Non-Penetrating Glaucoma Surgery
Dedication To the women in our lives: Loty Mermoud, Samia Nada, Marianne Mermoud, Ghada Ibrahim, Sophie Mermoud, and Heba Shaarawy. In recognition of their undivided love, their unlimited tenderness, and their uncompromising strength. We salute and thank you. André and Tarek
Non-Penetrating Glaucoma Surgery Edited by
André Mermoud
MD, PD
Head Glaucoma Unit Hôpital Ophtalmique Jules Gonin University of Lausanne Lausanne Switzerland
Tarek Shaarawy
MD
Head Glaucoma Unit Memorial Research Institute of Ophthalmology Giza Egypt Foreword by
Robert Ritch
MD
Chief Glaucoma Service The New York Eye and Ear Infirmary New York USA
Martin Dunitz
© 2001 Martin Dunitz Ltd, a member of the Taylor & Francis Group First published in the United Kingdom in 2001 by Martin Dunitz Ltd, The Livery House, 7–9 Pratt Street, London NW1 0AE This edition published in the Taylor & Francis e-Library, 2003. Tel.: +44 (0) 20 7482-2202 Fax.: +44 (0) 20 7267-0159 E-mail:
[email protected] Website: http://www.dunitz.co.uk All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher or in accordance with the provisions of the Copyright, Designs and Patents Act 1988 or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP. A CIP record for this book is available from the British Library. ISBN 0-203-21537-0 Master e-book ISBN Distributed in the United States by: ISBN 0-203-27180-7 (Adobe eReader Format) ISBN 1 84184 042 4 (Print Edition) Tel.: 1-800-215-1000 Distributed in Brazil by: Ernesto Reichmann Distribuidora de Livros, Ltda Tatuape 03440-000, São Paulo
Composition by Scribe Design, Gillingham, Kent, UK
Contents
1
2
3
4
5
Contributors
vii
Foreword
ix
Acknowledgements
xi
The history of filtering surgery Howard C Cohn
1
6
Experimental studies in non-penetrating glaucoma surgery 67 Christophe Nguyen and Tarek Shaarawy
7
Indications and contraindications for non-penetrating glaucoma surgery 87 Elie Dahan
8
Surgical technique 97 André Mermoud and Emilie Ravinet
9
Viscocanalostomy Robert Stegmann and Roberto G Carassa
Evolution of non-penetrating glaucoma surgery 13 André Mermoud Anatomical features of outflow pathway Farid Achache
21
How does non-penetrating glaucoma surgery work? 33 Douglas H Johnson and Mark Johnson Mechanisms of filtration in non-penetrating filtering surgeries 57 André Mermoud and Emilie Ravinet
10 Modulation of wound healing Tarek Shaarawy
109
117
11 Postoperative management of non-penetrating filtering surgery 125 Tarek Shaarawy
vi
Contents
12 Complications and reoperations 139 André Mermoud and Emilie Ravinet
15 Implants in non-penetrating filtering surgery 177 Corinne C Schnyder and Emilie Ravinet
13 Results of non-penetrating glaucoma surgery 161 Tarek Shaarawy
16 Erbium:YAG laser-assisted deep sclerectomy 185 Wolfgang E Lieb
14 Phacoemulsification combined with viscocanalostomy and deep sclerectomy 169 Fathi El-Sayyad
Index
195
Contributors Farid Achache, MD Hôpital Ophtalmique Jules Gonin, University of Lausanne, Lausanne, Switzerland.
André Mermoud, MD, PD Hôpital Ophtalmique Jules Gonin, University of Lausanne, Lausanne, Switzerland
Roberto G Carassa, MD Department of Ophtalmology and Visual Sciences, University S. Raffaele, Milano, Italy.
Christophe Nguyen, MD Hôpital Ophtalmique Jules Gonin, University of Lausanne, Lausanne, Switzerland.
Howard C Cohn, MD 45 Rue Vineuse, Paris 75016, France.
Emilie C M Ravinet, MD Hôpital Ophtalmique Jules Gonin, University of Lausanne, Lausanne, Switzerland.
Elie Dahan, MD University of the Witwatersrand and Oxford Eye Center, Johannesburg, South Africa. Fathi El-Sayyad, FRCS Magrabi Eye Hospital, Cairo, Egypt. Douglas H Johnson, MD Professor of Ophtalmology, Mayo Clinic, Rochester MN, USA. Mark Johnson, MD Biomedical Engineering Department, Northwestern University, Evanston, IL, USA. Wolfgang E Lieb, MD Professor of Ophthalmology, Julius-Maximilians-University, Würzburg, Germany.
Robert Ritch, MD Chief, Glaucoma Service, The New York Eye and Ear Infirmary, New York, USA. Corinne C Schnyder, MD Hôpital Ophtalmique Jules Gonin. University of Lausanne. Lausanne, Switzerland Tarek Shaarawy Hôpital Ophtalmique Jules Gonin, University of Lausanne, Lausanne, Switzerland Robert Stegmann, MBChB, Mmed Medical University of Southern Africa
Foreword Surgical procedures to lower intraocular pressure in glaucoma were first developed in the middle of the 19th century, although the role of a filtration bleb in successful surgery for open-angle glaucoma was not recognized initially. In 1869, DeWecker was the first to consider an anterior sclerotomy successful only if filtration continued postoperatively. Over the next half century, a number of procedures were described. Limbal trephination, introduced by Elliot in 1909, became the most popular operation until the 1940s, when it fell out of favor because the very thin conjunctival bleb predisposed to late endophthalmitis. Thermal cautery of the scleral wound edges with entry into the anterior chamber was described by Preziosi in 1924. Scheie’s modification, thermal sclerostomy, and posterior lip sclerectomy became the most widely used operations. These full thickness procedures were complicated by frequent flat anterior chambers, choroidal detachments, subsequent cataract formation, and late bleb leaks and endophthalmitis. Guarded filtration procedures were developed in the hope of avoiding these complications. In 1968, Cairns reported good success using microsurgical techniques to perform a trabeculectomy under a scleral flap, which was hinged either posteriorly in the sclera or anteriorly at the limbus. Although the incidence of late endophthalmitis was significantly reduced, the intraocular pressure was not lowered to as great an extent and certain subsets of patients, such as those with neovascular glaucoma, uveitis, and previous surgery had high rates of failure. Nevertheless, trabeculectomy rapidly became the procedure
of choice and has remained so for the past 30 years. The antifibrosis agents 5-fluorouracil and mitomycin C were introduced in an attempt to provide the pressure-lowering effect of fullthickness surgery with the safety of trabeculectomy. Surgical success rates were notably improved, particularly for complicated glaucomas, and they have become routinely used in virtually all filtration procedures. However, the thin, avascular blebs produced, particularly with mitomycin C, have led to a resurgence of chronic hypotony, hypotony maculopathy, bleb leaks, bleb dysesthesia, bleb infections, and endophthalmitis. There is an increasing consensus that new approaches to glaucoma surgery are needed. Ideally, these approaches should achieve a more physiological means of lowering intraocular pressure, be capable of reducing pressure to the low teens necessary to prevent progression of glaucomatous damage in most patients, and be free of the complications noted above. Non-penetrating surgery, originally attempted by Krasnov and by Zimmerman, with limited success, offers the promise of successful lowering of intraocular pressure with significant reduction in the frequency of the complications of shallow anterior chamber and hypotony. Two recently developed approaches, deep sclerectomy devised by André Mermoud, and viscocanalostomy, devised by Robert Stegmann, have seen increasing popularity among glaucoma surgeons. These procedures involve excising a portion of sclera under a larger scleral flap to make a scleral lake, the maintenance of which is enhanced by collagen
x
Foreword
implants or viscoelastics. The procedure is technically more difficult to perform compared to trabeculectomy and there is a longer learning curve. Reported results vary from equivalence in effect to trabeculectomy at the hands of the most experienced surgeons to a greater failure rate than trabeculectomy in many case series. Nevertheless, the potential opportunity to create a filtering procedure which successfully controls intraocular pressure in the absence of a bleb is a driving force to modify and improve upon non-penetrating glaucoma surgery. I believe that the momentum developing will provide the impetus for innovation in
techniques and methodology, probably in conjunction with newer developments in wound healing modification, to bring about a procedure which will take the place of trabeculectomy, just as trabeculectomy took the place of full-thickness procedures. This book, edited by André Mermoud and Tarek Shaarawy, is the first, and most likely not the last, to bring together a developing field in order to provide a comprehensive overview of concepts and techniques of non-penetrating glaucoma surgery. Robert Ritch, MD The New York Eye and Ear Infirmary
Acknowledgements In the rich, intricate tapestry of life people meet. These encounters can be brief or prolonged, but sometimes an everlasting effect emerges, and a person’s life will never be the same again. Therefore, first and foremost, we wish to thank our mentors, Professor T Murray, Professor C Gailloud, Professor JB Bourke, Professor T Souidan, Professor S Galal, Professor T El-Emary, Dr R Faggioni, Dr G Baerveldt, Dr J Salmon, and Dr A Azab, for their uncompromising principles, and the way in which the professoround integrity of each one of them has inspired our careers. We would also like to acknowledge with appreciation all the contributors for their excellent contributions. We are extremely
grateful and indebted to Professor Robert Ritch for writing the foreword. Also, we wish to thank our publisher Alan Burgess, of Martin Dunitz Ltd, for his patience and understanding, as well as Charlotte Mossop, the project editor, who saw this project through to timely completion. Not far behind the authors is a dedicated clan of supporters. They include our book coordinator Dr L Bolle, our talented illustrator Miss C Darphin, Mr M Curchod, the photography department of Jules Gonin Eye hospital, and our executive secretaries. This book is the joint effort of a group of people who enjoy what they do for a living. André Mermoud Tarek Shaarawy April 2001
1 The history of filtering surgery Howard C Cohn
The ancient Greeks and Romans did not differentiate between glaucoma and cataract. The term glaucoma in Greek was used to describe a general glazed appearance of the pupil. In the time of Hippocrates (460–377 BC) all maladies of the eye were attributed to ‘disturbed or ill humors’. The two conditions were not differentiated until the time of Celsius (25 BC–AD 50) and later Galen (AD 131–210): cataract was treatable, glaucoma was not (Ref. 1, page 380). For problems involving sight or the eye, from the second century BC various ‘magic’ eye drops were concocted containing zinc, copper, mydriatics, and other substances including albumin, saliva, mother’s milk, children’s urine, crocodile and lion bile. 2 The first person to suggest an association between raised intraocular pressure (IOP) and glaucoma was At Tabari in the 10th century, followed by Sams-ad-Din of Cairo in the 14th century who described ‘a migraine of the eye or headache of the pupil, an illness associated with pain in the eye, hemicrania and dullness of the humors, followed by dilatation of the pupil and cataract. If it becomes chronic, tenseness of the eye and blindness intervene’. But it was not until 1622 that the first original description of glaucoma as distinct from cataract was described by Richard Banister: ‘The humour settled in the hollow nerves, be growne to any solid or hard substance, it is not possible to be cured. If one feele the Eye by rubbing upon the
Eie-lids, that the Eye be growne more solid and hard than naturally it should be.’ Pierre and Antoine-Pierre Demours gave an excellent description of the association of raised IOP and glaucoma. In the early 19th century Guthry, Lawrence, and Donders described two separate conditions with raised IOP: acute inflammatory syndrome and non-congestive or simple glaucoma (Ref. 1, pages 381–83). In the late 18th century and early 19th century, the cause of glaucoma was ascribed to gouty iritis (Beer, 1792), serous choroïditis (Mackenzie, 1835), thickening of the sclera (Coccius, 1867), among others (Ref. 1, page 387).
Iridectomy and acute glaucoma Von Graefe3 in 1857 described the beneficial effect of an iridectomy in treating acute glaucoma but noted no improvement in cases of chronic glaucoma treated by the same method. He did describe grossly cystoid cicatrices in about 20% of eyes undergoing iridectomy, but thought they were less desirable than a smooth well-healed, scleral incision.4 At this period in history the difference between open-angle glaucoma and angleclosure glaucoma was in the process of being
2
The history of filtering surgery
discovered. Before the difference was recognized there were various theories proposed as to how an iridectomy lowered IOP. As reviewed by Lagrange,5 Donders proposed that iridectomy reduced secretion of aqueous humor by reflex pathways, whereas Fuchs and Axenfeld thought that the liquid went under the choroid to be resorbed. Bowman and Ulrich proposed that vitreous passed through the iridectomy into the anterior chamber to ‘disappear by osmosis’. Knies in 1876 and Weber in 1877 were on the right track when they described the high frequency of anteriorchamber obstruction in acute glaucoma. After the introduction of the gonioscope by Salzmann in 1914, Seidel, Curran, and Raeder in the 1920s introduced the concept of pupillary block, and the explanation of why iridectomy cured angle-closure glaucoma but not chronic simple (open-angle) glaucoma. Curran showed that a peripheral (and not sector) iridectomy was sufficient (Ref.1, page 387). Various surgical procedures have been proposed over time for the treatment of glaucoma according to our progressive understanding of the disease process.
Simple sclerotomy Mackenzie in 1835 was the first to do an invasive procedure—sclerotomy—designed to treat serous choroiditis, the cause he proposed for simple glaucoma; he later added a paracentesis.
Iridodesis Because a paracentesis rapidly closed, Critchett6 in 1858 proposed drawing a piece of iris
into the corneal wound to facilitate drainage by ‘iris inclusion’ or iridodesis. A broad needle was used to make a corneal incision at the limbus. The iris was drawn into the wound with a blunt hook and left as is with the protruding part excised. Kronfeld7 aptly called it ‘a technically simple but otherwise horrible method to relieve pupillary block... not likely to give rise to a lasting safe outlet for aqueous’.
Anterior sclerotomy Louis de Wecker may be regarded as the father of glaucoma filtering surgery. He was the first to realize that it was the scleral incision and not the excision of iris that was responsible for the pressure lowering of von Graefe’s iridectomy in some cases of glaucoma.8 He described the filtering cicatrix9 as a desirable result, and said that it was not the nuisance with risk of complications that von Graefe thought. De Wecker in 1869–71 described the anterior sclerotomy with a Graefe type knife (Ref. 1, page 528). An incision and counter incision were made just behind the limbus with the knife drawn up toward the limbus as for a cataract incision but leaving the limbus intact. The goal was to form a filtering bleb but the incision soon closed. De Wecker added an iridectomy and Dianoux in 1905 proposed prolonged massage, neither of which were successful.10
Small-flap sclerotomy In 1903, Major Herbert proposed a small-flap sclerotomy where a small incision was made into the anterior chamber through the sclera
Small-flap sclerotomy
(a)
(b)
(c)
(d)
3
Figure 1.1 Lagrange’s sclerecto-iridectomy (1906). (a) Graefe knife incision through the limbus; (b) Incision carried superiorly and posteriorly exiting the sclera 2–3 mm behind the limbus and exiting the conjunctiva another 3 mm further on; (c) Sclerectomy; (d) Sector iridectomy; (e) Final appearance: no sutures used. (e)
behind and parallel to the limbus. A small limbus-based sclera flap was then raised and iris was incarcerated into the wound. There was no resection of scleral tissue. Herbert also
tried infolding of the conjunctiva into scleral incisions but obtained few long-term successes.7
4
The history of filtering surgery
Sclerecto-iridectomy Lagrange5 described a sclerecto-iridectomy in 1906 in which a corneoscleral conjunctival flap was created with a Graefe knife. (Fig. 1.1) A sclerectomy of the anterior lip was done with scissors, followed by a basal iridectomy. No sutures were used. Aqueous humor then had free access to the subconjunctival space. It is interesting to see that his results are reported mainly in terms of visual function: visual acuity and the visual field, besides the presence of a filtering cicatrice. (The IOP could not yet be measured accurately and was described only as normal, slightly, or greatly elevated by digital testing.) Of 15 cases followed-up for at least 6 months, visual acuity was the same in 12 and improved in 3. The visual field was stable in 13 and improved in 2. O’Brian (Ref. 1, page 534) in 1947 described 85% success with Lagrange’s sclerecto-iridectomy. Lagrange5 summarized the mainstream of thought at the turn of the century in his 1907 article quoting Priestly-Smith ‘the only effective treatment of chronic glaucoma is the creation of a sclero-corneal subconjunctival fistula’, and Critchett who said: ‘the operator who finds the means of assuring formation of this filtering cicatrice will have rendered a great service to humanity.’
Iridencleisis Holth11 reported iridencleisis in 1908, after noting that well-healed incisions for iridectomies in chronic glaucoma did not work as well as did irregular wounds where there was often a piece of iris incarcerated and a filtering cicatrice. The iridencleisis procedure begins with a triangular 6 mm lance incision made in
the conjunctiva 5 mm behind the limbus, then the lance is advanced into the anterior chamber at the limbus. A sector iridectomy is done with incarceration of one or both iris pillars into the scleral wound. Various modifications of the procedure are described; no sutures were used. Holth describes the postoperative care: ‘Of course, the patient is left in the dark for the first 4 days after the operation to avoid constriction of the pupil...’ He had 85% success in 34 cases with normal IOP re-established as measured by the new Schiotz tonometer. All successful cases had persistent ‘conjunctival edema’.
Posterior trephination Argyll-Robertson, one of the first to introduce the concept of producing a filtering scar by sclerectomy, proposed posterior trephining in 1876 at the junction of the pars plana and ciliary body. Four cases with qualified success were reported. (Ref. 1, page 529).
Limbal trephination Elliot12 in 1909 described a technique of limbal trephination as an easier operation than Herbert’s small-flap sclerotomy or Lagrange’s sclerecto-iridectomy. The operation was done under cocaine and adrenaline local anesthesia. However, if necessary a ‘hypodermic of morphine’ was used. A conjunctival flap was dissected either at the superior limbus or inferior limbus (Fig. 1.2), Elliot noted that it was often easier to approach the inferior limbus of an eye of an anxious patient who tended to stay in upgaze. A trephine of 2 mm
Posterior-lip thermal sclerotomy
5
trephine hole too far posterior and entering the suprachoroidal space. If the bulging uveal tissue is excised vitreous may present. Elliot concluded: ‘To some the publication of the present paper may seem premature, the ideas embodied in it have been so long before the writer’s mind that he has wondered that others have not anticipated him. He is convinced that it is founded on sound principles and it is obviously very easy to perform the operation. He hopes that those who have been deterred from attempting the more difficult procedures of other surgeons will try this. Its technique is within the reach of all. Operative skill can scarcely be said to be required.’
Thermal sclerostomy
Figure 1.2 Elliot’s trephination procedure (1909).
diameter was used. A strong myotic, eserine, was used at the end of each operation. In 21 of his first 50 cases, an iridectomy was done, but only if the iris presented in the trephine hole at surgery. Two cases had to be reoperated because of iris incarceration. The anterior chamber had reformed the day after surgery in 37 of 50 cases. Intraocular pressure was lowered in all cases. Subconjunctival filtration was noted to be ‘very free’. No cases of ‘septic accident’ were described. The one complication Elliot mentioned was making the
Preziosi13 in 1924 proposed doing a thermal sclerostomy under a conjunctival flap. A galvano cautery was used in the absence of any knife incision to enter the anterior chamber, thereby creating a fistula.
Posterior-lip thermal sclerostomy Scheie14 described a posterior-lip thermal sclerostomy in 1958. He had observed that application of an electric cautery for hemostasis of the posterior lip of an iridectomy incision produced inadvertent filtering blebs. The thinking of the time is illustrated in his introduction: ‘The filtering cicatrice seemed to be best explained by slight retraction of the wound edges resulting from scleral shrinkage cause by the cautery. The fact the filtration
6
The history of filtering surgery
(a)
(b)
(c)
(d)
Figure 1.3 Scheie’s procedure (1958). (a) Ab externo incision perpendicular to sclera 1 mm behind the limbus; (b) Application of cautery to lips of scleral incision with gaping wound; (c) Prolapse of iris; (d) Iris root grasped for peripheral iridectomy. Reproduced by permission of the American Journal of Ophthalmology.)
occurred was surprising because many ophthalmic surgeons have cautioned against the use of cautery even for control of bleeding when performing a filtration operation.’ The technique was done under local anesthesia, (Fig. 1.3) A limbus base conjunctival flap was raised. After initial cauterization of the sclera at the limbus a small scratch
incision is made with the blade through the cauterized area 1 mm behind the limbus. Then cautery is progressively applied to the posterior lip of the incision, which is progressively deepened until the iris prolapses. An iridectomy is done and the conjunctiva and Tenon’s capsule are closed by 6.0 catgut in separate layers. No medication was instilled.
Sinusotomy Scheie presents data on 41 eyes of 30 patients, 14 with angle-closure glaucoma. In the 14 eyes operated for angle-closure glaucoma the IOP was regarded as controlled if repeat readings were under 30 mmHg by Schiotz tonometry. In 21 of 27 eyes with openangle glaucoma the IOP was controlled, although six of these eyes had hypotony. Hypotony was defined as a tension of 10 mmHg or less, but in no instance caused diminution of vision or visual difficulty. The longest follow-up was 14 months. Apart from two small hyphemas, no operative complications were encountered in his series. He did note that the bleb was usually thick and should be much less prone to developing infection compared with the thin polycystic blebs associated with corneo-scleral trephination. Scheie15 compared his thermal sclerostomy with iridencleisis and trephination. In a larger series of 111 eyes with open-angle glaucoma, the success rate of the Scheie procedure was 86% in controlling the IOP. The most frequent complication was flat or shallow anterior chamber: only two-thirds of eyes had a reformed chamber at 3 days. Hyphema occurred in 17 eyes, and hypotony in 20 eyes, but none had disk edema or loss of visual acuity. His success rate with iridencleisis was 83% in 141 eyes, with hypotony noted in just 3.5%. Delayed anterior chamber reformation occurred in seven eyes. The success rate of trephination in achieving IOP control was 98% in 69 eyes with hypotony in 21, none with visual loss. Hyphema occurred in 20 eyes. Scheie used a 1.0 mm or a 1.5 mm trephine. In comparison to the 2 mm trephine originally used by Elliott, the surface area of a 1 mm trephine is four times smaller. Scheie’s results with trephination were better than those of others. Leydhecker16 found the success of Eliott’s operations to be only about 60%; an unfavourable comparison with the 80–90% success of the Scheie procedure. Throughout the
7
1960s and early 1970s, the Scheie procedure was one of the most frequently performed filtering procedures.
Posterior-lip sclerectomy Iliff and Haas17 described a posterior-lip sclerectomy in 1962. Under a limbus-based conjunctival flap a 5 mm incision is made into the anterior chamber and a Holth scleral punch is used to make a scleral opening of 1 mm 3 mm. An iridectomy is done. Haas18 in 1967 described an 85% success with posterior-lip sclerectomy, although complications included flat blebs, flat chambers, and choroidal detachments.
Trabeculotomy Trabeculotomy was described by Harmes and Dannheim19 in 1969 with a 60% success rate for controlling IOP. Trabeculotomy was not designed as a filtering procedure as such, but was supposed to increase outflow facility. Of 300 cases, 12% developed a filtering bleb and 8% had a gross hyphema requiring anteriorchamber washout.
Sinusotomy Krasnov20 published his sinusotomy or externalization of Schlemm’s canal in the 1960s, assuming the site of obstruction to outflow was intrascleral beyond the outer wall of Schlemm’s canal. If the outer wall is opened leaving the inner wall intact, reduction of IOP should be obtained. A resection of a narrow
8
The history of filtering surgery
1.5 mm wide lamella of sclera directly over Schlemm’s canal is made from the 10 o’clock to 2 o’clock position. Krasnov said that care should be taken not to damage the inner wall of Schlemm’s canal, which is the trabecular zone. The moment of reaching Schlemm’s canal is crucial in sinusotomy. If the diagnosis of intrascleral glaucoma should prove correct there is a constant flow of fluid through the undamaged trabecular meshwork. In cases of ‘trabecular insufficiency’ the site over Schlemm’s canal will be more or less dry and a different surgical procedure should be used. Krasnov described 340 cases with a followup of 1–5 years with normalization of IOP in 83%. A prolonged normalization of IOP is usually associated with visible subconjunctival filtration. He concludes: ‘We have now almost completely abandoned conventional fistulising surgery in glaucoma apart from exceptional cases.’
Setons and shunts Many different materials have been implanted into the anterior chamber in an attempt to facilitate filtration. Rollett and Moreau in 1907 placed horse hair through corneal punctures in two cases of absolute glaucoma. Zorab used a silk loop through a keratome incision under the conjunctiva in a procedure he called aqueoplasty. Substances including gold leaf, platinum, various plastic rods and plates were placed in limbal wounds to act as wicks to keep a sclerostomy open. Results overall were poor (Ref. 1, page 543). Since the 1970s posterior tube shunts have achieved a certain degree of success in maintaining functional blebs in eyes where standard filtration procedures are not feasible or have failed. Molteno21 was the first to
report success with his episcleral plate joined to a plastic tube coming from the anterior chamber.
Trabeculectomy Trabeculectomy was first tried under a scleral flap by Sugar22 in 1961. A 2 mm section of the trabecular meshwork and Schlemm’s canal was removed with a punch forceps, and a peripheral iridectomy was done. The scleral flap was then tightly sutured and all cases failed. Sugar proposed the reason for failure: crushing action of the punch forceps closing off the edges of Schlemm’s canal. The first successful trabeculectomy technique was described by Cairns23 in 1968. The procedure excised a length of Schlemm’s canal, adjacent trabecular meshwork, tip of the scleral spur, and deep layers of the cornea; a fornix or limbal-based scleral flap was used. An iridectomy was done, and the scleral flap was sutured firmly. Only six of 300 cases had a flat anterior chamber, one of which lasted for over 3 days. Moderate uveitis was present in five cases. There were no hyphemas. Cairns24 reviewed 80 cases in 1972. Overall IOP control was 97.5%; 30% had no obvious drainage bleb; only 2.5% of those without a bleb needed medical therapy. Cairns proposed five possible modes of action: • creation of a fistula • aqueous humor drainage through the trabecular meshwork into Schlemm’s canal • aqueous drainage directly into collector channels • possible localized cyclodialysis • possible hyposecretion
Laser sclerostomy From the 1970s guarded filtration surgery by trabeculectomy became the standard operation for uncontrolled open-angle glaucoma as well as other types. In comparison to full thickness procedures the risk of hypotony was reduced by the protective scleral flap. Modifications of the technique, including laser suture lysis and releasable flap sutures, allowed even finer control of postoperative filtration. Trabeculectomy is the standard against which non-penetrating filtering procedures can be compared. Several investigators have looked at the long-term results of trabeculectomy. In 150 eyes followed-up for an average of 10 years (range 1–20 years) Watson et al25 reported that 90% had a final IOP of less than 20 mmHg. Once drainage was established there were excellent chances that IOP would remain controlled. Only three eyes had longterm hypotony and no cases of endophthalmitis or blebitis were reported, but visual fields remained stable or improved in only 41%. The progressive visual field loss in the other 59% may have been because a low enough ‘target pressure’ could not be achieved. Wilensky and Chen 26 looked at a follow-up after 15 years of 40 eyes that had an initially successful trabeculectomy at 1 year (defined as IOP < 21 mmHg or a drop of at least 33% if initial IOP was < 21 mmHg). The success rate of IOP control was 83% at 5 years, 73% at 10 years, and only 42% at 15 years. Nouri-Mahdavi et al27 looked at the optic disc and automated visual field changes to judge long-term stability of primary openangle glaucoma after trabeculectomy. In 78 eyes followed-up for 25–112 months the probability of a single operation giving successful IOP control was only 48% at 3 years and 40% at 5 years. Khalili et al28 looked at 700 trabeculectomies followed-up for 3–12 years. Success was defined as IOP below 21 mmHg throughout the
9
entire study period, no evidence of progressive disc or visual field deterioration, no drop in visual acuity, and no additional glaucoma surgery. The success rate was only 44%. In half of the failures the IOP began to rise within the first postoperative month. Molteno et al29 looked at first-time trabeculectomies on 289 eyes with primary open-angle glaucoma. The probability of obtaining an IOP of 21 mmHg or less was 0.93 at 5 years, 0.87 at 10 years, and 0.85 at 15 years, but mean visual acuity decreased significantly over the years leading to blindness in 47 eyes. The chances of retaining visual acuity greater than 20/400 and greater than 5° radius visual field was only 0.6 at 15 years. Eyes with better preoperative visual acuity had better chances of preserving useful vision. Johnson et al30 found that the probability of blindness in 86 eyes was 46% at 10 years after filtering surgery, mostly trabeculectomy. The resultant IOP (14 mmHg) in the group that went blind was similar to that which preserved visual function (15.4 mmHg), but the group that went blind had more advanced field loss at the time of surgery (scotomas above and below the horizontal axis). A conclusion one can draw is that an IOP of even 14 mmHg is too high long-term for an eye with advanced glaucomatous loss.
Laser sclerostomy Laser sclerostomy is a filtering procedure that has not gained widespread popularity. Various techniques have been described either ab interno31–33 or ab externo, most of which were done with the THC:YAG (holmium) laser that creates a 200 micron diameter lumen (onehundredth the surface area of Elliot’s 2 mm
10
The history of filtering surgery
trephine hole). With the holmium laser, Iwach et al34 reported at 30 months in 81 eyes a probability of success of 0.72 in low-risk eyes and 0.3 in high-risk eyes. Complications of this full-thickness procedure included more hypotony and flat or shallow anterior chambers than with the guarded filtration of trabeculectomy, as well as iris incarceration in the absence of an iridectomy. An additional reason for lack of acceptance of this technique was the concomitant rise in popularity of antimetabolites, which increased the success rate of standard trabeculectomy.
Non-penetrating trabeculectomy With recognition of the juxta canalicular meshwork and inner wall of Schlemm’s canal being the site of major resistance to outflow,35 procedures were devised to selectively remove this tissue, leaving a thin trabeculo-Descemetic membrane intact. Zimmerman et al36 first described a nonpenetrating trabeculectomy in 1984, followed by Koslov et al 37 who described non-penetrating deep sclerectomy in 1989. Viscocanalostomy was described by Stegmann37 in 1995. These new operations, the subject of this book, are the most specific target oriented procedures to date to treat open-angle glaucoma. In evaluating these new procedures it is wise to keep in mind the words of Duke-Elder (Ref. 1, page 528): ‘Any operation devised for the relief of glaucoma should ideally be such as to preserve the function of the eye, maintain its tension within normal limits, and retain the integrity of the globe. The number of operations advocated from time to time is evidence that this ideal has never been attained.’
Acknowledgment I thank Dr. Ana Bassols of Chauvin Laboratories for invaluable help in tracking down original old publications.
References 1. Duke-Elder S. System of Ophthalmology (Vol XI). London: Henry Kimpton, 1969. 2. Clavel J, Chausson JD. La Perennité d’une Vocation. Lausaunne: Fondation de l’Asile des Aveugles, 1993. 3. von Graefe A. Uber die Wirkung der Iridectomie bei Glaucom und uber den glaucomatosen Prozess. Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie 1857;3:456–555. 4. von Graefe A. Weitere Zusatze uber Glaukom und die Heilwirkung der Iridectomie. Arch Ophthalmol 1861;8:243–313. 5. Lagrange F. Nouveau traitement du glaucome chronique simple: iridectomie et sclerectomie combinées. Ann Ocul 1907;137:89–103. 6. Critchett G. Cases illustrative of a new method of treating deep-seated inflammation of the globe or acute glaucoma. J R Lon Ophthalmol Hosp 1858;1:57–66. 7. Kronfeld P. The rise of the filter operations. Surv Ophthalmol 1972;17:168–79. 8. de Wecker L. Die Sklerotomie als Glaukomoperation. Ber Ophthalmol Ges 1871;8:305–10. 9. de Wecker L. La cicatrice a filtration. Ann Ocul 1882;87:133–43. 10. Dianoux C. Glaucome et sclerotomie. Ann Ocul 1905;133:81–85. 11. Holth S. Iridencleisis antiglaucomatosa. Ann Oculist 1908;137:345–75. 12. Elliot RH. A preliminary note on a new operative procedure for the establishment of a filtering cicatrix in the treatment of glaucoma. Ophthalmoscope 1909;7:804–06.
References 13. Preziosi L. The electro-cautery in the treatment of glaucoma. Brit J Ophth 1924;8: 414. 14. Scheie HG. Retraction of scleral wound edges as a fistulizing procedure for glaucoma. Am J Ophthalmol 1958;45:220–29. 15. Scheie HG. Filtering operations for glaucoma: a comparative study. Am J Ophthalmol 1962;53:571–90. 16. Leydhecker W. Comparative study of late after-effects of glaucoma operations. In Glaucoma Tutzing Symposium, Karger, Basel/New York, 1967, 224–38. 17. Iliff CE, Haas JS. Posterior lip sclerectomy. Am J Ophthalmol 1962;54:688–93. 18. Haas JS. Symposium on Glaucoma. Trans N Orleans Acad Ophthal, 1967:175. 19. Harms H, Dannheim R. Epicritical consideration of 300 cases of trabeculotomy “ab externo”. Trans Ophthal Soc UK 1969;88:491–99. 20. Krasnov MM. Externalization of Schlemm’s canal (sinusotomy) in glaucoma. Br J Ophthal 1968;52:157–61. 21. Molteno ACB. New implant for glaucoma clinical trial. Br J Ophthalmol 1971;53:606. 22. Sugar HS. Experimental trabeculectomy in glaucoma. Am J Ophthalmol 1961, 51:623–27. 23. Cairns JE. Trabeculectomy—preliminary report of a new method. Am J Ophthalmol 1968;66:673–79. 24. Cairns JE. Surgical treatment of primary open angle glaucoma. Trans Ophthalmol Soc UK 1972;92:745–56. 25. Watson PG, Jakeman C, Ozturk M et al. The complications of trabeculectomy (a 20-year follow-up). Eye 1990;4:425–38. 26. Wilensky JT, Chen TC. Long term results of trabeculectomy in eyes that were initially successful. Trans Am Ophthalmol Soc 1996;94:147–59. 27. Nouri-Mahdavi K, Brigatti L, Weitzman M, Caprioli J. Outcomes of trabeculectomy for
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
11
primary open-angle glaucoma. Ophthal mology 1995;102:1760–69. Khalili MA, Diestelhorst M, Krieglstein GK. Long-term follow-up of 700 trabeculectomies. Klin Monatasbl Augenheilkd 2000;217:1–8. Molteno AC, Bosma NJ, Kittelson JM. Otago glaucoma surgery outcome study: long-term results of trabeculectomy, 1976 to 1995. Ophthalmology 1999;106:1742–50. Johnson DH, Parc CE, Oliver J et al. The long term outcome of glaucoma filtration surgery. Invest Ophthal Vis Sci 2000;41:S518. Jaffe GJ, Mieler WF, Radius RL et al. Ab interno sclerostomy with a high powered argon endolaser. Arch Ophthalmol 1989;107:1183–85. March WF, Bernitzky D, Gherezghiher T et al. Creation of filtering blebs with the YAG laser in primates and rabbits. Glaucoma 1985;7:43–45. Melamed S, Solomon A, Neumann D et al. Internal sclerostomy using laser ablation of dyed sclera in glaucoma patients: a pilot study. Br J Ophthalmol 1993;77:139–44. Iwach AG, Hoskins HD, Drake MV, Dickens CJ. Update of the sunconjunctival THC:YAG (holmium) laser sclerostomy ab externo clinical trial: 30 month report. Ophthalmic Surg 1994;25:13–21. Bill A, Svedbergh B. Scanning electron microscopic studies of the trabecular meshwork and the canal of Schlemm—an attempt to localize the main resistance to outflow of aqueous humor in man. Acta Ophthalmol 1972;50:295. Zimmerman TJ, Kooner KS, Ford VJ et al. Effectiveness of nonpenetrating trabeculectomy in aphakic patients with glaucoma. Ophthalmic Surg 1984;15:44–50. Stegmann RC. Viscocanalostomy: a new surgical technique for open angle glaucoma. An Inst Barraquer 1995;25:225–32.
2 Evolution of non-penetrating glaucoma surgery André Mermoud
Historical review of nonpenetrating filtering surgery In 1962 Kraznov performed the first sinusotomy. This operation consisted of removing a lamellar band of the sclera and opening Schlemm’s canal over 120° from 10 to 2 o’clock (Fig. 2.1).1–4 The inner wall of Schlemm’s canal was untouched and then the conjunctiva was closed. Kraznov believed that the aqueous outflow resistance in most cases of primary open-angle glaucoma was situated at the level of scleral aqueous-drainage veins and not in the trabeculum. He therefore developed a safe non-penetrating filtering surgery, leaving in place the trabeculum and the inner wall of Schlemm’s canal. When there was no percolation of aqueous through the trabeculum and Schlemm’s canal inner wall, Kraznov entered the anterior chamber and performed a peripheral iridectomy, creating a full-thickness procedure that was the standard filtering surgery at that time. Sinusotomy was definitely safer than full thickness surgery with almost no postoperative complication, which was certainly not the case with standard fullthickness procedures frequently leading to a major hypotony, followed in many cases by a flat anterior chamber, choroidal detachments, and cataract formation. Kraznov also reported
Figure 2.1 Schematic representation of sinusotomy Schlemm’s canal is unroofed. There is no superficial scleral flap to cover the sclerectomy. Inner wall of Schlemm’s canal is untouched.
14
Evolution of non-penetrating glaucoma surgery
that the filtering blebs were more diffuse after sinusotomy and that they tended to disappear with time. Sinusotomy never became popular because it was a difficult operation, it needed a surgical microscope and Schlemm’s canal had to be found, which was not easy. Moreover the surgical results were not convincing. Kraznov reported an 83% success rate but did not specify the success criteria, the number of patients followed-up or the period of followup. Postic and Stankov-Tomic4 have reported a 50% success rate in 12 glaucoma patients operated on by sinusotomy. These six successful patients had low intraocular pressure (IOP) with filtering blebs. The other 50% presented a primary drop in IOP after surgery, and then an IOP rise because of fibrosis of the filtering bleb. To my knowledge, there is no long-term report on the outcome of sinusotomy. In the late 1960s, and for the next three decades, trabeculectomy described by Sugar5 in 1961 and Cairns6 in 1968 became the standard technique for filtering surgery, providing a satisfactory IOP control with fewer postoperative complications than full-thickness filtering procedures. However, even with the many modifications proposed to the original trabeculectomy, the lack of a reproducible postoperative IOP reduction as well as the early postoperative complications led several surgeons to reconsider Kraznov’s work. Several techniques of non-penetrating filtering surgery based on sinusotomy have been described. Since the main aqueous outflow resistance may be located at the juxtacanalicular trabeculum and the inner wall of Schlemm’s canal, these two anatomical structures have to be removed. Ab externo trabeculectomy (Fig. 2.2) was first proposed by De Laage de Meux and Kantelip7 in 1976, and later by Zimmerman et al8,9 (Fig. 2.3) in 1984, by Arenas in 199110 and Tanibara et al11
Figure 2.2 Schematic representation of ab externo trabeculectomy. A deep sclerectomy unroofing Schlemm’s canal is covered by superficial scleral flap. Schlemm’s canal inner wall and juxtacanalicular trabeculum are removed.
in 1993. Another method to improve the aqueous outflow in a patient with a restricted posterior trabeculum clearance is to remove the corneal stroma behind the anterior trabeculum and the Descemet’s membrane (Fig. 2.4). This technique was first described
Ab externo trabeculectomy
15
Figure 2.3 Professor Thom J Zimmerman
by Fyodorov13 and Kozlov et al14 (Fig. 2.5) and later by Sanchez et al.15
Ab externo trabeculectomy Ab externo trabeculectomy is very similar to sinusotomy except for the presence of a superficial scleral flap and the removal of the inner wall of Schlemm’s canal and the juxtacanalicular trabeculum (Fig. 2.2).
Figure 2.4 Schematic representation of deep sclerectomy. Under superficial scleral flap, deep corneosclerectomy, unroofing Schlemm’s canal, is performed. Corneal tissue behind anterior trabeculum and Descemet’s membrane are removed.
Surgical technique The conjunctiva is opened either at the fornix or at the limbus in the superior quadrant. A 4 4 mm superficial scleral flap is created at the 12 o’clock position. The depth of this scleral flap corresponds to about one third of the full scleral thickness. A radial cut is made on the edge of the flap at the limbus to locate
the Schlemm’s canal. Once the Schlemm’s canal is found, it is unroofed in the same manner as Kraznov did in sinusotomy. At this stage, there is a 4 mm-long Schlemm’s river parallel to the limbus. Different techniques
16
Evolution of non-penetrating glaucoma surgery Figure 2.6 Professor Alain Bechtoille
Figure 2.5 Professor Valentin Kozlov Figure 2.7 Dr Elie Dahan have been proposed to remove the inner wall of Schlemm’s canal. A fine forceps with two small plates at the end may be used to grab the endothelium and to peel it off from one side to the other. Trabeculo-aspiration has been proposed by Bechtoille (Fig. 2.6) who uses a fine canula connected to a phaco infusionaspiration system (unpublished work). Dahan (Fig. 2.7) uses a fine diamond-coated spatula that allows the surgeon to scrape the endothelium (unpublished work). All of these manoeuvres have been grouped into so-called ab externo trabeculectomy because the juxtacanalicular trabeculum is removed and the corneoscleral and uveoscleral trabecula are left intact. Valtot, in an unpublished report, showed that the tissues removed corresponded to the endothelium of the Schlemm’s canal and the juxtacanalicular trabeculum; this was confirmed by Roy et al, who examined a large series of excised fragments by transmission and electronic microscopy (unpublished work). The outflow resistance of the remaining membrane formed by the posterior trabeculum has been studied by Rossier et al 16 They found that in enucleated human eyes the
mean outflow facility increased from 0.21 ± 0.6 to 2.03 ± 1.43 µL/min per mmHg after the removal of 4 mm of the Schlemm’s canal inner wall and the juxtacanalicular trabeculum.
Results Zimmerman and colleagues8,9 have reported good results of non-penetrating ab externo trabeculectomy in both phakic and aphakic patients. However, after their first two publications, they abandoned this technique because of
Deep sclerectomy surgical difficulties. Arenas currently continues to use ab externo trabeculectomy, and has reported a success rate of 88%.10 Other investigators who have been using similar techniques have reported satisfactorily controlled IOPs in 85.8% to 90% of patients.11,12 With regard to long-term results, Arenas, Valtot, and Bechtoille have all separately reported satisfactory IOP control over time, but no results have yet been published.
Deep sclerectomy Deep sclerectomy was first described by Fyodorov13 and Kozlov et al.14 The route of the aqueous outflow is different from the one described for sinusotomy and ab externo trabeculectomy where the postoperative drainage occurs through the posterior trabeculum. In deep sclerectomy, the main outflow occurs at the level of the anterior trabeculum and the Descemet’s membrane. This was shown by Vaudaux and Mermoud17 in an ex vivo model of deep sclerectomy. They reported that the mean outflow facility increased from 0.19 ± 0.03 to 24.5 ± 12.6 µL/min per mmHg after deep sclerectomy. In comparison with the same experiment performed in ex vivo ab externo trabeculectomy, the postoperative outflow facility increase is ten times higher after deep sclerectomy.16 To provide an aqueous outflow through the anterior trabeculum and the Descemet’s membrane, the corneal stroma behind these structures has to be removed (Fig. 2.4).
Surgical technique The conjunctiva may be opened either at the fornix or at the limbus. A 5 5 mm superfi -
17
cial scleral flap is made, including one third of the scleral thickness (300 µm). To be able to reach the Descemet’s membrane later in the dissection, the superficial scleral flap has to be cut 1–1.5 mm anteriorly into the clear cornea. A second deep scleral flap measuring 4 4 mm is dissected, leaving about 10% of the sclera over the choroid and the ciliary body. The second flap is usually started in its posterior part. The horizontal dissection is started posteriorly, moving anteriorly with a crescent blade. Near the limbus the Schlemm’s canal is automatically unroofed. The dissection is continued anteriorly with a blunt spatula or a sponge to find the natural cleavage plan between the Descemet’s membrane and the corneal stroma. When the Descemet’s membrane has been exposed for 1 mm, the second scleral flap is excised. At this stage, the aqueous is seen percolating through the anterior trabeculum and the Descemet’s membrane. To enhance the filtration, an ab externo trabeculectomy can be peformed as well at this stage. To keep the intrascleral space created patent, an implant may be used. Kozlov et al14 have proposed a collagen implant that resorbs itself within 6–9 months.18,19 Stegmann et al use high viscosity hyaluronic acid,20 and Sourdille and Dahan are using reticulated hyaluronic acid and Hema implants, respectively (unpublished work).
Results Kozlov et al14 have reported an 85% success rate, but no information regarding success criteria or follow-up is available. Demailly et al21 reported a mean decrease in IOP of 9.1 ± 7.1 mmHg after 219 deep sclerectomy procedures with collagen implants. They reported a success rate using Kaplan–Meier
18
Evolution of non-penetrating glaucoma surgery
survival analysis of 89% without glaucoma medication at 6 months and 75.6% at 16 months; with glaucoma medication their success rate increased to 97% at 6 months, and 79% at 16 months. Karlen et al22 have reported the mediumterm success rate (36 months) of 100 patients who underwent deep sclerectomy with collagen implant. The mean preoperative IOP was 27.8 ± 8.6 mmHg and dropped to 5.7 ± 4 mmHg on the first postoperative day and remained stable at 13 ± 3.5 mmHg during the entire follow-up period. Complete success, defined as an IOP lower than 21 mmHg without medication, was 44.6% at 36 months; qualified success, defined as an IOP lower than 21 mmHg with medication, was 97.7% at 36 months. Goniopuncture had to be performed on 41 of the patients, and 5-fluorouracil injections were given in 23 patients; cataract progression occurred in seven patients. When the different types of open-angle glaucoma were compared, no difference was found in terms of reduction in IOP, number of patients requiring antiglaucoma medication, or success rate. There was, however, a tendency for a lower success rate in patients with pseudoexfoliative or pseudophakic glaucoma. In comparison with the standard filtering trabeculectomy, deep sclerectomy offers a similar IOP drop with a lower rate of postoperative complications and a quicker recovery of visual acuity.23,24
Viscocanalostomy The assumed mechanism of filtration in viscocanalostomy is different from the one described in other non-penetrating filtering surgeries. Stegmann et al20 think that the aqueous filters through the trabeculo-
Descemet’s membrane to the scleral space as in deep sclerectomy, but that it does not form a subconjunctival filtering bleb since the superficial scleral flap is tightly closed with numerous nylon 10/0 sutures. From the scleral space, the aqueous humour is supposed to reach the Schlemm’s canal, which is open on either side of the deep sclerectomy, and then flows into the aqueous episcleral veins. Until now, no scientific study has been able to confirm this hypothesis, and, in my own hands, patients who underwent a viscocanalostomy presented in 50% of the cases a subconjunctival filtering bleb. Long-term follow-up of viscocanalostomy is reported to be satisfactory. In a prospective study involving 214 eyes of 157 patients, a postoperative IOP below 22 mmHg was observed in 82.7%. 20
Summary In comparison with full-thickness filtering procedures, sinusotomy showed an important progress in terms of postoperative complications. Unfortunately, the long-term IOP drop did not come up to expectations and trabeculectomy became the standard filtering surgery for the next three decades. Since sinusotomy has been modified over the past few years, non-penetrating filtering surgery has become of interest again, mainly because it provides a more reproducible early drop in IOP with fewer postoperative complications. The main changes made from sinusotomy in the new non-penetrating filtering surgeries are: the peeling of the inner wall of Schlemm’s canal (ab externo trabeculectomy) and the excision of the corneal stroma behind the anterior trabeculum and the Descemet’s membrane (deep sclerectomy). These new techniques compare favourably with
References trabeculectomy in terms of IOP drop. However, long-term follow-up is still needed to tell us whether this IOP control can be maintained or not.
References 1. Krasnov MM. Vestn Oftal 1964;77:37–41. 2. Krasnov MM. Externalization of Schlemm’s canal (sinusotomy) in glaucoma. Br J Ophthalmol 1968;52:157–61. 3. Krasnov MM. Symposium: microsurgery of the outflow channels–sinusotomy foundations, results, prospects. Trans Am Acad Ophthalmol Otolaryyngol 1972;76:368–74. 4. Postic S, Stankov-Tomic M. Sinusotomie d’après Krasnov dans le glaucome chronique simple. Bull Mém Soc Fr Ophtalmol 1967;80:716–26. 5. Sugar HS. Experimental trabeculectomy in glaucoma. Am J Ophthalmol 1961;51:623. 6. Cairns JE. Trabeculectomy: preliminary report of a new method. Am J Ophthalmol 1968;66:673–79. 7. de Laage de Meux M, Kantelip B. Surgical anatomy of corneoscleral limbus. Arch Ophthalmol (Paris) 1976;36:39–50. 8. Zimmerman TJ, Kooner KS, Ford VJ et al. Effectiveness of non penetrating trabeculectomy in aphakic patients with glaucoma. Ophthalmic Surg 1984;15:49–50 9. Zimmerman TJ, Kooner KS, Ford VJ et al. Trabeculectomy vs non penetrating trabeculectomy: a retrospective study of two procedures in phakic patients with glaucoma. Ophthalmic Surg 1989;15:734–40 10. Arenas E. Trabeculectomy ab-externo. Highlights Ophthalmol 1991;19:59–66 11. Tanibara H, Negi A, Akimoto M et al. Surgical effects of trabeculotomy ab externo on adults eyes with porimary open angle glaucoma and pseudoexfoliation syndrome.
19
Arch Ophthalmol 1993;111:1653–61. 12. Tavano G, Chabin T, Barrut JM. Hémitrabéculectomie non invasive. Bull Soc Ophtalmol Fr 1993;93:749–50. 13. Fyodorov SN. Non-penetrating deep sclerectomy in open-angle glaucoma. Eye Microsurg (Russian) 1989;52–5. 14. Kozlov VI, Bagrov SN, Anisimova SY et al. Deep sclerectomy with collagen. Eye Microsurg 1990;3:44–46 15. Sanchez E, Schnyder CC, Sickenberg M et al. Deep sclerectomy: results with and without collagen implant. Int Ophthalmol 1997;20:157–62. 16. Rossier A, Uffer S, Mermoud A. Aqueous dynamics in experimental ab externo trabeculectomy. Ophthalmic Res 2000 Jul–Aug;32:165–71 17. Vaudaux J, Mermoud A. Aqueous humor dynamics in non-penetrating filtering surgery. Ophthalmol Pract 1998;38:S1064. 18. Chiou AGY, Mermoud A, Hediguer S et al. Ultrasound biomicroscopy of eyes undergoing deep sclerectomy with collagen implant. Br J Ophthalmol 1996;80:541–44. 19. Chiou AGY, Mermoud A, Underdahl PJ, Schnyder CC. An ultrasound biomicroscopic study of eyes after deep sclerectomy with collagen implant. Ophthalmology 1998;105:104–08. 20. Stegmann R, Pienaar A, Miller D. Viscocanalostomy for open angle glaucoma in black African patients. J Cataract Refract Surg 1999;25:316–22. 21. Demailly P, Jeanteur-Lunel MN, Berkani M et al. Non penetrating deep sclerectomy associated with collagen device in primary open angle glaucoma: middle-term retrospective study. J Fr Ophtalmol 1996;19,11:659–66 22. Karlen M, Sanchez E, Schnyder CC et al. Deep sclerectomy with collagen implant: medium term results. Br J Ophthalmol 1999;83:6–11 23. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with
20
Evolution of non-penetrating glaucoma surgery
collagen implant and trabeculectomy in openangle glaucoma. Cataract Refract Surg 1999;25:323–331 24. Chiou AGY, Mermoud A, Jewelewicz DA.
Comparison of post-operative inflammation following deep sclerectomy with collagen implant versus standard trabeculectomy. Clin Exp Opthalmol 1998;236:593–6
3 Anatomical features of outflow pathway Farid Achache
Intraocular pressure (IOP) is important for maintainance of ocular rigidity and hence optimization of optical function. Aqueoushumor dynamics control IOP, which depends on the rate of aqueous-humor production (inflow) and the rate at which it leaves the eye through the anterior chamber angle (outflow). The pressure gradient from anterior chamber to episcleral veins is explained by a resistance to filtration in the way of the aqueous outflow. Since Morton Grant1,2 demonstrated that this site of aqueous-outflow resistance is located between the anterior chamber and Schlemm’s canal, it has been generally accepted that the greatest resistance is found in the internal wall of Schlemm’s canal between the trabecular meshwork and the canal. It has also been shown that the resistance might also relate to canal collapse. In addition, experimental studies on monkeys show that about 30% of this resistance is located within the intraocular outflow channels. Therefore the precise location of the resistance to outflow is not completely described and is still a matter for debate. Grant has estimated this resistance to be 3 mmHg/µL per min. On the basis of these findings it has been established that IOP may be expressed by a physical factor called the ‘Facility of outflow’ (0.3 µL/min per mmHg).
Outflow pathways The standard drainage route that conveys aqueous humor from within outwards consists of trabecular meshwork, Schlemm’s canal, collector channels, intrascleral venous plexus, aqueous veins, episcleral and conjunctival veins. This outflow pathway is regarded as the conventional route and accounts for 83–96% of aqueous outflow. The uveoscleral pathway, first described in monkeys by Anders Bill3 more than 30 years ago, is an accessory system also called the unconventional outflow route, which drains 5–15% of aqueous-humor production. The uveoscleral pathway allows free access from the anterior chamber to the supraciliary and suprachoroidal spaces via the collagen-containing spaces between the ciliary-muscle-fibre bundles. From there, the fluid crosses the sclera and the uveal vascular system. This secondary drainage pathway has given rise to a new treatment strategy that reduces IOP by enhancing uveoscleral outflow. In addition, we must take into account the transscleral outflow, which is a less wellknown drainage route (Fig. 3.1).4,5
22
Anatomical features of outflow pathway Figure 3.1 Drainage pathways of the aqueous humor.
Production of aqueous humor Aqueous humor crosses posterior and anterior chambers, providing the surrounding avascular structures with oxygen and other nutrients such as glucose, aminoacids, and polypeptide growth-modulating factors. At the same time, aqueous humor disposes of metabolic wastes coming from these structures. The pathway of aqueous flow involves active production of aqueous humor by the non-pigmented ciliary epithelium of the ciliary processes. Aqueous humor consists of a dilute
solution of the constituents of the plasma, in addition to the substances specifically secreted. Three mechanisms take place in aqueous humor production. Diffusion and ultrafiltration of lipids and water substances are passive processes in response to osmotic gradient and hydrostatic pressure, and account for 15–20% of the whole aqueous humor flow. Secretion is an active process based on the active transport of sodium through the non-pigmented layer of the ciliary epithelium. The sodium leaves the capillary wall easily and then crosses the outer basal lamina and the pigmented layer. Sodium overcomes the tight junctions of the non-
The ciliary body
23
Figure 3.2 The ciliary muscle with the longitudinal ciliary fibers (LCM) and the circular ciliary fibers (CCM). Vascularization of the ciliary body. MAC = major arterial circle, ACA = anterior ciliary arteries, LPCA = long posterior ciliary arteries. The ciliary epithelium (CE) with the outer pigmented and the inner non-pigmented layers.
pigmented layer by penetrating the apex of the cell under the carbonic anhydrase dependent reaction. Finally, sodium leaves the cell to reach the posterior chamber, by way of the sodium pump. This latter process needs the energy provided by Na+ K+ ATPase reaction, which transforms ATP into ADP. Secretion is an active sodium-dependent process, accounting for 95% of the whole aqueous humor production. The aqueous secretion rate is about 2.5 µL/min.5,6
The ciliary body The ciliary body is the medium part of the uvea, between iris and choroid; it extends 6 mm from the scleral spur anteriorly to the ora serrata posteriorly. The anterior 2 mm is the pars plicata or corona ciliaris, characterized by the radial ridges of the ciliary processes. The posterior 4 mm is called the pars plana or orbicularis ciliaris, because of its flat surface. On a sagittal section, the ciliary
24
Anatomical features of outflow pathway
body is three-sided and has a triangular shape; its outer side is adjacent to the supraciliaris space and the sclera. The base of the triangle is occupied by the root of the iris, which is inserted in the middle part of the anterior face of the ciliary body. The space between the root of the iris and the scleral spur is occupied by the ciliary body band, located in the apex of anterior chamber angle. On the inner surface of the ciliary processes, the zonulae fibers stretch from the top of the ciliary processes towards the equatorial area of the lens. The three components of the ciliary body are the ciliary processes, the ciliary vessels, and the ciliary muscle (Fig. 3.2). 7
The ciliary processes The 70 major and minor ciliary processes spread over 2 mm in the area called pars plicata or corona ciliaris, and extend into the posterior chamber. The minor ciliary processes are set between the major processes. Each ciliary process is limited by a two-layered ciliary epithelium that extends anteriorly to the iris. The outer layer, close to the stroma and the ciliary vessels, is a pigmented layer composed of cuboidal cells with numerous melanin granules that lies on a basement membrane. The inner layer is non-pigmented and faces the aqueous humor of the posterior chamber; it consists of columnar cells rich in mitochondria, and is limited by an internal basement membrane. The cells are linked to each other by tight junctions and interlock by interdigitations seen on their lateral and apical surfaces. The ciliary epithelium, particularly the non-pigmented layer, forms a barrier between blood and the ocular cavity, which is called the ciliary bloodaqueous barrier. The stroma occupies spaces between the epithelium and capillaries; it is sparse and composed of ground substance
consisting of mucopolysaccharides, proteins, a few collagen fibrils, and wandering cells of connective tissue and blood origin.4,7
The ciliary vessels Blood is provided by branches of the major arterial circle, resulting from the anastomoses of the anterior and posterior ciliary arteries. The anterior region of the ciliary process is irrigated by the anterior ciliary process arterioles and the posterior and central area is supplied by the posterior ciliary process arteries. The ciliary vessels branch into a dense capillary network along the base of the ciliary process. The endothelium of the capillaries is surrounded by pericytes and mural cells lying on a basement membrane and interrupted by fenestrae, which promotes a high permeability through the vascular wall. The whole capillary network joins the choroidal veins.7
The ciliary muscle The ciliary muscle is composed of two main parts: longitudinal and circular. The anterior tendons insert the ciliary body anteriorly by elastic-like tendons to the scleral spur and the corneoscleral meshwork as well as uveal trabeculae. The posterior tendons insert into Bruch’s membrane and the elastic network of the choroid. The muscle bundles of the ciliary body are meridional in the outer portion, reticular in the intermediate portion, and circular in the inner portion. The meridional fibers, also called Bruch’s muscle, stretch posteriorly to the supra choroidal lamina. The circular fibers, also known as Müller’s muscle, stand in the innermost area of the ciliary body. The spaces between the muscle bundles are occupied by sparse connective tissue made up
The scleral spur of collagen fibers and fibroblasts, and are wider in the longitudinal part. The width of the spaces depends on the state of muscular activity, enlarging during relaxation, which allows aqueous humor access, and closing during contraction. Muscle cells as well as muscle bundles are limited by a basal lamina and the capillaries have a continuous wall. The ciliary muscle is richly innervated with sensory nerve endings of trigeminal origin. The longitudinal portion is innervated by sympathetic fibers arising from the superior sympathetic ganglion and the circular portion receives parasympathetic fibers coming from the third nerve.7
Posterior and anterior chambers Posterior chamber Aqueous humor secreted by the ciliary processes pours down into the posterior chamber. The posterior chamber is a narrow space bounded posteriorly by the zonula fibers and the lens, peripherally by the ciliary processes, and anteriorly by the inner surface of the iris. The posterior chamber’s volume is about 0.06 mL.
25
lens. The volume of the anterior chamber is about 0.2 mL, and its depth ranges from 2.6 mm to 4.4 mm (average 3.15 mm), being shallower in young children, hyperopic eyes, and old people. The peripheral part of the anterior chamber is the angled recess called the drainage angle.8
Anterior chamber angle The different structures of the angle are located in a defined region of 1.5 mm diameter called the limbus, which is the transitional zone between the cornea and the sclera. The outer surface of the limbus is flat and extends 2 mm from the corneolimbal junction anteriorly to the sclerolimbal junction posteriorly. The corneoscleral junction is the apparent limbus from which we find the pars plicata, about 3 mm behind, and the pars plana, about 5 mm posteriorly. The sclerolimbal junction is the transition between white sclera and bluishgray limbus and represents the main landmark for the most conventional surgical procedures. The inner surface of the limbus is a groove called the scleral sulcus, the anterior boundary of which is Schwalbe’s line, the posterior boundary is the scleral spur. The scleral sulcus accommodates the Schlemm’s canal and the trabecular meshwork (Figs 3.3 and 3.4).
Anterior chamber
The scleral spur
Aqueous humor leaves the posterior chamber and flows through the pupil to reach the anterior chamber. The anterior chamber is limited anteriorly by the inner corneal face and posteriorly by the outer surface of the iris, and through the pupil by the anterior face of the
The scleral spur is in the deep limbus. The posterior ridge of the scleral spur, called the scleral roll, is a circular condensation made up of collagen fibers for the most part (80%) and a small amount of elastic tissue for the remainder (5%). Moses and Grodzki9 proved in 1977
26
Anatomical features of outflow pathway Figure 3.3 Structures of the anterior chamber angle; anatomical landmarks seen in gonioscopy. SL = Schwalbe’s line; TM = trabecular meshwork; SS = scleral spur; CBd = ciliary board; SC = Schlemm’s canal; I = iris; S = sclera; CB = ciliary body.
Figure 3.4 Outer projections of the deep limbus structures.
Trabecular meshwork
27
that there are myofibroblast-like cells within the scleral spur. The scleral spur is the mooring point of the anterior tendons of the longitudinal ciliary muscle. The contraction of the ciliary muscle pulls the scleral spur backwards and opens the trabecular meshwork spaces, improving the outflow.6
give contractile properties to the trabecular meshwork. The trabecular meshwork consists of three parts, from within outwards: the uveal meshwork, the corneoscleral meshwork; the juxtacanalicular meshwork, also called endothelial or cribriform meshwork (Fig. 3.4).8,10,11
Schwalbe’s line
Uveal meshwork
Schwalbe’s line is the transition area between the corneal endothelium, the end of Descemet’s membrane and the anterior part of the trabecular meshwork. This region has been called Zone S; it consists of circular arranged collagen and elastic fibers. 8
Trabecular meshwork The trabecular meshwork is a sieve-like band of connective tissue about 750 µm in width, closing the inner part of the scleral sulcus as a bridge joining Schwalbe’s ring to the scleral spur, providing a tubular duct: the Schlemm’s canal. The meshwork is made up of several sheets containing an extracellular collagen substance, elastic fibers, and mucopolysaccharides. Each trabecular beam is surrounded by a single layer of endothelial cells characterized by a strong phagocytory activity and linked by intercellular junctions; they are oriented in the long axis of the trabecular beam, with a bulging nucleus, and lie on a basement membrane. The cells produce glycoaminoglycanes as hyaluronic acid, collagen types I, III, IV, V, and VI, extracellular glycoproteins such as fibronectine and laminin, and fibrillar material. The core of the trabecular beam is made up of collagen and elastics fibrils that
The uveal meshwork is the inner portion neighbouring the aqueous humor of the anterior chamber; it consists of a superimposition of perforated sheets. The size of the apertures range from 25 µm to 75 µm. Posteriorly, the uveal meshwork is connected with the ciliary muscle whereas the anterior part joins the periphery of Descemet’s membrane, the inner part of the Schwalbe’s ring, and the corneal lamellae or the corneoscleral trabeculae. Elastic fibrils are dispersed in the center of the collagen core. 8,9
Corneoscleral meshwork The corneoscleral meshwork is the mid portion of the trabeculum; it extends from the scleral roll posteriorly close from the anterior tendons of the ciliary muscle to the anterior side of the scleral sulcus. This meshwork consists of about 15 layers, the openings of which decrease in size from within outwards with the range 5–50 µm. Elastic fibrils are dispersed around the collagen core. Each beam is surrounded by trabecular cells lying on a basal membrane. The trabecular cells are spindle-shaped and present long cytoplasmic processes; they contain numerous Golgi complexes, rough endoplasmic, ribosomes, lysosomes, and mitochondria.
28
Anatomical features of outflow pathway Figure 3.5 Schlemm’s canal and trabecular meshwork layers.
Juxtacanalicular meshwork The juxtacanalicular meshwork is involved in the increased resistance of outflow; it is adjacent to the inner wall of the Schlemm’s canal and represents the outermost portion of the trabecular meshwork. The meshwork is made up of a layer of connective tissue delineated by endothelium. The endothelial cells are spindle-shaped with long cytoplasmic processes and contain the same intracytoplasmic material. The juxtacanalicular meshwork is a band of 2–20 µm between the endothelium’s canal and the outermost sheet of the corneoscleral trabeculum. The structure of this portion is made up of two to five layers of
loosely arranged cells in a connective substance that is mostly collagen. The outer cells share a basal lamina with the inner wall of Schlemm’s canal. Some of the cells are joined by desmosomes and gap junctions whereas others are separated by pores of up to 10 mm in width through which aqueous humor can filter to reach the endothelium lining of Schlemm’s canal.8,11
Schlemm’s canal Schlemm’s canal was described by Schlemm in 1830; it drains aqueous humor from the
Schlemm’s canal
29
Figure 3.6 Schlemm’s canal: the vacuolar transcellular channels (from Tripathi, RC. Experimental Eye Research 1977;25:65–116).
trabeculum to the episcleral and conjunctival veins via the collector channels. The canal is a circular channel, single or multiple, 36–40 mm in length and 190–370 µm in diameter. The lumen is either elongated, oval, or triangular, and sometimes crossed by septa; it has a vessel-like structure limited by a layer of endothelial cells with a bulging nucleus in the lumen. The endothelial cells contain numerous Golgi complexes, lysosomes, and actine microfilaments; they are rich with hyaluronic acid synthesized by the Golgi complexes, and are dispersed on a basal lamina interrupted on the trabecular side. The inner wall of the canal is characterized by the presence of giant vacuoles in the endothelial cells; this specific feature is the result of progressive invagination of the
basal pole of the cells. The vacuoles contain aqueous humor and sometimes erythrocytes. In somes cases, vacuoles have apical openings that may connect trabeculum with the lumen of the canal, providing a transcellular channel (Figs 3.5 and 3.6). Studies have shown that the process of vacuole formation is pressure-dependent.12–19 The number and size of the vacuoles depend on IOP. There are many vacuoles at high pressure and few at low pressure. The outer wall of Schlemm’s canal consists of endothelial cells attached one to another by zonulae occludentes laying on a basal lamina more consistent than the one on the inner wall one.8,11,20 A pericanalicular connective tissue surrounds the canal.
30
Anatomical features of outflow pathway Figure 3.7 Collector channels. AC = anterior chamber; PC = posterior chamber; C = cornea; I = iris; S = sclera; CP = ciliary processes; SP = scleral spur; E = endothelium; L = Schwalbe’s line; CM = corneoscleral meshwork; UM = uveal meshwork; SC = Schlemm’s canal; ICC = interior collector channels; ECC = external collector channels; 1 = intrascleral venous plexus; 2 = deep scleral plexus; 3 = ciliary venous plexus; AV = aqueous vein; EPV = episcleral vein; CV = conjunctival vein; M1 = longitudinal fibers of ciliary muscle; M2 = circular fiber of ciliary muscle (modified from Tripathi RC. Experimental Eye Research 1977;25:65–116).
Collector channels Internal collector channels The internal collector channels are simple digitations of the inner wall of Schlemm’s canal, without connection with the anterior chamber and were first described by Sondermann in 1933.
External collector channels The external collector channels arise from the outer wall of the Schlemm’s canal and are
limited by the same endothelium lining surrounded by a fine connective tissue in a complex system of 25–35 vessels running in two directions. Those vessels that reach the episcleral plexus directly are called aqueous veins (up to eight); they were first described by Ascher in 1942 and they were linked to the Schlemm’s canal by Ashton in 1952. Aqueous veins are thick and terminate in episcleral and conjunctival veins in a laminated junction called the laminated vein of Goldman; they can be observed with a slitlamp, 2 mm from the limbus and more often in the inferonasal region. The other system consists of thinner vessels that drain indirectly
References into three staged venous plexuses; the deep and mid-scleral plexus and the episcleral plexus. The deep scleral plexus is made up of fine branches of the anterior ciliary veins and join the mid-scleral plexus to constitute the intrascleral plexus. The intrascleral plexus collects blood from the ciliary venous plexus, which flows into the episcleral plexus. The episcleral plexus receives blood from the conjunctival veins of the perilimbal conjunctiva and drains into the cavernous sinus via the anterior ciliary and superior ophthalmic veins (Fig. 3.7). The conjunctival veins join the superior ophthalmic or facial veins through the palpebral and angular veins.8
Innervation of the outflow pathway The innervation of the trabecular meshwork is provided by the supraciliary nerve plexus and the ciliary plexus, which takes place in the area of the scleral spur. Parasympathetic fibers arise from the ciliary ganglion and perhaps from the pterygopalatine ganglion and the ganglion cells of the ciliary body. Sympathetic fibers come from the superior sympathetic ganglion. Sensory nerves originate from the trigeminal ganglion.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
References 14. 1. Grant WM. Further studies on facility of flow through trabecular meshwork. Arch Ophthalmol 1958;60:324–44. 2. Grant WM. Experimental aqueous perfusion in enucleated human eyes. Arch Ophthalmol 1963;69:783–801. 3. Bill A. The routes for bulk drainage of
15.
16.
31
aqueous humour in the vervet monkey (Cercopithecus ethiops). Exp Eye Res 1966;5:55–7. Shields B. Textbook of glaucoma: aqueous humor dynamics 2nd edn. Baltimore: Kimberly Kist, 1987:5–20. Alm A, Kaufman PL, Kitazawa Y et al. Uveoscleral outflow: biology and clinical aspects. London: Mosby International Limited, 1998. Brubaker RF. The flow of aqueous humor in the human eye. Trans Am Ophthalmol Soc 1982;80:391–5. Cole DF. Aqueous and ciliary body. In: Graymore CN, ed, Biochemistry of the eye. New York: Academic Press, 1970:114–18. Bron AJ, Tripathi RC, Tripathi BJ. Anterior chamber and drainage angle. In: Wolff’s anatomy of the eye and orbit. 8th edn. London: Chapman and Hall, 1997:279–307. Moses RA, Grodzki WF Jr. The scleral spur and scleral roll. Invest Ophthalmol Vis Sci 1977; 16:925–28. Ritch R, Shields M, Krupin T. The Glaucomas. St. Louis: Mosby-Year Book Inc, 1996:71–131. Fine BS. Structure of the trabecular meshwork and the canal of Schlemm. Trans Am Acad Ophthalmol Otol 1966;70:777–83. Johnstone MA, Grant WM, Murray A. Pressure-dependent changes in structures of the aqueous outflow system of human and monkey eyes. Am J Ophthalmol 1973;75:365–82. Tripathi RC. Comparative physiology and anatomy of the aqueous outflow pathway. In: Davson H, ed., The Eye. New York and London. Academic Press. 1974;5:163–237. Tripathi RC. Pathologic anatomy of the outflow pathway of aqueous humour in chronic simple glaucoma. Exp Eye Res 1977;25:403–7. Grierson I, Lee WR. Acid mucopolysaccharides in the outflow apparatus. Exp Eye Res 1975;21:417–31. Grierson I, Lee WR. Light microscopic
32
Anatomical features of outflow pathway
quantitation of the endothelial vacuoles in Schlemm’s canal. Am J Ophthalmol 1977;84:234–45. 17. Grierson I, Lee WR. Pressure effects on flow channels in the lining endothelium of Schlemm’s canal. Acta Ophthalmol 1978;56:935–50. 18. Kayes J. Pressure gradient changes in the trabecular meshwork of monkeys. Am J Ophthalmol 1975;79:549–56.
19. Svedbergh B. Effects of artificial intraocular pressure elevation on the outflow facility and the ultrastructure of the chamber angle in the vervet monkey (Cercopithecus ethiops). Acta Ophthalmol 1974;52:829–44. 20. Speakman JS. Drainage channels in the trabecular wall of the Schlemm’s canal. Br J Ophthalmol 1960;44:513–23.
4 How does non-penetrating glaucoma surgery work?* Douglas H Johnson and Mark Johnson
The advent of a new surgical procedure for glaucoma often raises the question of how the procedure lowers intraocular pressure (IOP), especially given the conventional understandings of the site of outflow resistance in glaucoma. Trabeculectomy was introduced in 1968 as a means of bypassing the clogged trabecular meshwork, allowing aqueous to enter Schlemm’s canal directly through the cut ends of the canal.1 Later experience found trabeculectomy most successful in cases in which a filtration bleb developed, giving rise to the understanding that it functions as a “guarded” filtration procedure.2,3 IOP can be lowered in the absence of a visible filtration bleb, however, indicating that the procedure may well allow aqueous to enter Schlemm’s canal directly in some cases, or alternatively that subclinical transconjunctival filtration of aqueous can occur. Viscocanalostomy and deep sclerectomy are new operations for glaucoma that have been designed to avoid the complications of filtering blebs and also the shallow or flat anterior chambers sometimes seen after trabeculectomy.4,5 Both procedures involve fashioning a partial thickness scleral flap, removing a second layer of sclera deep to the initial flap, and exposing Descemet’s membrane. Descemet’s membrane is thought to act as a semipermeable layer of tissue, allowing aqueous to percolate through it. Schlemm’s
canal is also unroofed during the removal of the second, deep scleral layer. In viscocanalostomy, the cut ends of the canal are then expanded with a viscoelastic material such as Healon. Healon is also injected into the region of excised sclera, or “scleral lake”, to prevent healing. By never entering the anterior chamber or removing trabecular meshwork, complications such as hypotony and hyphema are said to be avoided.4 Deep sclerectomy is similar because Schlemm’s canal is unroofed and Descemet’s membrane exposed by the removal of the second, deeper layer of sclera. Variations of these procedures include removing the inner wall of Schlemm’s canal and adjacent meshwork, but leaving the inner meshwork intact, or placement of a collagen implant or drainage device in the filtration bed.5 Do these operations relieve the specific pathological problem of primary open-angle glaucoma (POAG)? Or do they function as simply another way to make a hole in the eye? This chapter is a summary of current thought on the pathophysiology of aqueous-outflow resistance in glaucoma, and in light of this provides an interpretation of the mechanism of pressure reduction created by these new surgeries. The chapter does not attempt to be comprehensive in scope, but rather to present a synopsis of current conventional wisdom regarding aqueous-outflow resistance.
*Reproduced with permission from Johnson DH, Johnson M. How does non-penetrating glaucoma surgery work? Aqueous outflow resistance and glaucoma surgery. J Glaucoma 10: 55–67 (Lippincott Williams & Wilkins)
34
How does non-penetrating glaucoma surgery work?
Aqueous-outflow resistance The increased IOP found in glaucoma is caused by an increase in aqueous-outflow resistance within the drainage pathways, and not excess secretion of aqueous humor.6,7 Thus, researchers for the past 100 years have concentrated on trying to understand the mechanism by which outflow resistance is generated in the normal eye, and how this resistance is increased in the glaucomatous eye. Aqueous humor passes from the anterior chamber through the outflow pathway as a bulk flow driven by a pressure gradient. Neither metabolic poisons8 nor temperature9 affect this bulk flow (outside of an effect on the viscosity of the fluid), and thus the outflow system does not involve active transport. There is consensus that the bulk of outflow resistance in the normal eye resides near or within the inner wall of Schlemm’s canal; however, there is no such consensus about where the increased outflow resistance characteristic of POAG is localized, although it appears not to reside in the aqueous veins. We begin by reviewing the basis for these conclusions.
Two pathways for aqueous drainage Two pathways for aqueous drainage have been found to exist. The “conventional” pathway through the trabecular meshwork was discovered first. Early experiments showed that dye injected into the anterior chamber enters the episcleral veins and can be seen exiting at the limbus. These limbal vessels on the surface of the eye, the “aqueous veins”, contain aqueous humor.10 On histological
examination, these aqueous veins originate as collector channels in the outer wall of Schlemm’s canal. Casting techniques show that the trabecular meshwork, Schlemm’s canal, the collector channels and aqueous veins, and the episcleral veins form a continuous pathway.11 The “unconventional” or uveoscleral pathway originates at the angle of the eye. Aqueous passes through the ciliary body and ciliary muscle, enters the supraciliary and suprachoroidal spaces, and finally passes through the sclera12–14 or is reabsorbed by the vortex veins.15,16 In lower animals, the origin of this pathway is the ciliary cleft. The ciliary cleft becomes progressively smaller as the ciliary muscle enlarges and accommodative ability of the animal increases. The unconventional outflow may account for 30% of aqueous outflow in young monkeys and young humans.17 With age, uveoscleral outflow becomes reduced, decreasing to perhaps 10% of the total outflow in both monkeys and humans.18 It is difficult to study the amount of aqueous drainage by this route, however, and studies must use radioactive tracers or microspheres to understand the dynamics of fluid flow within this pathway. In humans, these methods cannot be done, and studies must use tonography or aqueous-humor fluorophotometry. These methods make a number of assumptions in the calculation of uveoscleral flow, which make the interpretation of results tenuous.19
Trabecular meshwork (conventional outflow) As stated above, the conventional aqueousoutflow system in the human eye is comprised of the trabecular meshwork, Schlemm’s canal, and the aqueous veins (Fig. 4.1). The trabecu-
Aqueous-outflow resistance
35
Figure 4.1 Trabecular meshwork Schlemm’s canal (SC) appears as a large, single lumen channel in this section. Aqueous spaces between lamellae are visible. (U = uveal meshwork; C = corneoscleral meshwork; J = juxtacanalicular tissue. Light microscopy, toluidine blue stain, 400).
lar meshwork has two major regions, the uvealcorneoscleral region and the juxtacanalicular region. The resistance of the trabecular meshwork to aqueous outflow was assessed by Grant nearly half a century ago.20 In a now classic experiment, a modified scalpel was used to incise the trabecular meshwork of enucleated normal eyes, with measurements taken of outflow resistance before and after each cut. Incision of the uveal and proximal corneoscleral meshwork did not affect outflow resistance. A deeper incision through the entire meshwork and into Schlemm’s canal, however, eliminated 75% of the normal outflow resistance.20 Other investigators have confirmed the finding that an appreciable fraction of outflow resistance (25–50%) is distal to the trabecular meshwork and Schlemm’s canal.21,22 Not all studies agree, however; Mäpea and Bill, using a micropuncture technique to measure the pressure distribution in the outflow pathway,
concluded that less than 10% of the outflow resistance is distal to the inner wall. While this discrepancy has not been resolved, all studies do agree that at least half of aqueous outflow resistance is generated proximal to the aqueous veins. Grant20 also studied a series of eight enucleated glaucomatous eyes, and found that an incision through the meshwork into Schlemm’s canal eliminated all of the abnormal glaucomatous outflow resistance. The remaining scleral resistance was similar to that found in normal eyes. This finding of the abnormal outflow resistance of glaucoma residing proximal to the aqueous veins is supported by the success of trabeculotomy, goniotomy, and direct removal of the trabecular meshwork (goniocurettage) in adults with POAG.23–28 Although the success of these procedures may diminish with time, their initial success in lowering IOP shows that the meshwork is the site of the abnormally high
36
How does non-penetrating glaucoma surgery work?
Figure 4.2 Uveal meshwork as seen from anterior chamber in view similar to clinical gonioscopy. Round cords of tissue form the first layer of meshwork, nearest the anterior chamber. Deeper layers appear as wider, flatter lamellae. (C = cornea; Ir = iris. Scanning electron microscopy, 250).
Figure 4.3 Trabecular meshwork. Corneoscleral lamellae consist of broad, flat sheets of tissue with oval windows (ar rowheads) allowing aqueous to pass between layers. JCT has irregular arrangement of tissue, without organized lamellae. Two erythrocytes are present (arrow), probably displaced during tissue dissection. (CS = corneoscleral lamellae; SC = Schlemm’s canal. Scanning electron microscopy, 250).
outflow resistance in POAG. In addition, laser trabeculoplasty also reduces outflow resistance in the glaucomatous eye.29 Because no incision is made during laser trabeculoplasty, inadvertent fistula formation through the
sclera cannot occur, as could potentially happen after trabeculotomy. While it is unclear how laser trabeculoplasty actually works in the meshwork to lower this abnormal outflow resistance,30,31 its application to
Aqueous-outflow resistance
37
Figure 4.4 Juxtacanalicular tissue. Schlemm’s canal endothelial lining is a continuous sheet of cells, which contain giant vacuoles (GV). JCT appears as loose arrangement of extracellular matrix. Elastic tendons (E), which will insert into inner wall of canal, ar e visible (6 250).
the trabecular meshwork strongly suggests a local meshwork effect. Evidence exists for contraction of the meshwork around the lasered spots, stimulation of trabecular-cell replication, and induction of matrix metalloprotease enzymes that digest the extracellular matrix.32–35
Uveal and corneoscleral meshwork regions
The uveal and corneoscleral regions are composed of a series of sheets or lamellae of collagenous tissue covered by a nearly continuous lining of endothelial cells. The uveal meshwork contains small, thin, cord-like lamellae (Fig. 4.2). The corneoscleral meshwork has wider, flatter lamellae, which contain oval windows between layers (Fig. 4.3). The aqueous spaces between the uveal cords are very large and become smaller in the
subsequent layers of the corneoscleral region. The number and size of these openings are large enough that the uveal and corneoscleral meshwork can be expected to create negligible resistance to flow. Poiseuille’s law predicts that a single pore 100 µm long (the thickness of the trabecular meshwork from the anterior chamber to Schlemm’s canal) with a diameter of 20 µm can carry the entire aqueous outflow (2 µL/min) with a pressure drop of 5 mmHg. Since the uveal and corneoscleral regions of the meshwork have numerous openings this large and larger,36 we can conclude that the pressure drop through this region is negligible. Experimental support for this proposition was provided by Grant,20 who cut through the proximal regions of the meshwork and found no effect on outflow resistance. In the latter stages of open-angle glaucoma, collapse and fusion of the trabecular lamellae have been described in these
38
How does non-penetrating glaucoma surgery work?
Figure 4.5 Inner wall of Schlemm’s canal and underlying juxtacanalicular tissue. Schlemm’s canal has been unroofed by removing outer wall of canal. Inadvertent damage occurred to some inner-wall cells, removing them and exposing the elastic tendons of the JCT. Boundary of damaged cell layer indicated by arrowheads. (IW = inner-wall cells; ET = elastic tendons. Scanning electron microscopy, 1 800).
regions in some eyes, but are not generally considered a prominent feature of the glaucomatous process.37–40
Juxtacanalicular region
The second major region of the trabecular meshwork is the loose tissue near Schlemm’s canal, known as the juxtacanalicular connec-
tive tissue (JCT). This region contains fairly free cells within an extracellular matrix (Fig. 4.4). The cells are interconnected by thin arms to one another, to the cells of the inner wall of Schlemm’s canal, and to fine collagen and elastic fibrils and fibers found in this region. These cells differ from those of the endothelial lining of Schlemm’s canal, because they have a more a fibroblastic appearance and only patches of surrounding basal lamina.41 The
Aqueous-outflow resistance intervening extracellular matrix, or “ground substance”, contains basement membrane material, including collagen IV, laminin, fibronectin, proteoglycans, and glycosaminoglycans. Tendon-like extensions from the ciliary muscle pass through this region42 and insert into the wall of Schlemm’s canal (Fig. 4.5). These connections are responsible for the effects of ciliary muscle contraction on outflow resistance.43
Outflow resistance of JCT and role of extracellular matrix
With its small openings and tortuous flow pathways, the JCT is expected to be the principal site of outflow resistance.44,45 Using microcannulation techniques, Maepen and Bill found that most of the outflow resistance was localized in the JCT of living monkeys, within several µm of the inner wall endothelium.46 Although this technique is difficult and subject to artifact, the results are consistent with the expected site of outflow resistance reported by other studies.37–40,44,45 Studies are not unanimous, however, that the site of outflow resistance is in the JCT. Alternative sites have been proposed, including a layer of cells between the corneoscleral meshwork and JCT,47 fusion of the trabecular lamellae,48 or collapse of Schlemm’s canal.49 Theoretical calculations suggest the JCT contains too much optically empty space to account for the measured outflow resistance of the eye. 44,45,47,50 In addition, the configuration of the JCT changes with IOP, appearing collapsed at low pressures and expanded at higher pressures.50–52 This expansion of the JCT with higher pressures does not fit the measured increase in aqueous-outflow resistance that occurs as IOP is raised. This increase is about 1% per mmHg and is called
39
the outflow obstruction coefficient (Q).53 Conventional thought suggests that partial collapse of the canal accounts for this increase in resistance with increased pressure. Lens depression, which prevents canal collapse, eliminates Q.54 As mentioned above, theoretical calculations44,45,47,50 of the flow resistance of the JCT indicate that the aqueous channels, viewed with conventional electron microscopy, would generate an insignificant fraction of outflow resistance. If the JCT were filled with an extracellular matrix gel such as glycosaminoglycans and proteoglycans, sufficient outflow resistance would be created to match that measured in the eye.44 Glycosaminoglycans and proteoglycans are known to be present in the meshwork and JCT.55–60 Glycosaminoglycans generate flow resistance in other connective tissues,61 and they could create flow resistance in the eye. Proteoglycans consist of glycosaminoglycans attached to a core protein. They are poorly visualized with conventional histochemical techniques.62 Glycosaminoglycans are highly negatively-charged molecules that hold substantial amounts of water, and occupy large volumes of space as a consequence of their charge and hydration. Conventional histochemical preparation techniques employ cationic ions that collapse these macromolecules, and thus they are not well visualized on standard electron micrographs. The result would be large areas of empty space on electron micrographs. Large amounts (up to 40% of total area) of such empty spaces are seen in the JCT on conventional electron micrographs.44,45,47,50 Because it is not known if these empty regions were truly empty in life, before fixation and processing of the tissue, or were filled with proteoglycans that were lost in processing, the empty spaces are called “optically” empty spaces.
40
How does non-penetrating glaucoma surgery work?
Endothelial lining of Schlemm’s canal
While the JCT contains large areas of optically empty space as seen with electron microscopy, a continuous anatomic barrier to aqueous outflow does exist. This barrier is the endothelial lining of Schlemm’s canal. This endothelial lining has several unique aspects which appear to represent a specific engineering solution to a unique physiological situation: movement of fluid into the lumen of a vessel across an intact endothelium down a pressure gradient, without collapsing the lumen, rather than from the lumen into the surrounding tissue. Venous capillaries also have fluid movement into them, from the surrounding tissue, but this is because of the higher oncotic pressure within the lumen that draws the tissue fluid in. As in other endothelia, the endothelial cells that line Schlemm’s canal are attached to one another by tight junctions.41,63 One of the unique aspects of this endothelium is the appearance of out-pouchings or invaginations in the endothelial lining, called “giant vacuoles”41,64 (Figs. 4.4 and 4.6). Giant vacuoles can form within one cell or between neighboring cells. On their basal side is an opening which connects with the underlying aqueous spaces of the JCT. They are not formed by metabolic processes, and do not require energy to form.8,9 Current thought indicates that these structures form passively as a result of the pressure drop across the inner wall endothelium.64,65 Tripathi64 suggested that the giant vacuoles may be transient structures and proposed a process by which vacuolar formation could occur as a cyclic process. A second unique characteristic of the endothelium of Schlemm’s canal is the appearance of small pores passing through these cells (Fig. 4.6). The pores are predominantly intracellular, but a substantial fraction of them are also intercellular66 and may correspond to a
Figure 4.6 Inner wall of Schlemm’s canal. In contrast with figure 4.5, cellular monolayer is intact. Bulging structures are presumably giant vacuoles, although they cannot always be distinguished from prominent nuclei. Pores (arrowheads) are found in presumed giant vacuoles and also in flat areas between cells. (Scanning electron microscopy, 3 700).
paracellular pathway described by Epstein and Rohen.67 The tight junctions between the endothelial cells become less complex, with fewer junctional strands, as IOP increases.63 This “loosening” of the junctions could make the formation of intercellular pores more
Aqueous-outflow resistance
41
Table 4.1 Hydraulic conductivity (Lp) of endothelia19 Endothelium
Lp (cm2 s per g) 10
Reference
Not fenestrated Brain capillary Cornea Lung capillary Skeletal muscle capillary Aorta Mesentery, omentum Aqueous-outflow pathway
0.03 1.6 3.4 2.5–7 9 50 4 000–9 000
117 116 117 117 118 53,117 *
Fenestrated Intestinal mucosa Synovium (knee) Renal glomerulus
32–130 120 400–3 100
53,117 53 53,117
*Flow rate of 2 µL/min driven by a pressure drop of 5 mmHg through a cross-sectional area of between 0.054cm2 and 0.13 cm2 (canal width of 150–350µm; canal length around the eye of 3.6 cm3); note that this is not necessarily Lp for innerwall endothelium since this calculation is based on the entire pressure drop through the outflow pathway; Lp for inner-wall endothelium is probably higher than this value.
likely at higher pressures, and could serve as a means of “self-regulation” of IOP by the meshwork. The pores have an average diameter of about 1 µm (size range: 0.1 µm to greater than 3 µm),68 and their density in the inner wall varies between 1 000 and 2 000 pores/mm2, or roughly 0.2–1 pores/cell.68–74 Scanning electron microscopy shows that 13–29% of giant vacuoles have pores, consistent with this estimate.70,72,73 The nature of the pores is not fully understood, because the number of pores may increase if fixative is perfused through an eye, suggesting that some pores may be fixation artifacts.66,68 The finding of pores by numerous laboratories, however, using both scanning and electron microscopy, suggest they are real structures.64,66,69,71,73,74
Outflow resistance of endothelium
Another unique aspect of the endothelium of Schlemm’s canal is its “leakiness”: it has the highest hydraulic conductivity of any endothelium in the body (Table 4.1). Endothelia of high hydraulic conductivity are usually fenestrated (e.g. glomerulus), whereas the endothelium of Schlemm’s canal is not. Yet the hydraulic conductivity of the canal endothelium is almost ten times higher than that of the renal glomerulus. In addition, the intercellular spaces are composed of tight junctions, which greatly limit fluid flow through the intercellular spaces. In other endothelia, fluid passes between the cells through the cells’ junctional complexes,75 or through fenestra if the endothelia is fenes-
42
How does non-penetrating glaucoma surgery work?
trated.76 Compared with other endothelia that have tight junctions, the large hydraulic conductivity of the endothelium of Schlemm’s canal is even more exceptional. From this, we can conclude that fluid probably passes through the endothelium of Schlemm’s canal by a different mechanism to that which occurs in other endothelia. The pores of the endothelium of Schlemm’s canal appear to constitute this difference. Pores are numerous enough that the outflow resistance has been calculated to be, at most, 10% of the total outflow resistance of that measured in the eye. 71 Furthermore, it has been found that an increase in the IOP increases the number of inner wall pores.73,77 This is not associated with a decrease in outflow resistance53 which provides further evidence that the inner-wall endothelium accounts for only a small part of the outflow resistance. Giant vacuoles and pores allow the endothelium of the canal to function as a one-way valve, because they decrease greatly in number when the pressure within the canal becomes higher than that in the eye;51,73 this prevents reflux of blood from the canal into the eye during periods when the episcleral venous pressure is elevated, such as with bending or Valsalva maneuvers. This unique physiological requirement is also necessary in the drainage pathways for cerebrospinal fluid where giant vacuoles and pores are also seen.78–80 Although the pores of the endothelium of Schlemm’s canal are numerous enough that the endothelium is predicted to have a low resistance to aqueous outflow, disruption of the endothelium can nonetheless greatly decrease outflow resistance. Perfusion with agents that interfere with the cytostructural protein actin (cytochalasins, latrunculins) or that interfere with cell-to-cell contacts (EDTA) cause ruptures of the inner wall that reduce
outflow resistance.81–88 This change is more than would be predicted based on the calculated resistance of the inner-wall pores.66,68,71 Removal of these agents leads to a reversal of the inner-wall ruptures and a return of resistance toward baseline. On the assumption that these agents affect the cytoskeleton as expected and as shown by histological studies,84,86,87 four explanations are possible: 1) The inner and outer walls of Schlemm’s canal may have more resistance than theorized. 2) Disruption of the endothelial cells also disrupts the underlying basement membrane and extracellular matrix. 3) Cytoskeletal agents affect other cells within the meshwork, particularly the cells of the juxtacanalicular tissue. These cells have processes that connect with each other and with the endothelium of the canal. Disruption of these connections could “relax” the juxtacanalicularSchlemm’s canal network, loosening the tethering of the inner wall, expanding the canal wall and increasing the draining surface, permitting more extensive flow through the meshwork. 89 4) The inner wall acts in conjunction with the underlying extracellular matrix to modulate outflow resistance. In this potential hydrodynamic interaction, termed “funneling”, the endothelial pores themselves contribute negligible flow resistance, but since they force the fluid to “funnel” through those regions of the JCT nearest the pores, their number and size can greatly increase the effective outflow resistance of the JCT.90 Disruption of cells, or separation of innerwall cells from their underlying attachments, would eliminate the funneling effect and decrease outflow resistance. This may
Aqueous-outflow resistance explain the results of studies reporting that disruption of the inner-wall cells decreases outflow resistance.81,84,86,89
Collapse of Schlemm’s canal
Schlemm’s canal is a continuous channel oriented in a circumferential direction. The canal is oval in shape, with dimensions of about 280 30 µm at low IOP.50,80,91,92 A lumen of this size is too large to generate an appreciable outflow resistance. As IOP increases, the trabecular meshwork expands into the lumen of the canal, causing a concomitant narrowing of the lumen,51–80 raising the possibility that this collapse may cause a significant increase in outflow resistance. However, throughout the canal, especially near the collector channels, septae are present between the inner and outer walls. The proximity of these structures to the collector channels suggests they will prevent collapse of the canal and occlusion of the collector channels as IOP is increased.80,91,93 At an IOP of 40 mmHg, the canal is predicted to be largely collapsed except at the sites of the septae.91 Nesterov postulated that canal collapse could cause the elevated outflow resistance characteristic of POAG, and designed an operation to unroof the canal to remedy this problem.49 Although outflow resistance is elevated by collapse of the canal, resistance levels at high IOPs are not as high as found in glaucoma (in normal eyes, facility changed from a baseline of 0.40 µL/min per mmHg at 10 mmHg to a facility of 0.28 µL /min per mmHg at 50 mmHg; whereas the facility of eyes with POAG is usually less than 0.13).53,91 Assuming that Q, the obstruction of outflow with increasing IOP, is caused by collapse of the canal, the underlying problem in glaucoma must therefore involve more than just the collapse of
43
Schlemm’s canal. While collapse of Schlemm’s canal is not the primary cause of glaucoma, if it occurs it can make the problem of increased IOP worse. Pilocarpine, which decreases outflow resistance, increases ciliary muscle tone and acts to expand the trabecular meshwork and JCT, and may also open the canal.94
Collector channels and aqueous veins
After entering the canal, the aqueous humor travels circumferentially around the eye until it reaches one of the 30 or so collector channels that join Schlemm’s canal. Fluid flows from the collecting channels into aqueous veins that ultimately drain into the episcleral venous system. The aqueous veins have an average diameter of 50 µm and a length of about 1 mm.21 Use of Poiseuille’s law indicates that the resistance of the aqueous veins should be negligible, if the veins are neither collapsed nor compressed. Measurement of the pressure in Schlemm’s canal in live monkeys supports this conclusion.46,95 Experimental evidence in the human eye, however, indicates that some resistance does occur in the collector channels and aqueous vein system. The trabeculotomy experiments previously mentioned have shown that 25–50% of the total outflow resistance is distal to Schlemm’s canal, presumably in the aqueous veins. However, most studies suggest the abnormal increase in outflow resistance found in glaucoma is not found in the aqueous veins nor in Schlemm’s canal. 20,23,91
Abnormal outflow resistance in the glaucomatous eye Several pieces of evidence indicate the trabecular meshwork is the site of the abnormally
44
How does non-penetrating glaucoma surgery work?
increased outflow resistance of POAG, as discussed above. Surprisingly, histological examination of the meshwork does not show specific abnormalities or ultrastructural changes that could account for the elevation of pressure.37–40 The few changes found appear to be an exaggeration of aging changes found in the normal eye. A small increase in the amount of tendon and tendon-sheath material in the JCT is found, increasing from 15% to 22% in POAG when compared with aged normal eyes.37,38 This increase in tendon and tendonsheath material does not occur early in the disease process: IOP can be elevated even with normal amounts of the tendon and tendonsheath material.39 The increase in the tendon and tendon-sheath material is not enough to obstruct aqueous channels. 44,45,47,50 Studies have also examined glycosaminoglycans and proteoglycans of the meshwork in POAG, phagocytosis by trabecular cells, and the size of Schlemm’s canal.59,60,92,96 Elucidation of the pathophysiologic mechanism of POAG remains an area of intense research.
Uveoscleral outflow (unconventional pathway) The uveoscleral (or uveovortex) pathway originates at the angle of the eye, passes through the ciliary body and ciliary muscle, enters the supraciliary and suprachoroidal spaces, and finally passes through the sclera.12,13 Aqueous humor and aqueous proteins seep through sclera and episclera, passing into orbit, and are absorbed there by blood vessels. Aqueous may also be absorbed osmotically by the vortex veins.15,16 The unconventional outflow is relatively insensitive to IOP, increasing only a small amount with increases of pressure.14 The ciliary muscle probably represents a major site of flow resistance along this
pathway. Pilocarpine, which causes ciliary muscle contraction and decreases the size of spaces between the muscle bundles, decreases outflow through this pathway. Atropine, which relaxes the ciliary muscle, does the converse.97 Furthermore, prostaglandin F2, shown to increase unconventional outflow,98–101 may act by decreasing the extracellular matrix between ciliary muscle bundles.102
The mechanisms of laser and glaucoma surgery Laser trabeculoplasty Since laser trabeculoplasty was first described,29 it has been recognized that this procedure decreases outflow resistance by a mechanism other than simply making holes in the trabecular meshwork. When holes are created, they quickly heal shut (even with a YAG laser).29,103 Wise and Witter29 hypothesized that laser trabeculoplasty worked mechanically, either by shrinking collagen or through the formation of scar tissue that later contracts. Such contraction or shrinkage would lead to tension on the remaining trabecular meshwork, which then would open the intertrabecular spaces,29 and/or prevent collapse of Schlemm’s canal.31 Melamed and Epstein provided support for this hypothesis by showing that the actual sites of the laser burns appeared to be non-filtering, with aqueous flow being diverted to the remaining meshwork.104 Laser-induced shrinkage of the trabecular tissues does not lead to an immediate change in outflow facility in enucleated human
The mechanisms of laser and glaucoma surgery
45
Figure 4.7 Inner wall of Schlemm’s canal and ruptured septa. Canal has been unroofed by removing outer wall. One of the septae that bridge the inner and outer walls has been inadvertently damaged, revealing loose arrangement of cells within. (IW = inner wall cells; Sep = ruptured septa. Scanning electron microscopy, 3 700).
eyes.30,31 This finding is consistent with the clinical observation that it takes about 3–6 weeks after laser for outflow facility to improve.29 While it could be that scarring caused by the laser takes several weeks to occur and affect outflow facility, another mechanism of laser action could be a change in the activity of the trabecular cells.30,31,105,106 Increased phagocytic activity, activation of cells leading to altered metabolic activity, increased levels of cell division, or a tissue remodeling between the lasered spots may occur. However, no conclusive experiment has been done to elucidate the mechanism by which improvements in outflow facility occurs after laser treatment.
Schlemm’s canal. 1 Healing and fibrosis occur, however, and it is likely that the cut ends of the canal become closed with scar tissue.2,107,108 The development of a filtering bleb is strong evidence that aqueous humor bypasses both the meshwork and canal, and exits through the surgical fistula. The advantage to this procedure over a full-thickness filter is the prevention of low pressures provided by the scleral flap. Once the eye has healed, aqueous continues to seep through the fistula and enters the filtration bleb.
Non-penetrating surgery Viscocanalostomy
Trabeculectomy As mentioned above, the original concept of trabeculectomy was to bypass the trabecular meshwork and allow aqueous humor to enter
Unroofing Schlemm’s canal (removing the outer wall of the canal) can cause damage to the inner wall of the canal (Figs 4.5 and 4.7).109,110 The septae, which bridge the inner and outer walls, can easily damage the inner
46
How does non-penetrating glaucoma surgery work?
(a) (a)
(b)
(b)
Figure 4.8 Viscocanalostomy in necropsy eye. Dilated Schlemm’s canal is apparent, this region of canal had a cannula inserted. Rupture of anterior portion of canal wall apparent at low (Fig. 4.8a) and high (Fig. 4.8b) magnification. Arrow denotes ruptured anterior portion of canal wall. Note compaction of meshwork immediately underlying inner wall of canal, with loss of intertrabecular spaces. (SC = Schlemm’s canal; toluidine blue; Fig. 4.8a at 40, Fig. 4.8b at 400). AC = anterior chamber; CC = collector channel.
Figure 4.9 Viscocanalostomy showing portion of canal receiving viscoelastic material only. This portion of meshwork was distal to the end of the cannula. Note septum within lumen of canal. Dilation of canal, with rupture of anterior portion of canal wall is apparent at low (Fig. 4.9a) and high (Fig. 4.9b) magnification (long arrow). Fig. 4.9a also shows disruption of inner wall of canal (short arrow) from displacement of septum: note sharp kink in septum. Same eye as Fig. 4.8. (toluidine blue; Fig. 4.9a at 40, Fig. 4.9b at 400). AC = anterior chamber; SC = Schlemm’s canal.
The mechanisms of laser and glaucoma surgery wall when they are pulled away during the unroofing procedure. Injection of viscoelastic into the ends of Schlemm’s canal is designed to enlarge the canal,4 but it is likely that this injection ruptures both the inner and outer endothelial walls of the canal as shown in both the human eye (Figs 4.8 and 4.9) and the monkey.110 These ruptures probably extend into the JCT, and may also rupture some of the meshwork itself. The operation probably functions as a “gentle” trabeculectomy, allowing aqueous to bypass the site of abnormal outflow resistance, the JCT tissue and enter the canal through these presumed and inadvertent ruptures. In addition, excising a deep layer of sclera and exposing Descemet’s membrane may also create a route for aqueous drainage that bypasses the meshwork. Studies in the rabbit, however, indicate that Descemet’s membrane is not permeable enough to allow relief of the elevated IOP of glaucoma.111,112 If the ruptured regions of the JCT and canal heal with time, surgery may fail in those eyes that did not develop filtration blebs. There is no theoretical basis for relieving elevated IOP by expanding the lumen of Schlemm’s canal. Injection of viscoelastic will certainly dilate the canal lumen, but the viscoelastic itself will probably not remain in the canal long enough to prevent healing of the cut ends of the canal. The high-molecularweight viscoelastic used in this procedure has not been shown to retard healing. The creation of a “scleral lake” underneath the partial thickness scleral flap has no theoretical effect on the abnormal outflow resistance found in glaucoma. Ultrasonic measurements of the area of the lake found no relation to IOP in a series of human eyes.113 A small effect on pressure may occur by removing a segment of the aqueous veins in that region, eliminating some of the normal resis-
47
tance created by these aqueous veins (see above). If the cut ends of the aqueous veins did not heal shut, but remained open, this might lower IOP a few millimeters. Deep sclerectomy
During conventional trabeculectomy, many surgeons do not actually remove a piece of the trabecular meshwork at all, but rather a piece of cornea anterior to the meshwork. Deep sclerectomy takes this approach, with the exception of leaving Descemet’s membrane intact.5 Descemet’s membrane is semi-permeable, and therefore creates some resistance to aqueous drainage into the surgical fistula, but as indicated above, Descemet’s membrane is not permeable enough to relieve the elevated pressure of glaucoma.111,112 Deep sclerectomy appears to be another variation of a “guarded filter”, adding a second “guard” to that of the partial thickness scleral flap, which is still used in this surgery. The operation also unroofs Schlemm’s canal, and aqueous percolates through the remaining trabeculo-Descemet’s membrane. As mentioned above, damage to the inner wall of the canal is highly likely to occur during the removal of the outer wall when unroofing the canal (Figs 4.5 and 4.7). Such damage to the inner wall and underlying juxtacanalicular tissue would allow aqueous a new route into Schlemm’s canal.
Trabecular aspiration Application of high vacuum to the meshwork region is reported to lower IOP in pseudoexfoliative glaucoma, but has little effect on POAG.114,115 If the suction were strong enough to break the endothelial lining of the canal and rupture the JCT, IOP would be lowered; this would be expected to lower pressure in both
48
How does non-penetrating glaucoma surgery work?
POAG and pseudoexfoliation. If the suction were not that strong, it could function in removing the accumulations of pseudoexfoliative material that occur within the meshwork.116
Goniotomy While spectacularly successful in infant eyes with abnormally developed meshworks, goniotomy and trabeculotomy have been generally disappointing in adult eyes.3,117,118 Trabeculotomy and goniotomy make excellent theoretical sense, because the surgical incision through the trabecular meshwork and JCT into Schlemm’s canal allows aqueous to bypass the abnormal portions of the meshwork. Because the aqueous veins may have some outflow resistance, as discussed above, IOP should not drop too low. Clinical studies report that pressure stabilizes in the high teens.117,118 Blood reflux into the eye from Schlemm’s canal could occur if the episcleral venous pressure was increased during a cough or Valsalva maneuver, or if the patient were to bend over. This would probably be an acceptable, minor side-effect if the operation otherwise kept IOP normal, avoiding a filtration bleb and all of its attendant problems. Healing of the goniotomy incision has been the main problem in adult eyes.107,108,117–119 Why healing does not occur more frequently in the infant eye is unknown, but may relate to the elastic condition of infant eye tissues. Infant sclera retracts when cut, sometimes making it difficult to make a trabeculectomy scleral flap cover the surgical bed from which it was dissected. This same elastic property may be the reason that infant eyes become larger when IOP is elevated. Such buphthalmos occurs until about the age of 2 years, which is about the same age at which goniotomies are no longer
are effective in children. In the infant eye undergoing goniotomy, the cut ends of the meshwork may retract enough because of the elastic nature of the infantile tissues that the incision gapes, and does not heal together. In the adult eye, such elastic retraction would not occur, allowing the cut ends of the meshwork to lie in apposition to each other, and thus allow healing of the incision. A study of goniotomy in adults has suggested that incision of the meshwork near Schwalbe’s line, anterior to the usual site of incision in goniotomy, may overcome this problem. Success in adult eyes has been reported with this anterior incision.120
Goniocurettage Goniocurettage is an operation related to goniotomy that involves removal of the trabecular meshwork. By the use of a sharpened curette to scrape away the meshwork for about 90° of the circumference of the eye, Jacobi et al121–123 report success in lowering IOP. An ab interno incision is used, and a filtration bleb is not created. Damage to the collector channels during the removal of meshwork could limit the effectiveness of this procedure, which is the most promising of the new glaucoma surgeries. IOP is lowered into the high teens, and does not reach as low a level as after conventional filtration surgery.
Future goals Our understanding of aqueous-outflow mechanisms is incomplete, especially with regard to the pathogenesis of POAG. Although ideas abound, and a working hypothesis has been presented in this chapter,
References much remains unsolved. The ideal surgical procedure would address the as-yet-unknown site of pathology in glaucoma and leave the eye otherwise intact. In practice, however, any procedure that is effective in lowering IOP, that has minimal complications and sideeffects, and provides long-term success in pressure control would be helpful in the management of glaucoma. The current practice of filtration surgery, especially with the use of antifibrotic agents such as mitomycin C, creates eyes that can develop conjunctival leaks, infection, and problems from filtration blebs. We look forward to improvements in the surgical control of pressure in the new millennium.
Summary Histological, experimental, and theoretical studies of the aqueous-outflow pathways point toward the juxtacanalicular region and inner wall of Schlemm’s canal as the likely site of aqueous-outflow resistance in the normal eye. At least 50% of the aqueous-outflow resistance in the normal eye, and the bulk of the pathologically increased resistance in the glaucomatous eye reside in the trabecular meshwork and/or in the inner wall of Schlemm’s canal. The uveoscleral/uveovortex pathways, which account for perhaps 10% of the aqueous drainage in the normal aged human eye, can become major accessory routes for aqueous drainage after pharmacological treatment. Surgeries designed to incise or remove the abnormal trabecular meshwork of glaucoma address the pathological problem of the disease. Surgeries that unroof Schlemm’s canal or expand the canal, such as viscocanalostomy, probably cause inadvertent
49
ruptures of the inner wall and juxtacanalicular tissue, thus relieving the abnormal outflow resistance of glaucoma. This chapter is a summary of current thought on the pathophysiology of aqueous-outflow resistance in glaucoma, and in light of this provides an interpretation of the mechanism of pressure reduction created by these new surgeries.
References 1. Cairns JE. Trabeculectomy: preliminary report of a new method. Am J Ophthalmol 1968;66:673–79. 2. Spencer WH. Histologic evaluation of microsurgical glaucoma techniques. Trans Am Acad Ophthalmol Otolaryngol 1972;76:389–97. 3. Schwartz AL, Anderson DR. Trabecular surgery. Arch Ophthalmol 1974;92:134–38. 4. Stegmann R, Pienaar A, Miller D. Viscocanalostomy for open-angle glaucoma in black African patients. J Cataract Refract Surg 1999;25:316–22. 5. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with collagen implant and trabeculectomy in openangle glaucoma. J Cataract Refract Surg 1999;25:323–31. 6. Grant WM. Clinical measurements of aqueous outflow. Arch Ophthalmol 1951;46:113–31. 7. Brubaker RF. Measurement of aqueous flow by fluorophotometry. In: Ritch R, Shields MB, eds. The glaucomas, Vol I. St Louis, MO: Mosby, 1989:337–44. 8. Bárány EH. In vitro studies of the resistance to flow through the angle of the anterior chamber. Acta Soc Medicorum Upsaliensis 1953;59:260–76. 9. Van Buskirk EM, Grant WM. Influence of temperature and the question of involvement of cellular metabolism in aqueous outflow. Am J Ophthalmol 1974;77:565–72.
50
How does non-penetrating glaucoma surgery work?
10. Ascher KW. Aqueous veins. Am J Ophthalmol 1942;25:31–38. 11. Ujiie K, Bill A. The drainage routes for aqueous humor as revealed by scanning electron microscopy of corrosion casts. Scanning Electron Microscopy 1984;2:849–56. 12. Bill A. Conventional and uveo-scleral drainage of aqueous humour in the cynomolgus monkey (Macaca irus) at normal and high intraocular pressures. Exp Eye Res 1966;5:45–54. 13. Bill A, Phillips CI. Uveoscleral drainage of aqueous humor in human eyes. Exp Eye Res 1971;12:275–81. 14. Bill A. Blood circulation and fluid dynamics in the eye. Physiol Rev 1975;55:383–416. 15. Pederson JE, Gaasterland DE, MacLellan HM. Uveoscleral aqueous outflow in the rhesus monkey: importance of uveal reabsorption. Invest Ophthalmol Vis Sci 1977;16:1008–17. 16. Sherman SH, Green K, Laties AM. The fate of anterior chamber fluorescein in the monkey eye. I. The anterior chamber outflow pathways. Exp Eye Res 1978;27:159–73. 17. Toris CB, Yablonski ME, Wang YL, Camras CB. Aqueous humor dynamics in the aging human eye. Am J Ophthalmol 1999 127:407–12. 18. Gabelt BT, Kaufman PL. Uveoscleral outflow decreases in old rhesus monkeys. Invest Ophthalmol Vis Sci 2000;41:S253 (ARVO abstr). 19. Johnson M, Erickson K. Mechanisms and routes of aqueous humor drainage. In: Albert DM, Jakobiec FA, eds. Principles and practice of ophthalmology, Vol 4. Philadelphia: WB Saunders Co, 2000, 2577–95. 20. Grant WM. Experimental aqueous perfusion in enucleated human eyes. Arch Ophthalmol 1963;69:783–801. 21. Rosenquist R, Epstein D, Melamed S et al. Outflow resistance of enucleated human eyes at two different perfusion pressures and
different extents of trabeculotomy. Curr Eye Res 1989;8:1233–40. 22. Schuman JS, Chang W, Wang N et al. Excimer laser effects on outflow facility and outflow pathway morphology. Invest Ophthalmol Vis Sci 1999;40:1676–80. 23. Luntz MH, Livingston DG. Trabeculotomy ab externo and trabeculectomy in congenital and adult-onset glaucoma. Am J Ophthalmol 1977;83:174–79. 24. Tanihara H, Negi A, Akimoto M et al. Surgical effects of trabeculectomy ab externo on adult eyes with primary open angle glaucoma and pseudoexfoliation syndrome. Arch Ophthalmol 1993;111:1653–61. 25. Quaranta L, Hitchings RA, Quaranta CA. Ab-interno goniotrabeculotomy versus mitomycin C trabeculectomy for adult openangle glaucoma: a 2–year randomized clinical trial. Ophthalmology 1999;106:1357–62. 26. Jacobi PC, Dietlein TS, Krieglstein GK. Goniocurettage for removing trabecular meshwork: clinical results of a new surgical technique in advanced chronic open-angle glaucoma. Am J Ophthalmol 1999;127:505–10. 27. Jacobi PC, Dietlein TS, Krieglstein GK. Microendoscopic trabecular surgery in glaucoma management. Ophthalmology 1999;106:538–44. 28. Jacobi PC, Dietlein TS, Krieglstein GK. Technique of goniocurettage: a potential treatment for advanced chronic open angle glaucoma. Br J Ophthalmol 1997;81:302–07. 29. Wise JB, Witter SL. Argon laser therapy for open-angle glaucoma: a pilot study. Arch Ophthalmol 1979;97:319–22. 30. Van Buskirk EM, Pond V, Rosenquist RC, Scott TS. Argon laser trabeculoplasty: studies of mechanism of action. Ophthalmology 1984;91:1005–10. 31. Van Buskirk EM. Pathophysiology of laser trabeculoplasty. Surv Ophthalmol 1989;33:264–72.
References 32. Melamed S, Epstein DL. Alterations of aqueous humour outflow following argon laser trabeculoplasty in monkeys. Br J Ophthalmol 1987;71:776–81. 33. Bylsma SS, Samples JR, Acott TS, Van Buskirk EM. Trabecular cell division after argon laser trabeculoplasty. Arch Ophthalmol 1988;106:544–47. 34. Acott TS, Samples JR, Bradley JMB et al. Trabecular repopulation by anterior trabecular meshwork cells after laser trabeculoplasty. Am J Ophthalmol 1989;107:1–6. 35. Dueker DK, Norberg M, Johnson DH et al. Stimulation of cell division by argon and NA:YAG laser trabeculoplasty in cynomolgus monkeys. Invest Ophthalmol Vis Sci 1990;31:115–24. 36. Tripathi RC. Comparative physiology and anatomy of the aqueous outflow pathway. In: Davson H, Graham LT, eds. The eye: comparative physiology, Vol 5. London and New York: Academic Press, 1974:163–356. 37. Rohen JW. Why is intraocular pressure elevated in glaucoma? Ophthalmology 1983;90:758–65 38. Lütjen-Drecoll E, Shimizu T, Rohrbach M, Rohen JW. Quantitative analysis of ‘plaque material’ in the inner and outer wall of Schlemm’s canal in normal and glaucomatous eyes. Exp Eye Res 1986;42:443–55. 39. Gottanka J, Johnson DH, Martus P, LütjenDrecoll E. Severity of optic nerve damage in eyes with POAG is correlated with changes in the trabecular meshwork. J Glaucoma 1997;6:123–32. 40. Alvarado JA, Yun AJ, Murphy CG. Juxtacanalicular tissue in primary open angle glaucoma and in nonglaucomatous normals. Arch Ophthalmol 1986;104:1517–28. 41. Gong H, Tripathi RC, Tripathi BJ. Morphology of the aqueous outflow pathway. Microscopy Res Tech 1996;33:336–67. 42. Rohen JW, Futa R, Lütjen-Drecoll E. The fine structure of the cribriform meshwork in normal and glaucomatous eye as seen in
51
tangential sections. Invest Ophthalmol Vis Sci 1981;21:574–85. 43. Rohen JW, Lütjen E, Bárány E. The relationship between the ciliary muscle and the trabecular meshwork and its importance for the effect of miotics on aqueous outflow resistance. Albrecht v Graefes Arch Klin Exp Ophthal 1967;172:23–47. 44. Ethier CR, Kamm RD, Palaszewski BA et al. Calculations of flow resistance in the juxtacanalicular meshwork. Invest Ophthalmol Vis Sci 1986;27:1741–50. 45. Seiler T, Wollensak J. The resistance of the trabecular meshwork to aqueous humor outflow. Graefe’s Arch Clin Exp Ophthalmol 1985;223:88–91. 46. Mäpea O, Bill A. Pressures in the juxtacanalicular tissue and Schlemm’s canal in monkeys. Exp Eye Res 1992;54:879–83. 47. Murphy CG, Johnson M, Alvarado JA. Juxtacanalicular tissue in pigmentary and primary open angle glaucoma: the hydrodynamic role of pigment and other constituents. Arch Ophthamol 1992;110:1779–85. 48. Teng CC, Katzin HM, Chi HH. Primary degeneration in the vicinity of the chamber angle as an etiologic factor in wide-angle glaucoma. Am J Ophthalmol 1957;43:192–203. 49. Nesterov AP. Role of blockade of Schlemm’s canal in pathogenesis of primary open angle glaucoma. Am J Ophthalmol 1970;70:691–96. 50. Ten Hulzen RD, Johnson DH. Effect of fixation pressure on juxtacanalicular tissue and Schlemm’s canal. Invest Ophthalmol Vis Sci 1996;37:114–24. 51. Johnstone MA, Grant WM. Pressure dependent changes in the structures of the aqueous outflow system of human and monkey eyes. Am J Ophthalmol 1973;75:365–83. 52. Van Buskirk EM. Anatomic correlates of changing aqueous outflow facility in excised human eyes. Invest Ophthalmol Vis Sci 1982;22:625–32.
52
How does non-penetrating glaucoma surgery work?
53. Brubaker RF. The effect of intraocular pressure on conventional outflow resistance in the enucleated human eye. Invest Ophthalmol Vis Sci 1975;14:286–92. 54. Van Buskirk EM. Changes in the facility of aqueous outflow induced by lens depression and intraocular pressure in excised human eyes. Am J Ophthalmol 1976;82:736–40. 55. Acott TS, Westcott M, Passo MS, Van Buskirk EM. Trabecular meshwork glycosaminoglycans in human and cynomolgus monkey eye. Invest Ophthalmol Vis Sci 1985;26:1320–29. 56. Tschumper RC, Johnson DH, Bradley JMB, Acott T. Glycosaminoglycans of human trabecular meshwork in perfusion organ culture. Curr Eye Res 1990;9:363–69. 57. Johnson DH, Bradley J, Acott T. The effect of dexamethasone on glycosaminoglycan of human trabecular meshwork in perfusion organ culture. Invest Ophthalmol Vis Sci 1990;31:2568–71. 58. Johnson DH, Knepper PA. Microscale analysis of the glycosaminoglycans of the human trabecular meshwork: a study in perfusion cultured eyes. J Glaucoma 1994;3:58–69. 59. Knepper PA, Goosens W, Hvizd M, Palmberg PF. Glycosaminoglycans of the human trabecular meshwork in primary open angle glaucoma. Invest Ophthalmol Vis Sci 1996;37:1360–67. 60. Knepper PA, Goosens W, Palmberg PF. Glycosaminoglycan stratification of the juxtacanalicular tissue in normal and primary open angle glaucoma. Invest Ophthalmol Vis Sci 1996;37:2414–25. 61. Levick JR. Flow through interstitium and other fibrous matrices. Quart J Exp Physiol 1987;72:409–37. 62. Hascall VC, Hascall GK. Proteoglycans. In: Hay ED, ed. Cell biology of extracellular matrix. New York, NY: Plenum Press, 1981:39–63. 63. Ye W, Gong H, Sit A et al. Interendothelial junctions in normal human Schlemms’s canal
respond to changes in pressure. Invest Ophthalmol Vis Sci 1997;38:2460–68. 64. Tripathi RC. Ultrastructure of Schlemm’s canal in relation to aqueous outflow. Exp Eye Res 1968;7:335–41. 65. Johnstone MA. Pressure-dependent changes in nuclei and the process origins of the endothelial cells lining Schlemm’s canal. Invest Ophthalmol Vis Sci 1979;18:44–51. 66. Ethier CR, Coloma FM, Sit AJ, Johnson M. Two pore types in the inner-wall endothelium of Schlemm’s canal. Invest Ophthalmol Vis Sci 1998;39:2041–48. 67. Epstein DL, Rohen JW. Morphology of the trabecular meshwork and inner-wall endothelium after cationized ferritin perfusion in the monkey eye. Invest Ophthalmol Vis Sci 1991;32:160–71. 68. Sit AJ, Coloma FM, Ethier CR, Johnson M. Factors affecting the pores of the inner wall endothelium of Schlemm’s canal. Invest Ophthalmol Vis Sci 1997;38: 1517–25. 69. Kays J. Pore structure of the inner wall of Schlemm’s canal. Invest Ophthalmol Vis Sci 1967;6:381–94. 70. Bill A. Scanning electron microscopic studies of the canal of Schlemm. Exp Eye Res 1970;10:214–18. 71. Bill A, Svedbergh B. Scanning electron microscopic studies of the trabecular meshwork and the canal of Schlemm––an attempt to localize the main resistance to outflow of aqueous humor in man. Acta Ophthamol 1972;50:295–320. 72. Segawa K. Pore structures of the endothelial cells of the aqueous outflow pathway: scanning electron microscopy. Jpn J Ophthalmol 1973;17:133–39. 73. Lee WR, Grierson L. Pressure effects on the endothelium of the trabecular wall of Schlemm’s canal: a study by scanning electron microscopy. Graefe’s Arch Clin Exp Ophthalmol 1975;196:255–65. 74. Svedbergh B. Effects of intraocular pressure on the pores of the inner wall of Schlemm’s
References canal. Jpn J Ophthalmol 1976;50:127–35. (Suppl III) 75. Curry FE, Michel CC. A fiber matrix model of capillary permeability. Microvasc Res 1980;20:96–99. 76. Levick JR, Smaje LH. An analysis of the permeability of a fenestra. Microvasc Res 1987;33:233–56. 77. Grierson I, Lee WR. Pressure effects on flow channels in the lining endothelium of Schlemm’s canal. Acta Ophthalmologica 1978;56:935–52. 78. Tripathi RC. Tracing the bulk outflow route of cerebrospinal fluid by transmission and scanning electron microscopy. Brain Res 1974;80:503–06. 79. Tripathi RC, Tripathi BJ. Bulk flow of humors of the eye and brain through vacuolar transendothelial channels. Prog Appl Microcirc 1985;5:118–34. 80. Tripathi RC. The functional morphology of the outflow systems of ocular and cerebrospinal fluids. Exp Eye Res 1977;24:65–116 (suppl). 81. Johnson DH. The effect of cytochalasin D on outflow facility and the trabecular meshwork of the human eye in perfusion organ culture. Invest Ophthalmol Vis Sci 1997;38:2790–99. 82. Kaufman PL, Bárány EH. Cytochalasin B reversibly increases outflow facility in the eye of the cynomolgus monkey. Invest Ophthalmol Vis Sci 1977;16:47–53. 83. Kaufman PL, Bill A, Bárány EH. Effect of cytochalasin B on conventional drainage of aqueous humor in the cynomolgus monkey: the ocular and cerebrospinal fluids. Exp Eye Res 1977;25:411–14. 84. Bill A, Lütjen-Drecoll E, Svedbergh B. Effects of intracameral Na 2EDTA and EGTA on aqueous outflow routes in the monkey eye. Invest Ophthalmol Vis Sci 1980;19:492–504. 85. Kaufman PL, Erickson K. Cytochalasin B and D dose-outflow facility response relationships in the cynomolgus monkey. Invest Ophthalmol Vis Sci 1982;23:646–50. 86. Hamanaka T, Bill A. Morphological and
53
functional effects of Na 2EDTA on the outflow routes for aqueous humor in monkeys. Exp Eye Res 1987;44:171–90. 87. Svedbergh B, Lütjen-Drecoll E, Oberr M, Kaufman PL. Cytochalasin B-induced structural changes in the anterior ocular segment of the cynomolgus monkey. Invest Ophthalmol Vis Sci 1987;17:718–34. 88. Peterson JA, Tian B, Bershadsky AD et al. Latrunculin-A increases outflow facility in the monkey. Invest Ophthalmol Vis Sci 1999;40:931–41. 89. Sabanay I, Gabelt BT, Tian B et al. H-7 effects on the structure and fluid conductance of monkey trabecular meshwork. Arch Ophthalmol 1999;118:955–62. 90. Johnson M, Shapiro A, Ethier CR, Kamm RD. Modulation of outflow resistance by the pores of the inner wall endothelium. Invest Ophthalmol Vis Sci 1992;33:1670–75. 91. Johnson M, Kamm RD. The role of Schlemm’s canal in aqueous outflow from the human eye. Invest Ophthalmol Vis Sci 1983;24:320–25. 92. Johnson DH, Matsumoto Y. Schlemm’s canal becomes smaller after successful filtration surgery. Arch Ophthalmol 2000;118:1251–56 93. Hoffman F, Dumitrescu L. Schlemm’s canal under the scanning electron microscope. Ophthalmic Res 1971;2:37–45. 94. Van Buskirk EM. Anatomic correlates of changing aqueous outflow facility in excised human eyes. Invest Ophthalmol Vis Sci 1982;22:625–32. 95. Mäpea O, Bill A. The pressures in the episcleral veins, Schlemm’s canal and the trabecular meshwork in monkeys: effects of changes in intraocular pressure. Exp Eye Res 1989;49:645–63. 96. Matsumoto Y, Johnson DH. Trabecular meshwork phagocytosis in glaucomatous eyes. Ophthalmologica 1997;211: 147–52. 97. Bill A, Wålinder P-E. The effects of
54
How does non-penetrating glaucoma surgery work?
pilocarpine on the dynamics of aqueous humor in a primate (Macaca irus). Invest Ophthalmol 1966;5:170–75. 98. Crawford K, Kaufman PL, Gabelt BT. Effect of topical PGF 2a on aqueous humor dynamics in cynomolgus monkeys. Curr Eye Res 1987;6:1035–44. 99. Camras CB, Podos SM, Rosenthal JS et al. Multiple dosing of prostaglandin F2 or epinephrine on cynomolgus monkey eyes. I. Aqueous humor dynamics. Invest Ophthalmol Vis Sci 1987;28:463–69. 100. Camras CB, Bhuyan KC, Podos SM et al. Multiple dosing of prostaglandin F2 or epinephrine on cynomolgus monkey eyes. II. Slit-lamp biomicroscopy, aqueous humor analysis, and fluorescein angiography. Invest Ophthalmol Vis Sci 1987;28:921–26. 101. Kerstetter JR, Brubaker RF, Wilson SE, Kullersrand LJ. Prostaglandin F 2-1-isopropyl ester lowers intraocular pressure without decreasing aqueous humor flow. Am J Ophthalmol 1988;105:30–34. 102. Lütjen-Drecoll E, Tamm E. Morphological study of the anterior segment of cynomolgus monkey eyes following treatment with prostaglandin F2. Exp Eye Res 1988;47:761–69. 103. Melamed S, Teehasaened C, Epstein DL. Role of fibronectin in closure of YAG trabeculopuncture. Laser Light Ophthalmol 1989;2:233–41. 104. Melamed S, Epstein DL. Alterations of aqueous humour outflow following argon laser trabeculoplasty in monkeys. Br J Ophthalmol 1987;71:776–81. 105. Bylsma SS, Samples JR, Acott TS, Van Buskirk EM. Trabecular cell division after argon laser trabeculoplasty. Arch Ophthalmol 1988;106:544–47. 106. Acott TS, Samples JR, Bradley JMB et al. Trabecular repopulation by anterior trabecular meshwork cells after laser trabeculoplasty. Am J Ophthalmol 1989;107:1–6. 107. Bárány EH, Linnér E, Lütjen-Drecoll E,
Rohen JW. Structural and functional effects of trabeculectomy in cynomolgus monkeys. Albrecht v Graefes Arch Klin Exp Ophthal 1972;184:1–28. 108. Lütjen-Drecoll E. Electron microscopic studies on reactive changes of the trabecular meshwork in human eyes after microsurgery. Albrecht v Graefes Arch Klin Exp Ophthal 1972;183:267–85. 109. Sit AJ, Coloma FM, Ethier CR, Johnson M. Factors affecting the pores of the inner wall endothelium of Schlemm’s canal. Invest Ophthalmol Vis Sci 1997;38:1517–25. 110. Smit BA, Johnstone MA. Effects of viscocanalostomy on the histology of Schlemm’s canal in primate eyes. Invest Ophthalmol Vis Sci 2000;41:S578 (abstr). 111. Speigel D, Schefthaler, Kobuch K. Outflow facilities through Descemet’s membrane in rabbits. Invest Ophthalmol Vis Sci 2000;41:S578 (abstr). 112. Fatt I. Permeability of Descemet’s membrane to water. Exp Eye Res 1969;8:340–54. 113. Sannace C, Miserocchi E, Carassa RG et al. Viscocanalostomy: an ultrasound biomicroscopic study. Invest Ophthalmol Vis Sci 2000;41:S578 (abstr). 114. Jacobi PC, Krieglstein GK. Trabecular aspiration: clinical results of a new surgical approach to improve trabecular facility in glaucoma capsulare. Ophthalmic Surg 1994;25:641–45. 115. Jacobi PC, Dietlein TS, Krieglstein GK. Bimanual trabecular aspiration in pseudoexfoliation glaucoma: an alternative in nonfiltering glaucoma surgery. Ophthalmology 1998;105:886–94. 116. Gottanka J, Martus P, Johnson DH, LütjenDrecoll E. Correlation of pseudoexfoliative material and optic nerve damage in pseudoexfoliation syndrome. Invest Ophthalmol Vis Sci 1997;38:2435–46. 117. Luntz MH, Livingston DG. Trabeculotomy ab externo and trabeculectomy in congenital and adult-onset glaucoma. Am J Ophthalmol 1977;83:174–79.
References 118. Tanihara H, Negi A, Akimoto M et al. Surgical effects of trabeculotomy ab externo on adult eyes with primary open angle glaucoma and pseudoexfoliation syndrome. Arch Ophthalmol 1993;111:1653–61. 119. Melamed S, Pei J, Puliafito CA, Epstein DL. Q-switched neodymium-YAG laser trabeculopuncture in monkeys. Arch Ophthalmol 1985;103:129–33. 120. Quaranta L, Hitchings RA, Quaranta CA. Ab-interno goniotrabeculotomy versus mitomycin C trabeculectomy for adult openangle glaucoma: a 2–year randomized clinical trial. Ophthalmology 1999;106:1357–62.
55
121. Jacobi PC, Dietlein TS, Krieglstein GK. Goniocurettage for removing trabecular meshwork: clinical results of a new surgical technique in advanced chronic open-angle glaucoma. Am J Ophthalmol 1999;127:505–10. 122. Jacobi PC, Dietlein TS, Krieglstein GK. Microendoscopic trabecular surgery in glaucoma management. Ophthalmology 1999;106:538–44. 123. Jacobi PC, Dietlein TS, Krieglstein GK. Technique of goniocurettage: a potential treatment for advanced chronic open angle glaucoma. Br J Ophthalmol 1997;81:302–07.
5 Mechanisms of filtration in nonpenetrating filtering surgeries André Mermoud and Emilie Ravinet
To improve the reproducibility and safety of filtering procedures, several non-penetrating filtering surgeries have been described in the past few years.1–16 The principal common concept of non-penetration is to create filtration through a naturally occurring membrane that acts as an outflow resistance site, which allows a progressive drop in intraocular pressure (IOP) and avoids postoperative ocular hypotony. This membrane, the trabeculo-Descemet’s membrane, consists of the trabeculum and the peripheral Descemet’s membrane.1 To expose the membrane a deep sclerokeratectomy should be done, which also provides a postoperative scleral space. This space may act as an aqueous reservoir and as a filtration site, which may prevent the need for a large subconjunctival filtration bleb. In this way the risk of late bleb-related endophthalmitis can be reduced. Several studies have shown that in patients with various glaucomas such as primary openangle glaucoma, pseudoexfoliative, pigmentary, the main site of aqueous outflow resistance is located at the juxtacanalicular trabecular meshwork and the inner wall of Schlemm’s canal. By removing the internal wall of Schlemm’s canal and the juxtacanalicuar meshwork, the main outflow resistance in patients with openangle glaucoma can probably be relieved. This additional technique has been called ab
Figure 5.1 Schematic representation of the trabeculoDescemet’s membrane. (A) Posterior trabeculum, (B) anterior trabeculum and (C) Descemet’s membrane.
externo trabeculectomy and has been proposed by several investigators (Fig. 5.2).5–7 In primary and secondary closed-angle glaucoma and probably in congenital glaucoma, the outflow resistance is located before the trabecular meshwork. Thus nonperforating filtering surgery is probably not
58
Mechanisms of filtration in non-penetrating filtering surgeries
(a)
(b) Figure 5.2 Schematic representation of the anatomy before (a) and after (b) ab externo trabeculectomy. The arrows represent the site of aqueous humor passage. The pealing of the membrane corresponds to the ablation of the inner endothelium of Schlemm’s canal and the juxtacanalicular trabeculum (c) scanning electron microscopic view of the peeled membrane.
(c)
indicated for the treatment of these forms of glaucoma.
and safety of non-penetrating surgeries: aqueous humor flow through the trabeculoDescemet’s membrane and aqueous resorption after its passage through the trabeculoDescemet’s membrane.
Mechanisms of filtration after non-penetrating glaucoma surgery
Flow through trabeculoDescemet’s membrane
There are two sites of interest when one studies the mechanisms involved in efficiency
The trabeculo-Descemet’s membrane offers resistance to aqueous humor outflow. This
Mechanisms of filtration after non-penetrating glaucoma surgery
59
Figure 5.3 Continuous IOP recording during a deep sclerectomy. The slope of decrease is slow and corresponds to an average of 2.7 ± 0.06 mmHg/min.
resistance allows a slow decrease in IOP during surgery and accounts for the reliable and reproducible IOP on the first postoperative day. Thus the main advantage of the trabeculo-Descemet’s membrane is to reduce immediate postoperative complications, such as hypotony, flat anterior chamber, choroidal detachments, and induced cataract. In an experimental model, the gradual decrease in IOP was studied and the resistance of the trabeculo-Descemet’s membrane calculated. Experiments were done on enucleated human eyes unsuitable for keratoplasty. The mean rate of IOP decrease was 2.7 ± 0.6 mmHg/min (Fig. 5.3). The ocular aqueous outflow resistance dropped from a mean of 5.34 ± 0.19 mmHg µL per min pre-
operatively to a mean of 0.41 ± 0.16 mmHg µL per min postoperatively.1 The resistance of the trabeculo-Descemet’s membrane thus appears low enough to ensure a low IOP and yet high enough to maintain the anterior chamber depth and avoid the postoperative complications in relation to hypotony. In the same study, Vaudaux and Mermoud,1 on histological examination looked at the surgical site by ocular perfusion with ferritine. They were able to show that the main outflow through the trabeculo-Descemet’s membrane occurred at the level of the anterior trabeculum (Fig. 5.4). There was, however, some degree of outflow through the posterior trabeculum and Descemet’s membrane.
60
Mechanisms of filtration in non-penetrating filtering surgeries Aqueous humor resorption After aqueous humor passage through the trabeculo-Descemet’s membrane, four hypothetical mechanisms of aqueous resorption may occur: a subconjunctival filtering bleb; an intrascleral filtering bleb; a suprachoroidal filtration; an episcleral vein outflow via Schlemm’s canal (Fig. 5.5).
3 1
2
Subconjunctival bleb
Figure 5.4 Histology of a deep sclerectomy performed in an eye perfused with ferritine. Blue color represents the aqueous humor passage. Main aqueous humor flow takes place at the level of anterior trabeculum (1). There is also some passage through Descemet’s membrane (2) and the posterior trabeculum (3).
A
As observed after trabeculectomy almost all patients undergoing non-penetrating filtering surgeries have a diffuse conjunctival bleb on the first postoperative day. As shown by ultrasonic biomicroscopy (UBM) studies, successful cases have a low profile and diffuse subconjunctival filtering bleb even years after surgery (Fig. 5.6). However, this bleb tends to be smaller than the one seen after trabeculectomy (Fig. 5.7).
D
B
C
Figure 5.5 Schematic representation of deep sclerectomy with four hypothetical mechanisms of aqueous humor resorption after passage through the trabeculo-Descemet’s membrane. (A) the subconjunctival filtering bleb; (B) the intrascleral filtering bleb; (C) the subchoroidal passage; (D) the episcleral drainage via Schlemm’s canal ostia.
Mechanisms of filtration after non-penetrating glaucoma surgery
(a)
61
(b)
Figure 5.6 Photographic (a) and ultrasonic biomicroscopy (b) images of subconjunctival filtering bleb 5 years after deep sclerecomy with collagen implant.
(a)
(b)
Figure 5.7 Photographic (a) and ultrasonic biomicroscopy (b) images of subconjunctival filtering bleb 5 years after trabeculectomy.
Intrascleral bleb
During deep sclerectomy, a certain volume of sclera is removed ranging between 5 and
8 mm3. Provided the superficial scleral flap does not collapse, this scleral volume may be transformed into an intrascleral filtering bleb. Thus, in order to keep this intrascleral volume
62
Mechanisms of filtration in non-penetrating filtering surgeries
Figure 5.8 Ultrasonic biomicroscopy image of intrascleral filtering bleb 2 years after a deep sclerectomy with collagen implant. The echogenicity of the intrascleral bleb is very similar to that of a diffuse low grade subconjunctival filtering bleb.
upright, different devices such as the collagen implant have been tried. Hyaluronic acid or non-resorbable Hema implants have also been used. On UBM, an intrascleral bleb was observed in more than 90% of patients who received a collagen implant (Fig. 5.8) and the mean volume of the intrascleral bleb was 1.8 mm3 in an unpublished study done by D Kazakova and myself. In the intrascleral filtering bleb the aqueous resorption mechanism may be different to the one occurring in the subconjunctival space. The aqueous humor is probably resorbed by new aqueous drainage vessels, as shown by Delarive et al (unpublished data). In this study on rabbits Delarive et al showed that in the scleral space created after deep sclerectomy, irrespective of whether or not a collagen implant was used, new aqueous humor drainage vessels were growing and resorbing the aqueous flowing through the trabeculo-Descemet’s membrane (Fig. 5.9). Similar results have been obtained by Nguyen
(a)
(b)
Figure 5.9 (a) Histology of intrascleral filtration site 9 months after deep sclerectomy with collagen implant done in a rabbit. There are many newly formed aqueous humor draining vessels stained in blue by ferritine, which was injected into the anterior chamber before enucleation. (b) Same preparation in an unoperated eye. There are only a few drainage vessels.
Mechanisms of filtration after non-penetrating glaucoma surgery
63
(a) Figure 5.10 (a) Anterior segment fluoresceine and indocyanine green angiography representing normal rabbit limbal area with the Schlemm’s canal and a few collector canals.
(b) Figure 5.10 (b) Same area 6 months after deep sclerectomy with collagen implant. Numerous new aqueous-humor drainage vessels are present in surgical site.
and coworkers with the same model by anterior segment fluorescein and indocyanine green angiography (Fig. 5.10) (unpublished data). Subchoroidal space
By thinning the sclera by 90%, aqueous humor outflow into the suprachoroidal space may occur; in fact on UBM, it is possible to see fluid between the ciliary body and the remaining sclera (Fig. 5.11) in 45% of the patients studied years after the deep sclerectomy (D Kazakova et al, unpublished data). However, this observation could also indicate a chronic ciliary body detachment with subsequent reduction of the aqueous production. Further studies on aqueous dynamics following nonpenetrating filtering surgery are needed to better understand the exact mechanisms of aqueous drainage and their respective importance in terms of success and complications.
Figure 5.11 Ultrasonic biomicroscopy image of a subchoroidal aqueous humor passage after deep sclerectomy.
64
Mechanisms of filtration in non-penetrating filtering surgeries
Figure 5.12 Schlemm’s canal is dilated peroperatively during viscocanalostomy. The aqueous humor drainage may take this route to reach episcleal veins after operation.
Figure 5.13 The Hema implant inserted with both arms into the two ostia of Schlemm’s canal. This procedure may also promote resorption of aqueous humor through the episcleral veins via Schlemm’s canal.
Schlemm’s canal
of open-angle glaucoma. The immediate postoperative complication rate is low, and visual acuity is almost unaffected. The created trabeculo-Descemet’s membrane window allows a progressive drop in IOP while at the same time offering enough resistance to prevent the immediate postoperative complications. Downstream, there appears to be several hypothetical aqueous outflow mechanisms, namely, a subconjunctival filtering bleb, an intrascleral bleb with probable new aqueous drainage veins, a suprachoroidal passage with hypothetical increased uveoscleral outflow and/or decreased aqueous humor production by chronic ciliary body detachment, and a possible physiological route towards the episcleral veins via the two open ostia of Schlemm’s canal.
When the deep sclerectomy dissection is done, Schlemm’s canal is opened and unroofed. On either side of the deep sclerectomy the two surgically created ostia of Schlemm’s canal may drain the aqueous humor into the episcleral veins. This mechanism is probably more important after viscocanalostomy, during which the ostia and Schlemm’s canal are dilated with high viscosity hyaluronic acid (Fig. 5.12). It is probably also important with Hema implants because the two arms of the “T” are inserted into the two ostia of Schlemm’s canal, thereby preventing their collapse (Fig. 5.13). Research has still to be done to establish the importance of this mechanism.
Conclusion and summary Non-penetrating filtering surgeries as performed by several investigators offer a drop in IOP and a satisfactory long-term success rate for all types
References 1. Vaudaux J, Mermoud A. Aqueous dynamics after deep sclerectomy: ex-vivo study. Ophthalmic Pract 1998;16:204–09.
References 2. Sanchez E, Schnyder CC, Mermoud A. Résultats comparatifs de la sclérectomie profonde transformée en trabéculectomie et de la trabéculectomie classique. Klin Monatsbl Augenheilkd 1997;210:261–64. 3. Chiou AGY, Mermoud A, Jewelewicz DA. Post-operative inflammation following deep sclerectomy with collagen implant versus standard trabeculectomy. Graefe’s Arch Clin Exp Ophthalmol1998;236:593–96. 4. Sanchez E, Schnyder CC, Sickenberg M et al. Deep sclerectomy: results with and without collagen implant. Int Ophthalmol 1996–1997;20:157–62. 5. Zimmerman TJ, Kooner KS, Ford VJ. Effectiveness of non penetrating trabeculectomy in aphakic patients with glaucoma. Ophthalmic Surg 1984;15: 44–50. 6. Zimmerman TJ, Kooner KS, Ford VJ et al. Trabeculectomy vs nonpenetrating trabeculectomy: a retrospective study of two procedures in phakic patients with glaucoma. Ophthalmic Surg 1984;15:734–40. 7. Arenas E. Trabeculectomy ab-externo. Highlights Ophthalmol 1991;19:59–66. 8. Tanibara H, Negi A, Akimoto M. Surgical effects of trabeculotomy ab externo on adults eyes with primary open angle glaucoma and pseudoexfoliation syndrome. Arch Ophthalmol 1993;111:1653–61. 9. Stegmann RC. Viscocanalostomy: a new
10.
11.
12.
13.
14.
15.
16.
65
surgical technique for open angle glaucoma. An Inst Barraquer Spain. 1995;25:229–32. Kozlov VI, Bagrov SN, Anisimova SY. Deep sclerectomy with collagen. Eye Microsurgery 1990;3:44–46. Demailly P, Jeanteur-Lunel MN, Berkani M. Non penetrating deep sclerectomy associated with collagen device in primary open angle glaucoma: middle term retrospective study. J Fr Ophthalmol 1996;19:659–66. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with collagen implant and trabeculectomy in open angle glaucoma. J Cataract Refract Surg 1999;25:323–31. Karlen M, Sanchez E, Schnyder CC et al. Deep sclerectomy with collagen implant: medium term results. Br J Ophthalmol 1999;83:6–11. Shaarawy T, Karlen ME, Sanchez E et al. Long term results of deep sclerectomy with collagen implant. Acta Ophthalmol Scand 2000;78:323. Chiou AG, Mermoud A, Hediguer SE, Faggioni R. Ultrasound biomicroscopy of eyes undergoing deep sclerectomy with collagen implant. Br J Ophthalmol 1996;80:541–44. Chiou AGY, Mermoud A, Underdahl PJ, Schnyder CC. An ultrasound biomicroscopic study of eyes after deep sclerectomy with collagen implant. Ophthalmology 1998;105:104–08.
6 Experimental studies in non-penetrating glaucoma surgery Christophe Nguyen and Tarek Shaarawy
In primary open-angle glaucoma the main site of aqueous-outflow resistance is thought to be at the level of juxtacanalicular trabecular meshwork and the inner wall of Schlemm’s canal.1 In other types of secondary open-angle glaucoma—such as pseudoexfoliative glaucoma, pigmentary glaucoma, some types of
uveitic glaucoma, and traumatic glaucoma— the resistance of aqueous outflow is probably also in part located at the same site. By removing the internal wall of Schlemm’s canal and the juxtacanicular meshwork, the main outflow resistance in these patients can probably be relieved.
(b) (a) Figure 6.1a Normal anatomy of the angle. 1 = Descemet’s membrane, 2 = Anterior trabeculum, 3 = Posterior trabeculum, 4 = Schlemm’s canal.
Figure 6.1b Filtration membrane after ab externo trabeculectomy. 1= Descemet’s membrane, 2 = Anterior trabeculum, 3 = Posterior trabeculum. Inner wall of Schlemm’s canal and juxtacanalicular trabeculum have been removed.
68
Experimental studies in non-penetrating glaucoma surgery
Figure 6.2 Filtration membrane after deep sclerectomy (DS). 1 = Descemet’s membrane, 2 = Anterior trabeculum, 3 = Posterior trabeculum, 4 = Inner wall of Schlemm’s canal. Corneal stroma behind the anterior trabeculum and Descemet’s membrane is removed.
To decrease the outflow resistance without penetrating the anterior chamber, different surgical techniques have been described. In ab externo trabeculectomy (Fig. 6.1)2–7 the internal wall of Schlemm’s canal and the juxtacanicular meshwork are removed, whereas in deep sclerectomy described by Fyodorov8 and Kozlov et al9 (Fig. 6.2), and viscocanalostomy described by Stegmann et al,10 the corneal stroma behind the anterior trabeculum and Descemet’s membrane is removed. The common goal of these techniques is to improve the outflow facility but retain some residual outflow resistance by maintaining a membrane between the anterior chamber and the scleral dissection.
A great deal of interest has been directed towards the properties of this membrane. Outflow resistance of the membrane before and after surgery has been calculated. Histological studies have been done to visualize the site of filtration in pig, rabbit, and human eyes. With the description by Kozlov et al9 of the use of a collagen implant placed within the scleral bed to enhance the filtration of deep sclerectomy, another direction of research was opened. Some experimental studies have been done to better understand the mechanisms of filtration involved in deep sclerectomy with an implant and also to compare deep sclerectomy with and without an implant. Investigations have been done as well on the structural aspect of the juxtacanalicular trabeculum and the inner wall of Schlemm’s canal, and the variability of depth dissection after non-penetrating glaucoma surgery.
Model for experimental non-penetrating glaucoma surgery For in vivo studies on non-penetrating glaucoma surgery, the best model is the monkey. However, the rabbit model has been extensively used in glaucoma research and is certainly more practical. Deep sclerectomy is technically possible on the rabbit’s eye. The outflow canal in rabbits is a circular draining area without endothelium located behind the limbus and measures about 200 µm in diameter. In some areas the rabbit sclera may be thin and these sites should not be used for surgery. However, in appropriate locations, there is enough sclera for a deep sclerectomy. For in vitro research, enucleated pig eyes and enucleated human eyes unsuitable for penetrating keratoplasty are generally used.
Aqueous dynamics after non-penetrating glaucoma surgery
69
Figure 6.3 Schematic representation of aqueous outflow measurement. 1 = Balanced salt solution (BSS) reservoir; 2 = Manometer; 3 = Microsyringe pump; 4 = Electronic transducer; 5 = Pressure monitor; 6 = Chart recorder; 7 = 26-gauge needle.
Aqueous dynamics after non-penetrating glaucoma surgery Anterior trabeculum and Descemet’s membrane pathway after deep sclerectomy The method used in deep sclerectomy to improve aqueous outflow in a patient with
restricted posterior trabeculum clearance is to remove the corneal stroma behind the anterior trabeculum and Descemet’s membrane, creating a new pathway for aqueous drainage.8,9 The membrane made up of anterior trabeculum and Descemet’s membrane is called the trabeculo-Descemet’s membrane. The presence of this membrane offers a resistance to aqueous outflow from the anterior chamber to the subconjunctival space (Fig. 6.2) and leads to a progressive decrease in intraocular pressure (IOP) after deep sclerectomy.11,12
70
Experimental studies in non-penetrating glaucoma surgery
Vaudaux et al13 first studied in vitro the aqueous dynamics through this membrane. The aims of their study were to examine the resistance of the trabeculo-Descemet’s membrane, and to show, with histological examinations, the precise site of aqueous outflow after deep sclerectomy. Twelve enucleated pig eyes and nine enucleated human eyes unsuitable for penetrating keratoplasty (from an eye bank) were cannulated in the anterior chamber. Outflow facility was measured before and after deep sclerectomy, by the infusion with constant pressure method (Fig. 6.3).14–18 The trabeculoDescemet’s membrane resistance was then calculated as the reciprocal of the outflow facility. The outflow facility of trabeculoDescemet’s membrane outflow facility was found by substracting the preoperative outflow facility from the postoperative outflow facility. The mean preoperative outflow facility was 0.22 ± 0.03 µL/min per mmHg in pig eyes and 0.19 ± 0.01 µL/min per mmHg in human eyes. The mean postoperative outflow facility was 31.9 ± 12.0 µL/min per mmHg in pig eyes and 25.5 ± 12.6 µL/min per mmHg in human eyes.
Figure 6.4 Histology of the trabeculo-Descemet’s membrane dyed with ferritin. The dye is present in the anterior part of the trabecular meshwork (human eye).
The mean value of trabeculo-Descemet’s membrane outflow facility was 31.7 ± 12.0 µL/min per mmHg in pig eyes and 24.3 ± 12.6 µL/min per mmHg in human eyes. The investigators concluded that the trabeculo-Descemet’s membrane offers a sufficient intraoperative and postoperative aqueous outflow resistance to lower most of the surgically-induced complications of filter-
Table 1 Mean values of aqueous outflow facility and resistance: before (preoperative) and after deep sclerectomy (postoperative) and of the trabeculo-Descement’s membrane (TDM), in porcine and human eyes Parameter (µL/min/mm Hg)
Porcine (± SD; N = 9)
Human (± SD; N = 5)
Preoperative facility Postoperative facility TDM facility Preoperative resistance Postoperative resistance TDM resistance
0.22 ± 0.03 31.9 ± 12.0 31.7 ± 12.0 4.61 ± 0.76 0.22 ± 0.08 0.24 ± 0.09
0.19 ± 0.01 24.5 ± 12.6 24.3 ± 12.6 5.34 ± 0.19 0.41 ± 0.13 0.43 ± 0.14
Aqueous dynamics after non-penetrating glaucoma surgery
71
ing surgery. However, the resistance is small enough to induce postoperative low and controlled IOP (Table 1). They also showed that after deep sclerectomy, aqueous outflow occurs mainly at the anterior trabeculum level as seen on histological specimens dyed with ferritin (Fig. 6.4). The anterior trabeculum is much thinner compared to the posterior trabeculum. Furthermore, the anterior part of the trabecular meshwork is not used in physiological conditions and may not be damaged by glaucomatous disease to the same extent. By removing the corneal stroma behind the anterior trabeculum, deep sclerectomy allows aqueous humor to filter through trabeculum that has been less used. This remaining trabeculum is so thin (10 µm) that it could be regarded as diffused perforations. Further studies are needed to establish whether there is an aqueous flow through the Descemet’s membrane.
Posterior trabeculum pathway after ab externo trabeculectomy As the likely site of main aqueous outflow resistance,1 the juxtacanalicular trabeculum and the inner wall of Schlemm’s canal are removed in ab externo trabeculectomy.2–7 In this technique the filtration membrane is smaller in surface area and is formed by the posterior trabeculum alone, because the internal wall of Schlemm’s canal and the juxtacanicular meshwork are removed (Fig. 6.1). In comparison to deep sclerectomy, ab externo trabeculectomy is an easier and safer surgical technique. Although the risk of intraoperative perforation is reduced, no clinical or experimental study has reported any advantage of ab externo trabeculectomy over deep sclerectomy or viscocanalostomy.
Figure 6.5 Electron microscopy of the peeled membrane.
Rossier et al,19 using basically the same technique as Vaudaux et al13 studied aqueous dynamics after ab externo trabeculectomy (Fig. 6.1). The purpose of their research was to assess the decrease in IOP, the residual outflow resistance of the trabeculum, and the filtration site by histological examination in enucleated pig and human eyes after ab externo trabeculectomy. Rossier et al19 did experiments in two phases: first on 16 fresh enucleated pig eyes, and then on six enucleated human eyes that were regarded as unsuitable for corneal graft. The anterior chamber of each eye was cannulated and perfused at constant pressure. Measurement of the resistance to aqueous
72
Experimental studies in non-penetrating glaucoma surgery
(a)
(b)
Figure 6.6a Light microscopy of the corneoscleral angle after ab externo trabeculectomy. This section shows the postoperative human trabecular meshwork. 1= Cornea; 2 = residual trabecular meshwork; 3 = sclera; 4 = longitudinal ciliary muscle; 5 = iris.
Figure 6.6b Electron micrograph showing a part of a human trabeculum. Ferritin (little black dots) is visible in the trabeculular stroma (1) and lumen (2). 3 = Collagen fibrils; 4 = elastic fibers (electron-dense mass); 5 = endothelial cells. 25000.
outflow was done by the constant pressure method (Fig. 6.3), 14–18 which was done before and after external trabeculectomy. The surgery began by opening the Schlemm’s canal. At this stage, aqueous humor started to filter through
the remaining trabeculum and internal wall of Schlemm’s canal, and the first decrease in IOP was seen on the pressure monitor. The internal endothelium and juxtacanalicular trabeculum were then removed with a forceps (Fig. 6.5), at
Aqueous dynamics after non-penetrating glaucoma surgery
(a)
(b)
73
(c)
Figure 6.7 Comparison between sinusotomy, deep sclerectomy and ab externo trabeculectomy in pig eyes. (a) Sinusotomy. A lamellar band of the sclera is removed and Schlemm’s canal is opened over 120° from to 10 to 2 o’clock. The inner wall of Schlemm’s canal is not touched. (b) Ab externo trabeculectomy. (c) Deep sclerectomy. Arrows = aqueous dynamics.
which point a second decrease in IOP was seen on the recording chart. The remaining trabecular resistance was calculated and eyes were then prepared in order to study the site of aqueous filtration. The investigators injected ferritin into the anterior chamber and the eyes were fixed, sectioned, and then examined histologically. In pig eyes, they reported that the average rate of IOP decrease was 35.7 ± 27.2 mmHg/min after Schlemm’s canal had been opened and 47.3 ± 25.6 mmHg/min after ab externo trabeculectomy. In human eyes, the average rate in IOP decrease was 4.71 ± 2.80 mmHg/min after Schlemm’s canal had been opened and 14.8 ± 7.15 mmHg/min after ab externo trabeculectomy. Rossier et al19 reported a mean outflow facility of 0.31 ± 0.13 µL/min per mmHg in pig eyes and 0.24 ± 0.08 µL/min per mmHg in human eyes before surgery, and 79.0 ± 47. ) µL/min per mmHg in pig eyes and 6.33 ± 6.67 µL/min per mmHg in human eyes after ab externo trabeculectomy.
In pig eyes, the mean residual membrane outflow facility was 78.7 ± 50.0 µL/min per mmHg. 5 mm of residual trabecular membrane provided 99.6% of the total postoperative outflow. Ab externo trabeculectomy allowed the preoperative outflow facility to be multiplied by 255. When compared to pig eyes, human eyes had much lower residual membrane outflow facility (mean 6.10 ± 6.63 µL/min per mmHg). The residual membrane provided 96.4% of the postoperative filtration, and ab externo trabeculectomy multiplied the preoperative outflow facility value by only 26.4. The difference between the pig and the human outflow facility may be explained by the difference between their respective angle anatomies. When studied on radial histological slides, the trabeculum of pig eyes is wider and thinner, which may explain the higher outflow facility through that area. After the filtration site had been traced by cationized ferritin, histological examination allowed Rossier et al19 to show that aqueous
74
Experimental studies in non-penetrating glaucoma surgery
humor exited effectively through the residual posterior trabecular meshwork (Figs. 6.6a and 6.6b). They concluded that ab externo trabeculectomy significantly lowers IOP and improves outflow facility in enucleated pig and human eyes. By comparison of the two quite similar studies, deep sclerectomy seems to offer better reduction of outflow resistance than ab externo trabeculectomy in human eyes. However, in the pig eyes the results were the opposite probably because the anatomy of pig eyes is different and it is not possible to fully expose the anterior trabeculum and the Descemet’s membrane during deep sclerectomy. In the study by Vaudaux et al13 dissection of the pig eye was not really a deep sclerectomy but rather a sinusotomy (Fig. 6.7).20–23 In this operation, a lamellar band of the sclera is removed and Schlemm’s canal is opened over 120° from 10 to 2 o’clock. The inner wall of Schlemm’s canal is not touched and then the conjunctiva is closed.
Deep sclerectomy with a collagen implant Comparative study of deep sclerectomy with and without collagen implant In deep sclerectomy the simple removal of the scleral and corneal tissue overlying trabecular structures carries the risk of inducing secondary fibrosis and subsequent failure to control IOP. The sclerocorneal space created by non-penetrating glaucoma surgery acts as an aqueous decompression space that should be kept open. Kozlov et al9 first proposed the
Figure 6.8 Comparison of mean intraocular pressure (IOP) after deep sclerectomy with and without collagen implant during the 9 months of follow-up. After an initial significant decrease, the mean IOP returned to preoperative value after 2 months.
use of intrascleral collagen implant to maintain this newly created space. To better understand the usefulness and mechanisms of action of intrascleral implant, Delarive et al24 used an animal model and compared deep sclerectomy with and without a collagen implant. The issues of filtration, aqueous dynamics, evolution of the collagen implant with time, and scarring response of different adjacent tissues involved in deep sclerectomy with and without collagen implant were all addressed. Deep sclerectomy was done in both eyes on 18 pigmented rabbits, and a randomly selected eye received the collagen implant. This procedure was followed by measurements of IOP and aqueous outflow facility, together with light microscopy studies at different times after surgery. The follow-up of this study was 9 months.
Deep sclerectomy with a collagen implant
Figure 6.9 Comparison of preoperative and postoperative mean outflow facility (OF) (µL/min per mmHg) in deep sclerectomy with collagen implant (DSCI) and deep sclerectomy (DS). A significant increase in OF was observed during the entire 9 months of follow up in both DSCI and DS.
The investigators reported a significant decrease in IOP (p < 0.009) during the first 6 weeks after deep sclerectomy with collagen implant (mean IOP = 13.07 ± 2.95 mmHg preoperatively and 9.08 ± 2.25 mmHg at 6 weeks); deep sclerectomy without collagen implant showed a significant decrease in IOP at weeks 4 and 8 after surgery (mean IOP = 12.57 ± 3.52 mmHg at 8 weeks; p = 0.035; Fig. 6.8). After the initial decrease, the mean IOP returned progressively to its preoperative value in both groups. Outflow facility was significantly increased throughout the 9 months of follow-up in both deep sclerectomy with collagen implant and deep sclerectomy groups (p < 0.05; Fig. 6.9). The preoperative mean outflow facility was 0.15 ± 0.02 µL/min per mmHg. At 9 months, the mean outflow facility was 0.52 ± 0.28 µL/min per mmHg (p = <0.012) for deep sclerectomy with collagen
75
Figure 6.10 Number of aqueous drainage vessels in the scleral filtration space after deep sclerectomy with collagen implant (DSCI) and deep sclerectomy (DS), and in the non-operated site. The number of new drainage vessels in the operated sclera with or without collagen implant is significantly higher than the number of drainage vessels recorded in the non-operated sclera.
1 5 4
2 3
Figure 6.11 Histological specimen showing an irregular scleral canal at the operated site 1 month after deep sclerectomy. 1 = Cornea; 2 = Anterior chamber; 3 = Iris; 4 = Ciliary body; 5 = Scleral canal.
76
Experimental studies in non-penetrating glaucoma surgery
5
4
1
3
4
1
2 3
2
(a)
(b)
Figure 6.12 Histological specimens showing scleral canal at higher magnification. (a) 9 months after deep sclerectomy with collagen implant; (50). Spindle cells cover the entire wall of the canal. 1 = Scleral canal, 2 = Anterior chamber, 3 = Iris, 4 = Cornea. (b) 9 months after deep sclerectomy; (25). Irregular canal at the operated site without spindle cells. 1 = Iris; 2 = Ciliary body; 3 = Trabeculum, 4 = Irregular scleral canal, 5 = Cornea.
implant and 0.46 0.07 µL/min per mmHg (p <0.001) deep sclerectomy. Light microscopy studies showed the appearance of new aqueous drainage vessels in the sclera adjacent to the dissection site in both surgical groups. There were fewer of these vessels was less in deep sclerectomy than in deep sclerectomy with collagen implant, but this difference was not significant (Fig. 6.10). Because the collagen implant in deep sclerectomy dissolves in the conventional medium used for the preparation of sections (xylol), under light microscopy a canal inside the sclera was seen (Fig. 6.11). In both groups, a drainage canal was seen from 2 weeks up to the 9 months of follow-up. The main difference between sections from the two groups was the presence of spindle cells lining the wall of this canal in deep sclerectomy with collagen implant: these spindle cells were sparsely
present 2 months after deep sclerectomy with collagen implant and covered the entire wall of the canal 6 months postoperatively (Fig. 6.12a). In comparison, in deep sclerectomy sections an irregular canal without spindle cells was seen (Fig. 6.12b). There was no inflammatory reaction at any time after surgery, irrespective of the surgical procedure. Delarive et al24 concluded that deep sclerectomy with or without collagen implant provided a significant increase in outflow facility throughout the 9 months of follow-up, which was partly explained by new drainage vessels in the sclera surrounding the operated site. Microscopy studies revealed the appearance of spindle cells lining the collagen implant in deep sclerectomy with collagen implant after 2 months. The collagen implant was well tolerated by the rabbit’s tissues, since no inflammatory response was seen during the entire study.
Deep sclerectomy with a collagen implant
77
Figure 6.13 Anterior segment angiogaphy in normal rabbit’s eye.
Figure 6.14 Anterior segment angiography 1 month after deep sclerectomy with collagen implant with many new draining vessels.
Anterior-segment angiographies after deep sclerectomy with collagen implant
physiological aqueous drainage canal, and the deep sclerectomy site with the implant was clearly seen in the early phase after surgery. New aqueous draining vessels were starting to grow 3 weeks after the operation and continued to grow during the entire follow-up (Figs. 6.13, 6.14, 6.15).
To further investigate the role of new aqueous humor drainage vessels appearing close to the operative site after deep sclerectomy, Roy and colleagues25,26 developed a new technique using anterior-segment fluorescein and indocyanin green angiographies in the rabbit. This technique made it possible to visualize the normal rabbit aqueous outflow pathway. They used this technique to study aqueous pathway after deep sclerectomy with collagen implant. Fluorescein and indocyanin green was injected in the anterior chamber of rabbit’s eyes. Successive sessions of anterior segment angiographies, with filters for both fluorescein and indocyanin green, were done at various periods after surgery varying from 2 weeks to 9 months. The angiographies were done on eyes that underwent deep sclerectomy with collagen implant. The investigators were able to see the
Figure 6.15 Anterior segment angiography 3 months after deep sclerectomy with collagen implant with many new vessels.
78
Experimental studies in non-penetrating glaucoma surgery
Ultrasonic biomicroscopy after deep sclerectomy with collagen implant High-frequency ultrasound biomicroscopy (UBM) allows detailed anatomical assessment of the anterior eye segment. As with human eyes,11,12 UBM imaging has been done in rabbit eyes that underwent deep sclerectomy with collagen implant. With several UBM examinations it is possible to follow the evolution over time of the operated site. The persistence of the intrascleral bleb and the subconjunctival bleb over time are particularly observed as these blebs may reflect the success of the operation. Mermoud and colleagues (unpublished data) have shown that after deep sclerectomy with collagen implant in rabbit eyes, the intrascleral bleb and the collagen implant are very well
recognized 1 week postoperatively (Fig. 6.16). The collagen implant is completely resorbed after 3 months postoperatively. At that period the blebs are seen with difficulty but a retrotrabeculo-Descemet’s space is maintained and the iridocorneal angle is well opened.
Hyaluronic implant in non-penetrating glaucoma surgery Sourdille et al27 studied non-penetrating glaucoma surgery with another type of implant. They assessed experimentally the tolerance and efficacy of a reticulated hyaluronic acid implant (SK gel) after nonpenetrating filtering surgery. This material was
Figure 6.16 Ultrasound biomicroscopy showing intrascleral bleb 1 week after deep sclerectomy with collagen implant in rabbit’s eye. 1 = Cornea; 2 = Iris; 3 = Scleral flap; 4 = Ciliary body; 5 = Sclera; 6 = Collagen implant.
Dissection depth in deep sclerectomy known to be biocompatible and well tolerated and reticulation should greatly slow its degradation. 25 New Zealand white rabbits were included in their study. A control group of ten had surgery only (group A), while others also received the implant (group B). The surgical technique was identical in both groups: a deep sclerectomy was done in both groups and, in group B, the SK gel implant was placed. Slit-lamp observations, videorecordings, photography, IOP readings, and pachymetry were done weekly. 13 eyes were enucleated and fixed in 10% buffered formaldehyde on postoperative days 7, 14, 21, and 28 in group A and postoperative days 3, 7, 14, 28, 56, 84, and 180 in group B. The specimens were routinely stained with hematoxylin-eosin, periodic acid-Schiff, and Masson’s trichome. In all animals, central corneal thickness remained unchanged. Until postoperative day 14, slight conjunctival inflammation and peripheral corneal edema were noted at the site of surgery on both groups. They also found that the rabbit eyes with the implant followed a different healing process to the eyes without the implant. In group A, fibroblastic proliferation proceeded to completely fill the subscleral and preDescemetic cavity with fibrin clots and fibroblasts by postoperative day 7. In group B, fewer fibroblasts were seen in the cavity, which filled more slowly. The cavity location was detectable histologically at all periods. The implant was slowly bioabsorbed, and remnants were seen at the operative site (where the tissue was removed) up to postoperative day 56. This site was detectable at all histological study periods. The reduction in IOP lasted longer in these eyes than those in group A (5 months vs 3 weeks; p < 0.05). All the extraocular and intraocular structures and tissues remained normal with no signs of foreign-body reaction detected at any histology period.
79
The investigators finally reported that the results of the comparative experimental surgery showed excellent tolerance and efficacy in the rabbit eyes with a hyaluronic acid implant.
Dissection depth in deep sclerectomy One of the critical aspects of the surgical technique of deep sclerectomy seems to be the correct dissection of the inner scleral flap at the proper depth and the deroofing of Schlemm’s canal.28 More than in conventional trabeculectomy,29 the success of this mode of surgery is linked to the intraoperative anatomical outcome.10,28,30,31 Most surgeons who are experienced with this special kind of glaucoma surgery emphasize that the procedure is demanding, but proves reliable after completing a learning curve of 10–20 interventions.31–33 However, little was known about the morphological variability of the deep scleral flap, beyond the surgeon’s intraoperative impression, or the frequency with which the outer wall of Schlemm’s canal really was incorporated in the dissected scleral flap. To study the variability of dissection depth in deep sclerectomy, Dietlein et al34 analysed the morphology of the deep scleral flap and compared the morphological results with the intraoperative view of the tissue during dissection under the operating microscope. They performed deep sclerectomy in seven children and 22 adults with glaucoma and prepared the excised deep scleral flap for light microscopy. Serial sections of the complete specimens were meticulously studied at different magnifications to find signs of the outer wall of Schlemm’s canal, which is characterized by endothelial cells with a high frequency of pigmentation and
80
Experimental studies in non-penetrating glaucoma surgery
adhesive fragments of trabecular beams. To find the region of Schlemm’s canal they looked specifically for radial drainage vessels. Morphology was compared to the intraoperative appearance of the operated tissue. The results from Dietlein et al34 showed that morphological signs of removal of parts of the outer wall of Schlemm’s canal along with the deep scleral flap were found in serial sections from 15 (52%) of 29 patients. In five (17%) of these 15 patients noticeable remnants of the juxtacanalicular trabecular meshwork were also found, although in only one patient was this obvious during surgery. In 14 (48%) of 29 patients no evidence of the deroofing of Schlemm’s canal was found within the excised deep scleral flap, although intraoperatively the dissection seemed to have been too superficial in only five patients. No significant correlation was found by the investigators between dissection depth and patient’s age, state of ocular refraction or history of previous surgery. Dietlein et al34 concluded that deep sclerectomy, even when performed by an experienced glaucoma surgeon, produces biopsy material of great morphological variability that does not always correspond to the intraoperative appearance of the site of operation. More than in conventional trabeculectomy this variability may be of importance for the outcome of surgery.
Structural characteristics of the external portion of the trabecular meshwork in the glaucomatous patient The principal aim of ab externo trabeculectomy is a selective ablation of the external
portion of the trabecular meshwork, which is involved in aqueous outflow resistance (i.e. the inner wall of Schlemm’s canal and external trabecular layers, especially the cribriform trabecular meshwork).1,35 In fact, the peeling of the inner wall of Schlemm’s canal is also often made in deep sclerectomy, to enhance postoperative aqueous filtration. To better understand how filtration is increased when the inner wall of Schlemm’s canal is peeled, Hammard et al36 have tried to assess, with a confocal microscope, the structural characteristics of this part of the trabecular meshwork in patients with glaucoma. They obtained 36 external trabecular membranes from 33 consecutive patients with glaucoma (mean age = 56.5 ± 14.5 years, 26 white and seven black patients) and from four normal donors via necropsy (mean age 60.5 ± 7.7 years), who underwent a deep sclerectomy associated with an ab externo trabeculectomy according to the same surgical procedure. In glaucomatous eyes, 28 eyes had primary open-angle glaucoma, two had pigmentary glaucoma, five had exfoliative glaucoma, and one had traumatic glaucoma. Under conjunctival and scleral flaps, the roof of the Schlemm’s canal was opened and removed. A deeper dissection led to the removal of the inner wall of the Schlemm’s canal and the adjacent external trabecular membrane, which allowed a satisfactory aqueous flow through the remaining internal trabecular meshwork layers. In black patients (seven patients, ten eyes), a sponge soaked in mitomycin C (0.2 mg/mL for 3 min) was placed between the sclera and the Tenon’s capsule. In 13 cases, a sponge impregnated with 5–fluorouracil (50 mg/mL) was placed in the first scleral bed, before opening the roof of Schlemm’s canal. After fixation with acetone or triton 100 and immunostaining of extracellular matrix (fibronectin) or cytoskeleton
Conclusion (vimentin), the samples were analysed with a confocal microscope. Hammard et al36 found that the mean thickness of the external trabecular membranes was 34.4 ± 7.3 µm in glaucomatous eyes, and was not significantly different in the control eyes (39.0 ± 10.7 µm). The main characteristic of the glaucomatous external trabecular membrane was a paucicellularity compared to the controls (respectively, 21.6 ± 12.1 cells/area and 156.1 ± 28.8 cells/area). Their analysis with the confocal microscope showed that the cellular layer of Schlemm’s canal endothelium was well visualized in nearly all cases. The external trabecular membranes involved two different portions of the trabecular meshwork, which were on top of each other. The architectural characteristics of the outermost portion of the external trabecular membrane, with their star-shaped cells arranged in a homogenous extracellular matrix, suggested that it was the cribriform trabecular meshwork. The inner portion of the external trabecular membranes showed cells arranged regularly in a fibrillar extracellular matrix as it was described in the corneoscleral trabecular meshwork. The thickness of the removed external trabecular meshwork did not appear to depend on the patient’s race or type of glaucoma, because no great difference was reported. In addition, the number of trabecular cells was not statistically different when an antimitotic drug (5-fluorouracil or mitomycin C) was used. The results from Hammard et al36 confirmed the previous reported histopathological changes of the glaucomatous trabecular meshwork. Moreover, the thickness of the removed external trabecular meshwork and its structural characteristics showed that the ablation of the trabecular meshwork layers necessary to obtain a satisfactory aqueous
81
filtration not only involved the cribriform layers but also one layer of the inner part of the trabecular meshwork (i.e. the corneoscleral trabecular meshwork). In fact, the mean thickness of the removed trabecular meshwork was 34 µm in glaucomatous and control eyes, when the cribiform trabeculum usually varies from 7 µm to 15 µm.37,38 This observation suggested that the aqueous humor resistance not only involves the cribriform trabecular meshwork but also a part of the corneoscleral trabecular meshwork. Therefore it is probably necessary to also remove some of the inner part of the trabecular meshwork in case of confirmed advanced glaucoma, to guarantee a proper postoperative aqueous flow.
Conclusion In the past few years experimental surgery has been playing an important part in nonpenetrating glaucoma surgery. New techniques were practiced and assessed experimentally before clinical trials began. With the publication of medium-term and long-term clinical results of non-penetrating glaucoma surgery,39–41 experimental surgical studies have focused on identifying the functional mechanisms and the histopathological impact of this surgery. Outflow facility has been experimentally calculated in both ab externo trabeculectomy and deep sclerectomy, and the outflow facility increase is four times higher after deep sclerectomy.19 The site of aqueous filtration seems to be the posterior trabeculum in ab externo trabeculectomy, while the anterior trabeculum, and probably the Descemet’s membrane as well, are the main sites of aqueous filtration in deep sclerectomy. This can be explained by the anatomic nature of each surgery.
82
Experimental studies in non-penetrating glaucoma surgery
Experimental studies have also shown the morphological variability of excised tissue in deep sclerectomy, which underlines the difference between true morphology and intraoperative appearance of the operated site. This variability has a great bearing on the outcome of the surgery, and surgeons should pay attention to the efficiency of their dissection. The excised tissue in non-penetrating glaucoma surgery, namely corneoscleral tissue, Schlemm’s canal, and the juxtacanalicular trabecular meshwork, have been the subject of histopathological studies concerned with the pathology of glaucoma. The fibrillar and paucicellular aspect of the most inner part of the excised external trabecular meshwork seems to correspond to corneoscleral trabecular meshwork, which supports the hypothesis that in some glaucoma cases this inner part of the trabecular meshwork is also partly responsible for the raised resistance to aqueous flow, and thus to the increase in IOP. The use of implants in non-penetrating glaucoma surgery is a fairly new method in glaucoma therapy. Experimental studies have shown the beneficial effect of implants on the success rates of non-penetrating glaucoma surgery. Collagen implant seems to be associated with the formation of more new aqueous drainage vessels at the surgical site, which may explain why implants offers a better IOP control for longer periods. Unfortunately, no study has been done, as yet, to compare the efficacy of available implants, or to test the effect of size, shape, and material of the implant on the success rate. Further studies should determine ideal specifications of the implant to be used as well as assess the necessity, or otherwise, of the use of antimetabolites with implants in nonpenetrating glaucoma surgery. Finally, it is only fair to say that we are far from achieving perfection with glaucoma
surgery, but if this goal is to be reached, it would only be through keen interest and meticulous experimentation.
Appendix: Description of some experimental techniques Aqueous outflow pathway imaging in the rabbit Roy et al25,26 developed a new technique to visualize the rabbit aqueous outflow pathway using a digital imaging system with indocyanin green and fluorescein injection in the anterior chamber. After a paracentesis, 0.1 mL (1 mg/mL) of indocyanin green and 0.1 mL (0.1 mg/mL) of fluorescein are injected simultaneously into the anterior chamber. Several sequenced pictures are taken with a digital camera to visualize the distribution of the dyes within the outflow pathway so the dynamics of the outflow over time can be observed. The eyes should be treated with a daily application of antibiotic corticosteroid combination for 7 days. In the early phases, the shape of the outflow canal around the limbus can be clearly seen. Several collecting veins close to the recti muscles can also be identified. There is only slight fluorescein leakage during the early phases, allowing adequate visualization of the morphology of the outflow system. In the late phases, the sclera is stained with the fluorescein, and thus, no details are visible. The indocyanin green dye allows better recognition of the fine details of the outflow structure. This method is fairly simple, safe, and precise, and allows visualization of the details of the outflow pathways in the rabbit eyes.
References These results could be of great value in further assessment of the outcome of filtering surgeries in animal models.
Outflow facility measurements in the rabbit Under microscopic control, rabbit eyes are cannulated with a needle-guided silicon catheter, which is introduced through the corneal limbus into the anterior chamber. The puncture site is watertight. The catheter is positioned between the anterior plane of the iris and the posterior plane of the cornea without touching either of them. The catheter is connected via polyethylene tubing to a microsyringe pump that allow various flow rates ranging from 0.2 µL/h to 426 mL/h. The system is connected via a stopcock to an electronic pressure transducer that connects to an amplifier. The pressure measurements are printed continuously on a chart recorder. The system is filled with balanced salt solution. After cannulation, the pressure in the system is increased in successive 10 mmHg steps from 10 mmHg up to 40 mmHg. At each pressure level, a constant IOP is maintained with the appropriate infusion rate. We thus obtain for each eye four pressure levels. The infusion flow measurements are then plotted against pressure in the system. A regression line is obtained. The slope of this line represents the outflow facility. The outflow facility is then calculated by the Goldman equation: C = ∆I/∆IOP C ∆I
= outflow facility = (I2–I1), where I1 and I2 are successive inflow rates (µL/min) ∆IOP = (P2–P1), where P1 and P2 represent IOP at I1 and I2 respectively (mmHg)
83
References 1. Maepea O, Bill A. Pressures in the juxtacanalicular tissue and Schlemm’s canal in monkeys. Exp Eye Res 1992;54:879–83. 2. de Laage de Meux MP, Kantelip B. Surgical anatomy of corneoscleral limbs. Arch Ophtalmol (Paris) 1976;36:39–50. 3. Zimmerman TJ, Kooner KS, Ford VJ et al. Effectiveness of non-penetrating trabeculectomy in aphakic patients with glaucoma. Ophthalmic Surg 1984;15:44–50. 4. Zimmerman TJ, Kooner KS, Ford VJ et al. Trabeculectomy vs non-penetrating trabeculectomy: a retrospective study of two procedures in phakic patients with glaucoma. Ophthalmic Surg 1984;12:227–9. 5. Arenas E. Trabeculectomy ab externo. Highlights Ophthalmol 1991;19:59–66. 6. Tanibara H, Negi A, Akimoto M et al. Surgical effects of trabeculectomy ab externo on adults eyes with primary open angle glaucoma and pseudoexfoliation syndrome. Arch Ophthalmol 1993;111:1653–61. 7. Tavano G, Chabin T, Barrut JM. Hémitrabeculectomie ab externo non invasive. Bull Soc Ophthalmol Fr 1993;8–9:749–50. 8. Fyodorov SN. Non-penetrating deep sclerectomy in open angle glaucoma [in Russian]. Eye Microsurg 1989;1:52–55. 9. Kozlov VI, Bagrov SN, Anisimova SY et al. Deep sclerectomy with collagen [in Russian]. Eye Microsurg 1990;3:44–46. 10. Stegmann R, Pienaar A, Miller D. Viscocanalostomy for open-angle glaucoma in black African patients. J Cataract Refract Surg 1999;25:316–22. 11. Chiou AG, Mermoud A, Hediguer SE et al. Ultrasound biomicroscopy of eyes undergoing deep sclerectomy with collagen implant. Br J Ophthalmol 1996;80:541–44. 12. Chiou AG, Mermoud A, Underdahl JP, Schnyder CC. An ultrasound biomicroscopic study of eyes after deep sclerectomy with collagen implant. Ophthalmology 1998;105:746–50.
84
Experimental studies in non-penetrating glaucoma surgery
13. Vaudaux J, Uffer S, Mermoud A. Aqueous dynamics after deep sclerectomy: in vitro study. Ophthalmic Practice 1999; 16: 204–9 14. Barany EH, Scotchbrook S. Influence of testicular hyaluronidase on the resistance to flow through the angle of anterior chamber. Acta Physiol Scand 1954;30:240–48. 15. Becker B, Constant MA. The facility of aqueous outflow: a comparison of tonography and perfusion measurements in vivo and in vitro. Arch Ophthalmol 1956;55:305–12. 16. Hetland-Eriksen J, Odberg T. Experimental tonography on enucleated human eyes. I. The validity of Grant’s tonography formula. Invest Ophthalmol 1975;14:199. 17. Mermoud A, Baerveldt G, Minckler DS et al. Aqueous humor dynamics in rats. Graefes Arch Clin Exp Ophthalmol 1996;234(suppl 1):S198–203. 18. Dijkstra BG, Ruijter JM, Hoyng PF. Outflow characteristics of isolated anterior segments of human eyes. Invest Ophthalmol Vis Sci 1996;37:2015–21. 19. Rossier A, Uffer S, Mermoud A. Aqueous dynamics in experimental ab-externo trabeculectomy. Ophthal Res 2000;32:165–71. 20. Krasnov MM. Vest Offal 1964; 77: 37–41 21. Krasnov MM. Externalization of Schlemm’s canal (sinusotomy) in glaucoma. Br J Ophthalmol 1968;52:157–61. 22. Krasnov MM. Symposium on microsurgery of the outflow channels—sinusotomy: foundations, results, prospects. Trans Am Acad Ophthalmol Otolaryngol 1972;76:368–74. 23. Postic S, Stankov-Tomic M. Sinusotomie d’après Krasnov dans le glaucoma chronique simple. Bull Mem Soc Fr Ophthalmol 1967;80:716–26. 24. Delarive T, Mermoud A, Uffer S, Rossier A. Histological findings of deep sclerectomy with collagen implant in an animal model. Proceedings of the First International Congress on Non-penetrating Glaucoma Surgery. Lausanne, February 2001.
25. Roy S, Boldea R, Perez D et al. Visualisation du système d’écoulement de l’humeur aqueuse chez le lapin avec la fluorescéine et le vert d’indocyanine. Klin Monatsbl Augenheilkd 2000;216:305–08. 26. Roy S, Perez D, Curchod M et al. Filtration imaging after deep sclerectomy in the rabbit. Ophthal Res 1999; 31: S1–257 27. Sourdille P, Santiago PY, Ducourneau Y. Non-perforating surgery of the trabeculum with reticulated hyaluronic acid implant. J Fr Ophthalmol 1999;22:794–97. 28. Khaw PT, Siriwardena D. “New” surgical treatments for glaucoma. Br J Ophthalmol 1999;83:1–2. 29. Taylor HR. A histologic survey of trabeculectomy. Am J Ophthalmol 1976;82:733–5. 30. Sourdille P. Nonpenetrating trabecular surgery: it’s worth the change. J Cataract Refract Surg 1999;25:298–300. 31. Welsh NH, DeLange J, Wasserman P, Ziemba SL. The “deroofing” of Schlemm’s canal in patients with open-angle glaucoma through placement of a collagen drainage device. Ophthalmic Surg Lasers 1998;29:216–26. 32. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with collagen implant and trabeculectomy in openangle glaucoma. J Cataract Refract Surg 1999;25:323–31. 33. Sanchez E, Schnyder CC, Mermoud A. Comparative results of deep sclerectomy transformed to trabeculectomy and classical trabeculectomy. Klin Monatsbl Augenheilkd 1997;210:261–4. 34. Dietlein TS, Luke C, Jacobi PC et al. Variability of dissection depth in deep sclerectomy: morphological analysis of the deep scleral flap. Graefes Arch Clin Exp Ophthalmol 2000;238:405–09. 35. Seiler T, Wollensak J. The resistance of the trabecular meshwork to aqueous humor outflow. Graefes Arch Clin Exp Ophthalmol 1985;223:88–91. 36. Hammard P, Sourdille PH, Valtot F,
References Baudouin C. Deep non-penetrating sclerectomy with external trabeculectomy. An evaluation with the confocal microscope. Proceedings of the First International Congress On Non-Penetrating Glaucoma Surgery. Lausanne, February 2001. 37. Schlotzer-Schrehardt U, Naumann GO. Trabecular meshwork in pseudoexfoliation syndrome with and without open-angle glaucoma: a morphometric, ultrastructural study. Invest Ophthalmol Vis Sci 1995;36:1750–64. 38. Buller C, Johnson D. Segmental variability of the trabecular meshwork in normal and glaucomatous eyes. Invest Ophthalmol Vis Sci 1994;35:3841–51.
85
39. Demailly P, Lavat P, Kretz G, Jeanteur-Lunel MN. Non-penetrating deep sclerectomy (NPDS) with or without collagen device (CD) in primary open-angle glaucoma: middle-term retrospective study. Int Ophthalmol 1996–97;20:131–40. 40. Karlen ME, Sanchez E, Schnyder CC et al. A deep sclerectomy with collagen implant: medium term results. Br J Ophthalmol 1999;83:6–11. 41. Shaarawy T, Karlen ME, Sanchez E et al. Long term results of deep sclerectomy with collagen implant. Acta Ophthalmol Scand 2000;78:323.
7 Indications and contraindications for non-penetrating glaucoma surgery Elie Dahan
Indications In general, the indications for non-penetrating glaucoma surgery (NPGS) are wider and more inclusive than those for classical trabeculectomies, for two reasons: NPGS is safer but not less efficient than trabeculectomies,1 and NPGS is indicated in certain types of glaucoma where trabeculectomies normally fail or are not possible. Until the advent of NPGS, penetrating glaucoma surgery was generally regarded as the last resort in the treatment of glaucoma. When medical therapy and laser failed to lower intraocular pressure (IOP) to an acceptable level, glaucomatologists explained to their patients that an operation was necessary to halt the progression of the disease. NPGS with its lower complication rate can be offered earlier in the course of the disease. In fact, NPGS can be offered as a first-line treatment in cases where it is obvious that medical treatment will not lower the IOP to acceptable levels. This factor is particularly important in glaucoma patients under 50 years of age who have a longer life span. They obviously cannot be medically treated continuously for several decades. Furthermore, glaucoma surgery in general and NPGS in particular is more successful in glaucoma patients who were not exposed to medical treatment.2,3 The noxious effects of
topical medications on the conjunctiva are well documented.2,4 The conjunctival tissues undergo scarring processes when exposed to certain topical medications. Scarred conjunctiva, as found in patients who have been medically treated for years, is less amenable to the formation of a healthy diffuse bleb than a “virgin” conjunctiva. It is possible that even the trabeculum undergoes biochemical-structural changes after years of medical treatment, rendering it less responsive to NPGS. It is therefore logical to propose NPGS earlier rather than later when the chances of favourable outcomes are greater. The previous teaching of “first medical and laser treatment and then surgical treatment” has to be reviewed in the advent of the promising outcomes of NPGS.
Open-angle glaucoma Open-angle glaucoma is the commonest type of glaucoma (Fig. 7.1) and NPGS is the best treatment for it. Furthermore, NPGS targets the presumed site of pathology, namely the trabecular meshwork. During NPGS, the trabeculum is exposed and examined by the surgeon who can assess on site the amount of filtration in vivo. The experienced surgeon can compare the appearance of the trabeculum and its filtration capacity with the eyes he has
88
Indications and contraindications for non-penetrating surgery
Figure 7.1 Optic-nerve excavation in open-angle glaucoma.
Figure 7.2 Fundus image of high-myopic glaucoma patient.
operated on before. The site of aqueousoutflow resistance is presumed to be the juxtacanalicular trabeculum and the inner wall of Schlemm’s canal. During NPGS, the surgeon attempts to improve filtration by “reconditioning” the trabecular meshwork. Scraping, thinning out, and peeling the posterior trabeculum improve filtration. NPGS has the advantage of being less cataractogenic than trabeculectomy and, ideally, should replace it in all phakic open-angle glaucoma patients.5,6
Glaucoma patients with high myopia Conventional glaucoma surgery in patients with high myopia carries an especially high risk of complications because of the abnormal globe dimensions (Fig. 7.2). Choroidal detachments and consequent shallow anterior chambers (Fig. 7.3), occurs in 10–15% of
Figure 7.3 Choroidal detachment complicating trabeculectomy procedure for high-myopic glaucoma patient.
trabeculectomies performed in high-myopic glaucoma patients. 7–9 Hamel et al 10 studied NPGS in high-myopic glaucoma patients. 38% had an IOP below 21 mmHg without medication (complete
Pigmentary glaucoma
89
Figure 7.4 Pigmentary glaucoma: slit-like iris defects seen on transillumination.
Figure 7.5 Pigmentary glaucoma: open and heavily pigmented angle.
success rate) 48 months after surgery. 81% of patients achieved an IOP below 21 mmHg with or without medication (qualified success rate) 48 months after surgery. In their series, only two patients developed choroidal detachments, one of which was secondary to blunt trauma to the operated eye, 6 days after surgery. This incidence is lower than the 10–15% rate reported in classical trabeculectomy.7–9 The factors predisposing myopic eyes to choroidal effusion may be related to the larger intraocular volume of myopic eyes, to the thinner sclera, and to vulnerable choroidal blood vessels.7 NPGS appears to offer glaucoma patients with high myopia a safer outcome because of the gradual intraoperative IOP reduction.
resistant to medical treatment. Pigmentary glaucoma also occurs more frequently in young myopic male adults, and it is better to offer a safe surgical solution without depending on complex combination medical treatment. NPGS targets the site of pathology, namely the pigment-loaded trabecular meshwork (Fig. 7.5), which can be reconditioned to re-establish filtration. Pigmentary glaucoma used to be treated relentlessly with all manner of topical medications and laser treatment. The patient was offered surgery when his conjunctiva was severely scarred as a result of the copious topical treatment. Often, the trabeculum became brittle due to the repeated attempts to treat by laser. The outcome of glaucoma surgery in these circumstances is not favourable. NPGS has a better outcome in “virgin” eyes.2,3 The young myopic patient with pigmentary glaucoma will benefit especially from NPGS because it is safer than trabeculectomy. The dreaded choroidal detachments that occur more frequently in myopes11 after penetrating glaucoma surgery
Pigmentary glaucoma NPGS is the treatment of choice for pigmentary glaucoma (Fig. 7.4) because it is very
90
Indications and contraindications for non-penetrating surgery
are less likely to occur in NPGS because the IOP is lowered very gradually during surgery.
Pseudoexfoliation glaucoma Pseudoexfoliation glaucoma is a form of openangle glaucoma where there is accumulation of exfoliation material along all the aqueous outflow pathways. Since the exfoliation material is found especially in the trabeculum and Schlemm’s canal (Fig. 7.6), NPGS is the treatment of choice for this condition. To the NPGS-trained surgeon, opening Schlemm’s canal in the pseudoexfoliation patient is spectacular. Exfoliation material is found in abundance in the canal lumen. This material can be peeled away from the exposed trabeculum to re-establish filtration. IOP drops to acceptable levels for several years and when exfoliation material accumulates again, the site of filtration can be revised to restore filtration. NPGS can be done alone or in conjunction with cataract extraction according to
Figure 7.6 Pseudoexfoliation glaucoma.
patient age, cataract status, and refractive error. In their study on NPGS in pseudoexfoliation glaucoma, Shaarawy and Mermoud12 reported on the long-term results of NPGS with collagen implant. Complete success rate (IOP lower than 21 mmHg without medication), was 63.6% at 48 months postoperatively. At 60 months, 95.4% of the patients with pseudoexfoliation glaucoma still had an IOP below 21 mmHg with and without medication (qualified success rate). They concluded that NPGS provides good long-term control of IOP, with few immediate postoperative complications in patients with pseudoexfoliation glaucoma.
Aphakic and pseudophakic glaucoma Formerly, glaucomatologists relied heavily on medication to lower IOP to acceptable levels in aphakic glaucoma. Progressive loss of fields and eventual loss of vision were often the rule. Trabeculectomies were not regarded as a valid proposition because they necessitate peripheral iridectomies. In aphakic glaucoma, iridectomy is not desirable because the vitreous moves forwards through the iridectomy and blocks the filtration site. Extensive basal vitrectomy is needed to prevent blockage, but it is difficult to accomplish. The ever-present residual vitreous often finds its way to the filtration site and blocks it. Traction retinal detachment is not an uncommon complication in these combined vitrectomy-trabeculectomies. NPGS does not require iridectomy; therefore it is particularly indicated in aphakic glaucoma. The only drawback of NPGS in this particular glaucoma is the status of the trabeculum. When aphakia
Sturge–Weber syndrome
91
Figure 7.7 Pseudophakic glaucoma patient.
Figure 7.8 Axenfeld’s anomaly.
has been long-standing, the trabeculum is often collapsed and scarred; restoration of its function depends on its status and on the surgeon’s experience and skill. These cases should not be operated on by a novice but by an experienced NPGS surgeon. Shaarawy et al13 reported on NPGS in aphakic and pseudophakic glaucoma patients (Fig. 7.7). At 48 months, 55% of the patients had an IOP below 21 mmHg without medications, and 95% had an IOP below 21 mmHg with and without medication. This study showed the efficacy and safety of NPGS in aphakic and pseudophakic glaucoma patients.
practically the only treatment available for these patients and whenever possible NPGS should be tried first because of its low complication rate. The degree of success of NPGS is a function of the anatomical distortion of the angle structures and the surgeon’s experience. NPGS will succeed more in cases where the anatomy is less distorted. When NPGS fails, it is always possible to revert to penetrating glaucoma surgery, particularly in severely anatomically distorted cases. Tixier et al 14 reported on NPGS in congenital glaucoma; nine of 12 operated eyes had an IOP below 16 mmHg at 10 months without medication. They concluded that NPGS is at least as effective as trabeculectomy in congenital glaucoma but it carries fewer complications because the site of filtration is not perforated.
Congenital and juvenile glaucoma Congenital and juvenile glaucoma patients cannot rely on medication because they still have many years to live. Generally, their glaucoma is severe and results in rapid opticnerve damage and loss of vision. Surgery is
Sturge–Weber syndrome Sturge–Weber syndrome, a cutaneous haemangiomatous disorder, is often associated
92
Indications and contraindications for non-penetrating surgery patients with glaucoma. Since choroidal effusions following fistulizing surgery are known in these patients, NPGS offers a safer alternative.
Figure 7.9 Gonioscopy image of the same case as Fig. 7.8, strands of peripheral iris attached to posterior embryotoxon.
Aniridia and anteriorsegment dysgenesis syndromes
with congenital or developmental glaucoma. The greater number and tortuosity of the conjunctival blood vessels can be an indicator of glaucoma. Minor angle abnormalities, heterochromia, and choroidal haemangioma are often present in Sturge–Weber syndrome
The success of NPGS in aniridia (Fig. 7.10) and anterior-segment dysgenesis (Axenfeld’s anomaly) (Fig. 7.8) cases is dependent on the degree of anatomical distortion. Because these cases are rare and complicated, only an experienced NPGS surgeon should operate on these cases. Schlemm’s canal rudiments are often seen in these cases. In aniridia and anterior segment dysgenesis syndromes the trabeculum is abnormal and it is possible to predict intraoperatively which case will respond to NPGS according to the amount of filtration observed during the operation.
Figure 7.10 Aniridia with dislocated lens.
Figure 7.11 Essential iris atrophy.
Status post-laser trabeculoplasty
Figure 7.12 Uveitic glaucoma.
Figure 7.13 Narrow-angle glaucoma.
Glaucoma secondary to uveitis
Narrow-angle glaucoma
When elevated IOP persists after uveitis (Fig. 7.12) has been under control, glaucoma surgery is indicated. NPGS is indicated in these cases because it explores the site of resistance to aqueous outflow. During the inflammatory phases the trabeculum ultra-structures undergo changes, which interfere with their normal function. These changes are mostly temporary but when they are permanent, glaucoma results. The trabeculum can be reconditioned to improve filtration. Nevertheless, in cases where multiple peripheral anterior synechiae have occurred NPGS may not offer an efficient solution.
Relative contraindications The relative contraindications for NPGS depend on the status of the trabeculum because this surgery relies on the integrity of this structure for its outcome.
93
At present, most of the glaucomatologists have realized that the treatment of choice in narrow-angle glaucoma (Fig 7.13) is cataract/lens extraction. Laser iridotomy or surgical iridectomy is only a temporary measure and removal of the crystalline lens, irrespective of its transparency, deepens the anterior chamber and opens the angle of the eye. When narrow-angle glaucoma has persisted for some time, glaucoma surgery could be indicated in combination with lens extraction. In this case, NPGS should be attempted, even though the trabeculum may not respond to the surgery.
Status post-laser trabeculoplasty In eyes previously treated by laser trabeculoplasty, the trabeculum may be friable and
94
Indications and contraindications for non-penetrating surgery
Figure 7.14 Neovascular glaucoma.
Figure 7.15 Late neovascular glaucoma.
rupture during surgery. If iris prolapse is significant, NPGS is then converted to classical trabulectomy.
Post-trauma anglerecession glaucoma In angle-recession glaucoma, the trabeculum has been torn and NPGS is not possible. NPGS can be attempted, however, because damage to the trabeculum is not always complete.
Absolute contraindications: neovascular glaucoma In neovascular glaucoma, new blood vessels invade the angle (Figs. 7.14–7.16). NPGS will fail in these cases because the iridocorneal angle
Figure 7.16 Angiography of same case as Fig. 7.14.
is invaded by blood vessels. The trabeculum loses its filtering function because of the neovascularization. This type of glaucoma is the most difficult to treat and until now only implantation of silicone-tube valves has yielded results.15
References
References 8. 1. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with collagen implant and trabeculectomy in openangle glaucoma. J Cataract Refract Surg 1999;25:323–31. 2. Lavin MJ, Wormald RP, Migdal CS, Hitchings RA. The influence of prior therapy on the success of trabeculectomy. Arch Ophthalmol 1990;108:1543–48. 3. Dahan E, Drusedau MU. Nonpenetrating filtration surgery for glaucoma: control by surgery only. J Cataract Refract Surg 2000;26:695–701. 4. Jay JL, Allan D. The benefit of early trabeculectomy versus conventional management in primary open angle glaucoma relative to severity of disease. Eye 1989;3:528–35. 5. Henchoz L, Shaarawy T, Mermoud A. Surgery induced cataract and cataract progression following deep sclerectomy with collagen implant compared to trabeculectomy. In: American Academy of Ophthalmology. Dallas, 2000, USA. 6. Gandolfi S, Cimino L. Deep sclerectomy without absorbable implants and with unsutured scleral flap: prospective, randomized 2-year clinical trial vs trabeculectomy with releasable sutures. In: Association for Research in vision and Ophthalmology. 2000. Fort Lauderdale, USA. 7. Ruderman JM, Harbin TS Jr, Campbell DG. Postoperative suprachoroidal hemorrhage
9.
10.
11.
12.
13.
14.
15.
95
following filtration procedures. Arch Ophthalmol 1986;104:201–05. Bellows AR, Chylack LT, Hutchison BT. Choroidal effusion during glaucoma surgery in patients with prominent episcleral vessels. Arch Ophthalmol 1979;97:493–7. Bellows AR, Chylack LT Jr, Hutchinson BT. Choroidal detachment: clinical manifestation, therapy and mechanism of formation. Ophthalmology 1981;88:1107–15. Hamel M, ST, MA. Deep sclerectomy with collagen implant in glaucomatous patients with high myopia. In: European Glaucoma Society millennium meeting, 2000. London, UK. Mitchell P, Hourihan F, Sandbach J, Wang JJ. The relationship between glaucoma and myopia: the Blue Mountains Eye Study. Ophthalmology 1999;106:2010–15. Shaarawy T, Mermoud A. Long-term results of deep sclerectomy with collagen implant in pseudoexfoliative glaucoma. Ophthal Res 2000;32. Shaarawy T et al. Long-term results of deep sclerectomy with collagen implant in pseudophakic glaucoma patients. In: American Academy of Ophthalmology, 2000. Dallas, USA. Tixier J, Dureau P, Becquet F, Dutier JL. Sclérectomie profonde dans le glaucome congénital: résultats préliminaires. J Fr Ophtalmol 1999;22:545–48. Mermoud A, Salmon JF, Alexander P et al. Molteno tube implantation for neovascular glaucoma. Long term results and factors influencing the outcome. Ophthalmology 1993;100:897–902.
8 Surgical technique André Mermoud and Emilie Ravinet
When performing non-penetrating glaucoma surgery the surgeon, besides obtaining adequate control of intraocular pressure (IOP), has two aims. First, a trabeculoDescemet’s window needs to be created that allows a reproducible postoperative outflow resistance, thereby decreasing the immediate postoperative complication rate. Second, an intrascleral filtering bleb needs to be promoted at the expense of a subconjunctival bleb, thus reducing the risk of late hypotony and blebrelated endophthalmitis. According to results of medium-term and long-term studies on non-penetrating
glaucoma surgery, we know that over 50% of patients will need a Nd:Yag goniopuncture some time after surgery.1,2 Therefore it is important to create a thin but large enough Descemet’s window. To promote a functional intrascleral filtering bleb, a large and deep sclerectomy should be done. Several implants have been proposed in an attempt to avoid collapse of this created space. By describing the surgical procedure stepby-step, this chapter should help the surgeon to find Schlemm’s canal easily, dissect a correctly sized trabeculo-Descemet’s window, and obtain a large intrascleral filtering bleb.
(a)
(b)
Figure 8.1 a Limbal-based conjunctival incision. b Fornix-based conjunctival incision.
98
Surgical technique
All kinds of anaesthesia have been used successfully with non-penetrating glaucoma surgery. With peribulbar or retrobulbar block, the smallest amount of anaesthetic should be used to allow rotation of the globe and thus give good exposure for the deep sclerectomy dissection. 3–4 mL of a 50:50 solution of bupivacaine 0.75% and Xylocaine 4%, and hyaluronidase 50 U are usually sufficient for successful local anaesthesia. Topical and subconjunctival anaesthesia are also feasible in selected cases. A superior rectus muscle or an intracorneal traction suture is placed and the eyeball is rotated to expose the chosen site for the deep sclerectomy (usually the superior quadrant). The conjunctiva is opened either at the limbus or in the fornix (Figs. 8.1a and 8.1b). In the limbal-based conjunctival flap, the opening should be immediately below the superior rectus insertion to avoid haemorrhage from the muscular arteries. Tenon’s space is then opened. The incision is extended in both directions for a total length of about 8–10 mm. The conjunctival flap is held gently, the flap is undermined in all directions with scissors taking care not to buttonhole. When sufficient sclera is exposed under the conjunctival flap, the surface loose epithelial tissue is scraped with, for example, a Hockey blade. We tend to excise abundant Tenon’s capsule, when present. In limbal-based conjunctival flap, the scleral exposure should be as anterior as possible; minimal wetfield electrocoagulation cautery should be applied as necessary (Fig. 8.2). A superficial scleral flap—a third of the estimated total scleral thickness (about 300 µm), measuring 5 5 mm—should be delineated with a metal blade and then dissected with a crescent ruby blade (Huco vision SA, St-Blaise, Switzerland) (Figs 8.3a and 8.3b). It is important to continue the
Figure 8.2 Wetfield electrocoagulation cautery.
dissection forward into clear cornea for 1–1.5 mm (Figs 8.3c and 8.3d). A sponge soaked in mitomycin-C, 0.02%, is used at this stage, in patients at high risk of scarring. The sponge is placed in the scleral bed as well as on top of the scleral flap in the subconjunctival space for a total of 45–60 s, taking care not to touch the edges of the conjunctival incision (Fig. 8.4). After removing the sponge, the site is washed thoroughly with balanced salt solution (20–30 mL). An assistant holding the scleral flap with fine-toothed forceps without excessive traction, or a traction suture passing through the scleral flap, allows good exposure of the scleral bed for the deep scleral dissection. At this stage also, magnification is increased and maximum illumination used. The deep scleral flap measures 4 4 mm, leaving a margin of 0.5 mm on each side and slightly wider posteriorly. The two lateral incisions are made first, followed by the posterior one, using a diamond blade (0.5 3 mm) (Fig. 8.5). Complete perforation of the sclera in some parts of the incisions allows visualization of the ciliary
Surgical technique
(a)
(b)
(c)
(d)
99
Figure 8.3 a Superficial scleral flap (measures 5 5 mm). b Delineation of superficial scleral flap with metal blade, depth of incision about 300 µm. c Extension of superficial scleral flap into clear cornea for 1–1.5 mm. d Variation in shape of superficial scleral flap.
body anteriorly and the choroid body posteriorly (Fig. 8.6). In fact, complete perforation of the sclera down to the choroid in one of the two corners allows an estimation of the remaining scleral depth. From our experience, this has never led to any complication. The deep scleral flap is then dissected forward with
a crescent ruby blade (2 mm bevelled up) from one posterior corner keeping in the same plane and only leaving 5–10% of sclera above the uvea (Figs. 8.7a and 8.7b). By following these recommendations, the surgeon will find Schlemm’s canal; the dissection is then continued anteriorly. The scleral
100
Surgical technique
Figure 8.4 In patients at risk for postoperative fibrosis, a sponge soaked in mitomycin-C is placed in the surgical site and then washed with balanced salt solution.
Figure 8.6 Choroid or ciliary body vizualisation allows good estimation of total remaining scleral thickness.
(a)
Figure 8.5 Deep sclerectomy delineated with diamond blade (measures 4 4 mm when completed). (b) spur is also an excellent landmark for identifying Schlemm’s canal (Fig. 8.8). In the posterior part of the scleral dissection, scleral fibres are laid at random in multiple directions. Anteri-
Figure 8.7 a Beginning of deep sclerectomy horizontal dissection with a ruby blade. b Horizontal dissection in progress at halfway stage.
Surgical technique
101
Figure 8.8 Opening of Schlemm’s canal. The canal appears dark and contrasts with adjacent white scleral spur.
Figure 8.9 Complete opening of Schlemm’s canal.
Figure 8.10 Paracentesis.
Figure 8.11 Exposure of anterior trabeculum and Descemet’s membrane.
orly, they become more organized, eventually forming a ligament parallel to the limbus, namely the scleral spur, just posterior to Schlemm’s canal (Fig. 8.9). Before opening Schlemm’s canal along its posterior border, from the right side to the left in right-handed people, a paracentesis is done to decrease the
IOP and also allow anterior chamber reformation, should a large perforation occur (Fig. 8.10). After “deroofing” Schlemm’s canal, the sclerocorneal dissection is then carried forward for 1–1.5 mm to remove the sclerocorneal tissue in front of the anterior trabeculum and Descemet’s membrane. This part of the surgery
102
Surgical technique
Figure 8.12 Radial cut with metal blade held upside down to expose Descemet’s membrane.
is difficult because there is a high risk of anterior chamber perforation in inexperienced hands. To detach Descemet’s membrane two radial incisions are made in the corneal tissue (Fig. 8.11). To avoid perforation of Descemet’s membrane or anterior trabeculum, the metal blade can be used upside down while at the same time cutting anteriorly situated structures with the sharp side of the blade, continuing forward the lateral-radial cuts (Fig. 8.12). To complete the exposure of the trabeculoDescemet’s membrane between the two radial cuts, one can use a wet triangular sponge to press down while simultaneously lifting up the deep scleral flap with the other hand (various instruments may be used for this step, such as metal blade, diamond knife, spatula, depending on the surgeon’s preference (Figs. 8.13a and 8.13b). On completion of the anterior dissection, the deep scleral flap is removed first by precutting it very close to its base and on its posterior portion with a diamond blade (Fig. 8.14) and then by completing the excision with Galan’s scissors (Fig. 8.15). Percolation of aqueous through the trabeculo-Descemet’s
(a)
(b) Figure 8.13 a Dissection of Descemet’s membrane with sponge. b Dissection of Descemet’s membrane with ruby blade.
Figure 8.14 Final stage of anterior dissection of deep scleral flap by cuts with diamond or metal blade.
Surgical technique
Figure 8.15 Flap finally removed with Galan’s scissors.
window should already be present at this stage; this will be enhanced after ab externo trabeculectomy by restoring some physiological aqueous outflow through the posterior trabeculum. This additional procedure removes the juxtacanalicular trabeculum and Schlemm’s endothelium (Fig. 8.16) as has been shown on histological examination (see Chapter 5). Those structures are thought to offer the highest outflow resistance, in primary and probably other types of secondary openangle glaucoma. During ab externo trabeculectomy,3 Schlemm’s endothelium and juxtacanalicular trabeculum are peeled away with blunt forceps (Fig. 8.17; deep sclerectomy forceps 13.0 mm jaws, Huco vision SA). This procedure is easily done and often a continuous 4 mm membrane can be peeled off. However, it is crucial to dry the exposed inner wall of Schlemm’s canal before delicately holding it. Variations of ab externo trabeculectomy, as described here, using scraping (Dahan) or aspirating (Bechetoille) methods also exist. To maintain the patency of the deep scleral bed, any space-
103
Figure 8.16 Inner wall of Schlemm’s canal as seen with scanning electron microscope.
Figure 8.17 Peeling of inner endothelium of Schlemm’s canal and juxtacanalicular trabeculum, by the double plated forceps.
maintaining device, for example, implants or viscoelastics, may be used. In the case of a round collagen implant a single 10/0 nylon suture is used to secure it inside the scleral bed
104
Surgical technique
Figure 8.18 Collagen implant sutured in scleral bed. This implant will serve as a space maintainer to create an intrascleral space for aqueous humor filtration.
(Fig. 8.18). The collagen implant should rest anteriorly on the trabeculo-Descemet’s membrane. The superficial scleral flap is then repositioned into place, covering the spacemaintaining device and a loose 10/0 nylon suture, which is later buried, is placed at each posterior corner (Fig. 8.19). Conjunctiva and Tenon’s capsule are closed in two layers with a running, aqueous proof, 8/0 Vicryl suture (Fig. 8.20). The round collagen implant measures 4 mm in length and 0.5 mm in diameter; it is processed from lyophilized porcine scleral collagen, which is sterilized by radiation. The implant is highly hydrophilic, the water content of the hydrated device being 99%.4 After placement in the scleral space the implant absorbs ocular fluids and swells to about twice its dry state size. Ultrasound biomicroscopic studies have shown complete resorption of the collagen implant 6–9 months after surgery, leaving an intrascleral space for aqueous filtration.5 Other space-maintaining devices that have been used in non-penetrating glaucoma surgery include
Figure 8.19 Superficial scleral flap repositioned and sutured with two untied 10/0 nylon sutures.
Figure 8.20 Closure of Tenon’s capsule and conjunctiva with running 8/0 Vicryl suture.
high-viscosity hyaluronic acid which is also used in Stegmann’s viscocanalostomy (Healon GV, Pharmacia, Upsalla, Sweden; Fig. 8.21).6 Reticulated hyaluronic acid (SK gel implant, Corneal, Paris, France; Fig. 8.22) has been used
Trabeculo-Descemet’s perforation
Figure 8.21 Injection of Healon GV in the scleral space.
105
Figure 8.22 SK gel.
to maintain space, which may promote the patency of the scleral space for a longer period, as has a non-absorbable Hema implant (T-flux, IOLtech, La Rochelle, France; Fig. 8.23). For more information on implants see Chapter 10.
Trabeculo-Descemet’s perforation As with any surgery, there is a learning curve for non-penetrating glaucoma surgery, particularly because the dissection is done anteriorly. Initially, surgeons may quite often perforate the thin trabeculo-Descemet’s membrane.7,8 Management of this complication is fully detailed in Chapter 14. Briefly, it is important, when a substantial portion of the iris prolapses through the perforation, for the surgeon to do a peripheral iridectomy (Fig. 8.24) and tightly close the superficial scleral flap with six to eight 10/0 nylon sutures (Fig. 8.25). The scleral space
Figure 8.23 T-flux sutured with nylon 10/0.
should be filled with high-viscosity hyaluronic acid to increase the aqueous outflow resistance and prevent postoperative hypotony and shallow anterior chamber. If the anterior chamber is shallow intraoperatively, a viscoelastic substance may also be injected through the paracentesis into the anterior chamber under the
106
Surgical technique
Figure 8.24 Basal iridectomy after iris prolapse.
trabeculo-Descemet’s membrane to reposition the iris (Fig. 8.26). This injection should be done carefully to avoid postoperative ocular hypertension. Collagen or the other implants are usually not used in this situation. However, should the perforation of the trabeculoDescemet’s membrane be minimal, a collagen implant may be used and positioned so that it will cover the defect by swelling up. The addition of a viscoelastic substance will then depend on the size of the microperforation. Combined phacoemulsification and nonpenetrating glaucoma surgery are feasible and are detailed in Chapter 9.
Postoperative management and medication Although patients can be discharged from hospital on the same day as the surgical procedure takes place, we tend to extend their stay until the next day.
Figure 8.25 Tight superficial scleral flap closure with eight nylon 10/0 sutures to increase aqueous humor outflow resistance.
Figure 8.26 Careful injection of viscoelastic in the anterior chamber to separate the iris from Descemet’s membrane.
At first patients are treated with a topical combination of corticosteroid and antibiotic, three times a day for 2–3 weeks, followed by a non-steroidal anti-inflammatory drug, three times a day for at least 3 months postoperatively. Mydriatics are not routinely prescribed.
Postoperative management and medication
107
Figure 8.29 Diamond blade.
Figure 8.27 Huco instruments used in non-penetrating glaucoma surgery. Figure 8.30 Double-plated forceps for peeling the inner wall of Schlemm’s canal.
Figure 8.28 Ruby blade.
Figure 8.31 Dahan’s knife used to make two radial cuts in cornea to expose Descemet’s membrane.
108
Surgical technique
Management of bleb failures, goniopuncture and other early and late postoperative complications are discussed in Chapter 13.
2.
Instruments
3.
Several instruments have been specially designed to optimize surgical technique and facilitate each step of the procedure (Fig. 8.27). A ruby blade is used to dissect the horizontal plane of the superficial and deep scleral flap (Fig. 8.28). A sharp 30° diamond blade is used for deep scleral delineation and for excision of the deep scleral flap (Fig. 8.29). A doubleplated forceps (Fig. 8.30) is useful for peeling the inner wall of Schlemm’s canal and juxtacanalicular trabeculum (ab externo trabeculectomy). Dahan’s diamond knife has been designed to make the two radial cuts after opening the deep scleral flap in order to create the Descemet’s window (Fig. 8.31).
References 1.
Kozlov VI, Bagrov SN, Anisimova SY et al. Non penetrating deep sclerectomy with
4.
5.
6.
7.
8.
collagen. Ophthalmosurgery 1990;3: 44–46. Karlen M, Sanchez E, Schnyder CC et al. Deep sclerectomy with collagen implant: medium term results. Br J of Ophthalmol 1999;83:6–11. Arenas E. Trabeculectomy ab-externo. Highlights Ophthalmol 1991;19:59–66. Sanchez E, Schnyder CC, Sickenberg M et al. Deep sclerectomy: results with and without collagen implant. Int Ophthalmol 1997;20:157–62. Chiou AG, Mermoud A, Underdahl JP, Schnyder CC. An ultrasound biomicroscopic study of eyes after deep sclerectomy with collagen implant. Ophthalmology 1998;105:746–50. Stegmann RC. Viscocanalostomy: a new surgical technique for open angle glaucoma. An Inst Barraquer Spain 1995;25:229–32. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with collagen implant and trabeculectomy in openangle glaucoma. J Cataract Refract Surg 1999;25:323–31. Sanchez E, Schnyder CC, Mermoud A. Résultats comparatifs de la sclérectomie profonde transformé en trabéculectomie et de la trabéculectomie classique. Klin Monatsbl Augenheilkd 1997;210:261–64.
9 Viscocanalostomy Robert C Stegmann and Roberto G Carassa
Viscocanalostomy is a new surgical procedure developed for glaucoma surgery. The technique is non-penetrating and independent of external filtration, thus having many important potential advantages over standard trabeculectomy. The avoidance of anterior chamber opening reduces the risk of infection, cataract, hypotony, and flat anterior chamber. The absence of external filtration avoids bleb formation and related discomfort, minimizes the risk of late infections, and, most importantly, the success of the procedure is unaffected by conjunctival and episcleral scarring, which are the leading cause of failure in trabeculectomy. The procedure was first proposed in the early 1990s by Robert Stegmann, and, based on previous studies of non-penetrating trabeculectomies by Krasnov1 and by Zimmerman et al,2 viscocanalostomy aims to restore the natural outflow pathway by allowing the aqueous to leave the eye through Schlemm’s canal and the episcleral veins. The procedure creates a bypass by which aqueous humor can reach Schlemm’s canal, skipping the impaired trabecular meshwork. This bypass is created by the production of a “chamber” inside the sclera in direct communication both with Schlemm’s canal and with the anterior chamber through the intact Descemet’s membrane; a “window” is formed just anterior to the trabecular meshwork (Fig. 9.1). The aqueous enters the chamber by percolating through the membrane, and leaves it via Schlemm’s canal.
Figure 9.1 Ultrasound biomicroscopy (UBM) image showing intrascleral chamber and intact Descemet’s window.
Surgical technique To simplify the technique, a specific surgical set, comprising a 0.5 mm diamond knife, a 1 mm round steel bevel-up blade, and a 165 µm blunt needle to cannulate Schlemm’s canal (Alcon-Grieshaber, Switzerland) should be used. Viscocanalostomy is done under retrobulbar or peribulbar anesthesia and usually requires 25–40 minutes, depending on bleeding control. To avoid damage to outflow channels (such as Schlemm’s canal, aqueous
110
Viscocanalostomy Figure 9.2 Opening of Schlemm’s canal: by advancing the inner flap at correct plane of dissection, Schlemm’s canal is easily deroofed.
veins, collector channels), wetfield cautery is used as little as possible, and bleeding is reduced by frequent irrigation of the surgical area with vasoconstrictive solutions such as ornipressin (POR 8, Sandoz, Switzerland). To provide optimal visualization of the surgical site, a bridle suture should be passed either on the superior rectus or in clear cornea. The surgical technique can be divided into nine steps:
1: Conjunctival flap dissection Viscocanalostomy does not have to be accurate in the dissection of the conjunctival flap, and can be done in any quadrant, although the upper and temporal quadrants are most commonly chosen. The surgical field is prepared by creating a fornix-based
conjunctival flap, with as little wetfield cautery as possible.
2: Outer scleral flap dissection A 5 5 mm parabolic cut about 200 µm deep, is made with the diamond knife (the incision can be outlined by a calibrated diamond knife to ensure a constant depth of cut). After reaching the correct plane of cut the flap is dissected anteriorly in clear cornea by advancing the incision with the bevel-up spatula, which allows easier following of the plane.
3: Inner scleral flap dissection A 4 4 mm parabolic cut parallel to the outer incision is made beneath the outer flap. The
Surgical technique
111
Figure 9.3 Creation of Descemet’s window: by gently pulling inner flap and delicately pressing on Schlemm’s canal with cotton swab, the window is created and advanced in clear cornea.
choroidal plane must be almost reached, and this is revealed by the observation of a dark reflex at the bottom of the cut. With the specific beveled spatula a precise dissection is advanced until Schlemm’s canal is reached and deroofed (Fig. 9.2), leaving two patent openings on the lateral edges of the cut. To maintain the same plane of dissection and provide sharp lateral edges, progressive deepening of the lateral cuts is often needed.
4: Paracentesis A paracentesis should always be made to decrease intraocular pressure (IOP), to make cannulation of Schlemm’s canal easier, and to reduce bulging of Descemet’s membrane during its cleavage from the corneal stroma, which is at high risk of tear formation. To
avoid external pressure on the eye, the traction on the bridle suture should also be removed.
5: Cannulation of Schlemm’s canal With the specific 165 µm cannula, high molecular weight sodium hyaluronate (Healon GV, Pharmacia Corp, NJ, USA) is slowly injected into Schlemm’s canal by cannulating the two ostia at the lateral edges of the inner flap. To avoid damage to the canal endothelium, the insertion of the cannula should not go beyond 1–1.5 mm from the ostia. The injection of viscoelastic substance allows progressive atraumatic dilatation of Schlemm’s canal at a position of 1–2 o’clock from the ostia. Moreover, its hemostatic properties avoid bleeding and fibrin clot formation, thus limiting
112
Viscocanalostomy Figure 9.4 Schlemm’s canal and percolating window are evident at the bottom of the inner flap.
healing processes and scarring of Schlemm’s canal openings. The slow injection should be repeated six or seven times on each side.
then advanced in clear cornea for about 1 mm by a careful deepening of the lateral cuts with the round bevel-up spatula (Fig. 9.4).
6: Creation of Descemet’s window
7: Inner scleral flap excision
The alternative route by which aqueous humor bypasses the trabecular meshwork and reaches Schlemm’s canal is a window created right anterior to the side of the canal, and represented by the anterior portion of the trabecular meshwork and by the intact Descemet’s membrane. The window is created by gently pulling the inner scleral flap upwards and delicately depressing the floor of the canal and Descemet’s membrane with the tip of a cotton swab (Fig. 9.3). By carefully repeating the procedure, the membrane is progressively cleaved from the scleral flap. The flap itself is
The inner scleral flap is excised with very sharp Vannas’ scissors to avoid damage to Descemet’s membrane.
8: Outer scleral flap suture To seal the intrascleral chamber, the outer scleral flap should be tightly sutured by placing six or seven 10/0 nylon stitches. The step created by the different size of the two flaps allows a tight and better apposition of the external flap. Finally, to minimize bleeding
Results and prevent collapsing and scarring of the intrascleral chamber, high molecular weight sodium hyaluronate (Healon GV, Pharmacia Corp) is injected underneath the flap.
9: Closure of the conjunctiva The procedure ends by repositioning the conjunctiva with two lateral stitches, and by giving a subconjunctival injection of steroids and antibiotics. An intrascleral chamber has been thus created; its floor is made posteriorly by the thin scleral layer overlying the choroid, centrally by the deroofed Schlemm’s canal, and anteriorly by Descemet’s window over the anterior chamber. The roof of the chamber is formed by the tightly sutured outer flap, and two openings in Schlemm’s canal ostia are present in the lateral walls. The aqueous humor will by-pass the functionally impaired trabecular meshwork, by percolating through Descemet’s membrane into the intrascleral chamber. From there it will leave the eye through Schlemm’s canal ostia without need of external filtration; it is likely, however, that part of the aqueous humor could leave through the external flap under the conjunctiva or through the thin scleral layer into the suprachoroidal space. Experimental studies in animal models should better clarify the mechanism of the procedure.
Results Viscocanalostomy All published studies present high success rates with few and minor complications for visco-
113
canalostomy. In the report by Stegmann et al3 which had a follow-up of 64 months, 214 eyes of black patients had an IOP below 22 mmHg in 82.7% of cases without medical therapy and in 89% of the eyes with adjunctive use of beta-blockers. In a prospective pilot study on 33 eyes of white patients followed-up for 10 months, viscocanalostomy allowed an IOP below 21 mmHg in 86% of eyes and below 16 mmHg in 79% of the eyes without medication, with a mean postoperative pressure of 12.0 ± 3.0, range 7–18 mmHg.4 In a retrospective analysis presented at the 1998 meeting of the American Academy of Ophthalmology, in 26 eyes IOP decreased from 22.7 mmHg to 19.8 mmHg without additional medical therapy, and in a 30-month report on 80 consecutive cases, 66% of eyes reached an IOP below 20 mmHg.5,6 Drusedau et al 7 reported that 56 eyes followed-up for 1 year had a success rate, defined as the achievement of an IOP below 21 mmHg, of only 36%.
Viscocanalostomy versus trabeculectomy The only controlled randomized trial aimed at comparing viscocanalostomy with trabeculectomy is by Carassa and colleagues8,9 on 50 eyes followed-up for 24 months. Viscocanalostomy was performed as described, and trabeculectomies were done with no intraoperative antimetabolites but with the possibility of postoperative 5–fluorouracil injections and suture lysis. Mean initial and final IOPs were respectively 24.6 ± 10.6 and 14.0 ± 2.6 mmHg (p = .0000) in the trabeculectomy group (Group 1) and 22.3 ± 7.4 and 13.3 ± 3.8 mmHg (p = .0000) in the viscocanalostomy group (Group 2). Table 9.1 summarizes the IOPs reached over time in the
114
Viscocanalostomy
Table 9.1 Kaplan–Meier survival data for IOPs 21 mmHg with no medication with at least 20% IOP reduction from baseline. Month
Viscocanalostomy
Trabeculectomy
1 6 12 18 24
84% 80% 76% 76% 76%
100% 100% 96% 92% 92%
successful eyes; a significant difference, with Group 2 having lower IOP than Group 1, was achieved at 2, 3, and 6 months. This difference became smaller and non-significant thereafter because of an increasing trend in the trabeculectomy group. When considering the Kaplan–Meier cumulative probability of success, for an IOP of 6–21 mmHg with no medication, with at least 20% IOP reduction from baseline, the success of trabeculectomy was greater than viscocanalostomy at all time intervals, with a final success at 2 years of 92% and 76%, respectively (Table 9.1). The differences between the two groups were related to the initial failures in the viscocanalostomy group. In fact, four (16%) eyes were regarded as immediate failures: one for an intraoperative conversion, two for surgical revision after iris prolapse into a tear in Descemet’s membrane, and one for laser goniopuncture. If we were to exclude these eyes, thus considering only the eyes in which the procedure was successful, the success of the two techniques is the same throughout the follow-up with a final 90% for viscocanalostomy (Table 9.2). We can say that if viscocanalostomy is done properly and it is succesful in the first few
Table 9.2 Kaplan–Meier survival data for IOPs 21 mmHg with no medications with at least 20% IOP reduction from baseline. In the viscocanalostomy group only the initial successful eyes are considered (n = 21). Month
Viscocanalostomy
Trabeculectomy
1 6 12 18 24
100% 95% 90% 90% 90%
100% 100% 96% 92% 92%
days, its success is satisfactory and stable over time. Further studies should attempt to improve the surgical technique or try to select patients to ensure a more reproducible outcome, increasing the overall success rate by avoidance of initial failures. If we assume for success an IOP of 6–16 mmHg, trabeculectomy is more successful than viscocanalostomy, with a success rate of 75% versus 46% (Table 9.3). This overall advantage of trabeculectomy can be related to the adjunctive use of antimetabolites.
Table 9.3 Kaplan–Meier survival data for IOPs 16 mmHg with no medications with at least 20% IOP reduction from baseline. Month
Viscocanalostomy
Trabeculectomy
1 6 12 18 24
80% 56% 52% 52% 46%
92% 88% 84% 75% 75%
References On the other hand, viscocanalostomy allowed easier postoperative management with fewer visits, and gave rise to less eye discomfort than trabeculectomy, which might be expected given the absence of the filtering bleb. Viscocanalostomy had few complications, mostly related to intraoperative problems such as small microperforations in Descemet’s membrane that gave rise to an iris plug in three cases, and to a transient 1 mm hyphema in three other eyes. In the trabeculectomy group, more relevant complications occured, such as transient hypotony with peripheral choroidal detachment in three eyes, transient IOP spike in two eyes, small hyphema in one eye, and punctate keratopathy secondary to 5–fluorouracil injections in three cases. Although viscocanalostomy should avoid bleb formation, some external filtration was found by biomicroscopy or by ultrasound biomicroscopy (UBM) in one-third of the eyes.
Conclusions Viscocanalostomy seems a promising surgical technique for lowering IOP in glaucomatous eyes. The procedure has several potential advantages over trabeculectomy, the main one being the absence of external filtration and thus if conjunctival and episcleral scarring occurs, it does not influence the surgical outcome. When considering final IOPs of 16–21 mmHg, the rate of failure over time is similar between the two procedures. Viscocanalostomy is nonetheless still affected by initial failures related to intraoperative complications. Improvements in the technique and in the instrumentation are needed to increase the final success rate. Nevertheless, the procedure is affected by few and minor
115
complications, postoperative management is easy and it induces less eye discomfort than trabeculectomy. Results from basic research aimed at delineating the exact mechanism will inevitably improve in the surgical technique.
References 1.
2.
3.
4.
5.
6.
7.
8.
Krasnov MM. Sinusotomy: foundations, results, prospects. Trans Am Ophthalmol Otolarygol 1972;76:369–74. Zimmerman TJ, Kooner KS, Ford VJ et al. Trabeculectomy vs non-penetrating trabeculectomy: a retrospective study of two procedures in phakic patients with glaucoma. Ophthalmic Surg 1984;15:734–40. Stegmann R, Pienaar A, Miller D. Viscocanalostomy for open-angle glaucoma in black African patients. J Cataract Refract Surg 1999;25:316–22. Carassa RG, Bettin P, Fiori M, Brancato R. Viscocanalostomy: a pilot study. Eur J Ophthalmol 1998;8:57–61. Milauskas AT, Coleman AL. The evaluation of viscocanalostomy, a new glaucoma surgery. Final Program of 1998 AAO Meeting, San Francisco, American Academy of Ophthalmology 1998: 174. Sunaric-Megevand G, Desmangles P, Leuenberger PM. Follow-up after viscocanalostomy, a non-perforating glaucoma filtering surgery. Invest Ophthalmol Vis Sci 1999;40:S270 (abstr). Drusedau MU, von Wolff K, Bull H, von Barsewisch B. Viscocanalostomy for primary open-angle glaucoma: the Gross Pankow experience. J Cataract Refract Surg 2000;26:1367–73. Carassa RG, Bettin P, Fiori M et al. Viscocanalostomy vs trabeculectomy: a 12–month randomized prospective trial. Invest Ophthalmol Vis Sci 2000;41:S744 (abstr).
116 9.
Viscocanalostomy
Carassa RG. Viscocanalostomy vs trabeculectomy: 2–year follow-up. Glaucoma 2000: cutting-edge diagnosis and therapy. San Francisco: American Academy of Ophthalmology, 2000: 91–96.
10 Modulation of wound healing Tarek Shaarawy
Wound healing The wound healing process starts immediately1,2 after the surgical trauma, and can be classified into three stages:
Inflammatory stage Vascular permeability occurs, producing inflammatory exudates and edema, giving blood elements access to the wound. These blood elements clear the wound of debris by phagocytosis and enzyme activity, such as removal of damaged collagen by collagenase; they induce clotting and are the stimulus for a variety of cell developments. 3
Fibroblastic proliferative stage Fibroblastic activity intensifies after 12–14 h, possibly as a result of mitogenic stimuli from platelets. Fibroblasts produce collagen, initially as monomers that undergo polymerization in the extracellular space. These collagen fibers serve as support for angiogenesis. New vessels start as capillary buds that extend onto the collagen matrix, providing nutrients for more fibroblasts, which, in turn, lay down more immature collagen, allowing further growth of the new vessels. This process continues until the wound is bridged by this fragile
network of new vessels, usually occurring in a few days in a clean surgical wound.3
Connective tissue synthesis stage At least 11 different enzymes participate in the production of at least seven different forms of collagen. As this collagen is laid down, crosslinking and contraction occur, producing the acute scar. Over time, the collagen is remodelled, more or less to the approximate characteristics of the original tissue.3 In glaucoma filtering surgery it is the production, contraction, and remodelling of collagen that cause the failure of most blebs.
Factors influencing outcome Several factors interact to influence the outcome of glaucoma filtering procedures.
Characteristics of patients Age
Several studies have shown that filtration surgery is more likely to be successful in older patients.4–7 The success of filtration surgery in
118
Modulation of wound healing
the elderly has been attributed to “senescence of wound healing with aging”.8 Race
A tendency toward filtration failure has been reported in African-American patients,9–11 patients from East Africa,12,13 and the Caribbean.14 It is not known if the tendency for failure is related to biological or to socioeconomic factors such as the accessibility to health care. Experienced surgeons treating black patients from the USA and the Caribbean, who had good access to health care, still reported a pronounced tendency for filtration surgery to fail, suggesting inherent biological differences in wound healing.8 It is not known whether these differences are related to a predilection for hypertrophic scarring or keloids. Type of glaucoma
Some types of glaucoma are associated with higher incidence of early filtration failure with closure of the fistula and scarring of the bleb within the first few postoperative weeks. In neovascular glaucoma, fibrovascular tissue over the surface of the ciliary body, iris, and angle contribute to wound healing at the filtration site.15 Patients with active intraocular inflammation tend to experience early scarring as well.16 A high rate of filtration failure has also been reported in aphakic eyes.17
Conjunctival inflammation Long-term conjunctival inflammation and scarring predispose to a poor outcome after filtering surgery. Topical glaucoma medications have been implicated in causing chronic inflammatory reactions,18 and have thus
predisposed to failure of filtering surgery. The incidence of failure after deep sclerectomy has been four times higher in patients previously treated with topical glaucoma medications than in “virgin eyes”.19 Whether these changes are related to the medications themselves, the preservatives in the commercial preparations, or duration of treatment, is not yet known.
Pharmacological modulation of wound healing Three pharmacological strategies have been under intense investigation during the past few years, namely, the preoperative and postoperative administration of glucocorticoids to retard scarring, the local intraoperative or postoperative administration of antimetabolites to reduce fibroblast proliferation, and the use of agents that interfere with the synthesis of normal collagen, such as -aminopropionitrile, and penicillamine.
Antimetabolites The use of antimetabolites as intraoperative or postoperative adjuncts to glaucoma surgery has had a profound impact on success rates, particularly in glaucomas that have traditionally had very poor surgical prognosis. Antimetabolites improve the success rate of filtering surgery in eyes with a poor prognosis, and many surgeons use them when operating on patients with aphakia, a history of failed filtering surgery, penetrating ocular trauma, inflammatory or neovascular glaucoma. Antimetabolites have also been used with success in patients undergoing primary filtering surgery.20–23
Pharmacological modulation of wound healing 5-fluorouracil 5-fluorouracil is a pyrimidine analog that inhibits fibroblastic proliferation by acting selectively on the S (synthesis) phase of the cell cycle (S-phase specific). 5-fluorouracil is enzymatically converted to the nucleotide, 5-fluoro-2This deoxyuridylate monophosphate.24 competitively inhibits thymidylate synthetase, which catalyses the conversion of deoxyuridine phosphate to thymidine phosphate and affects DNA synthesis.8 5-fluorouracil may be converted to its corresponding ribophosphate, which is incorporated into RNA. Defective protein synthesis results from altered translation from mRNA and abnormal ribosomes. In a pilot study in monkeys, Gressel et al25 attempted to assess the ability of 5-fluorouracil to inhibit cicatrization at the filtering site. None of the control eyes developed blebs, but 75% of treated eyes developed blebs. The difference between intraocular pressures (IOP) in eyes treated with 5-fluorouracil and control eyes was also significant (p < = 0.05). Subsequent studies have shown the efficacy of 5fluorouracil as an adjunct to filtering surgery in different types of refractory glaucomas.26–33 5fluorouracil is frequently given as a subconjunctival injection of 5.0 mg for several days in the immediate postoperative period. Intraoperative application of 5-fluorouracil has also been shown to be effective.34,35 Animal studies have shown high concentrations of 5fluorouracil in both aqueous and vitreous humors after subconjunctival administration.36–38 The route of entry seems to be via the tear film of the cornea. Many regimens for administration, as well as different doses of 5-fluorouracil, have been reported. Initially, twice-daily administration via subconjunctival injection was given. A dose of 5.0 mg of 5-fluorouracil in 0.5 mL of saline is injected via a 30-G needle, 180° from
119
the site of filtering surgery. Injections start on the day of the surgery, or the day after, for 7 days.33 This protocol proved demanding for both the patient and the surgeon. Good results have been reported with less frequent administration, and lower concentrations.39,40 In fact, surgeons should attempt to titrate both dose and frequency according to individually variable clinical circumstances. Early postoperative complications of 5fluorouracil are usually corneal. Corneal erosions and even corneal ulcers are not an uncommon occurrence.41,42 Leaks from the conjunctival wound and suture tract occur in about one-third of patients.3 The likelihood of leaks occurring can be reduced by meticulous wound closure with tapered, non-spatulated, non-cutting needles. Other complications reported, but not as common, are choroidal hemorrhages,33 late bleb leaks, hypotonic maculopathy,43 and endophthalmitis.44
Mitomycin C Mitomycin C, an antitumor antibiotic, is a product of Streptomyces caespitosus.8 Mitomycin C inhibits DNA synthesis, and its inhibitory effects do not depend on the phase of the cell cycle; short exposures are sufficient to suppress proliferation.45 Histopathological examination of the surgical site after application of mitomycin C showed hypocellularity of fibroblasts and disruption of the normal architecture, as well as increased melanolipofuscin granules, vacuolated cytoplasm, and disrupted mitochondria of the ciliary body epithelium underlying the site of application.46 The wide use of intraoperative mitomycin C started in the early 1990s, especially after Chen et al reported the results of a 9-year prospective assessment of mitomycin C,47 and Palmer48 published his encouraging results of
120
Modulation of wound healing
the efficacy and safety of mitomycin C. Of special interest was the lack of corneal toxicity, which is a troublesome complication of 5fluorouracil, as well as the convenience of intraoperative administration for both the patient and the surgeon. There are many recommendations for the concentration of mitomycin C as well as the duration of application. Doses range from as low as 0.2 mg/mL for 30 s up to 0.5 mg/mL for 5 mins.49–52 A dose-response effect of the concentration of mitomycin C and IOP was reported, suggesting that the higher concentration leads to lower IOP.53 Higher doses are probably more effective but they render the patient more prone to complications. Except for a decreased incidence of corneal toxicity,54,55 complications from mitomycin C are similar to those seen with 5-fluorouracil.
5-fluorouracil or mitomycin C? It is difficult to compare two drugs that can be administered in different concentrations, different periods of application, different frequency, and different time of application (intraoperatively vs postoperatively). Nevertheless, the weight of evidence seems to lean in favour of mitomycin C. In multiple clinical trials results with mitomycin C equalled or surpassed those seen with 5-fluorouracil.55–58
Use of antimetabolites in non-penetrating glaucoma surgery There have been very few studies presented on the use of antimetabolites as adjuncts to non-
penetrating glaucoma surgery. Karlen et al.59 used postoperative subconjunctival injections of 5-fluorouracil in failing blebs after deep sclerectomy with collagen implant (DSCI). 23 (23%) patients received 0.1 mL of a 50 mg/mL solution of 5-fluorouracil. The mean number of injections per patient was 2.9 (SD 2.1), which was argued to be fairly low, and the mean time between operation and 5fluorouracil administration was 1.9 (SD 2.7) months. Hamard et al60 compared DSCI to deep sclerectomy with intraoperative application of 5-fluorouracil (50 mg/mL for 5 mins). IOP dropped in both groups, and success rates as well as incidence of complications were similar. These investigators concluded that the use of 5-fluorouracil is a safe procedure that can be favourably compared to the use of implants in non-penetrating glaucoma surgery. Shaarawy et al61 prospectively studied the success rates and the incidence of complications of DSCI and mitomycin C in refractory glaucoma cases. Their series of 54 eyes (54 patients) included prior failed filtering surgery, pseudophakic, inflammatory, and traumatic glaucoma. The investigators used a 0.2 mg/mL concentration of mitomycin C; a sponge soaked with mitomycin C was placed under the superficial flap for 2 mins. With a mean follow-up of 46 (SD 14) months, complete success rate (IOP ≤ 21 mmHg, without medical treatment) was 50%, while the qualified success rate (IOP ≤ 21 mmHg, with or without medical treatment) was 92%. With such a shortage of studies on the effect of antimetabolites on the success rates of nonpenetrating glaucoma surgery with antimetabolites, it is difficult to form a definite opinion about their role. Should their use be restricted to refractory cases, do they substitute implants or do they only increase the chances of an implant’s success? With mounting interest in
References this type of glaucoma surgery, at least some of those questions should soon be answered.
The future Antimetabolites, as useful as they are, do pose a risk of potentially blinding complications. Therefore, there remains a need for safer agents. Perhaps the most promising new approach appears to be in the use of molecular-based therapies, such as fully human neutralizing monoclonal antibodies, designed to target specific molecules in the scarring response.62 The effects of antibodies to transforming growth factor (TGF)-2 on in vitro and in vivo conjunctival scarring and after glaucoma filtration surgery were investigated.63 The outcome of glaucoma filtration surgery was improved in an animal model of aggressive conjunctival scarring compared to its control and was clinically safe, non-toxic, and well tolerated after subconjunctival administration. When the effect TGF-2 on the application site is histologically compared to that of mitomycin C, TGF-2 was found to be much less destructive to local tissue. Combination therapies may also afford an improved therapeutic index. It is anticipated that future therapies can offer safer, more specific, focal and titratable treatment, with far-reaching clinical applications.
References 1 Shoshan S. Wound healing. Int Rev Connect Tissue Res 1981;9:1–26. 2 Skuta GL, Parrish RK. Wound healing in glaucoma filtering surgery. Surv Ophthalmol 1987;32:149–70.
121
3. Stamper RL, Lieberman MF, Drake VF. Becker-Shaffer’s diagnosis and therapy of the glaucomas. 7th edn. St Louis: Mosby, 1999: 571–5. 4. Cadera W, Pachtman MA, Cantor LB et al. Filtering surgery in childhood glaucoma. Ophthalmic Surg 1984;15:319–22. 5. Gressel MG, Heuer DK, Parrish RK. Trabeculectomy in young patients. Ophthalmology 1984;91:1242–6. 6. Kwitko ML. Secondary glaucoma in infancy and childhood: a review. Can J Ophthalmol 1969;4:231–46. 7. Kwitko ML. Congenital glaucoma. A clinical study. Can J Ophthalmol 1967;2:91–102. 8. Parrish R, Folberg R. Wound healing in glaucoma surgery. In: Ritch R, Shields MB, Krupin T, eds. The glaucomas. St Louis: Mosby, 1996, 1633–51. 9. Freedman J, Shen E, Ahrens M. Trabeculectomy in a black American glaucoma population. Br J Ophthalmol 1976;60:573–4. 10. Merritt JC. Glaucoma blindness in African Americans: have 55 years of therapies, technologies, and talent altered blindness rates? J Natl Med Assoc 1996;88:809–19. 11. Miller RD, Barber JC. Trabeculectomy in black patients. Ophthalmic Surg 1981;12:46–50. 12. Bakker NJ, Manku SI. Trabeculectomy versus Scheie’s operation: a comparative retrospective study in open-angle glaucoma in Kenyans. Br J Ophthalmol 1979;63:643–45. 13. Kietzman B. Glaucoma surgery in Nigerian eyes: a five-year study. Ophthalmic Surg 1976;7:52–58. 14. Hilgers JH. Glaucoma surgery in Caribbean Negroes on Curaçao. Ophthalmic Surg 1980;11:808–10. 15. Allen RC, Bellows AR, Hutchinson BT et al. Filtration surgery in the treatment of neovascular glaucoma. Ophthalmology 1982;89:1181–87. 16. Liesegang TJ. Clinical features and prognosis in Fuchs’ uveitis syndrome. Arch Ophthalmol 1982;100:1622–26.
122
Modulation of wound healing
17. Heuer DK, Gressel MG, Pasrrish RK et al. Trabeculectomy in aphakic eyes. Ophthalmology 1984;91:1045–51. 18. Sherwood MB, Grierson I, Millar L et al. Long-term morphologic effects of antiglaucoma drugs on the conjunctiva and Tenon’s capsule in glaucomatous patients. Ophthalmology 1989;96:327–35. 19. Dahan E, Drusedau MU. Nonpenetrating filtration surgery for glaucoma: control by surgery only. J Cataract Refract Surg 2000;26:695–701. 20. Liebmann JM, Ritch R, Marmor M et al. Initial 5-fluorouracil trabeculectomy in uncomplicated glaucoma. Ophthalmology 1991;98:1036–41. 21. Ophir A, Ticho U. A randomized study of trabeculectomy and subconjunctival administration of fluorouracil in primary glaucomas. Arch Ophthalmol 1992;110:1072–75. 22. Rothman RF, Liebmann JM, Ritch R. Lowdose 5-fluorouracil trabeculectomy as initial surgery in uncomplicated glaucoma: longterm follow-up. Ophthalmology 2000;107:1184–90. 23. Costa VP, Moster MR, Wilson RP et al. Effects of topical mitomycin C on primary trabeculectomies and combined procedures. Br J Ophthalmol 1993;77:693–7. 24. Tahery MM, Lee DA. Review: pharmacologic control of wound healing in glaucoma filtration surgery. J Ocul Pharmacol 1989;5:155–79. 25. Gressel MG, Parrish RK, Folberg R. 5fluorouracil and glaucoma filtering surgery. I. An animal model. Ophthalmology 1984;91:378–83. 26. Egbert PR, Williams AS, Singh K et al. A prospective trial of intraoperative fluorouracil during trabeculectomy in a black population. Am J Ophthalmol 1993;116:612–6. 27. Goldenfeld M, Krupin T, Ruderman JM et al. 5-fluorouracil in initial trabeculectomy: a prospective, randomized, multicenter study. Ophthalmology 1994;101:1024–29.
28. Taniguchi T, Kitazawa Y, Shimizu U. Longterm results of 5-fluorouracil trabeculectomy for primary open-angle glaucoma. Int Ophthalmol 1989;13:145–49. 29. Rockwood EJ, Parrish RK, Heuer DK, Skuta GL. Glaucoma filtering surgery with 5fluorouracil. Ophthalmology 1987;94:1071–78. 30. Ruderman JM, Welch DB, Smith MF et al. A prospective, randomized study of 5fluorouracil and filtration surgery. Trans Am Ophthalmol Soc 1987;85:238–53. 31. Ruderman JM, Welch DB, Smith MF, Shoch DE. A randomized study of 5-fluorouracil and filtration surgery. Am J Ophthalmol 1987;104:218–24. 32. Fluorouracil Filtering Surgery Study Group. Five-year follow-up of the Fluorouracil Filtering Surgery Study. Am J Ophthalmol 1996;121:349–66. 33. Heuer DK, Parrish RK, Gressel MG et al. 5fluorouracil and glaucoma filtering surgery. II. A pilot study. Ophthalmology 1984;91:384–94. 34. Dietze PJ, Feldman RM, Gross RL. Intraoperative application of 5-fluorouracil during trabeculectomy. Ophthalmic Surg 1992;23:662–65. 35. Mora JS, Nguyen N, Iwach AG et al. Trabeculectomy with intraoperative sponge 5fluorouracil. Ophthalmology 1996;103:963–70. 36. Rootman J, Ostry A, Gudauskas G. Pharmacokinetics and metabolism of 5fluorouracil following subconjunctival versus intravenous administration. Can J Ophthalmol 1984;19:187–91. 37. Rootman J, Tisdall J, Gudauskas G et al. Intraocular penetration of subconjunctivally administered 14C-fluorouracil in rabbits. Arch Ophthalmol 1979;97:2375–78. 38. Fantes FE, Heuer DK, Parrish RK et al. Topical fluorouracil: pharmacokinetics in normal rabbit eyes. Arch Ophthalmol 1985;103:953–55. 39. Loane ME, Weinreb RN. Reducing corneal
References
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
toxicity of 5-fluorouracil in the early postoperative period following glaucoma filtering surgery. Aust N Z J Ophthalmol 1991;19:197–202. Weinreb RN. Adjusting the dose of 5fluorouracil after filtration surgery to minimize side effects. Ophthalmology 1987;94:564–70. Knapp A, Heuer DK, Stern GA et al. Serious corneal complications of glaucoma filtering surgery with postoperative 5-fluorouracil. Am J Ophthalmol 1987;103:183–87. Lee DA, Hersh P, Kersten D. Complications of subconjunctival 5-fluorouracil following glaucoma filtering surgery. Ophthalmic Surg 1987;18:187–90. Altan T, Temel A, Bavbek T et al. Hypotonic maculopathy after trabeculectomy with postoperative use of 5-fluorouracil. Ophthalmologica 1994;208:318–20. Parrish R, Minckler D. “Late endophthalmitis”—filtering surgery time bomb? Ophthalmology 1996;103:1167–68. Khaw PT, Doyle JW, Sherwood MB et al. Prolonged localized tissue effects from 5minute exposures to fluorouracil and mitomycin C. Arch Ophthalmol 1993;111:263–67. Nuyts RM, Felten PC, Pels E et al. Histopathologic effects of mitomycin C after trabeculectomy in human glaucomatous eyes with persistent hypotony. Am J Ophthalmol 1994;118:225–37. Chen CW, Huang HT, Bair JS, Lee CC. Trabeculectomy with simultaneous topical application of mitomycin-C in refractory glaucoma. J Ocul Pharmacol 1990;6: 175–82. Palmer SS. Mitomycin as adjunct chemotherapy with trabeculectomy. Ophthalmology 1991;98:317–21. Annen DJ, Sturmer J. Follow-up of a pilot study of trabeculectomy with low dosage mitomycin C (0.2 mg/ml for 1 minute): independent evaluation of a retrospective
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
123
nonrandomized study. Klin Monatsbl Augenheilkd 1995;206:300–02. Kitazawa Y, Suemori-Matsushita H, Yamamoto T, Kawase K. Low-dose and highdose mitomycin trabeculectomy as an initial surgery in primary open-angle glaucoma. Ophthalmology 1993;100:1624–28. Lee JJ, Park KH, Youn DH. The effect of low- and high-dose adjunctive mitomycin C in trabeculectomy. Korean J Ophthalmol 1996;10:42–47. Cheung JC, Wright MM, Murali S, Pederson JE. Intermediate-term outcome of variable dose mitomycin C filtering surgery. Ophthalmology 1997;104:143–49. Mietz H, Krieglstein GK. Three-year followup of trabeculectomies performed with different concentrations of mitomycin-C. Ophthalmic Surg Lasers 1998;29:628–34. Joos KM, Bueche MJ, Palmberg PF et al. One-year follow-up results of combined mitomycin C trabeculectomy and extracapsular cataract extraction. Ophthalmology 1995;102:76–83. Skuta GL, Beeson CC, Higginbotham EJ et al. Intraoperative mitomycin versus postoperative 5-fluorouracil in high-risk glaucoma filtering surgery. Ophthalmology 1992;99:438–44. Khaw PT, Doyle JW, Sherwood MB et al. Effects of intraoperative 5-fluorouracil or mitomycin C on glaucoma filtration surgery in the rabbit. Ophthalmology 1993;100:367–72. Lamping KA, Belkin JK. 5-fluorouracil and mitomycin C in pseudophakic patients. Ophthalmology 1995;102:70–75. Prata Jnr JA, Minckler DS, Baerveldt G et al. Trabeculectomy in pseudophakic patients: postoperative 5-fluorouracil versus intraoperative mitomycin C antiproliferative therapy. Ophthalmic Surg 1995;26:73–77. Karlen ME, Sanchez E, Schnyder CC et al. Deep sclerectomy with collagen implant: medium-term results. Br J Ophthalmol 1999;83:6–11.
124
Modulation of wound healing
60. Hamard P, Plaza L, Kopel J et al. Deep nonpenetrating sclerectomy and open angle glaucoma: intermediate results from the first operated patients. J Fr Ophtalmol 1999;22:25–31. 61. Shaarawy T, Achache T, Schnyder CC, Mermoud A. Deep sclerectomy with mitomycin C in refractory glaucoma patients. In: Mermoud A, Shaarawy T, eds. Proceedings of the First International Congress on Non-penetrating Glaucoma
Surgery, 1st edn. (Lausanne, Switzerland, 2001) 62. Cordeiro MF, Siriwardena D, Chang L, Khaw PT. Wound healing modulation after glaucoma surgery. Curr Opin Ophthalmol 2000;11:121–6. 63. Cordeiro MF, Gay JA, Khaw PT. Human anti-transforming growth factor-beta2 antibody: a new glaucoma anti-scarring agent. Invest Ophthalmol Vis Sci 1999;40:2225–34.
11 Postoperative management of nonpenetrating filtering surgery Tarek Shaarawy
The early postoperative period is a direct consequence of events that took place intraoperatively. Of great influence are the circumstances affecting the success of surgery, such as age and race of the patient; preoperative ocular states such as aphakia, or previous ocular surgery, are also factors of paramount importance. An accurate assessment of the patient’s postoperative condition, as well as correct decision-making, influence whether surgery will be successful or not. It used to be commonly believed that the course of filtering surgery was governed by forces beyond the control of the surgeon, but experience accumulated in postoperative control of the eye has reversed this impression. Much can be done after surgery to attain a successful outcome, such that postoperative management can be as important as preoperative evaluation, and surgical technique.
General postoperative care Because of the inherent instability of aqueous humour dynamics in an eye that has just undergone surgery, patients should not be allowed full physical activity during the early
postoperative period. Patients should also be advised to avoid situations that cause Valsalva’s manoeuvre, such as lifting heavy objects, coughing, and straining. In addition, if comfortable, the patient should sleep with the head slightly elevated and, so far as possible, avoid sleeping on the side of the operated eye. Bumping and rubbing the eye must be avoided at all times. The eye should be examined with the slitlamp frequently so that timely and appropriate adjustments in the aqueous production/runoff balance can be made through the judicious application of postoperative adjuncts.
Postoperative medications Topical medications used in the immediate postoperative period include cycloplegics, corticosteroids, and antibiotics. Cycloplegics have three useful functions. First, paralysis of the ciliary muscle tightens the zonular-irislens diaphragm and maximally deepens the anterior chamber. Second, the blood–aqueous barrier is maintained, thereby limiting proteinaceous exudation and cellular infiltration into the anterior chamber. Third, there is
126
Postoperative management of non-penetrating filtering surgery
a symptomatic relief from postoperative ciliary spasm. Generally, atropine or hyoscine (scopolamine) are used because of their long action. These drugs may also reduce the risk of postoperative malignant glaucoma. Nonetheless, many surgeons prefer not to prescribe a cycloplegic postoperatively in non-penetrating glaucoma surgery. They argue that because the eye is not penetrated, the postoperative inflammation is highly reduced from what would normally be the case with trabeculectomy. Studies comparing flare levels after non-penetrating surgery and trabeculectomy have proven this argument to be true.1 A broad-spectrum antibiotic may be used for 1–3 weeks as a theoretical prophylaxis against bleb infection and endophthalmitis. If there is corneal toxicity from 5-fluorouracil, antibiotics should be continued at a reduced frequency until it has cleared. There is no consensus on the use of prolonged antibiotic treatment; however, some practitioners prescribe them indefinitely once a day, whereas others do not. Corticosteroids, when used in adequate amounts, slow the rate of conjunctival epithelization, angiogenesis, and collagen synthesis. Glaucoma surgery success is improved by the topical use of these agents,2 the influence of which is greatest during the inflammatory phase in the first 3 days after surgery. If administration of corticosteroids is delayed until the fourth or fifth postoperative day, their effect is reduced. The effect of steroids is dose-related, therefore initial high doses should be tapered after a few weeks. There is no documented advantage of systemic versus local steroid administration in glaucoma surgery, although some surgeons use systemic steroids in the more complicated cases. The effect of steroids on anterior-segment inflammation is well documented.3,4 Topical
application every 15 min is more effective than hourly administration or periocular injection. Hourly administration is more effective than four-times-daily dosage. Combined periocular injection with frequent topical administration may be the most effective route. For topical administration, the acetate base appears to be more effective; prednisolone acetate 1.0% is recommended by one source as being the most potent.5 Steroids may raise postoperative intraocular pressure (IOP), even in some eyes with functioning filtering bleb, and their discontinuation may lower IOP.6 A 4–week course of topical steroid therapy is usually sufficient. If used over a longer period, a steroid-induced ocular hypertension may ensue. Non-steroidal anti-inflammatory drugs (NSAIDs) such as indomethacin and diclofenac sodium also reduce inflammation.7 These NSAIDs may be additive to the effect of corticosteroids in restoring the integrity of the blood–aqueous barrier after surgery.5 NSAIDs are also beneficial in controlling postoperative pain. The mechanism of action of NSAIDs on ocular pain is twofold—there is a direct effect on nerve fibres both by reducing sensation,8 and by inhibiting the pain-amplifying effect caused by prostaglandins. Several studies have indicated the value of NSAIDs in the prevention of postsurgical cystoid macular oedema.9 Some investigators argue that NSAIDs are at least as effective if not more so than corticosteroids for treating postoperative traumatic inflammation, and advocate their use as firstline therapy.10 With growing evidence of efficiency and compliability of combination therapies versus single compound drops, the surgeon can now choose from a range of combination antibiotics/corticosteroids and antibiotics/NSAIDs to simplify the treatment regimen, thus ensuring better patient compliance.
Assessment
Antifibrotic agents The use of intraoperative mitomycin C and 5fluorouracil has greatly improved the results of glaucoma surgery in less favourable cases. This category includes young patients, eyes with previous failed filtration surgery, trauma, aphakic and pseudophakic eyes, and eyes with neovascular glaucoma or active iridocyclitis. In addition, 5-fluorouracil is beneficial at the time of initial surgery in phakic eyes with open-angle glaucoma. Adjunctive 5fluorouracil increases success rate, decreases the level of postoperative IOP, and reduces the need for postoperative antiglaucoma medication.11,12 The more potent the antifibrotic agent, the greater the potential for both help and harm. Although 5-fluorouracil is less effective than mitomycin C in developing successful filtering blebs,13 the long-term side-effects of 5fluorouracil also are less serious than those of mitomycin C. Consequently, there is a theoretic justification in matching the type and dose of antifibrotics with the anticipated likelihood of failure of the procedure. The greater the likelihood of failure, the more potent the agent to be used. However, although it is apparent that these agents work more effectively in some patients than in others, it is not yet known how to identify patients in whom 5-fluorouracil or mitomycin C has its greatest effect.
127
anterior chamber, or a needling procedure to break early adhesions impeding outflow, are done at this time. Most surgeons have their own regimen for the frequency of evaluation. The regimen we follow, and that can be used as a guideline, is to examine the patient on the first and third postoperative day, followed by a weekly examination for the first month, monthly examinations for the following 6 months, and, finally, once every 6 months. Naturally, the frequency of evaluation differs greatly with different patients, being more frequent with complicated cases and vice versa.
Assessment The following factors should be assessed: •
Bleb: extent, height, vascularity, limbal cysts, micro cysts, and possible wound leaks (Fig. 11.1)
Frequency of evaluation The first two postoperative days are vital for ascertaining the adequacy and extent of filtration. Any critical modifications, such as injection of a viscoelastic agent to reform a flat
Figure 11.1 Typical filtration bleb after deep sclerectomy, diffuse and slightly elevated.
128
Postoperative management of non-penetrating filtering surgery Figure 11.2 Shallow anterior chamber—Grade 1: peripheral iris touch.
Figure 11.3 Shallow anterior chamber—Grade 2: iris touch up to the pupillary margin.
Figure 11.4 Shallow anterior chamber—Grade 3: lens-cornea appositition.
•
• •
Anterior chamber: hyphaema, hypopyon, and depth if shallow. Grade 1: peripheral iris touch (Fig. 11.2). Grade 2: iris touch up to the papillary margin (Fig. 11.3). Grade 3: lens-cornea apposition (Fig. 11.4) Cornea: clarity, epithelial erosion IOP
• • •
Presence of choroidal detachment or suprachoroidal haemorrhage Optic disc and macula Ultrasonic biomicroscopy (UBM) postoperatively is a useful instrument in assessing the condition of the surgical site.14 (Figs. 11.5 and 11.6)
Needling procedure
129
Assessment also includes detection of potential complications after surgery. Management of these complications is reviewed in a separate chapter.
Needling procedure
Figure 11.5 Ultrasonic biomicroscopy image of the anterior segment after deep sclerectomy with collagen implant. 1 = cornea; 2 = anterior chamber; 3 = iris; 4 = pupil; 5 = trabeculo-Descemet’s membrane; 6 = collagen implant; 7 = suture; 8 = subconjunctival bleb; 9 = localised suprachoroidal drainage.
Figure 11.6 Ultrasonic biomicroscopy image of filtration site in early postoperative period, collagen implant can be identified, as well as single stitch fixing it to the sclera and causing central narrowing of implant.
A major problem in achieving successful control of IOP after non-penetrating glaucoma surgery is the development of a fibrous capsule around the surgical site (Fig. 11.7), which may severely restrict the reabsorption of aqueous humour. One method of improving flow through poorly functioning filtering blebs is needling revision with or without adjunctive antimetabolite (5-fluorouracil or mitomycin C) injections. Needling revision was first described in 194115 and has been reported in several studies to be successful in restoring adequate function to fibrosed or encapsulated filtering blebs in 68% to 93% of cases.16–18 This procedure allows the surgeon to create an opening or openings in the wall of an encapsulated bleb or raise a flattened bleb at the slitlamp or in the operating room via
Figure 11.7 Encysted bleb after deep sclerectomy.
130
Postoperative management of non-penetrating filtering surgery
Figure 11.9 Needling procedure in maximally exposed surgical site.
Figure 11.8 Encysted bleb.
subconjunctival manipulation with a smallgauge needle. The needling technique involves instillation of topical anaesthetic drops and apraclonidine 0.5% or 1% (for vasoconstriction), a povidoneiodine solution applied to the eyelids and periorbital area, and a lid speculum placed in the eye. The needling can take place under the naked eye, or preferably, under the slit-lamp. The patient is instructed to look in the direction that would ensure maximum exposure to the surgery site—downwards in the case of a 12 o’clock surgical site (Fig. 11.8). A 30-gauge needle should be mounted on a 1 mL syringe. The syringe can be left straight or bent with a blade breaker to a “bent bayonet” shape;19 this shape may make needling easier for the physician. The needle should be introduced beneath the conjunctiva near the surgical site. A small
amount of local anaesthetic can be injected subconjunctivally (0.1 mL Xylocaine) to ensure a painless procedure (Fig. 11.9). The needle should then be advanced a short distance (about 3–5 mm) subconjunctivally with care taken to avoid perforation of blood vessels or the conjunctiva overlying the needle track. The tip of the needle should be carefully moved to penetrate the fibrous capsule or Tenon’s cyst, and several tears should be made in the capsule. The needle is then withdrawn in the same track. Frequently the clinical appearance of the bleb can be seen to change over the needling site, either enlarging in cases of fibrosis or becoming less tense in cases of encapsulation (Fig. 11.10). Slit-lamp examination and tonometry as well as Seidel testing should usually be done 15 min after the procedure. Needling can be repeated later, if deemed necessary. Some eyes could undergo multiple needling procedures when an initial improvement is followed by recurrence of fibrosis. Up to seven separate needlings are reported in published work.19
Needling procedure
Figure 11.10 Bleb is more diffuse and less tense after needling.
131
Antimetabolites could be injected during the needling procedure. Some surgeons prefer to inject antimetabolites in the quadrant opposite to the site of surgery. I prefer to inject antimetabolites, if needed, at the same site of the needling, which, in my experience, is a safe and mostly effective procedure. Common complications include conjunctival haemorrhage and transient wound leak. Ocular hypotony may occur, and, although rare, choroidal effusion has been reported.20 Aqueous humor may occasionally leak from the needle-entry site for several days but does not usually require any intervention. The main advantages of the procedure include its ease, minimal anaesthesia, and few complications.17 Figure 11.11 Intraocular pressure before and after goniopuncture.24
132
Postoperative management of non-penetrating filtering surgery
Nd:YAG goniopuncture Clinical trials have clearly shown the importance of Nd:YAG goniopuncture as an adjunctive to non-penetrating glaucoma surgery. Karlen et al21 in their medium term results of deep sclerectomy with collagen implant (DSCI) published in 1999 (mean follow-up 17.8 [SD 8.7] months) reported 41% of patients in their series had goniopunctures, while Shaarawy et al, 22 in the long-term study published in 2000 (mean follow-up 43.2 [SD 14.3] months), reported 47% of patients in their series needed goniopunctures. Shaarawy et al 22 performed goniopunctures with the Nd:YAG laser on 48 patients. The mean time between laser and DSCI was 13.5 (SD 13.0) months, the mean IOP before goniopuncture was 20.6 (SD 6) mmHg and the mean IOP after goniopuncture was 10.7 (SD 6.3) mmHg. This difference was statistically significant (p < 0.001). The success rate was 91.6%. Kozlov et al,23 also reported early and late postoperative ocular hypertension, and also performed goniopunctures with the Nd:YAG laser, and they achieved a success rate ranging from 85% to 94%. Mermoud et al, 24 used the Nd:YAG laser as well, and reported a success rate of 83% (Fig. 11.11). Demailly et al,25 in their series of 219 DSCI procedures, performed 90 goniopunctures with the Nd:YAG laser, of which 61 (68%) were successful. Goniopuncture is performed shortly after DSCI if there is insufficient percolation of aqueous humour at the trabeculo-Descemet’s membrane, probably due to the lack of surgical dissection. Goniopunctures may be required at a later time (more than 9 months after initial surgery), if low filtration occurs as a result of fibrosis of the trabeculo-Descemet’s
membrane, because goniopuncture results in increased filtration of aqueous humour and decreased IOP.
Procedure for goniopuncture With a gonioscopy contact lens (Fig. 11.12), the aiming beam is focused on the semi-transparent trabeculo-Descemet’s membrane (Fig. 11.13). In the free-running Q-switched mode, with a power of 4–5 mJ, two to 15 shots are applied (Fig. 11.14). This should result in the formation of microscopic holes through the
Figure 11.12 Haag-Streit CGAL goniolens, used to visualize the trabeculo-Descemet’s membrane.
Procedure for goniopuncture
Figure 11.13 Trabeculo-Descemet’s membrane visualized by goniolens.
133
trabeculo-Descemet’s membrane, allowing a direct passage of aqueous humor from the anterior chamber to the subconjunctival space. The success of goniopuncture depends mainly on the thickness of the trabeculo-Descemet’s membrane, hence the importance of sufficiently deep intraoperative dissection. The exact mechanisms by which goniopunctures reduce IOP are not yet fully understood. UBM studies are currently being conducted in an attempt to discover and explain these mechanisms (Figs. 11.15a and 11.15b). By opening the trabeculo-Descemet’s membrane, however, goniopuncture transforms a non-penetrating filtration procedure
Figure 11.14 Diagram showing laser beam creating microscopic hole in the trabeculoDescemet’s membrane.
134
Postoperative management of non-penetrating filtering surgery
(a)
(b)
Figure 11.15 (a) Ultrasound biomicroscopy of trabeculo-Descemet’s membrane before goniopuncture. (b) After goniopuncture, note the hole in the membrane and the increase in the size of the intrascleral and the subconjunctival bleb.
into a microperforating one. It is argued that since goniopuncture is commonly performed after deep sclerectomy, the surgery is in fact a perforating procedure done in two stages (47% of deep sclerectomy patients in the long-term study had goniopunctures). The fact remains that prevention of perforation intraoperatively and in the early postoperative period seems to lower the incidence of complications. Although the potential risk of late bleb-related endophthalmitis may be increased after goniopuncture, no such case has been reported.
Digital pressure Although digital pressure, also known as ocular massage, has been shown to be a useful technique in salvaging and retaining failing
filtering blebs after trabeculectomy, it is strictly contraindicated in non-penetrating glaucoma surgery. Digital pressure can cause IOP to rise, rupturing the delicate trabeculo-Descemet’s membrane, thus potentially causing a rapid drop in IOP. Rupture of trabeculo-Descemet’s membrane can also allow the iris to prolapse into the surgically created scleral space, blocking filtration, and increasing IOP.
Reoperations When reoperations are necessary, the original surgical site usually has substantial conjunctival scarring and increased vascularity. It is often easier and more successful to reoperate at an adjacent area with less fibrosis; if performed at the same operative site, a fornix-
References based flap is technically easier. If a limbusbased flap is desired, sharp dissection is preferable to scissors dissection. It is easy to buttonhole the adherent conjunctiva with scissors. The extent of the conjunctival scarring can often be delineated by gently lifting the conjunctiva with a non-toothed forceps. Another technique is to raise the conjunctiva by injecting balanced salt solution through a sharp 30-gauge needle away from the surgical region. Because postoperative scarring is even more prevalent with reoperations, an antifibrotic agent is recommended. In addition, large fibrotic-thickened sheets may demand a tenonectomy. This issue is discussed in detail in Chapter 14.
Neuroprotective strategies The concept of neuroprotection has become the new frontier of glaucoma research in the past few years. Neuroprotection can be defined as: protection of cell bodies of axons that are damaged by a primary stimulus (such as IOP) and protection of adjacent axons from noxious agents released as a result of primary damage (prevention of secondary damage). A number of potential strategies, which might interrupt the process of excitotoxicity secondary to excess concentrations of glutamate, have been contemplated. Included within these potential strategies are: glutamate inhibition, NMDA (N-methyl-D-aspartate) receptor blockade, exogenously applied neurotrophins, open-channel blockers, antioxidants, protease inhibitors, and gene therapy. A host of potential agents are currently under clinical investigation. Many of these agents are the result of work done on other diseases of the central nervous system such as Parkinson’s disease and amyotrophic lateral sclerosis. The
135
current thrust of therapeutic regimens is aimed at either pre-NMDA mechanisms or at the NMDA binding site. Pre-NMDA agents include riluzole and lifarazine, both of which act to attenuate the release of glutamate. Other current therapeutic strategies are aimed at blocking the binding site on the NMDA receptor or at blocking the open channel, which results from NMDA stimulation. Agents being investigated as NMDA-receptor antagonists include: CGS19755 and felbamate. Both magnesium and memantine are being investigated as open-channel blockers. In addition to these agents, gene therapy has been proposed. Since the process of apoptosis is thought to be a result of the checks and balances between survival genes and suicide genes, the alteration of the genetic material through gene therapy may allow for the exaggerated expression of the survival gene and the inhibition of the apoptotic process. Although gene therapy is still in its infancy, it is not inconceivable that glaucoma patients may benefit from this technology. Other potential therapies to inhibit apoptosis include smallmolecule drugs designed to turn off expression of apoptosis-related genes and injectible molecules targeted towards modulators of apoptosis.
Conclusion The goal is still the elimination of the devastating visual loss caused by glaucoma and postoperative management of glaucoma remains an important part of the surgeon’s endeavour.
References 1. Chiou AG, Mermoud A, Jewelewicz DA. Post-operative inflammation following deep
136
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Postoperative management of non-penetrating filtering surgery
sclerectomy with collagen implant versus standard trabeculectomy. Graefes Arch Clin Exp Ophthalmol 1998;236: 593–96. Roth SM, Spaeth GL, Starita RJ et al. The effects of postoperative corticosteroids on trabeculectomy and the clinical course of glaucoma: five-year follow-up study. Ophthalmic Surg 1991;22:724–29. Dunne JA, Travers JP. Topical steroids in anterior uveitis. Trans Ophthalmol Soc UK 1979;99:481–84. Dunne JA, Travers JP. Double-blind clinical trial of topical steroids in anterior uveitis. Br J Ophthalmol 1979;63:762–67. Leibowitz HM, Kupferman A. Drug interaction in the eye: concurrent corticosteroid-antibiotic therapy for inflammatory keratitis. Arch Ophthalmol 1977;95:682–85. Wilensky JT, Snyder D, Gieser D. Steroidinduced ocular hypertension in patients with filtering blebs. Ophthalmology 1980;87:240–44. Gwin TD, Stewart WC, Gwynn DR. Filtration surgery in rabbits treated with diclofenac or prednisolone acetate. Ophthalmic Surg 1994;25:245–50. Sun R, Gimbel HV. Effects of topical ketorolac and diclofenac on normal corneal sensation. J Refract Surg 1997;13:158–61. IDS Group. Efficacy of diclofenac eyedrops in preventing postoperative inflammation and long-term cystoid macular edema: Italian Diclofenac Study Group. J Cataract Refract Surg 1997;23:1183–89. Othenin-Girard P, Tritten JJ, Pittet N et al. Dexamethasone versus diclofenac sodium eyedrops to treat inflammation after cataract surgery. J Cataract Refract Surg 1994;20:9–12. Ophir A, Ticho U. Encapsulated filtering bleb and subconjunctival 5-fluorouracil. Ophthalmic Surg 1992;23:339–41. Wilson RP, Steinmann WC. Use of
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
trabeculectomy with postoperative 5fluorouracil in patients requiring extremely low intraocular pressure levels to limit further glaucoma progression. Ophthalmology 1991;98:1047–52. Skuta GL, Beeson CC, Higginbotham EJ et al. Intraoperative mitomycin versus postoperative 5-fluorouracil in high- risk glaucoma filtering surgery. Ophthalmology 1992;99:438–44. Chiou, AG, Mermoud A, Underdahl JP, Schnyder CC. An ultrasound biomicroscopic study of eyes after deep sclerectomy with collagen implant. Ophthalmology 1998;105:746–50. Ferrer H. Conjunctival dialysis in the treatment of glaucoma recurrent after sclerectomy. Am J Ophthalmol 1941;24:788–90. Pederson JE, Smith SG. Surgical management of encapsulated filtering blebs. Ophthalmology 1985;92:955–58. Shin DH, Juzych MS, Khatana AK et al. Needling revision of failed filtering blebs with adjunctive 5-fluorouracil. Ophthalmic Surg 1993;24:242–48. Ewing RH, Stamper RL. Needle revision with and without 5-fluorouracil for the treatment of failed filtering blebs. Am J Ophthalmol 1990;110:254–59. Chen PP, Palmberg PF. Needling revision of glaucoma drainage device filtering blebs. Ophthalmology 1997;104:1004–10. Published erratum: Ophthalmology 1997;104:1532. Potash SD, Ritch R, Liebmann J. Ocular hypotony and choroidal effusion following bleb needling. Ophthalmic Surg 1993;24:279–80. Karlen ME, Sanchez E, Schnyder CC et al. Deep sclerectomy with collagen implant: medium term results. Br J Ophthalmol 1999;83:6–11. Shaarawy T, Karlen ME, Sanchez E et al. Long term results of deep sclerectomy with collagen implant. Acta Ophthalmol Scand 2000;78:323.
References 23. Kozlov V, Bagrov SN, Anisimova SY et al. Deep sclerectomy with collagen. Eye Microsurg 1990;3:44–46. 24. Mermoud A, Karlen ME, Schnyder CC et al. Nd:Yag goniopuncture after deep sclerectomy with collagen implant. Ophthalmic Surg Lasers 1999;30:120–25.
137
25. Demailly P, Lavat P, Kretz G et al. Nonpenetrating deep sclerectomy (NPDS) with or without collagen device (CD) in primary open-angle glaucoma: middle-term retrospective study. Int Ophthalmol 1996;20:131–40.
12 Complications and reoperations André Mermoud and Emilie Ravinet
Even if the main advantage of non-penetrating filtering surgeries is the low rate of complications, glaucoma surgeons should be aware of intraoperative as well as early and late postoperative complications. A thorough understanding of each specific condition helps in making
the diagnosis and in subsequent management. In case of complete failure of a nonpenetrating filtering surgery, the surgeon may have to contemplate reoperation. This particular situation will also be discussed briefly and several options proposed.
Figure 12.1 Schematic representation of perforation at level of anterior trabeculum or Descemet’s membrane.
140
Complications and reoperations Figure 12.2 Schematic representation of perforation or tear at level of Schwalbe’s line.
Peroperative or intraoperative complications Perforation of trabeculoDescemet’s membrane Perforation may frequently occur in the learning phase of the surgery. In our own experience, perforation of the trabeculo-Descemet’s membrane (TDM) arose in 30% of the ten initial operations, whereas it was encountered in only 3% of the subsequent 100 procedures.1 The TDM is very thin and fragile. The perforation is almost always located in the thinnest portion, namely the anterior trabeculum and Descemet’s membrane (Fig. 12.1). The junction between these two anatomical structures is probably the weakest point and corresponds to Schwalbe’s line on gonioscopy (Fig. 12.2). A perforation at this level will usually lead to the formation of a long tear, which is usually followed by immediate iris prolapse (Fig. 12.3).
Figure 12.3 Tear between Descemet’s membrane and anterior trabeculum at level of Schwalbe’s line (iris is prolapsed).
Peroperative or intraoperative complications
(a)
141
(b) Figure 12.4 (a) Small perforation with deep anterior chamber. (b) Small perforation with shallow or flat anterior chamber without iris prolapse. (c) Small or large perforation with shallow or flat anterior chamber and iris prolapse.
(c) Appropriate and specific management should be given to perforations in the TDM according to their size, site, and associated subsequent complications. Small perforation and deep anterior chamber (Fig. 12.4a) is mainly observed with a small Descemet’s perforation, usually done
with the knife during the anterior deep dissection. No particular attention is given to the perforation and the surgery can be continued normally. Small or large perforation and shallow or flat anterior chamber with no iris prolapse (Fig. 12.4b) usually occurs with a large
142
Complications and reoperations
Descemet’s membrane hole or an anterior trabeculum hole. To avoid subsequent iris prolapse or peripheral anterior synaechia formations, viscoelastic should be injected into the anterior chamber under the trabeculoDescemet’s window to reposition the iris back. The smallest amount of low molecular weight hyaluronic acid should be used to avoid a postoperative ocular pressure spike. In addition, an implant resting on the perforation site may be used; by swelling up it should help to tamponade the hole. The superficial scleral flap should also be tightly sutured with 6–8 10/0 nylon sutures. Small or lage perforation, anterior chamber either formed or not, and prolapsed iris (Fig. 12.4c) calls for a peripheral iridectomy, because it is technically difficult to reposition the prolapsed iris (giving greater risk of surgical failure). Provided the perforation is small, the rest of the procedure should be as described in the previous paragraph. If the perforation is large or if the anterior chamber stays shallow or flat, an intracameral injection of low molecular weight hyaluronic acid should be done. The superficial scleral flap should be tightly closed and the viscoelastic should be injected as well in the created scleral space to increase the outflow resistance (Fig. 12.5). Any perforation of the TDM during deep sclerectomy transforms a non-penetrating filtering surgery into a penetrating one. Because the scleral space left after deep sclerectomy decreases the aqueous humor outflow resistance, a very tight superficial scleral-flap closure is then of great importance. This operation can be compared to a trabeculectomy with an additional deep sclerectomy.2 In a series of 20 patients who underwent an intraoperative perforation of the TDM, Sanchez et al 2 reported an increased incidence of postoperative complications such as flat
Figure 12.5 Tight superficial scleral flap closure after trabeculo-Descemet’s membrane perforation with iris prolapse and peripheral iridectomy.
anterior chamber, hypotony, choroidal detachment, inflammation, and hyphema. In this series, however, the superficial scleral flap was not tightly closed. In our experience, by closing the scleral flap with six to eight 10/0 nylon sutures, especially if done in combination with viscoelastics under the scleral flap, hypotony-related complications can be avoided in most of the cases.
Malignant glaucoma Malignant glaucoma may occur after any type of intraocular eye surgery. In non-penetrating filtering surgeries, this complication should, theoretically, be less frequent because the eye is not opened. However, malignant glaucoma has even been reported after laser therapy such as iridotomies or argon-laser trabeculoplasty and can also occur spontaneously. Malignant
Peroperative or intraoperative complications
(a)
143
(b)
Figure 12.6 (a) An ultrasound biomicroscopic image of a malignant glaucoma due to anterior ciliary body rotation with ciliolenticular block. (b) Resolution of the block after cycloplegia treatment.
glaucoma is characterized by a shallow anterior chamber and an elevated intraocular pressure (IOP); it must be differentiated from pupillary block and suprachoroidal hemorrhage. This condition may be secondary to an aqueous-humor misdirection into the vitreous cavity. A more frequent mechanism is the anterior ciliary body rotation, inducing a lensciliary block with subsequent entrapment of aqueous humor in the posterior chamber followed by a forward displacement of the lens-iris diaphragm. In our experience we have had one patient with narrow-angle glaucoma who, after deep sclerectomy with collagen implant, developed such a complication.3 An ultrasound biomicroscopy (UBM) image showed a ciliary block mechanism and cycloplegia treatment resolved the complication (Fig. 12.6a and 12.6b). Therefore, prophylactic cycloplegia and mydriatic therapy should be administered intraoperatively and postop-
eratively in all patients at risk of developing malignant glaucoma.
Hemorrhages Intraoperative hemorrhage may occur either at the conjunctivoscleral level or in the uveal tissue and can also be seen secondary to blood reflux at Schlemm’s canal ostia. Both blood and cautery induced thermal injury may stimulate fibroblast proliferation and be harmful to the filtration bleb survival. During the conjunctival and then the scleral dissection, major bleeding should be treated with light cauterization. The wet-field cautery is probably the safest tool to be used and allows minimal scleral burns. Strong topical vasoconstrictors may be useful to restrain the use of cautery.
144
Complications and reoperations
Figure 12.7 Decompression retinopathy. The intraretinal hemorrhages are located within the deep retinal layer and are caused by to rapid intraocular pressure decrease.
Intraocular bleeding is rare and may originate from any uveal or retinal vessels; it is usually secondary to the drop in IOP. The nonpenetration technique should prevent this complication because the IOP drop is slower and more progressive than it is during penetrating surgery. However, in patients with fragile vessels, arterial hypertension and/or anticoagulation therapy, especially if preoperative IOP is exceedingly elevated, intraocular bleeding may still occur. While small iris hemorrhages are insignificant, major retinal or choroidal bleeding may be sight threatening (Fig. 12.7). Blood reflux from Schlemm’s canal ostia may occur when the episcleral venous pressure is higher than the IOP. This complication makes dissection of TDM difficult. Blood reflux may be stopped in two ways: either by injecting high molecular weight viscoelastic into both Schlemm’s canal ostia using the Grieshaber canula (Fig. 12.8), or by injecting
(a)
(b) Figure 12.8 (a, b) High molecular weight hyaluronic acid injected into Schlemm’s canal for blood reflux, using the Grieshaber viscocanalostomy canula.
a balanced salt solution into the anterior chamber, thereby increasing the IOP. This reflux may also occur after surgery during Valsalva episodes. The phenomenon may be observed on gonioscopy through the Descemet’s window.
Early postoperative complications
Early postoperative complications
145
Hyphema
Early postoperative complications are uncommon after non-penetrating filtering surgeries compared to those after trabeculectomy.4 Table 12.1 shows a comparison of the most frequent postoperative complications between deep sclerectomy and trabeculectomy.5
Wound leak Wound leak or positive Seidel test occurs with the same frequency after penetrating and nonpenetrating filtering surgeries; it is usually seen after insufficient wound closure. The leak often stops by itself after discontinuation of steroid therapy. In rare cases a second wound closure has to be done under topical anesthesia.
Hyphema is a rare complication after nonpenetrating filtering procedures. The blood in the anterior chamber may originate either from a rupture of a small iris vessel such as in Fuchs heterochromia (Amsler sign), or from a leak of red blood cells through the TDM. Usually the erythrocytes are diffusely present within the anterior chamber. Leveled hyphema is very rare. No particular treatment is needed in this situation, and the anterior chamber clears itself within a few days with the routinely prescribed topical anti-inflammatory treatment.
Inflammation Because the anterior chamber is not opened during non-penetrating filtering surgery, the degree of inflammation is low compared with
Table 12.1 Early postoperative complications of 44 patients who underwent deep sclerectomy with collagen implant (DSCI), compared with a matched group of 44 patients who underwent trabeculectomy.5
Perforation of the trabeculo–corneal membrane Seidel Hyphema Flat anterior chamber Inflammation Choroidal detachment Macular edema Postoperative ocular hypotension Postoperative ocular hypertension Dellen *ns, not significant
DSCI n = 44
%
Trabeculectomy % n = 44
p
5 5 1 0 0 2 1 6 0 1
10 11 2 0 0 5 2 14 0 2
– 3 15 8 10 9 2 16 2 2
– ns 0.00011 0.006 0.001 0.05 ns 0.03 ns ns
– 7 344 18 231 20 5 36 5 5
146
Complications and reoperations Figure 12.9 Comparison of anteriorchamber flare measurements in the postoperative period in patients who underwent deep sclerectomy or trabeculectomy.
that after any kind of penetrating surgeries. In a prospective and randomized study, Chiou et al6 measured the anterior-chamber flare with a laser flare-cell meter in patients who underwent either deep sclerectomy with collagen implant or trabeculectomy.6 They reported that in the deep sclerectomy group the anterior-chamber inflammation was mild in the initial postoperative days and that the mean flare values corresponded to preoperative values after the first postoperative week. In the trabeculectomy group, the anteriorchamber inflammation was more severe and prolonged (Fig. 12.9). This difference can be attributed to the opening of the anterior chamber as well as the iridectomy done in the trabeculectomy group. Surgical anteriorchamber penetration and iridectomy induce a breakdown of the blood–aqueous humor barrier together with the release of inflammatory mediators in the aqueous humor. These mediators as well as inflammatory cells stimulate fibroblast proliferation in the surgical site
and lead to filtration failure. Even if the postoperative inflammation rate is low after non-penetrating filtering surgery, anti-inflammatory treatment should still be given routinely for at least 3 months after the operation. Inflammation may be more pronounced in patients with pseudoexfoliative glaucoma, pigmentary glaucoma, and glaucoma associated with uveitis or trauma. When nonpenetrating filtering surgeries are combined with cataract surgery the rate of inflammation is higher. However, less inflammation after phaco-deep sclerectomy than after phaco-trabeculectomy has been reported.7 Again, this difference may be due to the iridectomy done during the trabeculectomy operation. In combined surgeries, especially when iris retractors have been used, the routinely given anti-inflammatory regimen (e.g. topical diclofenac and antibiotics) might not be sufficient and should be changed accordingly.
Early postoperative complications
Figure 12.10 Heterochromia due to high flare in the anterior chamber after a tiny choroidal detachment.
147
Choroidal detachment
Figure 12.11 Ultrasound biomicroscopy image of subtle peripheral choroidal and ciliary body detachment.
So-called choroidal detachment corresponds to fluid accumulation between the sclera and the choroid; 8–10 it occurs secondarily to severe and prolonged hypotony or to compression of the vortex veins. After filtering surgeries, choroidal detachment is usually seen when the preoperative IOP is elevated to more than 40 mmHg and/or when the postoperative IOP drops to less than 3 or 4 mmHg. Another risk factor for this complication is hypermetropia or small eyes with thick sclera. In these patients, drainage of vortex veins is often insufficient, and the surgery may compromise it further. Choroidal detachment may be followed by defects in the peripheral visual field and may even produce an amaurosis when the detachment is complete (kissing choroidals). This situation may be painful for the patient and may require additional antiinflammatory therapy and analgesics. There is a concomittant variable uveal blood–aqueous barrier breakdown that, in rare cases, can be so severe as to be associated with heterochro-
mia (Fig. 12.10) As discussed above under postoperative inflammation, the elevated flare may stimulate fibroblast proliferation in the surgical site and threaten filtration. Choroidal detachment can also induce a ciliary body detachment with a subsequent decreased aqueous humor production, leading to an increased ocular hypotony and the risk of a more severe choroidal detachment. By increasing the IOP in patients with this complication, the vicious circle can be broken. After non-penetrating filtering surgeries, choroidal detatchment is rare and usually not significant. In different series this complication has been reported to occur in 2–5% after nonperforating filtering surgery1,5,11–21 whereas after trabeculectomy it was encountered in up to 20%.8 In our experience, after deep sclerectomy, with an UBM assessment, it was possible to diagnose very subtle peripheral choroidal and ciliary body detachments in almost 40% of the patients (Fig. 12.11).22 This
148
Complications and reoperations
may not be solely attributable to the low postoperative IOP obtained after nonpenetrating filtering surgeries but also to direct aqueous humor passage from the scleral space to the suprachoroidal space through the very thin scleral layer remaining in place after the deep sclerectomy. This phenomenon may be prolonged for months or even years and may partially explain the IOP reduction mechanism after non-penetrating filtering surgeries. No specific treatment is needed for small choroidal detachment because it resolves spontaneously within a few days. When the detachment is substantial, the patient should be treated with cycloplegic and anti-inflammatory agents. Drainage of the suprachoroidal space is only required when the detachment is so severe as to carry the risk of subsequent retinal detachment and/or corneal damage. The surgical drainage should preferably be done in the inferior nasal or temporal quadrant, 8–10 mm away from the limbus and with an anterior-chamber maintainer to compensate for the loss of ocular volume. Suprachoroidal fluid should be removed as completely as possible. The sclerotomy can be left open to allow residual fluid to drain.
Hypotony The mean IOP after non-penetrating filtering surgeries has been reported to be 5 ± 4.0 mmHg on the first postoperative day.1,5 Thus 50% of the patients present with an early ocular hypotony for some days. If short-lived and not associated with any secondary complication, ocular hypotony should not be regarded as a worrying complication. From our experience, even with an IOP of 0–2 mmHg on the first postoperative day, a progressive increase in IOP occurs in all patients in the next few days, without specific treatment. In fact, early
hypotony without any perforation is an excellent indicator of good surgical dissection. According to Vaudaux’s and Mermoud’s experimental work on the ouflow resistance of the TDM, the immediate postoperative IOP should be very low because the outflow facility, after deep sclerectomy, is about 130 times higher.23 Yet at the same time the TDM offers enough resistance to avoid anterior-chamber collapse; no flat anterior chamber has yet been clinically reported after non-penetrating filtering surgery.
Ocular hypertension Ocular hypertension, secondary to the superficial scleral flap being too tightly closed, is regularly observed after trabeculectomy and needs either laser suture lysis and/or ocular massage, unless releasable sutures have been used.4 Because the main site of postoperative aqueous humor outflow resistance after nonpenetrating filtering surgery is located at the TDM level, this complication should not occur if the dissection of the membrane has been done properly. Early postoperative IOP spikes can be due to •
•
•
• •
Insufficient surgical dissection, which is most common after non-penetrating filtering surgeries by surgeons lacking experience Hemorrhage in the scleral bed, which usually spontaneously resorbs within a few days Excess of viscoelastic remaining in the anterior-chamber, mainly after combined surgeries or anterior-chamber reformation following perforation of the TDM Malignant glaucoma Postoperative rupture of the TDM with iris prolapse, secondary to increased IOP from eye rubbing, Valsalva’s manoeuver, etc.
Early postoperative complications
Figure 12.12 Dellen formation after deep sclerectomy with collagen implant with voluminous subconjunctival bleb.
Figure 12.13 Blebitis after a deep sclerectomy in immunodeficient patient treated for pulmonary tuberculosis.
•
Infection and blebitis
•
Peripheral anterior synechia formation at the site of the filtering window, often secondary to a peroperative microperforation Steroid response within the first postoperative weeks.
Overall, IOP spikes are rare postoperative complications and should be treated according to each specific cause.
Dellen Because the superficial scleral flap is loosely closed, the subconjunctival filtering bleb is initially usually larger after non-penetrating filtering surgeries. Subsequently it rapidly decreases in size and becomes more diffuse. Should the bleb remain prominent for an extended period, or should the patient be teardeficient, a Dellen may be observed at the limbus (Fig. 12.12). With adequate lubrication this complication can be easily treated.
149
This complication is rare but may occur secondary to peroperative bacterial contamination and may be of great concern in immunocompromised patients. A broadspectrum topical antibiotic is routinely prescribed postoperatively to prevent this complication. In our own experience in 2000 operations we saw a blebitis only once, in a 75-year-old patient treated for pulmonary tuberculosis (Fig. 12.13). Intensive hourly topical antibiotics were sufficient to treat this complication. So far we have never had a case of endophthalmitis. The TDM probably acts as a barrier preventing bacterial organisms from reaching the anterior chamber.
Shallow or flat anterior chamber A completely flat anterior chamber has never been reported after non-penetrating filtering
150
Complications and reoperations rhages (Fig. 12.14). This complication usually has no long-term deleterious effects unless the bleeding occurs at the foveola, and it normally resolves within a few days.
Decreased visual acuity
Figure 12.14 Decompression retinopathy observed on first postoperative day in 30-year-old patient with traumatic glaucoma and preoperative IOP of 45 mmHg.
surgeries. This may explain why no surgically induced cataract has been reported after nonpenetrating filtering surgeries. Shallowing of the anterior chamber may be seen in patients presenting a large drop in IOP. In this situation, the anterior-chamber shallowing is usually secondary to choroidal detachment and spontaneously resolves with resorption of the choroidals. In the presence of a shallow anterior chamber, the surgeon should rule out perforation of the TDM, conjunctival wound leak, suprachoroidal hemorrhage, pupillary block, and malignant glaucoma.
Decompression retinopathy Decompression retinopathy is a rare complication after any filtering procedure and occurs mainly in young patients with a very high preoperative IOP. The rapid peroperative decrease in IOP is probably responsible for the multiple intraretinal, small, round hemor-
Because there is no postoperative cycloplegia, no basal iridectomy, and very little anteriorchamber inflammation, the drop in visual acuity is mild after non-penetrating filtering procedures. The central as well as the peripheral visual acuities are usually decreased by a mean of two Snellen lines for the first postoperative week. This is probably due to the readaptation of the retina and choroid blood flow to the new pressure gradient. Astigmatismrelated changes might also account for the decrease in visual acuity. In several studies, investigators have reported that the mean visual acuity usually returned to the preoperative value 7 days postoperatively (Fig. 12.15).1,5,11–21
Cataract formation No surgically induced cataract has yet been reported after non-penetrating filtering surgeries. This is probably due to the stability of the anterior-chamber depth and the low degree of inflammation.
Implant displacement Almost any type of intrascleral implant used to maintain the scleral space may migrate if they are not sutured peroperatively. Even if a subconjunctival migration does not have a direct negative effect, intracameral displacement due to rupture of the TDM may require reoperation for implant removal.
Late postoperative complications
151
Figure 12.15 Graph of mean visual acuity of patients who underwent deep sclerectomy with collagen implant.
Late postoperative complications
operated tissues and that the surgical procedure as such does not influence this process.
Unlike immediate postoperative complications, late postoperative complications occur with the same frequency in penetrating and nonpenetrating filtering surgeries. This may be explained by the fact that late complications are often related to excessive scarring of the
Fibrosis of conjunctival bleb Mermoud et al5 showed that there was more bleb fibrosis and flattening in the deep sclerectomy group than in the trabeculectomy group, although the difference was not significant (Table 12.2). This may be explained by the
Table 12.2 Late complications of 44 Patients who underwent deep sclerectomy with collagen implant (DSCI), compared with a matched group of 44 patients who underwent trabeculectomy.5
Bleb fibrosis Bleb encapsulation Cataract formation: – Surgery related cataract – Progression of preexisting cataract – Total *ns, not significant.
DSCI n = 44
%
Trabeculectomy % n = 44
p
10 5
23 11
8 5
18 11
ns ns
0 4 4
0 9 9
6 5 11
14 11 25
0.026 ns 0.04
152
Complications and reoperations
Figure 12.16 Ultrasound biomicroscopic image of a welldefined intrascleral filtering bleb with no or very subtle subconjunctival bleb.
Figure 12.17 Ultrasound biomicroscopic image of an encysted filtering bleb.
presence of an intrascleral filtering bleb, which reduces the need for subconjunctival filtration (Fig. 12.16). When the subconjunctival filtering bleb is flat but the IOP is adequate, no treatment should be given. However, when the bleb is flat and/or vascularized with an elevated IOP, an antimetabolite should be used to reduce the scarring process. Either 5-fluorouracil or mitomycin C may be injected subconjunctivally. Increased anti-inflammatory topical therapy should also be prescribed; for detailed technique readers should refer to Chapter 13 on postoperative management.
In non-penetrating filtering surgeries, the increased pressure inside the pseudocyst may force aqueous humor into the anterior portion of the surgical dissection and induce a localized or more extensive Descemet’s detachment. Diagnosis and management of this particular complication will be detailed later in this chapter. Treatment of an encysted bleb consists of medical therapy with anti-inflammatory and antiglaucomatous agents and/or in needling with or without injection of antimetabolites.
Encysted bleb
Increased intraocular pressure
Encysted bleb occurs when a fibrotic wall entraps the aqueous humor into a cyst-like structure in the subconjunctival space. The aqueous humor filters adequately through the TDM but cannot flow further. As a consequence, the IOP usually rises and may stimulate further thickening of the cyst wall (Fig. 12.17).
Late increased IOP may be caused by steroid response or fibrosis of any of the following: the TDM, the intrascleral space, the superficial scleral flap, or the conjunctival bleb. A UBM assessment is of great help in determining the site of aqueous obstruction. To exclude a decreased flow through the TDM, an Nd:YAG
Late postoperative complications
153
reported to be fibrosis of the superficial scleral flap. Usually the intrascleral space remains patent and the filtration is restored after superficial scleral flap reopening.
Thin bleb
Figure 12.18 Subconjunctival ischemic bleb after use of preoperative antimetabolites.
When antimetabolites have been used at the time of surgery or in the early postoperative period, the conjunctival bleb may become thin and ischemic (Fig. 12.18). This condition may, as is the case with trabeculectomy, lead to late chronic ocular hypotony and related complications as well as increase the risk of late blebrelated ocular infections.
Late rupture of the trabeculoDescemet’s membrane
laser goniopuncture should be tried first. This procedure is described in detail in Chapter 13. If the IOP after the goniopuncture remains uncontrolled, medical antiglaucoma therapy should be reintroduced; a revision of the filtering surgery may also be needed. After reoperated failed deep sclerectomies, the main cause of aqueous humor drainage failure has been
The risk of membrane rupture decreases with time because the postmembrane outflow resistance builds up slowly for several weeks after surgery. However, rupture can occur after severe ocular trauma (Fig. 12.19). Often there
(a)
(b)
Figure 12.19 Anterior segment as well as gonioscopic view of traumatic TDM rupture. Pupil is decentered, iris prolapse can be seen under the scleral flap.
154
Complications and reoperations Figure 12.20 Schematic representation of Descemet’s membrane detachment with aqueous humor passage from scleral space to subDescemet’s space at anterior edge of Descemet’s window.
is a concomitant iris prolapse with a decentered pupil and darkening of the subconjunctival area. If the IOP remains under control, no further treatment is needed. However, if the iris prolapse blocks the aqueous humor outflow and the IOP rises, medical or surgical therapy should be considered.
Descemet’s detachment Descemet’s membrane detachment is a rare complication after non-penetrating filtering surgeries. We estimate it to occur in about one out of 250–300 operated eyes. The pathogenesis depends on the type of surgery, although there appears to be an anatomical predisposition. With viscocanalostomy, detachment is related to the viscoelastic injection into the
artificial ostia of Schlemm’s canal, the canula probably being slightly misdirected. The detachment is noticed during the procedure or generally shortly afterwards. After other nonpenetrating filtering surgeries, this complication may be explained by the passage of aqueous humor from the scleral space to the sub-Descemet space at the anterior edge of the Descemet’s window (Fig. 12.20), secondary to an increased intrableb pressure because it may occur after trauma, encysted bleb, and vigorous ocular massage. Detachment is then usually diagnosed 4–8 weeks postoperatively. After viscocanalostomy, the cornea associated with the Descemet’s detachment remains clear probably because of the intact endothelium, Descemet’s complex and the chemical properties of sodium hyaluronate. However, after non-penetrating filtering surgeries, patients
Late postoperative complications
(a)
155
(b)
Figure 12.21 (a, b) Descemet’s detachment after deep sclerectomy with collagen implant.
Figure 12.22 UBM image of Descemet’s detachment.
Figure 12.23 Retro-Descemet’s space filled with blood after needling for encysed subconjunctival bleb.
complain of decreased visual acuity if the detachment extends over the visual axis because of adjacent corneal edema. Clinically this complication is diagnosed as a superior localized or more extensive separation of the Descemet’s membrane from its overlying stroma; it may be planar or not, and as
mentioned above, there may be concomitant corneal edema and decreased visual acuity (Figs. 12.21 and 12.22). Blood may also be seen in the retro-Descemet’s space (Fig. 12.23). From our experience, whether or not patients were being operated on for the repair
156
Complications and reoperations
of this complication, they all had final visual acuities similar to preoperative ones and good final ocular pressure control. If no descemetopexy was attempted, follow-up examinations showed that the appearance of the corneal inclusion of sodium hyaluronate after viscocanalostomy was essentially unchanged with only a slight tendency to resorption. Although no sign of corneal damage could be noted, longer term follow-up is needed. After other non-penetrating filtering surgeries, the detachment usually reattaches spontaneously shortly after treatment of the causative factor and its associated high intrableb pressure.
Peripheral anterior synechia The iris may adhere to the trabeculoDescemet’s window and form a peripheral anterior synechia (PAS) after the following situations: intraoperative microperforation with a microiris prolapse; iris entrapment into a goniopuncture hole, which usually occurs very soon after laser treatment; rupture of the TDM (blunt trauma or Valsalva) with subsequent iris prolapse. There may be an associated increase in IOP if there is insufficient aqueous humor flow through the membrane. A laser PAS lysis may be attempted to reposition the iris back; if it fails, a medical or secondary surgical treatment should be considered.
Blebitis or endophthalmitis Theoretically, late postoperative bleb infection should be less frequent with non-penetrating filtering surgeries than with trabeculectomy because of the reduced size of the conjunctival flap. By use of antimetabolites peroperatively or postoperatively with trabeculectomy, a prevalence of 2–6% bleb-related endoph-
thalmitis has been reported.24 As mentioned previously, we have not yet had a case of endophthalmitis following non-penetrating filtering surgery.
Cataract progression Henchoz et al,25 who compared three matched groups of patients who underwent either deep sclerectomy or trabeculectomy, or had no operation, reported that the incidence of cataract progression was not influenced by deep sclerectomy, whereas it was by trabeculectomy. As mentioned for immediate postoperative cataract induction, the absence of cataract progression after non-penetrating filtering surgery is probably due to the maintenance of the anterior-chamber depth and the low rate of inflammation induced postoperatively.
Chronic hypotony Late chronic ocular hypotony followed, in some cases, by hypotony-related maculopathy, has been observed in some patients who underwent deep sclerectomy with mitomycin C. The hypotony was caused by a very thin ischemic conjunctival bleb. The treatment of this condition is achieved through restoration of the filtering bleb by performing an excision of the ischemic tissue followed by a conjunctival graft.
Scleral ectasia In patients with thin sclera, the surgery may be followed by a scleral ectasia, which is of particular concern in patients with high myopia, chronic uveitis, and especially in those with associated rheumatoid or juvenile arthritis.26 The use of antimetabolites peroper-
Late postoperative complications
157
Figure 12.24 Scleral ex pack during reoperation (superficial scleral flap is fibrosed).
Figure 12.25 Opening of a previous scleral flap (deep sclerectomy is still patent).
atively or postoperatively may also increase the risk of this complication.
influenced by age, racial origin, previous use of topical medication, history of ocular operation or trauma, history of and previous or present inflammation. The filtration failure may also occur several months or even years after the initial operation. However, the filtration impairment may not be total, and then a simple topical glaucoma therapy can usually control the IOP. If not, reoperation should be considered. The following options are available, depending on the surgeon’s preference and the patient’s condition.
Reoperation Failures of filtration and IOP control after nonpenetrating filtering surgeries has been reported to occur in 5–10% of patients in medium-term follow-up studies.1,5,11–21 Depending on the surgeon’s preference, patients are either treated conservatively with glaucoma medications or reoperated on.27 In our experience, the reoperation rate is about 1% per year of follow-up. Elie Dahan, who does not use glaucoma medication, has a much higher rate of reoperation, in the range of 10% per year of followup.27 The major cause of filtration failure is fibrosis external to the TDM. The peroperative and/or postoperative use of antimetabolites may decrease the rate of filtration failure. However, even with the best surgical technique and follow-up care, some patients still experience filtration failure. The failure usually occurs in the first few postoperative months because of rapid excess scarring of the tissues, which is
Reopening primary surgical site Reopening is the easiest procedure to propose, when the failure is due to either subconjunctival fibrosis or too tight a closure of the superficial scleral flap, and when the Descemet’s window is transparent on gonioscopy. The conjunctiva should be carefully opened, avoiding the creation of a buttonhole. After the opening of the scleral flap (Fig. 12.24), the deep sclerectomy site is usually patent and aqueous humor can be seen to percolate through the TDM (Fig. 12.25). Antimetabo-
158
Complications and reoperations
lites as well as an intrascleral implant can be used to avoid subsequent filtration failure.
Another non-penetrating filtering surgery close to primary site When the scleral space is suspected to be closed because the Descemet’s window is not transparent on gonioscopic examination, the preferred option is to perform a second nonpenetrating filtering surgical procedure on either side of the first operation. The nasal side is usually preferred because temporal filtering blebs are often less comfortable for the patients.
Conversion of non-penetrating filtering surgery into trabeculectomy When the main site of outflow obstruction is suspected to be at the level of the TBM the surgeon may transform the non-penetrating filtering surgery into a trabeculectomy. In this situation, the superficial scleral flap should then be tightly closed because of the presence of a deep sclerectomy, which offers a decreased outflow resistance and a subsequent increased filtration.
Installation of artificial drainage shunt Filtration devices such as Molteno, Ahmed and Baerveldt tubes, should only be used in patients with known high-risk factors for bleb failure such as neovascular glaucoma, two or three previous failed standard filtering operations, or severe conjunctival scarring.
Cycloablation therapy This therapeutic option is usually proposed to patients with blind eyes or who are unwilling to undergo a second filtering operation.
References 1. Karlen ME, Sanchez E, Schnyder CC, Sickenberg M, Mermoud A. A deep sclerectomy with collagen implant: medium term results. Br J Ophthalmol 1999;83:6–11. 2. Sanchez E, Schnyder CC, Mermoud M. Résultats comparatifs de la sclérectomie profonde transformée en trabéculectomie et de la trabéculectomie classique. Klin Monatsbl Augenheilkd 1997;210:261–64. 3. Chiou AG, Mermoud A, Hediguer SE. Malignant ciliary block glaucoma after deep sclerotomy–ultrasound biomicroscopy imaging. Klein Monatsbl Augenheilkd 1996;208:279–81. 4. Watson PG, Jakeman C, Ozturk M et al. The complications of trabeculectomy: a 20-year follow-up. Eye 1990;4:425–38. 5. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with collagen implant and trabeculectomy in openangle glaucoma. J Cataract Refract Surg 1999;25:323–31. 6. Chiou AG, Mermoud A, Jewelewicz DA. Post-operative inflammation following deep sclerectomy with collagen implant versus standard trabeculectomy. Graefes Arch Clin Exp Ophthalmol 1998;236:593–96. 7. Gianoli F, Schnyder CC, Bovey E, Mermoud A. Combined surgery for cataract and glaucoma: phacoemulsification and deep sclerectomy compared with phacoemulsification and trabeculectomy. J Cataract Refract Surg 1999;25:340–46. 8. Brubaker RF, Pederson JE. Ciliochoroidal detachment. Surv Ophthalmol
References 1983;27:281–89. 9. Gressel MG, Parrish RK II, Heuer DK. Delayed nonexpulsive suprachoroidal hemorrhage. Arch Ophthalmol 1984;102:1757–60. 10. Ruderman JM, Harbin TS Jr, Campbell DG. Postoperative suprachoroidal hemorrhage following filtration procedures. Arch Ophthalmol 1986;104:201–05. 11. Zimmerman TJ, Kooner KS, Ford VJ et al. Effectiveness of nonpenetrating trabeculectomy in aphakic patients with glaucoma. Ophthalmic Surg 1984;15:44–50. 12. Zimmerman TJ, Kooner KS, Ford VJ et al. Trabeculectomy vs non-penetrating trabeculectomy: a retrospective study of two procedures in phakic patients with glaucoma. Ophthalmic Surg 1984;15:734–40. 13. Arenas E. Trabeculectomy ab-externo. Highlights of Ophthalmol 1991;19:59–66. 14. Gierek A, Szymanski A. Results of deep sclerectomy for open-angle glaucoma. Folia Ophthalmol (Leipzig) 1987;12:227–9. 15. Tavano G, Chabin T, Barrut JM. Hémitrabéculectomie ab externo non invasive. Bull Soc Ophtalmol Fr 1993;93:749–50. 16. Tanihara H, Negi A, Akimoto M et al. Surgical effects of trabeculotomy ab externo on adult eyes with primary open angle glaucoma and pseudoexfoliation syndrome. Arch Ophthalmol 1993;111:1653–61. 17. Stegmann RC. Visco-canalostomy: a new surgical technique for open angle glaucoma. An Inst Barraquer Spain 1995;25:229–32. 18. Kozlov VI, Bagrov SN, Anisimova SY et al. Non-penetrating deep sclerectomy with collagen. Eye Microsurg (Russian) 1990;3:44–46.
159
19. Fyodorov SN. Non penetrating deep sclerectomy in open-angle glaucoma. Eye Microsurg (Russian) 1989;2:52–55. 20. Sanchez E, Schnyder CC, Sickenberg M et al. Deep sclerectomy: results with and without collagen implant. Int Ophthalmol 1997;20:157–62. 21. Demailly P, Jeanteur-Lunel MN, Berkani M et al. Non penetrating deep sclerectomy associated with collagen device in primary open angle glaucoma: middle-term retrospective study. J Fr Ophtalmol 1996;19:659–66. 22. Chiou AG, Mermoud A, Hédiguer SE et al. Ultrasound biomicroscopy of eyes undergoing deep sclerectomy with collagen implant. Br J Ophthalmol 1996;80:541–44. 23. Vaudaux J, Mermoud A. Aqueous dynamics after deep sclerectomy: ex-vivo study. Ophthalmic Pract 1998;16:204–09. 24. Freedman J, Gupta M, Bunke A. Endophtalmitis after trabeculectomy. Arch Ophtalmol 1978;96:1017–8. 25. Henchoz L, Schnyder C, Shaarawy T et al. Surgery-induced cataract and cataract progression following deep sclerectomy with collagen implant compared to trabeculectomy. Ophthalmology 2000;107:205. 26. Milazzo S, Turut P, Malthieu D, Leviel MA. Scleral ectasia as a complication of deep sclerectomy. J Cataract Refract Surg 2000;26:785–87. 27. Dahan E, Drusedau MU. Nonpenetrating filtration surgery for glaucoma: control by surgery only. J Cataract Refract Surg 2000;26:695–701.
13 Results of non-penetrating glaucoma surgery Tarek Shaarawy
The widespread practice of non-penetrating filtering surgery has been hindered by the lack of sufficient studies reporting on the efficacy, safety, and reproducibility of non-penetrating filtering surgery. Fortunately, in the past two years we have seen a steady influx of papers published and presentations given, igniting interest in the glaucoma community. This chapter aims to summarize these studies and deal with the challenges to come.
Deep sclerectomy converted to trabeculectomy The preservation of an intact trabeculoDescemet’s membrane is of great importance to the success of deep sclerectomy. In the early learning phase of deep sclerectomy, many cases will be complicated with perforation of the membrane. Karlen et al1 reported three perforations in the first ten cases, and a further three in the next 100 cases, while Dahan and Drusedau2 reported perforation in one in three operations in the first 6 months of operating, and one in 20 thereafter. In the case of a perforation, many surgeons opt to convert the deep sclerectomy to a standard trabeculectomy.
In a landmark study, Sanchez et al3 compared a group of 19 patients who had their planned deep sclerectomy converted to trabeculectomy because of perforation of the trabeculo-Descemet’s membrane, with another group of 19 patients who had undergone trabeculectomy. They reported similar rates of complications and similar long-term results between the two groups. The study should be encouraging to surgeons willing to start doing deep sclerectomy, because it clearly shows that even in the case of a perforation, and conversion to trabeculectomy, the patient will not be subjected to statistically lower success rates than had trabeculectomy been done from the outset.
Viscocanalostomy Stegmann et al4 have published results regarding viscocanalostomy on black African patients with open-angle glaucoma. This prospective study was done on 214 patients who were poorly controlled by medical therapy. Stegmann et al reported an intraocular pressure (IOP) of 22 mmHg or less in 82% of eyes without medical therapy (complete success rate). If a beta-blocker was prescribed for cases not achieving 22 mmHg or less postoperatively, the success rate increased to
162
Results of non-penetrating glaucoma surgery
89.0% (qualified success rate). With a mean follow-up of 35 months (range, 6–64 months), this study shows that IOP can be controlled satisfactorily with viscocanalostomy. Shaarawy et al5 have reported their longterm results of viscocanalostomy; a nonrandomized prospective trial in which 68 eyes of 68 white patients were consecutively enrolled. The mean follow-up was 33.2 (SD 12) months. Complete success rate (IOP ≤ 21 mmHg, without medication) was 55%, whereas qualified success rate (IOP ≤ 21 mmHg, with or without medication) was 89%. They concluded that viscocanalostomy provides reasonable long-term IOP control with few immediate postoperative complications. The two previously mentioned trials show similar qualified success rates but significant differences of complete success rate. It is not quite clear where this difference stems from, but it is, perhaps, appropriate to point out the differences between the two studies.
Figure 13.1 Superficial flap loosely sutured with a two 10/0 stitches at the angles.
Stegmann’s patients were black Africans while Shaarawy’s were white Europeans. The two teams used different upper limits for their success criteria. Stegmann et al4 advocate a technique of tight closure of the scleral flap, in an attempt to force the percolating aqueous humor into the two surgically created ostia of Schlemm’s canal, whereas Shaarawy et al5 describe a technique where the superficial flap is loosely sutured (Fig. 13.1), so as not to prevent subconjunctival and intrascleral blebs from occurring.
Results of Deep Sclerectomy Deep sclerectomy without an implant The surgical procedure of deep sclerectomy without the use of any adjunctive treatment such as antimetabolites or an implant has been studied by various investigators. Bas and Goethals6 studied the preliminary results of the procedure and reported retrospectively on 29 patients (34 eyes) who had undergone deep sclerectomy. The mean follow-up period was 5.3 months and their complete success rate was 92% (IOP ≤ 20 mmHg, without medication). The main complication encountered during the learning period was perforation of the trabeculo-Descemet’s membrane, occurring in nine of the 34 eyes. In another medium-term retrospective study, Massy et al7 studied the same technique with a mean follow-up of 14.2 months. The success rate was 81% (IOP ≤ 21 mmHg, without medication) and 50% (IOP ≤ 16 mmHg, without medication). The complication rate
Results of deep sclerectomy was low: 2% choroidal detachment, 2% hypotony, no hyphaema, no endophthalmitis. The investigators compared their results with similar studies on trabeculectomy and concluded that deep sclerectomy offers similar success with much lower complication rates. Dahan and Drusedau2 have published their retrospective long-term results of deep sclerectomy. 46 patients (86 eyes) were followed for a mean of 46 months. They further divided their patients into high-risk (black Africans, Asian Indians, and whites younger than 45 years) and low-risk categories. Of 86 eyes, 38 were newly diagnosed with primary openangle glaucoma (POAG) and did not receive antiglaucoma treatment preoperatively. These 38 “virgin” eyes were chosen for deep sclerectomy as first-line treatment. The mean drop in IOP postoperatively was 50% at 46 months (from a mean of 30.4 mmHg preoperatively to a mean of 15.3 mmHg postoperatively). When IOP rose above 20 mmHg postoperatively, the investigators revised the filtration site to reestablish filtration, rather than prescribe medical treatment. The reoperation rate was 4.7 times higher in previously treated patients than in untreated patients. They concluded that deep sclerectomy offers reasonable success rates, and is more rewarding in previously untreated eyes.
Deep sclerectomy with an implant To enhance the filtration of deep sclerectomy, Kozlov et al8 described the use of a collagen implant placed within the scleral bed. The idea was for the collagen implant to occupy the surgically created intrascleral bleb under the superficial flap during the early postoperative period when the healing process is at its peak. The collagen implant is later absorbed, leaving behind a patent space to which aqueous
163
Figure 13.2 Intrascleral space left after absorption of collagen implant.
humor percolates and is then resorbed (Fig. 13.2). Other implants were later introduced by other investigators.9
Short and medium-term results
Demailly et al10 published their medium-term retrospective results of deep sclerectomy with collagen implant (DSCI). Their series included 159 patients (219 eyes), with a mean followup of 8 (range 3–20) months and a maximum follow-up of 20 months. They reported a probability success rate (IOP ≤ 20 mmHg) of 89% at 6 months, and 75.6% at 16 months. They concluded that DSCI is an excellent alternative to trabeculectomy. Karlen et al1 published the first prospective trial of DSCI, in which 100 patients (100 eyes)
164
Results of non-penetrating glaucoma surgery
were consecutively enrolled. Their mean follow-up was 17.8 ± 8.7 months. The mean preoperative IOP was 27.8 ± 8.6 mmHg; the mean postoperative IOP was 5.7 ± 4.0 at day 1, 11.2 ± 4.6 at month 1, 14.0 ± 3.5 at month 12, and 13.0 ± 3.8 at month 36. Complete success rate (IOP ≤ 21 mmHg, without medication) was 44.6% at 36 months, qualified success rate (IOP ≤ 21 mmHg, with or without medication) was 97.7% at 36 months. Early postoperative complications included hyphaema in seven patients, wound leak in ten patients, and subtle choroidal detachment in 11 patients. The results were both satisfactory in terms of success rates as well as incidence of complications.
Long-term results
Shaarawy et al11 have presented their longterm follow-up of DSCI involving 105 eyes of 105 patients with medically uncontrolled primary and secondary open-angle glaucoma. The mean follow-up period was 43.2 ± 14.3 months. The mean preoperative IOP was 26.8 ± 7 mmHg; the mean postoperative IOP was 5.1 ± 3 mmHg at day 1 and 11.8 ± 3 at month 60. Qualified success rate (IOP ≤ 21 mmHg, with or without medication) was 95.1% at 60 months. Complete success rate (IOP ≤ 21 mmHg, without medication) was 63% at 60 months. Patients with an IOP lower than 18 mmHg with medication had a success rate of 32.1% at 60 months. Forty-nine (46.6%) patients achieved an IOP equal to or lower than 15 mmHg, without medication at 60 months. The mean number of medications per patient was reduced from 2.3 to 0.49. These were indeed encouraging results in terms of the long-term efficacy and safety of DSCI.
Comparative studies Comparative studies between trabeculectomy and nonpenetrating glaucoma surgery One of the most intriguing questions about non-penetrating surgery is how well it fares in comparison to trabeculectomy, the latter being regarded, for many decades, as the gold standard against which all new glaucoma procedures were tested. Carrassa12 prospectively compared two groups of randomized patients, one group had trabeculectomy and the other viscocanalostomy. Each group included 25 patients of comparative male/female ratio and mean age. He concluded that although viscocanalostomy has a less significant rate of complications, in terms of incidence and gravity, trabeculectomy achieved lower mean IOP. Gandolfi and Cimino13 presented a prospective randomized 2-year trial of deep sclerectomy without implants and with unsutured scleral flap versus trabeculectomy. Thirtythree eyes (33 patients) were enrolled, 16 patients had trabeculectomy and 17 patients had deep sclerectomy. The investigators found better IOP control with trabeculectomy which seemed to offer lower IOP than deep sclerectomy, however, deep sclerectomy had a lower incidence of complications, in particular a lower 2-year progression incidence of lens nuclear opacities. Surgery-induced cataract has been one of the most disastrous complications of trabeculectomy, and to examine the incidence of surgery-induced cataracts and cataract progressions after surgery, Henchoz et al14 undertook a prospective trial involving 70 patients (70 eyes) who underwent DSCI,
Comparative studies
165
compared with 69 patients (69 eyes) who underwent trabeculectomy. The trabeculectomy group had a significantly higher incidence of surgery-induced cataract (16% vs 1% in DSCI, p = 0.002), choroidal detachments, hypotonies, and shallow anterior chambers. Both groups showed similar incidence of cataract progression. In another study, El Sayyad et al15 prospectively compared the efficacy and safety of deep sclerectomy without implants to trabeculectomy in bilateral primary open-angle glaucoma. Thirty-nine patients (78 eyes) were included in this study; patients were randomly assigned to receive deep sclerectomy in one eye and trabeculectomy in the other eye. There was no statistical significance between the two groups with regard to both the complete and qualified success rates. There was, however, a significantly lower incidence of complications for the deep sclerectomy group compared with the trabeculectomy group. In the studies by Gandolfi et al13 and El Sayyad et al,15 neither group used any implants with deep sclerectomy, whereas Mermoud et
al16 prospectively compared two well-matched groups of DSCI and trabeculectomy. In this study, each group constituted 44 eyes of 44 patients and were followed for up to 24 months. The mean follow-up for both groups was 14 (SD 6.3) months. The complete success rates (IOP ≤ 21 mmHg, without medications) were 57% for the trabeculectomy group versus 69% for the DSCI group. The number of postoperative medications was significantly lower in the DSCI group (p = .047). The investigators concluded that success rates of both DSCI and trabeculectomy were similar, however, there was a lower rate of complications in the DSCI group.
Figure 13.3 Complete success Kaplan-Meier survival analysis of DSCI versus deep sclerectomy.17
Figure 13.4 Complete success Kaplan-Meier survival analysis of DSCI versus deep sclerectomy.17
Comparative studies between deep sclerectomy with and without implant In deep sclerectomy the simple removal of the scleral and corneal tissue overlying trabecular structures risks inducing secondary fibrosis
166
Results of non-penetrating glaucoma surgery
and subsequent failure to control IOP. The sclerocorneal space created by non-penetrating glaucoma surgery acts as an aqueous decompression space that should be kept open. Kozlov et al8 first proposed the use of an intrascleral collagen implant to maintain this newly created space, and since then there has been continuing controversy over the value of implants. Very few studies have addressed this vital issue. Sanchez et al17 compared deep sclerectomy with and without collagen implants, and reported on their short-term results (mean follow-up of 9.7 ± 6.5 months). They prospectively compared 86 patients (86 eyes) who underwent DSCI with 82 patients (82 eyes) who underwent deep sclerectomy without an implant. They reported better complete and qualified success rates (Fig. 13.3) with the use of a collagen implant. They also reported a lower incidence of bleb fibrosis when the collagen implant was used, and that the need for postoperative glaucoma medications was significantly lower in the DSCI group (p = 0.0038). Shaarawy et al18 have presented their randomized prospective trial comparing deep sclerectomy with and without an implant. The trial involved 104 patients (104 eyes) in which half received a collagen implant with their deep sclerectomies, while the other half did not. The patients were followed for up to 60 months with a mean follow-up of 44 (SD 14) months for both groups. The two groups were well matched with regard to age, sex, race, preoperative medication, preoperative IOP, and causative factors. Complete success rate (IOP ≤ 21 mmHg, without medication) was 34.6% (18/52 patients) at 48 months for the deep sclerectomy group (Fig. 13.4), and 63.4% (33/52 patients) for the DSCI group. Qualified success rate (IOP ≤ 21 mmHg, with or without medication) was 78.8% (41/52
patients) at 48 months and 94% (49/52 patients) for the DSCI group. The mean number of medications was reduced from 2.1 (SD 0.8) to 1 (SD 1) after deep sclerectomy, and was reduced from 2.2 (SD 0.7) to 0.4 (SD 0.6) in the DSCI group (p = 0.001). The investigators concluded that the use of collagen implant seems to offer a significant advantage in non-penetrating glaucoma surgery. Although these studies point to the value of implant use, no studies are available, as yet, on the effect that material, shape, and size of implant have on the success rates of the surgery. Other implants, of different shapes and materials, are currently being clinically tested. If the value of these implants is ascertained, as is likely, there will be an influx of different implant designs, such as occurred with intraocular lenses.
Combined surgery Encouraged by the low complication rate of non-penetrating glaucoma surgery, investigators have assessed the results of this type of surgery combined with cataract surgery.
Combined viscocanalostomy and phacoemulsification The primary concern with regard to cataract and glaucoma surgery in the same eye is how and where to position the cataract operation in the management scheme of the patients’ condition. Is it wiser to choose one sequence and type of surgery before the other or to combine the two procedures? This remains a matter of unresolved controversy.
References Viscocanalostomy, in combined surgery for cataract and glaucoma, has been studied by Gimbel et al.19 They compared two techniques: a shared scleral incision versus a clear corneal incision separate from the viscocanalostomy site. The drop in IOP was significantly low in both groups. There were low complication rates, with no hypotony, choroidal detachment, or postoperative cystoid macular oedema in either group. They concluded that combined viscocanalostomy, cataract extraction, and intraocular-lens implantation was safe and efficacious in lowering IOP whether a shared scleral incision or a separate clear corneal incision was made for phacoemulsification.
Combined deep sclerectomy and phacoemulsification Gianoli and Mermoud20 reported on the results of phacoemulsification with deep sclerectomy compared with phacoemulsification with trabeculectomy. After a mean follow-up of 12.5 (SD 6) months, mean IOP and visual acuity were similar in both groups. The group assigned to phacoemulsification with deep sclerectomy experienced significantly less inflammation and hyphaema than the phacoemulsification with trabeculectomy. The investigators concluded that deep sclerectomy combined with cataract surgery is a better option, giving a lower rate of complications thus better ambulatory care.
Summary and conclusion From the weight of evidence that we currently possess, we can draw the following conclusions. Non-penetrating glaucoma surgery
167
certainly is safer than trabeculectomy. Nonpenetrating glaucoma surgery offers similar IOP control to trabeculectomy. The use of implants in non-penetrating glaucoma surgery offers better IOP control for longer durations, thus enhancing success rates. Surgery in the past has been regarded as a last resort for the treatment of glaucoma, which is largely due to the high complication rates commonly encountered with trabeculectomy. With non-penetrating glaucoma surgery, a safer surgery for the treatment of glaucoma, patients are offered the option of surgery earlier than usual. It is conceivable that some surgeons may offer the surgery as a first-line treatment for their patients. The studies that are currently available are hindered by the fact that the various investigators are using different design strategies and surgical techniques, and, more importantly, different success criteria; thus it is difficult to make comparisons between these studies. A prospective, randomized, multicentre study is needed, before a final conclusion can be made on how non-penetrating surgery fares, compared with trabeculectomy, and which technique of the many variations that are available is the best. Once this is achieved, a search for the ideal implant might ensue.
References 1. Karlen ME, Sanchez E, Schnyder et al. Deep sclerectomy with collagen implant: medium term results. Br J Ophthalmol 1999;83:6–11. 2. Dahan E, Drusedau MU. Nonpenetrating filtration surgery for glaucoma: control by surgery only. J Cataract Refract Surg 2000;26:695–701. 3. Sanchez E, Schnyder CC, Mermoud A. Comparative results of deep sclerectomy transformed to trabeculectomy and classical
168
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Results of non-penetrating glaucoma surgery
trabeculectomy. Klin Monatsbl Augenheilkd 1997;210:261–64. Stegmann R, Pienaar A, Miller D. Viscocanalostomy for open-angle glaucoma in black African patients. J Cataract Refract Surg 1999;25:316–22. Shaarawy T, Nguyen C, Achache F et al. Long-term results of viscocanalostomy in Caucasians. In: American Academy of Ophthalmology. Dallas, USA, 2000. Bas JM, Goethals MJ. Non-penetrating deep sclerectomy preliminary results. Bull Soc Belge Ophtalmol 1999;272:55–9. Massy J, Gruber D, Muraine M et al. Nonpenetrating deep sclerectomy in the surgical treatment of chronic open-angle glaucoma: mid-term results. J Fr Ophtalmol 1999;22:292–98. Kozlov V, Bagrov SN, Anisimova SY et al. Nonpenetrating deep sclerectomy with collagen. Eye Microsurg (In Russian) 1990;3: 44–6. Sourdille P, Santiago PY, Villain F et al. Reticulated hyaluronic acid implant in nonperforating trabecular surgery. J Cataract Refract Surg, 1999;25:332–39. Demailly P, Jeanteur-Lunel MN, Berkani M et al. Non-penetrating deep sclerectomy combined with a collagen implant in primary open-angle glaucoma: medium-term retrospective results. J Fr Ophtalmol 1996;19:659–66. Shaarawy T, Karlen ME, Sanchez E et al. Long term results of deep sclerectomy with collagen implant. Acta Ophthalmol Scand 2000;78:323 Carassa R. Viscocanalostomy versus trabeculectomy: a 12 months prospective randomized study. In: Proceedings of the American Society of Cataract and Refractive Surgery. Boston, USA, 2000. Gandolfi S, Cimino L. Deep sclerectomy without absorbable implants and with
14.
15.
16.
17.
18.
19.
20.
unsutured scleral flap: prospective, randomized 2-year clinical trial vs trabeculectomy with releasable sutures. In: Association for Research in Vision and Ophthalmology (ARVO) meeting. Fort Lauderdale, USA, 2000. Henchoz L, Shaarawy T, Mermoud A. Surgery induced cataract and cataract progression following deep sclerectomy with collagen implant compared to trabeculectomy. In: Proceedings of the American Academy of Ophthalmology meeting. Dallas, USA, 2000. El Sayyad F, Helal M, El-Kholify H et al. Nonpenetrating deep sclerectomy versus trabeculectomy in bilateral primary openangle glaucoma. Ophthalmology 2000;107:1671–4. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with collagen implant and trabeculectomy in openangle glaucoma. J Cataract Refract Surg 1999;25:323–31. Sanchez E, Schnyder CC, Sickenberg M et al. Deep sclerectomy: results with and without collagen implant. Int Ophthalmol 1996;20:157–62. Shaarawy T, Nguyen C, Achache F et al. Deep sclerectomy with and without collagen implant: long-term prospective study. In: First International Congress on Non-Penetrating Glaucoma Surgery. Lausanne, Switzerland, 2001. 1st ed. Gimbel HV, Penno EE, Ferensowicz M. Combined cataract surgery, intraocular lens implantation, and viscocanalostomy. J Cataract Refract Surg 1999;25:1370–5. Gianoli F, Mermoud A. Cataract-glaucoma surgery: comparison between phacoemulsification combined with deep sclerectomy or trabeculectomy. Klin Monatsbl Augenheilkd 1999;210:256–60.
14 Phacoemulsification combined with viscocanalostomy and deep sclerectomy Fathi El-Sayyad
The existence of cataract and glaucoma in the same eye requires special consideration when surgical intervention is needed. If filtering surgery is done first, subsequent cataract surgery may jeopardize an already functioning filtering bleb.1 If cataract surgery is done first, postoperative elevation of intraocular pressure (IOP) may occur and threaten further loss of visual field.2 Combined surgical management of coexisting cataract and glaucoma has gained popularity because of several advantages, including decreased risk and easier management of early postoperative IOP spikes, better long-term glaucoma control with respect to IOP or medications, and the need to do only one surgical procedure to manage both cataract and chronic glaucoma. 3 The technological innovations of smallincision cataract surgery have dramatically changed the outcome of the combined procedure in those eyes with coexisting cataract and glaucoma.4–6 Phototrabeculectomy provides IOP control statistically similar to two-stage surgery (glaucoma surgery followed by cataract extraction) and has the additional advantages of requiring only one operation with earlier visual rehabilitation. 7 Although trabeculectomy is currently regarded as the standard surgical procedure combined with cataract surgery,4,8 it can cause undesirable postoperative complications includ-
ing hyphema, excessive filtration leading to shallow or flat anterior chamber, choroidal detachment, hypotony maculopathy, suprachoroidal hemorrhage,9–13 bleb-related problems, and increased risk of endophthalmitis.14–16 Non-penetrating glaucoma surgery, including deep sclerectomy and viscocanalostomy, is a viable alternative to trabeculectomy with advantages of decreased postoperative complications.17–24 Both procedures can be combined with phacoemulsification in managing patients with coexisting cataract and glaucoma with very satisfactory results with regard to visual outcome and IOP control.25,26
Surgical technique Topical anesthesia can be used in combined viscocanalostomy or deep sclerectomy and phacoemulsification (0.5% tetracaine and 4% lidocaine). This technique is probably more comfortable for the patient than combined phacotrabeculectomy done under topical anesthesia; this may be because no peripheral iridectomy is needed during the non-penetrating procedure. Alternatively, peribulbar, retrobulbar, or general anesthesia can be used. A 7/0 superior peripheral corneal silk or vicryl suture is used for traction; this is more convenient than superior rectus traction
170
Phacoemulsification combined with viscocanalostomy
Figure 14.1 A fornix-based conjunctival flap and a superficial scleral flap are fashioned.
Figure 14.2 Dissection of the deep flap is performed.
suture, particularly if topical anesthesia is used. In addition, it eliminates the possibility of subconjunctival hemorrhage, buttonholing, and postoperative ptosis that can be inadvertently caused by the bridle suture. The combined procedure can be done by two techniques: a shared or one-site incision through which viscocanalostomy and phacoemulsification are performed, or clear corneal incision for phacoemulsification that is separate from the viscocanalostomy site, as done in phacotrabeculectomy.27 A 8–9 mm wide fornix-based conjunctival flap is prepared in the desired location, usually superiorly. Alternatively, the conjunctiva and Tenon’s capsule are opened in the upper fornix and the sclera is exposed. A fornix-based flap may be preferred, because it is easier and gives better exposure during phacoemulsification and fashioning of the superficial and deep scleral flaps (Fig 14.1). Minimal coagulation is advisable to preserve the function of episcleral vessels for aqueous drainage in the viscocanalostomy procedure. A Weck-Cel sponge (Solan, Jacksonville, FL,
USA) soaked in vasopressin may be applied in contact with bleeders to help in hemostosis. Light cautery also avoids excessive scarring and retraction of scleral tissue. Various shapes and dimensions of the superficial scleral flaps have been described in combined surgery.25,28,29 A larger flap is advisable to increase the surface area of drainage when the deeper scleral flap is excised. The scleral flap is fashioned between collector channels and the shape is usually parabolic or square, measuring 5 5 mm (Fig. 14.1). The outline of the superficial flap is deepened to about 300 µm, then planar dissection starts at the angle and advances forward with a 69 Baever blade (Alcon Surgical, Fort Worth, Texas, USA) or Grieshaber mini spoon blade (Alcon-Grieshaber, Schaffhausen, Switzerland). It is important to preserve the integrity of the superficial flap and to avoid shredding as conversion to trabeculectomy may be needed. The blade, which is held parallel to the scleral flap, is advanced forward with light pressure against the scleral bed. The flap is lifted just enough for exposure and is dissected
Surgical technique
Figure 14.3 Dissection is stopped near the Schlemm’s canal to allow for phacoemulsification.
forward 1 mm into clear cornea. A triangular deeper scleral flap is fashioned under the superficial one (Fig. 14.2) and the dissection is started very deep, as close as possible to the choroid. The uvea should appear glistening through the very thin scleral rim. The dissection is carried out in a forward direction until Schlemm’s canal is revealed, and, if the flap is dissected very deep, the roof of the canal will be included in the deeper flap and the canal will appear as a dark line anterior to scleral spur. It is advisable not to proceed with further dissection of the scleral flap into clear cornea until after phacoemulsification has been completed (Fig. 14.3), because the thin Descemet’s window can be blown out by irrigation pressure. If same-site phacosclerectomy or viscocanalostomy is intended, the deeper flap is repositioned and a tunnel incision is made under the superficial flap to allow phacoemulsification between the two flaps before completion of the non-penetrating glaucoma surgery (Fig. 14.4). In separate-site procedure, however, the operating microscope is rotated 90 degrees to
171
Figure 14.4 Same-site phacoemulsification incision under the superficial scleral flap.
Figure 14.5 Separate-site phacoemulsification incision through temporal cornea.
position the surgeon on the temporal side of the patient. The diamond knife is used to create a temporal clear corneal tunnel incision (Fig. 14.5). Viscoelastic material is then introduced and a paracentesis is performed to allow bimanual phacoemulsification. After anterior capsulorhexis and hydrodissection, the lens
172
Phacoemulsification combined with viscocanalostomy
Figure 14.6 Phacoemulsification is completed and foldable intraocular lens is inserted.
Figure 14.7 Blunt dissection with a sponge is used to create a Descemet’s window.
nucleus is removed by phacoemulsification and the remaining cortex by manual irrigation and aspiration. The incision is widened to 3.5 mm and a foldable intraocular lens is introduced into the capsular bag (Fig. 14.6). Intracameral acetylcholine chloride (MiocholT) is used for miosis and the microscope is rotated to the primary position. The sides of the deep block are dissected to create ostia of Schlemm’s canal if deroofing has not already been included within the deeper flap. The dissection of the deep scleral flap is continued towards the limbus to expose the sclerocorneal trabecular meshwork, which appears granular. The dissection is extended 1 mm into clear cornea, blunt dissection with a sponge is used to create Descemet’s window (Fig. 14.7). During viscocanalostomy dissection and after Schlemm’s canal is exposed, and IOP is lowered, there is often a reflux of aqueous-vein blood into Schlemm’s canal and into the surgical field. High molecular weight sodium hyaluronate (Healon GV, Pharmacia Corp, NJ, USA) is
injected into the open ends of Schlemm’s canal (Fig. 14.8) with a fine canula (AlconGrieshaber, Schaffhausen, Switzerland). The high molecular weight viscoelastic mechanically dilates the lumen of Schlemm’s canal,30 which restores the physiological outflow pathway of aqueous through the canal and the collector channels. The juxtacanalicular trabecular membrane is removed together with the inner wall of Schlemm’s canal. At this stage, aqueous humor is seen percolating through the remaining thin trabeculo-Descemet’s membrane. If percolation is inadequate, further stripping is needed (Fig. 14.9). Care should be taken to avoid perforation of the window. If this occurs, the procedure may be completed as trabeculectomy with internal block dissection and peripheral iridectomy. The deeper flap is excised with the diamond knife or Vannas scissors after adequate aqueous percolation has been achieved through the trabeculo-Descemet’s membrane. Healon GV is placed under the external flap
Augmented deep sclerectomy with phacoemulsification
173
Figure 14.8 Viscoelastic is slowly injected into the opening of Schlemm’s canal.
Figure 14.9 If aqueous percolation is inadequate, further stripping is needed.
before it is securely sutured with multiple interrupted 10/0 nylon sutures. Forcible injection of viscoelastics under the closed scleral flap should be avoided because it may cause stripping of Descemet’s window. An intrascleral space is created to act as a reservoir for the aqueous percolation from the anterior chamber through the thin sclerocorneal trabecular meshwork and Descemet’s window. Aqueous flows through the dilated ends of Schlemm’s canal into the collector channels and the episcleral veins when the conjunctiva is closed with interrupted nylon sutures at each periotomy with a horizontal suture in-between.
sub-Tenon’s filtration. The surgical technique is similar to viscocanalostomy with regard to the preparation of the scleral bed, fashioning of the superficial and deep scleral flaps, creation of Descemet’s window, and removal of the deeper flap. Identification and dilation of Schlemm’s canal are not parts of deep sclerectomy procedure. The superficial flap is closed less tightly with 10/0 nylon suture. Since deep sclerectomy produces subconjunctival filtration, it can be augmented by intraoperative or postoperative antimetabolites such as mitomycin C or 5-fluorouracil. Many studies report the benefit of mitomycin C in combined surgery.31–35 The use of antimetabolites is strongly recommended in complicated glaucomas associated with cataract in young people, repeated surgery, secondary glaucoma, and chronic uveitis). Mitomycin C or 5-fluorouracil can be applied during deep sclerectomy in the same way as in trabeculectomy. A sponge soaked with mitomycin C (0.3 mg/mL) or 5fluorouracil (50 mg/mL) is placed for 3–5 min
Augmented deep sclerectomy combined with phacoemulsification Unlike viscocanalostomy, deep sclerectomy is regarded as a filtering surgery that produces
174
Phacoemulsification combined with viscocanalostomy
between the conjunctiva and the superficial scleral flap or under the superficial scleral flap. Alternatively, in high risk patients, the sponge can be placed simultaneously under conjunctiva and under the superficial flap.34 It is not advisable to apply the antifibrotic agent under the deep scleral flap because it may be absorbed profusely through the very thin remaining sclera and trabeculo-Descemet’s membrane, causing severe hypotony and intraocular toxicity. 5-fluorouracil may be used postoperatively as adjunctive therapy in cases of failing blebs in the form of subconjunctival injections (5 mg/mL) to promote subconjunctival filtration.20,23,35 Alternatively, a collagen implant in the scleral bed may be used to promote filtration,19,21,23 with better surgical outcome. The device, which is positioned radially under the scleral flap, provides a support for the elimination route of aqueous humor and acts like a sponge; it also maintains the space between the deep scleral plane and the superficial scleral flap, allowing aqueous humor to flow freely subconjunctivally.
Complications The most common complication encountered during viscocanalostomy or deep sclerectomy is the perforation of Descemet’s window. If a microperforation occurs, this can be disregarded and the surgeon can proceed with no adverse complications. However, if a substantial perforation occurs, the procedure may be completed as phacotrabeculectomy with internal block dissection and peripheral iridectomy. The surgical outcome of patients undergoing sclerectomy with accidental perforation and
conversion to trabeculectomy was analysed by Sanchez et al, 36 who found that the long-term success rate of those eyes was similar to trabeculectomy. Failure to identify Schlemm’s canal is most commonly due to inadequate deep dissection almost down to the choroid. The surgeon has to restart deeper dissection 2–3 mm posterior to the scleral spur to be able to identify the Schlemm’s canal. A torn superficial scleral flap may occur, particularly in the shared incision in combined viscocanalostomy and phacoemulsification.25 The torn scleral flap needs extra suturing to eliminate the formation of subconjunctival bleb and allow aqueous percolation in the intrascleral space. Early postoperative complications may include IOP spikes, which have been noted in shared-incision combined phacoemulsification and viscocanalostomy in 6%25 and in combined deep sclerectomy and phacoemulsification in 13.3%. 26 Iris prolapse through trabeculo-Descemet’s membrane has been reported, caused by blunt trauma in the immediate postoperative period.26 Transient hypotony, hyphema, and shallow anterior chamber are rare complications, following combined deep sclerectomy and phacoemulsification.26 An increased resistance to aqueous outflow through the trabeculo-Descemet’s membrane may take place as a result of gradual thickening of the membrane. Meticulous dissection of the sclera, Schlemm’s canal, and the peripheral cornea during surgery allows aqueous percolation through the remaining thick membrane, which decreases the incidence of progressive thickening. Goniopuncture by Nd:YAG laser to disrupt the trabeculo-Descemet’s membrane can be used to convert sclerectomy into a perforating procedure. Previous reports show the need for goniopunture after sclerectomy.21,23,35,37
References
References 1. Seah SKL, Jap A, Prata JA et al. Cataract surgery after trabeculectomy. Ophthalmic Surg Lasers 1996;27:587–94. 2. Krupin T, Feitl M, Bishop KL. Postoperative intraocular pressure rise in open-angle glaucoma patients after cataract or combined cataract-filtration surgery. Ophthalmology 1989;96:579–84. 3. Shields MB. Another reevaluation of combined cataract and glaucoma surgery. Am J Ophthalmol 1993;115:806–11. 4. Munden PM, Alward WLM. Combined phacoemulsification, posterior chamber intraocular lens implantation, and trabeculectomy with mitomycin-C. Am J Ophthalmol 1995;119:20–29. 5. Gous PTJ, Roux P. Preliminary report of sutureless phacotrabeculectomy through a modified self-sealing scleral incision. J Cataract Refract Surg 1995;21:160–69. 6. El Sayyad F, El-Maghraby A. The contribution of phacoemulsification to combined cataract and glaucoma surgery. Curr Opin in Ophthalmol 1998;9:95–100. 7. El Sayyad FF, Helal MH, Khalil MM, ElMaghraby MA. Phacotrabeculectomy versus two-stage operation: a matched study. Ophthalmic Surg Lasers 1999;30:260–65. 8. Lyle WA, Jin JC. Comparison of a 3 and 6 mm incision in combined phacoemulsification and trabeculectomy. Am J Ophthalmol 1991;111:189–96. 9. Brubaker RF, Pederson JE. Ciliochoroidal detachment. Surg Ophthalmol 1983;27:281–89. 10. Gressel MG, Parrish RK II, Heuer DK. Delayed nonexpulsive suprachoroidal hemorrhage. Arch Ophthalmol 1984;102:1757–60. 11. Ruderman JM, Harbin TS Jr, Campbell DG. Postoperative suprachoroidal hemorrhage following filtration procedures. Arch Ophthalmol 1986;104:201–05.
175
12. Stewart WC, Shields MB. Management of anterior chamber depth after trabeculectomy. Am J Ophthalmol 1988;106:41–44. 13. Kao SF, Lichter PR, Musch DC. Anterior chamber depth following filtration surgery. Ophthalmic Surg 1989;20:332–36. 14. Wedrich A, Menapace R, Radax U, Papapanes P. Long-term results of combined trabeculectomy and small incision cataract surgery. J Cataract Refract Surg 1995;21:49–54. 15. Greenfield DS, Miller MP, Suner IJ, Palmberg PF. Needle elevation of the scleral flap for failing filtration blebs after trabeculectomy with mitomycin C. Am J Ophthalmol 1996;12:195–204. 16. Wilson MR, Kotas-Neutmann R. Free conjunctival patch for repair of persistent late bleb leak. Am J Ophthalmol 1994;117:569–74. 17. Zimmerman TJ, Kooner KS, Ford VJ et al. Effectiveness of nonpenetrating trabeculectomy in aphakic patients with glaucoma. Ophthalmic Surg 1984;15:44–50. 18. Zimmerman TJ, Kooner KS, Ford VJ et al. Trabeculectomy vs nonpenetrating trabeculectomy: a retrospective study of two procedures in phakic patients with glaucoma. Ophthalmic Surg 1984;15:734–40. 19. Fyodorov SN, Kozlov VI, Timoshkina NT et al. Nonpenetrating deep sclerectomy in open angle glaucoma. Ophthalmosurgery 1990;3:52–55. 20. Demailly P, Jeanteur-Lunel MN, Berkani M, et al. La sclérectomie profonde non perforante associée á la pose d’un implant de collagène dans le glaucome primitif à angle ouvert: résultats rétrospecitifs à noyen terme. J Fr Ophtalmol 1996;19:659–66. 21. Sanchez E, Schnyder CC, Sickenberg M et al. Deep sclerectomy: results with and without collagen implant. Int Ophthalmol 1996–97;20:157–62. 22. Chiou AG, Mermoud A, Jewelewicz DA et al. Post-operative inflammation following deep sclerectomy with collagen implant versus
176
23.
24.
25.
26.
27.
28.
29.
30.
Phacoemulsification combined with viscocanalostomy
standard trabeculectomy. Graefes Arch Clin Exp Ophthalmol 1998;236:593–96. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with collagen implant and trabeculectomy in open angle glaucoma. J Cataract Refract Surg 1999;25:323–31. Stegmann R, Pienaar A, Miller D. Viscocanalostomy for open-angle glaucoma in black African patients. J Cataract Refract Surg 1999;25:316–22. Gimbel HV, Anderson Penno EE, Ferensowicz M. Combined cataract surgery, intraocular lens implantation, and viscocanalostomy. J Cataract Refract Surg 1999;25:1371–74. Gianoli F, Schnyder C, Bovey E, Mermoud A. Combined surgery for cataract and glaucoma: phacoemulsification and deep sclerectomy compared with phacoemulsification and trabeculectomy. J Cataract Refract Surg 1999;25:340–46. El Sayyad F, Helal M, El-Maghraby A et al. One-site versus 2–site phacotrabeculectomy: a randomized study. J Cataract Refract Surg 1999;25:77–82. Maloney WF. Combined cataract and glaucoma gains greater efficiency with new techniques. Ocular Surgery News 1998;24:12–14. Singer HW. Viscocanalostomy: effective glaucoma technique without troublesome blebs. Ocular Surgery News (Int Ed) 1998;9:18. Allingham RR, de Kater AW, Ethier CR. Schlemm’s canal and primary open angle
31.
32.
33.
34.
35.
36.
37.
glaucoma: correlation between Schlemm’s canal dimensions and outflow facility. Exp Eye Res 1996;62:101–09. Shin DH, Kim YY, Sheth N et al. The role of adjunctive mitomycin C in secondary glaucoma triple procedures as compared to primary glaucoma triple procedure. Ophthalmology 1998;105:740–45. Carlson DW, Alward WL, Barad JP et al. A randomized study of mitomycin augmentation in combined phacemulsification and trabeculectomy. Ophthalmology 1997;104:719–24. Cohen JS, Greff LJ, Novack GD, Wind BE. A placebo-controlled, double-masked evaluation of mitomycin C in combined glaucoma and cataract procedures. Ophthalmology 1996;103:1934–42. El Sayyad F, Belmekki M, Helal M et al. Simultaneous subconjunctival and subscleral mitomycin-C application in trabeculectomy. Ophthalmology 2000;107:298–302. El Sayyad F, Helal M, El-Kholify H et al. Nonpenetrating deep sclerectomy versus trabeculectomy in bilateral primary openangle glaucoma. Ophthalmology 2000;107:1671–74. Sanchez E, Schnyder CC, Mermoud A. Résultats comparatifs de la sclérectomie profonde transformée en trabéculectomie et de la trabeculectomie classique. Klin Monatsbl Augenheilkd 1997;210:261–64. Mermoud A, Karlen ME, Schnyder CC et al. Nd:Yag goniopuncture after deep sclerectomy with collagen implant. Ophthalmic Surg Lasers 1999;30:120–25.
15 Implants in non-penetrating filtering surgery Corinne C Schnyder and Emilie Ravinet
To avoid a secondary collapse of the created intrascleral space after non-penetrating filtering surgery, surgeons have attempted to insert devices in the scleral space. These devices differ in material being used, size, shape, design, consistency, hydration ability, and whether or not absorbable.
Collagen implants The cylindrical collagen implant (AquaflowT, Staar Surgical AG, Nidau, Switzerland) is processed from lyophilized, highly purified, porcine scleral collagen, which is gamma sterilized and processed to be biologically inert so that there is no systemic reaction. Micro-
Figure 15.1 AquaflowT collagen glaucoma drainage device in dry state.
biological analyses on 30 g of raw collagen have not shown the presence of bacteria, virus, mycological organisms, or parasites. This cross-linked collagen-based biocompatible material measures about 4.0 0.5 0.5 mm in its dry state (Figs. 15.1 and 15.2). The water content of the hydrated device is 99%, and it is intended to swell to about two to three times its dry size as it absorbs fluids, after placement in the subscleral space and its subsequent exposure to ocular fluids (Fig. 15.3). Chiou and colleagues1,2 have reported ultrasound biomicroscopic (UBM) findings that are consistent with lowering of intraocular
Figure 15.2 AquaflowT collagen glaucoma drainage device sutured to scleral bed.
178
Implants in non-penetrating filtering surgery
Figure 15.3 AquaflowT collagen glaucoma drainage device in swollen state. Figure 15.5 Ultrasound biomicroscopy of filtering site after resorption of Aquaflow T collagen glaucoma drainage device 6 years after surgery.
Figure 15.4 Ultrasound biomicroscopy of collagen implant 3 months after surgery.
Figure 15.6 Square collagen implant. pressure (IOP) by aqueous filtration through the thin trabeculo-Descemet’s membrane to an area under the scleral flap that has been maintained patent by the collagen implant. By measuring both thickness and length of the collagen implant at different times postoperatively, complete resorption time of the collagen implant was found to be 6–9 months (Fig. 15.4). After complete resorption, the scleral space left remains patent and allows aqueous
flow from the anterior chamber to the subconjunctival space (Fig. 15.5). The collagen implant may also be shaped into a square (Fig. 15.6); it is then just put under the scleral flap without being sutured. Square implants are currently being studied to
Reticulated sodium hyaluronate implants
Figure 15.7 Ultrasound biomicroscopy of square collagen implant 2 weeks after surgery.
assess their safety and efficacy, taking into account IOP, visual acuity, complications, tolerance, and UBM findings (Fig. 15.7). The square shape may offer the possibility of an increased scleral bed; it also may be useful to prevent high aqueous outflow after trabeculoDescemet’s perforation.
Reticulated sodium hyaluronate implants Sourdille and colleagues3,4 have developed a biosynthetic crosslinked hyaluronate implant as an adjuvant to non-penetrating filtering surgery. This implant is available from Corneal Laboratoires, Paris, France. The choice of biosynthetic sodium hyaluronate decreases the risk of potential viral and uncon-
179
ventional infections compared to sodium hyaluronate of animal origin. Sodium hyaluronate is the sodium salt of hyaluronic acid, which is a glycosaminoglycan present in humans in the intercellular fundamental substance. Sodium hyaluronate was chosen because of its excellent biocompatibility and non-toxicity. Because of its rheological properties in solution, sodium hyaluronate is commonly used in ophthalmic surgery to protect ocular tissues and maintain the volume of the anterior chamber. Another use is to prevent adhesions from occurring between dissected or damaged tissue surfaces. However, in view of its rapid elimination, it had to be modified in order to prolong its presence in the tissues and its resistance to the various degradation phenomena. Sodium hyaluronate has therefore been crosslinked. Crosslinking is done by interchain grafting with polyfunctional crosslinking agents, which creates stable bonds between polymeric chains. This chemical crosslinking modifies the physical and rheological properties of sodium hyaluronate while preserving the biocompatibility of the basic polymer. Crosslinked sodium hyaluronate is a solid gel the consistency of which is a function of the crosslinking percentage—i.e. the number of interchain connections and therefore the network created. However, this gel, although insoluble in water, can inflate in contact with water. The gel is stable to heat and can therefore be autoclaved. The biocompatibility of the SK gel implants has been verified in tests of the sensibilization, cytotoxicity, genotoxicity and concentration in bacterial endotoxins. The implant has been used in rabbits to assess its efficacy and safety.5 The degradation of crosslinked gels is identical to the one of hyaluronic acid but slower. SK gel implants are sterile with a pH between 7 and 7.5. There are two types of implants that
180
Implants in non-penetrating filtering surgery Figure 15.8 SK gel 3.5 implant.
Figure 15.9 SK gel 4.5 implant.
can be placed in the scleral bed without being sutured: SK gel 3.5 and SK gel 4.5. SK gel 3.5 (Fig. 15.8) is an equilateral triangle 500 µm thick with 3.5 mm sides. The implant is inserted in the scleral bed, and is completely covered by the superficial scleral flap. SK gel 4.5 (Fig. 15.9) is a 4.5 3.0 0.5 mm trapezium 500 µm thick. The implant is inserted
under the superficial scleral flap but extends posteriorly beyond the scleral bed under the cover of the conjunctival flap. Prospective multicentre studies are currently being done to assess the safety and efficacy of these implants in humans, taking into account visual acuity, IOP, tolerance, complications, and UBM findings (Figs 15.10 and 15.11).
Figure 15.10 Ultrasound biomicroscopy of filtering site 1 day after SK gel 3.5 implant.
Figure 15.11 Ultrasound biomicroscopy of filtering site 3 months after surgery with SK gel 4.5 implant.
Poly-Megma implant
181
Figure 15.12 T-flux implant.
Poly-MegmaT implant T-flux implant (IOLtech Laboratoires, La Rochelle, France) is the first non-absorbable glaucoma implant; it should maintain a permanently patent intrascleral space. T-flux has a T-shape designed for optimal implantation, with a 4-mm arm length, 2.75 mm body height, and 0.1–0.3 mm thickness; the design of the drain creates an evacuating canal along the foot, in which there is a suturing hole. Each arm of the T-shape (Fig. 15.12) is inserted into one of the artificial openings of Schlemm’s canal, for better stability; it is then secured with a 10–0 nylon suture in the foot’s hole, to stabilize the drain in the scleral bed. Made of Poly-MegmaT, a highly hydrophilic acrylic, the T-flux also provides active draining by means of capillarity and osmosis. PolyMegma is a highly biocompatible material, it has low scarring effect, cannot be colonized by cells, and stimulates the formation of new aqueous veins. Theoretically, the T-flux implant should have all the expected qualities of a glaucoma
Figure 15.13 Ultrasound biomicroscopy of the T-flux implant 24 months after surgery.
implant: it is biocompatible and comfortable for the patient, creates a permanent intrascleral space (Fig. 15.13) and prevents adhesions between the scleral flap and the trabecular
182
Implants in non-penetrating filtering surgery
meshwork, keeps the artificial ostia of Schlemm’s canal open, and stabilizes the trabeculoDescemet’s membrane. These implants are under study at present and appear promising. However, they are slightly difficult to handle.
Viscoelastic substances Viscoelastic substances may also be used in non-penetrating filtering surgery. Sodium hyaluronate is a polysaccharide (disaccharide units linked by glycosidic bonds) that occurs naturally in the human body (e.g. in the vitreous body, aqueous humour, and on the corneal endothelium). The presence of sodium hyaluronate in the eye is necessary to maintain ocular homeostasis. Healon® (Pharmacia Corp., NJ, USA) is a sterile, biocompatible, non-pyrogenic, and non-inflammatory solution of the highly purified, high molecular weight fraction of sodium hyaluronate at a concentration of 1% (10 mg/mL) dissolved in a physiological buffer completely compatible with the eye. Healon GV ® has a higher molecular weight and the concentration of sodium hyaluronate is 1.4% (14 mg/mL). Except perhaps for Healon 5®, Healon GV® seems to be preferable to the other viscoelastics in respect of its better capacity to maintain space, act as a soft instrument, and travel through a small-bore canula. Healon GV® has greater dynamic viscosity, elasticity, rigidity, pseudoplasticity, and cohesion than other viscoelastics. 6,7 Healon 5® has an even higher molecular weight and the concentration of sodium hyaluronate is 2.3% (23 mg/mL). Healon®, Healon EV ® and Healon 5® may be placed in the scleral bed; both Healon® and Healon GV® are resorbed in 4–5 days and they may be used alone or in conjunction with another device.
Figure 15.14 Injection of Healon GV ® into the ostia of Schlemm’s canal.
During viscocanalostomy,8,9 Healon GV ® is injected with a special canula (AlconGrieshaber, Schaffhausen, Switzerland) into the two surgically created ostia of Schlemm’s canal, with the aim of dilating both the ostia
Figure 15.15 Ultrasound biomicroscopy of viscocanalostomy 12 months after surgery.
References and the canal (Fig. 15.14); it is then also placed in the scleral bed. Because subconjunctival filtration is not involved after viscocanalostomy as described by Stegmann,8 the superficial scleral flap is tightly sutured to keep the viscoelastic substance inside the scleral bed, thereby maintaining space (Fig. 15.15), and also to force the aqueous humor percolating through the trabeculo-Descemet’s membrane into the ostia of Schlemm’s canal. The use of other ophthalmic viscoelastics has not yet been reported in the published work in non-perforating filtering surgery.
183
References 1.
2.
3.
Conclusion
4.
To facilitate aqueous filtration after nonperforating filtering surgery, deep sclerectomy, or viscocanalostomy, different kinds of devices may be inserted in the scleral bed, to maintain the scleral space. Collagen implants during deep sclerectomy and Healon GV® in viscocanalostomy have been used for the longest time and results of studies are presented in corresponding chapters of this book. SK gel implants and the T-flux implant are under prospective clinical studies at present, and results of first studies are also presented in corresponding chapters of this book. Large prospective, comparative, multicentre studies, with long-term follow-up, will provide results showing the advantages and drawbacks of the above-mentioned implants.
5.
6.
7.
8.
9.
Chiou AGY, Mermoud A, Hédiguer SEA et al. Ultrasound biomicroscopy of eyes undergoing deep sclerectomy with collagen implant. Br J Ophtalmol 1996;80:541–44. Chiou AGY, Mermoud A, Underdahl JP, Schnyder CC. An ultrasound biomicroscopic study of eyes after deep sclerectomy with collagen implant. Ophtalmology 1998;105:746–50. Sourdille P, Santiago PY, Ducournau Y. Non perforating surgery of the trabeculum with reticulated hyaluronic acid implant: why, how, what results? J Fr Ophtalmol 1999;22:794–97. Sourdille P. Non penetrating trabecular surgery: it’s worth the change. J Cataract Refract Surg 1999;25:298–300. Sourdille P, Santiago PY, Villain F et al. Reticulated hyaluronic implant in nonperforating trabecular surgery. J Cataract Refract Surg 1999;25:332–39. Hütz WW, Eckhardt B, Kohnen T. Comparison of viscoelastic substances used in phacoemulsification. J Cataract Refract Surg 1996;22:955–59. Arshinoff S. The physical properties of ophthalmic viscoelastics in cataract surgery. Ophthalmic Pract 1991;9:2–7. Stegmann RC. Viscocanalostomy for open angle glaucoma. An Inst Barraquer Spain 1995;25:229–32. Stegmann RC, Pienaar A, Miller D. Viscocanalostomy for open angle glaucoma in black African patients. J Cataract Refract Surg 1999;25:323–31.
16 Erbium:YAG laser-assisted deep sclerectomy Wolfgang E Lieb
The Erbium:YAG laser has a wavelength of 2 940 nm. Being close to the maximal absorption of water, the Erbium:YAG laser is ideal for ablating tissue with a high water content. In ophthalmology, several applications of the Erbium:YAG lasers have been tried or are currently under investigation: corneal ablation in refractive surgery,1–8 lacrimal intervention with or without endoscopy,9,10 plastic and cosmetic eyelid surgery with skin resurfacing,11–14 cataract surgery8,15–18 and vitrectomy.8,18,19–23 Three major areas of glaucoma surgery have been investigated with various lasers, especially the 2.94 µm Erbium:YAG laser, over the last decade: penetrating laser sclerostomy ab externo and ab interno; trabecular ablation ab interno; lamellar sclerectomy ab externo. The pulsed Erbium:YAG laser has the advantage of being capable of very fine selective tissue ablation without creating significant thermal damage. Almost a decade ago several authors started to investigate the use of the Holmium laser,24 Nd:YAG25,26 Erbium:YAG laser27–29 and the Excimer laser30–32 in an attempt to create a minimally invasive glaucoma procedure. This procedure—laser sclerostomy—has been performed ab interno and ab externo, but long-term follow-up showed it to be inferior to standard trabeculectomy. Complications were initial hypotony (to be expected in non-
guarded full-thickness filtering procedures), obliteration of the ostium with iris prolapse, late failure due to scar formation at the level of Tenon’s capsule and conjunctiva and scarring of the sclerostomy site itself.33 Therefore, some investigators have advocated the routine use of antimetabolites 5-fluorouracil and mitomycin C with Nd:YAG and Erbium:YAG laser sclerostomy ab externo.34,35 In a study by Hill et al36 laser trabecular ablation was attempted with the Erbium:YAG (2.94 µm), Erbium:YSGG (2.79 µm), and Holmium:YSGG (2.1 µm) lasers. They found that the Erbium:YAG-laser (2.94 µm) is most suitable for the ablation of fine trabecular structures. Some investigators have shown ab interno ablation of the trabecular meshwork in animals and humans to be effective.37–44 Initial preclinical and clinical studies by Funk and Schlunck have supported this.45 Deep sclerectomy is becoming more and more popular as a non-penetrating filtering procedure. The technique was first introduced by Krasnov (sinusotomy) in 196946,47 followed by various modifications by Zimmerman and colleagues in 198448,49 and others later.50,51 The results are controversial. Several studies have shown that the outcome of deep sclerectomy is similar, and others, that it is even more successful compared with standard trabeculectomy in the treatment of open-angle glaucoma. All investigators have
186
Erbium:YAG laser-assisted deep sclerectomy
described fewer complications with the nonpenetrating technique.48,50,52,53 An 83.7% success rate with deep sclerectomy compared to 70.3% with trabeculectomy after 1 year was published by Zimmerman et al,48 who described the standard operation that is mainly used today. Stegmann et al51 completed deep sclerectomy by viscocanalostomy, publishing long-term success rates without medication of more than 80% in black African patients with primary openangle glaucoma. For several years collagen implants were used to prevent or reduce scarring postoperatively. Most investigators could show an increased success rate with a collagen drainage device and/or high-viscosity sodium hyaluronate.54–58 Seiler et al59 performed a partial excimer laser external trabeculectomy, which we based our own work on. To simplify the technically difficult procedure of deep sclerectomy with or without viscocanalostomy, we started to use the Erbium:YAG laser as a surgical tool to ablate scleral and corneal tissue layer by layer. We decided to use a pulsed Erbium:YAG laser Adagio® (Wavelight AG, Erlangen, Germany) —which has a narrow thermal damage zone and therefore lower risk of postoperative scarring—in animal cadaver eyes and then in clinical studies for the ablation of the deep corneoscleral window (Fig. 16.1).60,61 In 2000, a different approach to a laser-assisted partial thickness procedure, the laser grid trabeculectomy, was described by Jacobi et al. 62
Initial cadaver eye studies Using cadaver pig eyes, we surgically prepared a superficial scleral flap with an area of 5 5 mm as for a trabeculectomy procedure. The next step was the preparation of the deep
Figure 16.1 Zirconium fluoride fiberoptic with a quartz fiber tip (320 µm core diameter). 61
lamella with an area of 4 3 mm. Preparation was done with a pulsed Erbium:YAG Laser (Aesculap Meditec, Jena, Germany; model: modified Phacolase and MCL 29 Dermablate). The Phacolase works with a wavelength of 2 940 nm, a pulse duration of 200 µs, an energy of 10–180 mJ and a frequency of 10–100 Hz. The MCL 29 Dermablate has the same wavelength, a pulse duration of 250 µs, an energy of 100–1 200 mJ, and a frequency of 1–15 Hz. We ablated the scleral fibers of the deep lamella with an energy of 40–100 mJ, a frequency of 1–10 Hz and spot sizes of 500 µm and 1 mm (Fig. 16.2). A zirconium fluoride fiberoptic with a quartz fiber tip was used (Fig. 16.1) in combination with the Phacolase and a mirror joint arm with a quartz fiber tip in combination with the MCL 29 Dermablate. The beam was divergent. For maintaining intraocular pressure (IOP) at 40–60 mmHg an infusion into the anterior chamber was used. The eyes were histologically analysed (Fig. 16.3).
Initial cadaver eye studies
187
Figure 16.2 Ablation of the deep scleral lamella in a pig eye with the Erbium:YAG laser.61
Figure 16.3 Low-power photomicrograph of the iridocorneal angle after Erbium:YAG laser ablation of the deep lamella (hematoxylin and eosin, original magnification 10).61
With increasing experience it was possible to ablate the remaining deep corneoscleral lamella with the Erbium:YAG laser without penetrating the anterior chamber. The mean thickness of the sclera after preparing the superficial flap was 400 ± 120 µm. After laser ablation the remaining trabeculo-Descemet’s membrane was 220 ± 40 µm in the cadaver pig eyes. An energy of 70–85 mJ was used to ablate the harder tissue (sclera). 40–60 mJ was
used when deeper layers (trabecular meshwork and cornea, especially Descemet’s membrane) were reached. Ablation was stopped when the ablation reached Descemet’s membrane and the trabecular meshwork was visible. Because the wavelength of the Erbium:YAG laser is 2 940 nm, which is near the absorption maximum of water, the ablation stopped when the percolation of aqueous started. With spot
188
Erbium:YAG laser-assisted deep sclerectomy
sizes of 500 µm and 1 mm, we needed to continue ablating to get a continuous percolation in all areas of the inner lamella. We, therefore, had to dry the layers where the percolation had started with a swab and continue with the laser treatment until the area (4 3 mm) had the same depth, which was the end point of our operation. When we increased the energy to 100 mJ larger amounts of tissue were ablated, but control was more difficult. A smaller spot size of 500 µm resulted in a more precise preparation of the deep lamella. The ablation rate is safely controllable with frequencies of 1–10 Hz, with 10 Hz it is less time consuming. The laser procedure took a mean time of 5 min 42 s (±1 min 44 s).
Surgical technique in humans
Figure 16.4 Intraoperative view at Schlemm’s canal during laser removal of the outer wall.
• •
In a pilot study with 12 patients (eight men and four women with an average age of 67.3 ± 11 years) we performed Erbium:YAG laser-assisted deep sclerectomy (seven patients had pseudoexfoliation glaucoma, four had primary open-angle glaucoma and one patient had pigment-dispersion glaucoma). Surgery consists of: • • • • •
•
Fornix-based conjunctival flap 8 mm Minimal episcleral cautery Dissection of a superficial corneal groove Dissection of a 6 4 mm rectangular half thickness lamellar scleral flap Erbium:YAG laser ablation of deep sclerectomy site with deroofing of Schlemm’s canal, 43 mm (Fig. 16.4) Healon GV® or Healon 5® deposit at the sclerectomy site
Closure of scleral flap with three 10.0 nylon sutures Conjunctival closure with a running 10.0 nylon suture
We used the Erbium:YAG laser Adagio®, which has a wavelength 2 940 nm, pulse energy up to 100 mJ, pulse duration between 50 and 250 µs, and a pulse repetition frequency up to 100 Hz. A modified phacoemulsification handpiece with a 320 µm articulated zirconium fluoride fiber (ZrF4) and spot size of 500 µm proved to be most convenient for surgery (Fig. 16.1). For corneoscleral ablation we needed on average 1054 ± 251 laser pulses. The initial preparation was done with 85 mJ and 10 Hz, which was reduced to 60 mJ 10 Hz during deeper ablation. Total laser time was about 5 min (mean 319 ± 112 s) and the applied total energy was 58 ± 31 J. The preliminary results of IOP control are shown in Figure 16.5.
Conclusion
Figure 16.5 Preoperative and postoperative IOP after Erbium:YAG laser-assisted deep sclerectomy in 12 patients. (d) = day.
Conclusion Postoperative complications like choroidal detachment, corneal edema, or bleb encapsulation do occur in non-penetrating glaucoma surgery,63,64 however, most investigators report a lower complication rate,52,58,63–65 especially in the immediate postoperative period. There is less postoperative hyphema, shallow anterior chamber, anterior chamber inflammation, hypotony, and choroidal detachment. Increased resistance to aqueous outflow through the area of the trabecular meshwork and Descemet’s membrane may occur,64,66 which may be present immediately after deep sclerectomy as a result of insufficient surgical dissection and removal of the outer trabecular meshwork or later caused by fibrosis of the trabeculo-Descemet’s membrane.66 In these cases a Nd:YAG goniopuncture is necessary,
189
which transforms the non-penetrating procedure into a penetrating operation. Although the mechanism of action for the non-penetrating surgery is not only subconjunctival filtration but transscleral drainage, one advantage of non-penetrating surgery is the potentially safer use of antimetabolites.56 On the other hand, conjunctival and Tenon’s scar formation occurs, as in standard trabeculectomy. The main difficulty of learning the technique is preparation of the deep lamella and unroofing of Schlemm’s canal, without perforating the trabeculo-Descemet’s window. Here the Erbium:Yag laser might offer an advantage over manual preparation. Our ongoing study in patients with openangle glaucoma shows that Schlemm’s canal can act as a landmark, even with the Erbium:YAG laser. We could see a reflux of blood when we ablated the roof of Schlemm’s canal. The reflux of aqueous humor and blood stops the laser ablation, so that only the roof is destroyed and the canal itself is still visible as a landmark. After initial trials with a dermatological Erbium:YAG laser—using the modified phacolaser Adagio (Wavelight AG)—the procedure has become much more practical. The handling is easier with a flexible fiber optic; a small handpiece and lower energy levels can be used. With a frequency of 10 Hz the procedure is less time consuming and still safe. In our hands a spot size of 500 µm is ideal for the delicate preparation. With regard to costs, the laser technique certainly cannot compete with conventional microsurgical procedure. The use of an Erbium:YAG laser for this non-penetrating surgery is of course more expensive in comparison with a standard surgical deep sclerectomy, but the laser should be regarded as a multifunctional instrument. The Erbium phaco laser Adagio (Wavelight AG) that we
190
Erbium:YAG laser-assisted deep sclerectomy
use is also suitable for cataract surgery, anterior and pars plana vitrectomy, and periorbital plastic surgery. Erbium:YAG laser-assisted deep sclerectomy offers a safe and efficient alternative to microsurgical preparation of the deep corneoscleral lamella. The laser may help to reduce the number of unintended perforations during preparation of the deep corneoscleral window and deroofing of Schlemm’s canal. The ongoing study in patients with open-angle glaucoma will answer any concerns about the efficacy and safety of this modification of deep sclerectomy.
References 1. Behrens A, Küchle M, Seitz B et al. Stromal thermal effects induced by nonmechanical (2.94–µm) Erbium:YAG laser corneal trephination. Arch Ophthalmol 1998;116:1342–48. 2. Belgorod BM, Ediger MN, Weiblinger RP, Erlandson RA. Tangential corneal surface ablation with 193- and 308-nm excimer and 2936-nm Erbium-YAG laser irradiation. Arch Ophthalmol 1992;110:533–36. 3. Mrochen M, Meinhard FP, Semshichen V et al. Zur Photoablation der Hornhaut mit dem Erbium:YAG laser. Ophthalmologe 1999;96:387–91. 4. Mrochen M, Semshichen V, Funk RH, Seiler T. Limitations of Erbium:YAG laser photorefractive keratectomy. J Refract Surg 2000;16:51–59. 5. Peyman GA, Badaro RM, Khoobehi B. Corneal ablation in rabbits using an infrared (2.9-microns) Erbium:YAG laser. Ophthalmology 1989;96:1160–70. 6. Peyman GA, Beyer C, Kuszak J et al. Longterm effect of Erbium-YAG laser (2.9 microns) on the primate cornea. Int Ophthalmol 1991;15:249–58.
7. Peyman GA, Katoh N. Effects of an Erbium:YAG laser on ocular structures. Int Ophthalmol 1987;10:245–53. 8. Tsubota K. Application of Erbium:YAG laser in ocular ablation. Ophthalmologica 1990;200:117–22. 9. Kuchar A, Novak P, Pieh S et al. Endoscopic laser recanalisation of presaccal canalicular obstruction. Br J Ophthalmol 1999;83:443–47. 10. Meyer-Rüsenberg H, Emmerich KH, Luchtenberg M, Steinhauer J. Endoskopische Laserdakryoplastik. Methodik und Ergebnisse nach drei Monaten. Ophthalmologe 1999;96:332–34. 11. Alster TS. Clinical and histologic evaluation of six Erbium:YAG lasers for cutaneous resurfacing. Lasers Surg Med 1999;24:87–92. 12. Perez MI, Bank DE, Silvers D. Skin resurfacing of the face with the Erbium:YAG laser. Dermatol Surg 1998;24:653–58. 13. Ratner D, Tse T, Marchell N et al. Cutaneous laser resurfacing. J Am Acad Dermatol 1999;41:365–89. 14. Stuzin JM, Baker TJ, Baker TM. CO2 and Erbium:YAG laser resurfacing: current status and personal perspective. Plast Reconstr Surg 1999;103:588–91. 15. Gailitis RP, Patterson SW, Samuels MA et al. Comparison of laser phacovaporization using the Er-YAG and the Er-YSGG laser. Arch Ophthalmol 1993;111:697–700. 16. Höh H, Fischer E. Pilot study on Erbium laser phacoemulsification. Ophthalmology 2000;107:1053–61. 17. Ross BS, Puliafito CA. Erbium-YAG and Holmium-YAG laser ablation of the lens. Lasers Surg Med 1994;15:74–82. 18. Stevens GJ, Long B, Hamann JM, Allen RC. Erbium:YAG laser-assisted cataract surgery. Ophthalmic Surg Lasers 1998;29:185–9. 19. Margolis TI, Farnath DA, Destro M, Puliafito CA. Erbium-YAG laser surgery on experimental vitreous membranes. Arch Ophthalmol 1989;107:424–28. 20. Mrochen M, Petersen H, Wullner C, Seiler T.
References
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
Experimentelle Ergebnisse zur Erbium:YAG Laser Vitrektomie. Klin Monatsbl Augenheilkd 1998;212:50–54. Mrochen M, Riedel P, Kempe A, Seiler T. Erbium:YAG laser Vitrektomie: Temperaturmessungen in verschiedenen Austauschmedien. Ophthalmologe 2000;97:181–85. Petersen H, Mrochen M, Seiler T. Comparison of Erbium:yttrium-aluminumgarnet-laser vitrectomy and mechanical vitrectomy: a clinical study. Ophthalmology 2000;107:1389–92. Wesendahl T, Janknecht P, Ott B, Frenz M. Erbium:YAG laser ablation of retinal tissue under perfluorodecaline: determination of laser-tissue interaction in pig eyes. Invest Ophthalmol Vis Sci 2000;41:505–12. Schuman JS, Stinson WG, Hutchinson BT et al. Holmium laser sclerectomy. Success and complications. Ophthalmology 1993;100:1060–65. March WF, Gherezghiher T, Koss MC, Nordquist RE. Experimental YAG laser sclerostomy. Arch Ophthalmol 1984;102:1834–36. March WF, Gherezghiher T, Koss MC et al. Histologic study of a neodymium-YAG laser sclerostomy. Arch Ophthalmol 1985;103:860–63. Wetzel W, Haring G, Brinkmann R, Birngruber R. Laser sclerostomy ab externo using the Erbium:YAG laser: first results of a clinical study. Ger J Ophthalmol 1994;3:112–25. Wetzel W, Otto R, Falkenstein W et al. Development of a new Er:YAG laser conception for laser sclerostomy ab externo: experimental and first clinical results. Ger J Ophthalmol 1995;4:283–88. Wetzel W, Schmidt EU, Haring G et al. Laser sclerostomy ab externo using two different infrared lasers: a clinical comparison. Ger J Ophthalmol 1995;4:1–6. Brooks AM, Samuel M, Carroll N et al. Excimer laser filtration surgery. Am J Ophthalmol 1995;119:40–47.
191
31. Campos M, Lee PP, Trokel SL et al. Transconjunctival sinusotomy using the 193nm excimer laser. Acta Ophthalmol Copenh 1994;72:707–11. 32. Traverso CE, Murialdo U, Di LG et al. Photoablative filtration surgery with the excimer laser for primary open-angle glaucoma: a pilot study. Int Ophthalmol 1992;16:363–65. 33. Jacobi PC, Dietlein TS, Krieglstein GK. Prospective study of ab externo Erbium:YAG laser sclerostomy in humans. Am J Ophthalmol 1997;123:478–86. 34. Schmidbauer JM, Höh H, Jahnig T, Daberkow I. Antiproliferative Therapie mit 5Fluorouracil bei Erbium:YAG laser Sklerostomie ab externo. Ophthalmologe 1996;93:569–75. 35. Schmidt EU, Wetzel W, Droge G, Birngruber R. Mitomycin-C in laser sclerostomy: benefit and complications. Ophthalmic Surg Lasers 1997;28:14–20. 36. Hill RA, Baerveldt G, Ozler SA et al. Laser trabecular ablation (LTA). Lasers Surg Med 1999;11:341–46. 37. Dietlein TS, Jacobi PC, Krieglstein GK. Erbium:YAG laser ablation on human trabecular meshwork by contract delivery endoprobes. Ophthalmic Surg Lasers 1996;27:939–45. 38. Dietlein TS, Jacobi PC, Krieglstein GK. Ab interno infrared laser trabecular ablation: preliminary short-term results in patients with open-angle glaucoma. Graefes Arch Clin Exp Ophthalmol 1997;235:349–53. 39. Dietlein TS, Jacobi PC, Krieglstein GK. Erbium:YAG laser trabecular ablation (LTA) in the surgical treatment of glaucoma. Lasers Surg Med 1998;23:104–10. 40. Dietlein TS, Jacobi PC, Schröder R, Krieglstein GK. Experimental Erbium:YAG laser photoablation of trabecular meshwork in rabbits: an in-vivo study. Exp Eye Res 1997;64:701–06. 41. Jacobi PC, Dietlein TS, Krieglstein GK. Effects of Er:YAG laser trabecular ablation
192
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
Erbium:YAG laser-assisted deep sclerectomy
on outflow facility in cadaver porcine eyes. Graefes Arch Clin Exp Ophthalmol 1996;234(suppl 1):S204–8. Jacobi PC, Dietlein TS, Krieglstein GK. Photoablative Nd:YAG-laser goniotomy enhancing trabecular outflow facility in porcine cadaver eyes. Ger J Ophthalmol 1996;5:154–59. McHam ML, Eisenberg DL, Schuman JS, Wang N. Erbium:YAG laser trabecular ablation with a sapphire optical fiber. Exp Eye Res 1997;65:151–5. McHam ML, Eisenberg DL, Schuman JS, Wang N. Erbium:YAG laser sclerectomy with a sapphire optical fiber. Ophthalmic Surg Lasers 1997;28:55–58. Funk J, Schlunck G. Endoskopisch kontrollierte Erbium-YAG-laser-Goniotomie. Erste praeklinische Versuche. Ophthalmologe 1998;95:33–36. Krasnov M. Microsurgery of glaucoma: indications and choice of technique. Am J Ophthalmol 1969;67:857–64. Krasnov MM. Current technic of sinusotomy (externalization of Schlemm’s canal) without resection of the sclera. Vestn Oftalmol 1988;104:102–104. Zimmerman TJ, Kooner KS, Ford VJ et al. Trabeculectomy vs nonpenetrating trabeculectomy: a retrospective study of two procedures in phakic patients with glaucoma. Ophthalmic Surg 1984;15:734–40. Zimmerman TJ, Kooner KS, Ford VJ et al. Effectiveness of nonpenetrating trabeculectomy in aphakic patients with glaucoma. Ophthalmic Surg 1984;15: 44–50. Carassa RG, Bettin P, Brancato R. Viscocanalostomy: a pilot study. Acta Ophthalmol Scand Suppl 1998;23:51–2. Stegmann R, Pienaar A, Miller D. Viscocanalostomy for open-angle glaucoma in black African patients. J Cataract Refract Surg 25:1999;316–22. Mermoud A, Schnyder CC, Sickenberg M et al. Comparison of deep sclerectomy with
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
collagen implant and trabeculectomy in openangle glaucoma. J Cataract Refract Surg 1999;25:323–31. Schwenn O, Dick B, Pfeiffer N. Trabekulotomie, tiefe Sklerektomie und Viskokanalostomie. Nicht fistulierende mikrochirurgische Glaukomoperationen ab externo. Ophthalmologe 1998;95:835–43. Chiou AG, Mermoud A, Jewelewicz DA. Post-operative inflammation following deep sclerectomy with collagen implant versus standard trabeculectomy. Graefes Arch Clin Exp Ophthalmol 1998;236:593–96. Chiou AG, Mermoud A, Underdahl JP, Schnyder CC. An ultrasound biomicroscopic study of eyes after deep sclerectomy with collagen implant. Ophthalmology 1998;105:746–50. Demailly P, Lavat P, Kretz G, Jeanteur LM. Non-penetrating deep sclerectomy (NPDS) with or without collagen device (CD) in primary open-angle glaucoma: middle-term retrospective study. Int Ophthalmol 1996;20:131–40. Sanchez E, Schnyder CC, Sickenberg M et al. Deep sclerectomy: results with and without collagen implant. Int Ophthalmol 1996;20:157–62. Welsh NH, DeLange J, Wasserman P, Ziemba SL. The “deroofing” of Schlemm’s canal in patients with open-angle glaucoma through placement of a collagen drainage device. Ophthalmic Surg Lasers 1998;29:216–26. Seiler T, Kriegerowski M, Bende T, Wollensak J. Partielle externe Trabekulektomie. Klin Monatsbl Augenheilkd 1989;195:216–20. Klink T, Lieb W, Grehn F. Erbium:YAG laser-assisted deep sclerectomy. Invest Ophthalmol Vis Sci 1999;40:272. Klink T, Lieb W, Grehn F. Erbium:YAG laser-assisted preparation of deep sclerectomy. Graefes Arch Clin Exp Ophthalmol 2000;238:792–96. Jacobi PC, Dietlein TS, Colling T, Krieglstein
References GK. Photoablative laser grid trabeculectomy in glaucoma filtering surgery: histology and outflow facility measurements in porcine cadaver eyes. Ophthalmic Surg Lasers 2000;31:49–54. 63. Gianoli F, Schnyder CC, Bovey E, Mermoud A. Combined surgery for cataract and glaucoma: phacoemulsification and deep sclerectomy compared with phacoemulsification and trabeculectomy. J Cataract Refract Surg 1999;25:340–46. 64. Karlen ME, Sanchez E, Schnyder CC et al.
193
Deep sclerectomy with collagen implant: medium term results. Br J Ophthalmol 1999;83:6–11. 65. Chiou AG, Mermoud A, Hediguer SE et al. Ultrasound biomicroscopy of eyes undergoing deep sclerectomy with collagen implant. Br J Ophthalmol 1996;80:541–44. 66. Mermoud A, Karlen ME, Schnyder CC et al. Nd:YAG goniopuncture after deep sclerectomy with collagen implant. Ophthalmic Surg Lasers 1999;30:120–25.
Index
Amsler sign 145 anaesthesia 98, 109 angle, normal anatomy 67 angle-recession glaucoma, indications/contraindications for NPFS 94 animal models, experimental studies in NPFS 68 aniridia syndromes, indications/contraindications for NPFS 92 anterior chamber 25 shallow, grades 128 shallow or flat, postoperative 149–50 anterior chamber angle 25, 26 anterior chamber flare, DSCI vs trabeculectomy 146 anterior segment dysgenesis 91, 92 anti-inflammatories 126 antibiotics 126 antifibrotic agents 127 antimetabolites 120–1, 173 needling 131 wound healing 118 aphakic and pseudophakic glaucoma, indications/contraindications for NPFS 90–1 apraclonidine 130 aqueous humor drainage pathways conventional (trabecular) 21, 34–44 unconventional (uveoscleral) 21, 34, 44 experimental dynamics after NPFS 69–74 production 22–3 resorption, filtration mechanisms post NPFS 60–4 aqueous outflow pathway imaging, rabbit 82–3 aqueous veins 43 aqueous-outflow resistance 21, 33–44 abnormal in glaucomatous eye 43–4 drainage pathways 34–43 endothelium 41–3 see also Schlemm’s canal; trabeculo-Descemet’s membrane future goals 48–9 measurement, schema 69 artificial drainage shunt, at reoperation 158 Axenfeld’s anomaly 91, 92
bleb, postoperative blebitis 149, 156 encysted 152 thin, ischemic 153 ultrasonic biomicroscopy (UBM) 152 see also specific types Bruch’s muscle 24 Bruch’s membrane 24 cataract postoperative absence 150 progression, postoperative 156 surgery see phacoemulsification surgery-induced 164–5 CGS19755, NMDA receptor blockade 135 choroidal detachment, postoperative 147–8 ciliary body muscle 24–5 processes 24 vessels 23, 24 collagen implants in NPFS 120, 163–4, 177–9 results with/without implant 165–6 surgical technique 103–4 collector channels 43 and aqueous veins 30, 43 external 30–1 internal 30 combined surgery 166–7 phacoemulsification and deep sclerectomy 167 and viscocanalostomy 166–7 complications see peroperative complications; postoperative complications, early and late congenital glaucoma, indications/contraindications for NPFS 91 conjunctival bleb, fibrosis, postoperative 151–2 conjunctival inflammation, wound healing 118 conjunctival veins 31 connective tissue stage, wound healing 117 contraindications for NPFS see indications/contraindications for NPFS corneoscleral meshwork 27, 37–8
196
Index
corneoscleral regions 37–8 corticosteroids 126 cycloablation therapy 158 cycloplegics 125–6 prophylactic 143 cytochalasins 42 decompression retinopathy 150 deep sclerectomy 162–4 aqueous-outflow resistance 47 with collagen implant (DSCI) 120, 163–4 long, short and medium-term results 163–4 vs trabeculectomy, complications 145, 151, 164 with/without implant, results 165–6 with collagen implant, experimental 74–8 anterior-segment angiographies after surgery 77 comparative study with/without collagen implant 74–7 ultrasonic biomicroscopy after surgery 78 without collagen implant 162–3 combined with phacoemulsification 167 augmented with antimetabolites 173–4 converted to trabeculectomy 142, 158, 161 dissection depth, experimental 79–80 erbium:YAG laser-assisted 185–93 experimental, compared with trabeculectomy and sinusotomy 73 first, surgical technique and results 17–18 incisions 97–8 intact Descemet’s membrane 47 morphology of deep scleral flap 79–80 with phacoemulsification 167 and viscocanalostomy 169–76 principle 47 results, with/without implant 165–6 surgical technique 98–100 laser-assisted 185–93 variations 33 Dellen formation 149 Descemet’s membrane anatomy 15, 17 with anterior trabeculum, surgical technique 101–2 detachment, postoperative 154–6 function 33 see also trabeculo-Descemet’s membrane Descemet’s space, retro 155 Descemet’s window, creation, viscocanalostomy 112 digital pressure 134 drainage shunt, at reoperation 158 DSCI see deep sclerectomy, with collagen implant
electrocoagulation cautery 98, 143 endophthalmitis antibiotics 126 postoperative 156 endothelia, hydraulic conductivities 41 endothelial vacuolation cycle 29 endothelium, outflow resistance 41–3 erbium:YAG laser-assisted deep sclerectomy 185–93 initial cadaver eye studies 186–8 surgical technique in humans 188–9 evolution of NPFS 13–20 experimental studies in NPFS 67–85 aqueous outflow pathway imaging in rabbit 82–3 deep sclerectomy anterior trabeculum and Descemet’s membrane pathway 69–71 dissection depth 79–80 external trabecular meshwork 80–1 models 68 outflow facility measurements in rabbit 83 posterior pathway after ab externo trabeculectomy 71–4 extracellular matrix juxtacanalicular connective tissue 39 role in juxtacanalicular connective tissue (JCT) 39 felbamate, NMDA receptor blockade 135 fibroblastic proliferative stage, wound healing 79, 117 fibrosis postoperative conjunctival bleb 151–2 postoperative scleral flap 98, 100 filtering bleb formation encystment 129–30 intrascleral 60, 61–3 post deep sclerectomy 97, 127 post trabeculectomy 45 post viscocanalostomy 18 sub-conjunctival 60 filtration devices 158 filtration mechanisms post NPFS 57–65 aqueous humor resorption 60–4 intrascleral bleb 61–3 Schlemm’s canal unroofing 64 subchoroidal space 63 subconjunctival bleb 60–1 flow through trabeculo-Descemet’s membrane 58–60 filtration membrane formation post ab externo trabeculectomy 67 formation post deep sclerectomy 68
Index 5-fluorouracil 127, 173–4 vs mitomycin C, wound healing 119–20 Fuchs’ heterochromia 145 glycosaminoglycans 39 Goldman equation, outflow facility 83 goniocurettage 35, 48 goniotomy 35, 48 with anterior incision 48 IOP before/after 131 Nd:YAG laser 132–4, 152–3 healing see wound healing Healon GV see hyaluronate Hema implant 64, 105 hemorrhages, peroperative 143–4 high myopia, indications/contraindications for NPFS 88–9 history of filtering surgery 1–11, 13–20 evolution of NPFS 13–20 deep sclerectomy 17–18 viscocanalostomy 18 iridectomy 1–2 iridencleisis 4 iridodesis 2 limbal trephination 4–5 posterior trephination 4 sclerotomy 2–3 sinusotomy 7–8, 13–14 trabeculectomy, non-penetrating 10, 15–17 trabeculotomy 7, 8–9 hyaluronate corneal inclusion 155, 156 implants, experimental 78–9 injection 104–5, 111–12, 172 reticulated, implants 179–80 hydraulic conductivity, endothelia 41 hyphema, postoperative 145 hypotony chronic, late postoperative 156 postoperative, early 148 implants in NPFS 177–83 benefits 82 collagen 177–9 see also deep sclerectomy, with collagen implant displacement 150 Hema 64, 105 Poly-Megma 181–2 reticulated sodium hyaluronate 179–80
197
viscoelastic substances 182–3 incisions 97–8 indications/contraindications for NPFS 87–95 absolute contraindications 94 aniridia and anterior segment dysgenesis syndromes 92 glaucoma aphakic and pseudophakic 90–1 congenital and juvenile 91 with high myopia 88–9 narrow-angle 93 neovascular 94 open-angle 87–8 pigmentary 89–90 post-trauma angle-recession 94 pseudoexfoliation 90 secondary to uveitis 93 relative contraindications 93 status post-laser trabeculoplasty 93–4 Sturge–Weber syndrome 91–2 infants, goniotomy 35, 48 infection, postoperative 149 inflammation, anterior chamber flare, DSCI vs trabeculectomy 145–6 inflammatory stage, wound healing 117 instruments 107–8 intraocular bleeding 143–4 intraocular lens 172 intraocular pressure (IOP) 21, 152–3 and choroidal detachment 147–8 continuous recording during deep sclerectomy 59 experimental studies 74–6 facility of outflow 21, 43 hypotony 148, 156 increased postoperatively 152–3 intrascleral bleb aqueous humor resorption 61–3 promotion 97 iridectomy 1–2 post iris prolapse 105–6 iridencleisis 4 iridodesis 2 iris, aniridia 92 iris atrophy 92 iris prolapse 140, 141, 153–4, 174 juvenile glaucoma, indications/contraindications for NPFS 91 juxtacanalicular tissue (JCT) 28, 37, 38, 39 outflow resistance and role of ECM 39
198
Index
Lagrange sclerecto-iridectomy 3, 4 laminated vein of Goldman 30 laser goniopuncture 132–4, 152–3, 174 laser grid trabeculectomy 186 laser sclerostomy 9–10 laser trabeculoplasty 36–7, 44–5, 93–4 laser-assisted deep sclerectomy 185–93 initial cadaver eye studies 186–8 surgical technique in humans 188–9 latrunculins 42 limbal trephination 4–5 limbus, structures 26 malignant glaucoma, peroperative/postoperative 142–3 medications, postoperative 106–8, 125–6, 143 meridional fibers see Bruch’s muscle mitomycin C 119, 120, 127, 173 wound healing 119–20 models, experimental studies in NPFS 68 Muller’s muscle 24 mydriatics, prophylactic 143 myopia, high, NPFS 88–9 narrow-angle glaucoma 93 Nd:YAG laser goniotopuncture 132–4, 152–3, 174 needling revision 129–31 neovascular glaucoma, absolute contraindications for NPFS 94 neuroprotective strategies 135 NMDA receptor blockade 135 non-penetrating filtering surgery (NPFS) see deep sclerectomy; results of NPFS; surgical method; trabeculectomy NSAIDs 126 ocular hypertension, postoperative 148–9 ocular massage (digital pressure) 134 open-angle glaucoma 87–8 ornipressin 110 outflow facility, Goldman equation 83 outflow facility measurements, rabbit 83 outflow pathway anatomy 21–32 anterior chamber 25 ciliary body 23–5 collector channels 30–1 imaging in rabbit 82–3 innervation 31 posterior chamber 25 production of aqueous humor 22–3
Schlemm’s canal 28–9 Schwalbe’s line 27 scleral spur 25–7 trabecular meshwork 27–8 see also aqueous-outflow resistance outflow pathway imaging, rabbit 82–3 outflow resistance see aqueous-outflow resistance paracentesis, pre viscocanalostomy 101, 111 perforation of Descemet’s membrane see trabeculoDescemet’s membrane peripheral anterior synechia (PAS) 156 peroperative complications hemorrhages 143–4 malignant glaucoma 142–3 perforation of trabeculo-Descemet’s membrane 105–6, 140–2 phacoemulsification combined with augmented deep sclerectomy 173–4 combined with deep sclerectomy 167, 169–76 surgical technique 169–73 combined with viscocanalostomy 166–7 complications 146, 174 surgical technique 169–73 same-site vs separate-site 171 phacotrabeculectomy 170 pigmentary glaucoma 89–90 indications/contraindications for NPFS 89–90 pilocarpine 43 POAG, see also aqueous-outflow resistance Poiseuille’s law 37, 43 Poly-Megma implants in NPFS 181–2 posterior chamber 25 postoperative complications, early 145–51 cataract formation 150 choroidal detachment 147–8 decompression retinopathy 150 decreased visual acuity 150 Dellen formation 149 hyphema 145 hypotony 148 implant displacement 150 infection and blebitis 149 inflammation 145–6 malignant glaucoma 142–3 ocular hypertension 148–9 shallow or flat anterior chamber 149–50 wound leak 145 postoperative complications, late 151–7 blebitis or endophthalmitis 156
Index cataract progression 156 chronic hypotony 156 Descemet’s detachment 154–6 encysted bleb 152 fibrosis of conjunctival bleb 151–2 increased ocular pressure 152–3 peripheral anterior synechia 156 scleral ectasia 156–7 thin bleb 153 postoperative management 125–37 antifibrotic agents 127 assessment 127–9 evaluation frequency 127 digital pressure 134 general care 125 medication 106–8, 125–6 Nd:YAG goniopuncture 132–4 procedure 132–4 needling procedure 129–31 neuroprotective strategies 135 postoperative rupture of Descemet’s membrane 153–4 proteoglycans 39 pseudoexfoliative glaucoma indications/contraindications for NPFS 90 trabecular aspiration 47–8 race, and wound healing 118 reoperations 134–5, 157–8 conversion of NPFS into trabeculectomy 142, 158, 161, 174 cycloablation therapy 158 installation of artificial drainage shunt 158 reopening primary surgical site 157–8 second NPFS close to primary site 158 results of NPFS 161–8 combined surgery 166–7 comparative studies 164–6 trabeculectomy and NPFS 164–5 deep sclerectomy 162–4 deep sclerectomy with/without implant 165–6 viscocanalostomy 161–2 reticulated sodium hyaluronate implants 179–80 retinopathy, decompression 150 Scheie’s procedure 5–7 Schlemm’s canal 26, 28–9 anterior trabeculum or Descemet’s membrane 101–2 bleeding 143–4 cannulation 111–12
199
collapse 36, 43 collector channels and aqueous veins 30, 43 dilatation 64 endothelial lining 40–1 giant vacuoles 40, 41 hydraulic conductivity 41 and juxtacanalicular trabeculum 103 outflow resistance 41–3 pores 40, 41 vacuolation cycle 29 externalization (sinusotomy) 7–8, 13–14 inner wall 40, 45, 103 juxtacanalicular tissue (JCT) 28, 37, 38, 39, 103 paracentesis 101, 111 trabecular meshwork layers 15–16, 35 unroofing 100–1 aqueous humor resorption 64 trabeculectomy 15–16 see also viscocanalostomy Schwalbe’s line 26, 27 tear 140 scleral bed devices 177–83 scleral ectasia, postoperative 156–7 scleral flap see superficial scleral flap scleral plexus, deep 31 scleral space creation 47 implants/devices 177–83 scleral spur 25–7 sclerecto-iridectomy 3, 4 sclerectomy posterior-lip 7 see also deep sclerectomy sclerolimbal junction 25 sclerostomy 5–7, 9–10 laser 9–10 posterior-lip thermal 5–7 thermal 5 sclerotomy anterior 2 simple 2 small-flap 2–3 Seidel test 145 setons and shunts 8 shunts filtration 8 installation at reoperation 158 sinusotomy 7–8, 13–14 experimental, compared with trabeculectomy and deep sclerectomy 73
200
Index
SK gel implant 104–5, 179–80 spindle cells 76 steroids 126 Sturge–Weber syndrome, indications/contraindications for NPFS 91–2 subchoroidal space, aqueous humor resorption 63 subconjunctival bleb, aqueous humor resorption 60–1 superficial scleral flap 98–103 ciliary and choroid body visualization 98–100 closure, perforation of trabeculo-Descemet’s membrane 105–6, 140–2 deep sclerectomy 98–100 delineation with metal blade 98–9 extension into clear cornea 98–9 fibrosis 98, 100 repositioned and sutured 104 tear 174 variation in shape 98–9 surgical method 97–108 collagen implant 103–4 Descemet’s membrane or anterior trabeculum 101–102 fornix-based conjunctival incision 97–8 Hema implant 105 hyaluronate injection 104–5 identifying Schlemm’s canal 100 instruments 107–8 limbal-based conjunctival incision 97–8 postoperative management and medication 106–8 superficial scleral flap 98–103 Tenon’s capsule and conjunctiva closure 104 trabeculo-Descemet’s perforation 105–6, 140–2 wet-field electrocoagulation cautery 98 synechia, peripheral anterior, postoperative 156 T-flux implant 181–2 TDM see trabeculo-Descemet’s membrane Tenon’s capsule closure 104 excision 98 Tenon’s space, opening 98 trabecular aspiration 47–8 trabecular meshwork (conventional outflow) 34–43 aqueous veins and collector channels 43 corneoscleral lamellae 36 corneoscleral meshwork 27–8 drainage of aqueous humor 21, 34–44 endothelium 41–3
external portion, glaucomatous patient 80–1 juxtacanalicular meshwork 28, 38–9 uveal and corneoscleral regions 37–8 see also Schlemm’s canal trabeculectomy 8–9, 33, 45 ab externo 15–17, 58 experimental, compared with deep sclerectomy and sinusotomy 73 filtration membrane 67 technique 103–4 variations 103 conversion of NPFS at reoperation 142, 158, 161, 174 vs deep sclerectomy with collagen implant (DSCI), complications 145, 151 filtering bleb 45 first non-penetrating 10, 15–17 modes of action 8 vs NPFS, comparative studies 164–5 unroofing of Schlemm’s canal 15–16 vs viscocanalostomy 113–15 trabeculo-Descemet’s membrane 57–8 goniopuncture 132–4 outflow resistance measurements 69–71 perforation 105–6, 140–2 basal iridectomy after iris prolapse 105–6 conversion to trabeculectomy 142, 158, 161, 174 incidence 161 injection of viscoelastic 105–6 small/large, anterior chamber type 141 tight closure of superficial scleral flap 105, 142 resistance 58–9 see also Descemet’s membrane trabeculo-Descemet’s membrane pathway, anterior, post experimental deep sclerectomy 69–71 trabeculotomy 7 trabeculum, posterior pathway, post experimental ab externo trabeculectomy 71–4 transscleral outflow 21–2 trauma, angle-recession glaucoma 94 trephination 4–5 ultrasonic biomicroscopy (UBM) 78, 128–9, 143, 152 uveal meshwork 27, 36, 37–8 from anterior chamber 36 uveitic glaucoma 93 uveoscleral outflow (unconventional pathway) 21, 34, 44
Index vasopressin 170 viscocanalostomy 45–7, 109–116, 161–2 anaesthesia 109 and deep sclerectomy, with phacoemulsification 169–76 as ‘gentle’ trabeculectomy 47 history 18 necropsy eye 46 and phacoemulsification 166–7 postoperative detachment of Descemet’s membrane 154–6 results and complications 113–15 Stegmann’s 104, 109 success rate 161–2 surgical technique 33, 109–115 conjunctival flap dissection 110 conjunctival repositioning/closure 113 creation of Descemet’s window 112 inner scleral flap excision 112 outer scleral flap suture 112–13 outer/inner scleral flaps, dissection 110–11
paracentesis 101, 111 Schlemm’s canal, cannulation 111–12 vs trabeculectomy 113, 113–15 see also deep sclerectomy viscoelastic implants in NPFS 182–3 visual acuity, postoperative 150, 151 wet-field electrocoagulation cautery 98, 143 wound healing 117–24 age, race and type of glaucoma 117–18 conjunctival inflammation 118 future therapies 121 pharmacology 118–20 antimetabolites 118, 120–1 5-fluorouracil 119 mitomycin C 119–20 stages connective tissue stage 117 fibroblastic proliferative stage 117 inflammatory stage 117 wound leak, postoperative 145
201
