunction and gastroparesis in cirrhosis. J Clin Gastroenterol 2004;38:72–6. [100] Plauth M, Schutz ET. Cachexia in liver cirrhosis. Int J Cardiol 2002;85:83–7.

PALLIATIVE CARE FOR ESLD

219

[101] Strasser F, Bruera ED. Update on anorexia and cachexia. Hematol Oncol Clin North Am 2002;16:589–617. [102] Marchesini G, Bianchi G, Merli M, et al. Nutritional supplementation with branched-chain amino acids in advanced cirrhosis: a double-blind, randomized trial. Gastroenterology 2003;124:1792–801. [103] Fukushima H, Miwa Y, Ida E, et al. Nocturnal branched-chain amino acid administration improves protein metabolism in patients with liver cirrhosis: comparison with daytime administration. J Parenter Enteral Nutr 2003;27:315–22.