- Email: [email protected]
Rescue of exposed scleral buckles with cadaveric pericardial patch grafts
Rescue of Exposed Scleral Buckles with Cadaveric Pericardial Patch Grafts David J. Weissgold, MD,1 Robert H. Millay, MD,1 Thomas A. Bochow, MD, MPH2 Purpose: To describe a new method for salvaging externally exposed silicone scleral exoplants (buckles) to avoid removal and the consequent risk of retinal redetachment. Design: A retrospective, noncomparative, interventional case series. Participants: Four patients with exposed, clinically uninfected scleral buckles after initial surgery for the treatment of rhegmatogenous retinal detachment. Intervention: Processed human donor pericardium patch grafts (Tutoplast; Innovative Ophthalmic Products, Inc., Costa Mesa, CA) were used to cover exposed areas of scleral buckles in concert with conjunctivoplasty. Main Outcome Measures: Scleral buckle preservation was the goal of this new treatment strategy. Results: In four eyes treated with pericardial patch grafts to cover segments of exposed scleral buckles, three (75%) were managed successfully with one surgery. One eye (25%) had this treatment method fail and required scleral buckle removal. Conclusions: Processed human donor pericardium patch grafting is one useful way to avoid removing exposed scleral buckles, and consequently, sparing patients the risk of recurrent retinal detachment. Ophthalmology 2001;108:753–758 © 2001 by the American Academy of Ophthalmology. Extrusion and transscleral erosion (intrusion) of silicone exoplant and implant elements are uncommon postoperative complications of scleral buckling procedures used in the management of rhegmatogenous retinal detachment (RRD). When transconjunctival erosion occurs, a further and potentially devastating problem, scleral buckle infection, may follow. Scleral buckle infections, although rare, can be sight and eye threatening by virtue of their potential for causing scleral necrosis and panophthalmitis. Measures are normally taken to avoid such infections. Traditionally, exposed scleral buckles have been managed by removal. Alternatively, when the area of exposure is small and an active buckle infection is not suspected, long-term prophylactic topical antibiotics and/or conjunctivoplasty may be used. Although broad areas of exposure and conjunctival scarring may make conjunctivoplasty impossible, removal of buckling elements may result in retinal redetachment. We have devised a surgical method that uses the human cadaveric pericardial allografts for covering relatively broad areas of exposed silicone buckle elements. To our knowledge, this method has not been previously reported. We have treated three patients successfully with this approach.
Originally received: April 14, 2000. Accepted: November 29, 2000. Manuscript no. 200244. 1 Retina Service, Department of Ophthalmology, University of Vermont College of Medicine, Burlington, Vermont. 2 Eyecare Associates of East Texas, Tyler, Texas Presented, in part, at Nantucket Retina, Nantucket, Massachusetts, July 28, 2000. Reprint requests to David Weissgold, MD, University of Vermont, Department of Surgery, Division of Ophthalmology, 1 South Prospect St., Burlington, VT 05401. © 2001 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
Methods All patients diagnosed with exposed scleral buckle at our two centers during 1998 and 1999 were included in this retrospective case series. No patients with clinically infected scleral buckles were included. Informed consent, including the use of human cadaveric donor pericardium, was obtained from each subject before surgery. Each patient was followed postoperatively according to their clinical course and at the discretion of the treating surgeon. The operating surgeon in each case delivered the entirety of perioperative and longitudinal follow-up care. At surgery, the conjunctiva was undermined around the areas of buckle exposure. An attempt was made at anterior conjunctival mobilization, but the size of the exposed areas prohibited this. A piece of donor pericardium (Tutoplast; Innovative Ophthalmic Products, Inc., Costa Mesa, CA) was trimmed and inserted beneath the conjunctival edges posteriorly (Fig 1B). The anterior edge of the patch was sewn to the sclera a few millimeters posterior to the limbus with multiple interrupted No. 7-0 Vicryl sutures, and the conjunctival edges were approximated both to themselves and to the surface of the pericardial patch, also with No. 7-0 Vicryl suture (Fig 2B). The posterior edge of the pericardial patch was sewn onto the solid silicone buckle element in some eyes with multiple interrupted No. 7-0 Vicryl sutures and left unsutured in other cases. This left a well-covered buckle element, with a small area of exposed sclera just posterior to the limbus and uncovered donor pericardium anterior to the anterior conjunctival suture line.
Case Reports Case 1 A 62-year-old man was referred with an inferior RRD in his pseudophakic left eye. Snellen visual acuity was 20/25. The conjunctiva appeared healthy. The posterior segment examination ISSN 0161-6420/00/$–see front matter PII S0161-6420(00)00659-X
753
Ophthalmology Volume 108, Number 4, April 2001
Figure 2. Case 2, postoperative course. A, pericardial patch, in situ, 10 days postoperatively, without complete conjunctival epithelialization. B, complete healing of conjunctiva over grafted area, 10 months postoperatively.
Figure 1. Case 1, perioperative course. A, scleral buckle exposure at commencement of pericardial patch graft surgery. B, patch (arrow) in place, before conjunctival mobilization to cover most of its surface. C, complete healing of conjunctiva over grafted area, 5 weeks postoperatively.
revealed a macula-sparing, moderately bullous inferotemporal RRD. The detachment was treated with scleral buckling techniques with external drainage of subretinal fluid and transscleral cryoretinopexy in the areas of the breaks. A segmental No. 276 solid silicone exoplant buckle was used, and the eye was encircled with a silicone band. All No. 5-0 Mersilene anchoring suture knots were trimmed and rotated posteriorly, and the conjunctiva was closed
754
using several interrupted No. 6-0 plain gut sutures about the limbus. Ten weeks postoperatively, there was some epiphora, increasing conjunctival injection, and two broad areas of conjunctival dehiscence with resultant exposure of two anterior portions of the buckle. Owing to a minimum of conjunctival injection, a complete lack of purulent discharge, an absence of eyelid swelling, and no reported pain, we believed the scleral buckle was not infected. Because the buckle was placed relatively recently, treatment options aimed at buckle preservation were discussed with the patient. The subject consented to the use of donor pericardium as a patching material. Surgery was performed as outlined in the “Methods” section (Fig 1A, B). Postoperatively, the patient was maintained on topical ciprofloxacin drops every 3 hours. The conjunctival injection diminished, the surface of the exposed pericardial patch and then the anterior exposed sclera epithelialized slowly (Fig 1C), and complete healing was evident by 8 weeks postoperatively. The patient’s visual acuity remained 20/30.
Case 2 A 56-year-old monocular, pseudophakic woman was referred with a macula-sparing, bullous superotemporal RRD. Snellen visual
Weissgold et al 䡠 Rescue of Exposed Scleral Buckles acuity was 20/80. The conjunctiva appeared largely healthy and only minimally injected. Retinal breaks were treated with cryoretinopexy, external drainage of subretinal fluid was performed, and segmental scleral buckling and encircling were performed in a fashion similar to the procedure in case 1. Five weeks postoperatively, the patient was seen with a large area of temporal conjunctival dehiscence and scleral buckle exposure. There was no frank clinical evidence of buckle infection. Treatment options aimed at buckle preservation were discussed with the patient, and an attempt at patching with donor pericardium was chosen. Surgery was performed as outlined in the “Methods” section. The patient was placed on topical steroid drops postoperatively. The surrounding conjunctiva was moderately injected by 10 days postoperatively (Fig 2A). Re-epithelialization of the patch graft commenced slowly. Some loose Vicryl sutures were removed, and the previously exposed pericardial patch slowly epithelialized over the course of the ensuing 3 weeks. The conjunctiva and pericardial patch have remained stable since, without re-exposure. The visual acuity was 20/30 at the 10-month postoperative evaluation (Fig 2B).
Case 3 A 55-year-old woman was referred with hand motions vision and an RRD in her pseudophakic right eye. The conjunctiva and remainder of the anterior segment structures appeared unremarkable (except for the clear, well-positioned posterior chamber intraocular lens). Cryoretinopexy, external drainage of subretinal fluid, and segmental scleral buckling and encircling were performed in a fashion similar to the procedure in cases 1 and 2. Adjunctive pneumatic tamponade was used. Postoperatively, retinal reattachment was evident within days, and the conjunctiva healed normally within 2 weeks. However, the intermediate-term postoperative course was complicated by the development of a stage 4 idiopathic macular hole and several small peripheral inferior retinal tears. The tears were treated with argon laser retinopexy, but the patient declined surgery for the treatment of the macular hole. Eight weeks postoperatively there were some pigmented vitreous strands with increased vitreoretinal traction evident inferiorly, but the retina remained attached. At 10 weeks postoperatively, the patient required a second surgery. The retina had redetached completely and grade C3 proliferative vitreoretinopathy was evident. Standard three-port pars plana vitrectomy was performed. The sclerotomies were closed, and the conjunctiva was reapproximated with No. 7-0 Vicryl suture. The eye was maintained on topical tobramycin, steroid, and atropine drops for 10 weeks postoperatively. Thirteen months after the second surgery, an 8-mm ⫻ 8-mm area of conjunctival dehiscence was evident superotemporally, and the silicone buckle was exposed. Once again, there were no overt clinical signs or symptoms of buckle infection. Because of the history of proliferative vitreoretinopathy, buckle-preserving management strategies were discussed with the patient, and she opted for donor pericardial patching. Surgery was performed as outlined in the “Methods” section. Tobramycin drops were used four times daily postoperatively. The surface of the pericardial patch graft slowly epithelialized, and complete healing was evident by 3 weeks postoperatively. Eleven months after the patching surgery, the vision remained counting fingers at 6 feet (the macular hole never closed), the retina remained attached, and all of the conjunctival wounds were wellhealed.
Case 4 A 49-year-old man with a history of medically controlled chronic open-angle glaucoma was referred for management of an acute, macula-involving RRD in his phakic right eye. Grade B proliferative vitreoretinopathy was present. The conjunctiva and anterior segment structures appeared healthy in both eyes (despite chronic pilocarpine use on the right side). Snellen visual acuity was 20/ 100. Cryoretinopexy, external drainage of subretinal fluid, and segmental scleral buckling and encircling were performed in a fashion similar to the procedure in cases 1 to 3. The retina has remained completely reattached since the first postoperative day. Topical tobramycin, steroids, and scopolamine were used during the first postoperative month, in addition to timolol, brimonidine tartrate, and latanoprost drops. At no time in the postoperative course was pilocarpine used. The visual acuity recovered to 20/60. Nine months postoperatively, the patient was seen with diffuse conjunctival injection and a small bulbar conjunctival dehiscence temporally with exposure of the buckle, but no additional clinical evidence of buckle infection. Topical ciprofloxacin drops were commenced, and a pericardial patch graft was scheduled. Surgery was performed as outlined in the “Methods” section. A small anterior triangle of pericardium was left sutured down (to the episclera) but uncovered anteriorly, because wound tension prohibited conjunctival mobilization to the degree that would have been needed to completely cover the patch graft. Postoperatively, topical ciprofloxacin drops and pressure-lowering drops were continued. The anterior conjunctival wound edge retracted somewhat, the pericardial patch appeared to melt and, by 3 weeks postoperatively, the silicone buckle was exposed once again. A repeat pericardial patch was offered, but the patient declined, opting for removal of the scleral buckle. This was performed successfully, along with additional prophylactic laser retinopexy. It was possible to completely reapproximate the conjunctival edges anteriorly. Three months later, the conjunctiva was completely healed and the retina attached.
Results In three of four of our cases (75%), single surgeries to cover exposed scleral buckles with pericardial patches were successful, preventing the need for buckle removal for infection prophylaxis. In one case (25%), such surgery failed in the intermediate postoperative period. This patient’s scleral buckle was subsequently removed, and no retinal redetachment developed. In the three successfully grafted cases, reepithelialization of the pericardial surface took between 3 and 5 weeks. Postoperative conjunctival injection took between 3 and 8 weeks to abate.
Discussion Exposed implanted prosthetic materials, including scleral buckles, are at risk for becoming infected. Fortunately, scleral buckle infection is extremely uncommon.1–5 However, when it does occur, explantation of the prosthetic buckling material is generally required, because the response to local and/or systemic antibiotic therapy is usually poor. This may be related to the presence of a glycocalix biofilm secreted by bacteria surrounding the buckle element.6 Biofilm may protect the bacteria from host responses and may limit the penetration of antimicrobial drugs.
755
Ophthalmology Volume 108, Number 4, April 2001 Table 1. Cumulative Summary of Data Available in Literature Regarding the Risk of Recurrent Retinal Detachment after Removal of Exposed, but Clinically Uninfected, Scleral Buckles
Study Authors, YearReference No.
Number of Eyes with Recurrent Retinal Detachments after Removal of Scleral Buckles*
Number of Eyes with Exposure of Clinically Uninfected Scleral Buckles
Rate of Recurrent Retinal Detachment after Removal of Exposed Scleral Buckles (%)
Hadden, 19861 Ulrich and Burton, 19742 Smiddy et al, 19934 Shukla and Bhandari, 19957 Stratford, 19728 Yoshizumi, 19809 Wiznia, 198310 Lindsey et al, 198311 Hilton and Wallyn, 197812 Schwartz and Pruett, 197713 Deutsch et al, 199214 Total
5 8 9 2 12 1 0 1 1 2 5 46
14 24 31 26 46 16 3 9 23 23 61 276
36 33 29 8 26 6 0 11 4 9 8 17
*Eyes with recurrent retinal detachments that were present at the time of diagnosis of scleral buckle exposure are excluded from this analysis.
Although scleral buckle removal can often be accomplished without much surgical difficulty, at least in cases in which exoplants were used, retinal redetachment may result. Combining the results of a number of previously published series of scleral buckle outcomes, the overall retinal redetachment rate in this setting is approximately 17% (Table 1).1,2,4,7–14 A relatively short interval between initial RRD repair and the onset of scleral buckle exposure,7,11–13 persistent vitreoretinal traction evident at the time of detection of buckle extrusion,11 the presence of proliferative vitreoretinopathy at the time of initial RRD repair,11 and the lack of detection of retinal hole(s) at the time of the original RRD repair14 may all increase the risk of retinal redetachment at the time of scleral buckle explantation in a clinically uninfected eye. Strategies that might allow for preservation of preexisting buckle hardware in eyes with buckle exposure would be expected to greatly reduce the chance of retinal redetachment. Long-term prophylactic treatment of exposed scleral buckles with topical broad-spectrum antibiotic ophthalmic solutions is widely regarded by retinal surgeons as inadequate treatment in these cases.15,16 There is, however, one reported case of such treatment over an 11-year course, with no resultant clinically evident scleral buckle infection.17 We believe that topical antibiotic treatment alone is appropriate management for an exposed scleral buckle only in the short term while awaiting more definitive surgical management. Conjunctivoplasty is an uncommonly used management strategy in cases involving exposed scleral buckles.16 One must be certain that the exposed buckle is not infected, because a conjunctival flap will likely fail if it is. This technique is not useful when the area of exposure is large, because edge-to-edge conjunctival reapproximation will yield undesirable degrees of wound tension, and rotation of forniceal conjunctival flaps carries with it all of the problems inherent to conjunctival forniceal foreshortening. We agree with Kittredge and Conway16 that conjunctivoplasty is best viewed as a temporizing strategy in these cases rather than a permanent solution.
756
Autogenous fascial grafts have been successfully used as patches to cover areas of exposed scleral buckles.18 Temporalis fascia and fascia lata are two potential donor site materials that can be used in these cases. Although autogenous grafts carry lower risks of transmission of infection and immunogenic rejection than do heterogenous grafts, autogenous grafting procedures have the disadvantage of requiring a second surgical site for harvesting donor material. These second surgical sites are not generally anatomic regions where most ophthalmologists are comfortable operating, so additional surgical specialists may be needed. Donor scleral patching has been the most widely reported grafting technique for covering exposed scleral buckles.19 –21 Originally described in the mid-1970s,19 this technique generally uses preserved (sometimes irradiated) cadaveric human sclera. Full- or partial-thickness patches of sclera may be sutured to the recipient eye’s sclera anterior to the exposed buckle element. After being draped posteriorly over the exposed buckle element, the remaining edges of the scleral patch may be sutured or left unsutured posteriorly. Undermined, mobilized surrounding conjunctiva is then advanced anteriorly onto the exposed surface of the patch graft and sutured to it. In the intermediate postoperative period, viable conjunctiva will generally advance anteriorly to cover the entire surface of the scleral patch, ultimately preventing reexposure and infection, resulting in preservation of the original buckle. Watzke20 reported a 57% success rate in seven eyes treated with scleral patch grafts and followed for at least 6 months. Murdoch et al21 recently reported a 100% success rate in four cases treated with scleral patches overlaid with autogenous labial mucous membrane grafts and followed for 11 to 19 months postoperatively. Kittredge and Conway16 point out that conjunctival epithelium may grow around the edges of scleral patch grafts, undermining and potentially limiting success in some cases. Although there is not much discussion of this in the literature, we believe that gentle, but extensive, debridement of the surface of the exposed buckle and of the surrounding
Weissgold et al 䡠 Rescue of Exposed Scleral Buckles exposed sclera should be performed before any patching of these areas. Processed human cadaveric pericardium is widely available. According to its manufacturer,22 this tissue has been harvested from donors subjected to routine serologic testing (for human immunodeficiency virus-1 and -2, hepatitis B and C viruses, syphilis, human T-lymphocyte virus-1, and other transmissible pathogens). The pericardium is then reduced to its collagen and mineral components by passage through denaturation and inorganic solvents, preserved with organic solvents and, finally, sterilized with gamma-irradiation. The packaged final product, Tutoplast, is thin, pliable, easy to manipulate and trim, and has excellent tensile strength. In our experience, pericardium is much easier to work with than banked sclera, which is thicker and less pliable. Pericardium has been used to cover tube shunts in glaucoma surgeries,23 repair corneoscleral fistulas,24 repair necrotic sclera after pterygium surgery,25 promote reepithelialization of a penetrating keratoplasty donor button in a subject with ocular cicatricial pemphigoid,25 and close a leaking scleral tunnel wound after phacoemulsification and two earlier (nonpatching) surgical attempts at closure.25 In our cases 1, 2 and 3, short duration between the initial scleral buckling surgeries and the onset on buckle exposure was the primary reason for attempting to preserve the integrity of the buckles rather than remove them. In cases 1, 3, and 4, the patients had significant peripheral RRD and/or lattice retinal degeneration in their fellow eyes, and were felt to be at particularly high risk for redetachment with buckle removal. In case 2, the patient was monocular and we were quite concerned about the consequences of potential retinal redetachment. In case 1, our first attempt at this technique and the case whose success prompted attempts at pericardium use for patching in our subsequent three cases, the patient refused autogenous fascia grafting. Case 4 was our only patch failure. We postulate three possible explanations for this: 1. Lack of primary conjunctival coverage of the entire graft surface at the time of surgery. This was done, however, with complete success in cases 1 and 2. It would seem that this factor alone should not lead to graft failure. However, it is plausible that complete conjunctival coverage of pericardial patches may be necessary when other factors, such as lack of conjunctival health (see No. 2 below), tear film abnormalities, and topical drop toxicities, are at play in a particular case. 2. Conjunctival health. Although the conjunctiva in our patient appeared grossly healthy preoperatively, it is certainly possible that his long-term premorbid use of pressure-lowering medications may have damaged the conjunctival epithelium’s self-replicative capabilities. Pilocarpine, in particular, is well known to be toxic to the conjunctiva. Perhaps primary edge-toedge conjunctival reapproximation with coverage of the entire pericardial patch surface is required in eyes with less robustly healing conjunctiva.
3. Subclinical scleral buckle infection. This could have led to patch graft failure. We acknowledge the limitations of our study. This is a relatively small, nonrandomized, uncontrolled series. The duration of follow-up in case 1 is short, because the patient was lost to follow-up. Finally, we did not attempt to obtain cultures from the surfaces of the exposed buckles in our cases, as others have done,21 so we cannot be certain that our patients’ exposed buckles were not also infected. However, clinical signs of buckle infection other than exposure were absent in all four cases. Moreover, we believe that subclinical buckle infection would likely have biased our results toward patch failure, not the success that we are reporting. We do believe that it is advisable, however, to routinely culture exposed scleral buckle elements at the time of diagnosis and/or at the time of any surgical management. In conclusion, we believe that patch grafting of clinically uninfected exposed silicone scleral exoplants with processed cadaveric pericardium is relatively easy to perform and highly successful. We also believe that it is safe, although the theoretic risk of transmission of viral and other infectious agents always exists when heterogenous tissue is used. Pericardium may be preferable to banked sclera, because it is easier to use and has a similar, if not better, rate of long-term success. Although autogenous patch grafting techniques used with temporalis fascia and fascia lata completely eliminate the theoretic risk of transmission of infectious agents, they involve the added complexity and potential morbidity associated with second surgical sites. Processed cadaveric pericardium is readily available commercially. It is thin in comparison with banked sclera, which may facilitate conjunctival epithelialization over the grafts. On the basis of these factors, coupled with the 17% risk of retinal redetachment when removing a scleral buckle, we consider patching of exposed solid silicone exoplant buckles with donor pericardium to be an excellent management option for patients with clinically uninfected, exposed scleral buckles.
References 1. Hadden OB. Infection after retinal detachment surgery. Aust N Z J Ophthalmol 1986;14:69 –73. 2. Ulrich RA, Burton TC. Infections following scleral buckling procedures. Arch Ophthalmol 1974;92:213–5. 3. Flindall RJ, Norton EWD, Curtin VT, Gass JDM. Reduction of extrusion and infection following episcleral silicone implants and cryopexy in retinal detachment surgery. Am J Ophthalmol 1971;71:835–7. 4. Smiddy WE, Miller D, Flynn HW Jr. Scleral buckle removal following retinal reattachment surgery: clinical and microbiologic aspects. Ophthalmic Surg 1993;24:440 –5. 5. McMeel JW, Naegele DF, Pollalis S, et al. Acute and subacute infections following scleral buckling operations. Ophthalmology 1978;85:341–9. 6. Holland SP, Pulido JS, Miller D, et al. Biofilm and scleral buckle-associated infections. A mechanism for persistence. Ophthalmology 1991;98:933– 8. 7. Shukla M, Bhandari V. Management of extruded explants
757
Ophthalmology Volume 108, Number 4, April 2001
8. 9. 10. 11. 12. 13. 14. 15. 16.
after retinal detachment surgery. Afro-Asian Journal of Ophthalmology 1995;14:29 –32. Stratford TP. Fate of the reattached retina following removal of silicone elements. In: Pruett RC, Regan CJ, eds. Retina Congress, New York: Appleton-Century-Crofts, 1974; 623– 8. Yoshizumi MO. Exposure of intrascleral implants. Ophthalmology 1980;87:1150 – 4. Wiznia RA. Removal of solid silicone rubber exoplants after retinal detachment surgery. Am J Ophthalmol 1983;95:495–7. Lindsey PS, Pierce LH, Welch RB. Removal of scleral buckling elements. Causes and complications. Arch Ophthalmol 1983;101:570 –3. Hilton GF, Wallyn RH. The removal of scleral buckles. Arch Ophthalmol 1978;96:2061–3. Schwartz PL, Pruett RC. Factors influencing retinal redetachment after removal of buckling elements. Arch Ophthalmol 1977;95:804 –7. Deutsch J, Aggarwal RK, Eagling EM. Removal of scleral explant elements: a 10-year retrospective study. Eye 1992; 6(Pt6):570 –3. Wilkinson CP, Rice TA. Michels Retinal Detachment, 2nd ed. St. Louis: Mosby, 1997;1019,1054. Kittredge KL, Conway BP. Management of the exposed scleral explant. Semin Ophthalmol 1995;10:53– 60.
758
17. Schmidt CW, Cohen HB. Exposed scleral buckle: a case report in an eleven-year course. Ophthalmic Surg 1983;14: 238 –9. 18. Dresner SC, Boyer DS, Feinfield RE. Autogenous fascial grafts for exposed retinal buckles. Arch Ophthalmol 1991; 109:288 –9. 19. Wilson RS, Parker JC. Scleral patch for exposed silicone buckles. Ophthalmic Surg 1975;6:83–5. 20. Watzke RC. Scleral patch graft for exposed episcleral implants. Arch Ophthalmol 1984;102:114 –5. 21. Murdoch JR, Sampath R, Lavin MJ, Leatherbarrow B. Autogenous labial mucous membrane and banked scleral patch grafting for exposed retinal implants. Eye 1997;11(Pt 1):43– 6. 22. Tutoplast威 Processed Pericardium. [package insert]. Costa Mesa, CA: Innovative Ophthalmic Products, Inc. 23. Raviv T, Greenfield DS, Liebmann JM, et al. Pericardial patch grafts in glaucoma implant surgery. J Glaucoma 1998;7:27– 32. 24. Sibayan SAB, Latina MA. The use of processed pericardium in the repair of corneo-scleral fistulas. Ophthalmic Surg Lasers 1997;28:334 –5. 25. Schein OD. The use of processed pericardial tissue in anterior ocular segment reconstruction. Am J Ophthalmol 1998;125: 549 –52.
Originally received: April 14, 2000. Accepted: November 29, 2000. Manuscript no. 200244. 1 Retina Service, Department of Ophthalmology, University of Vermont College of Medicine, Burlington, Vermont. 2 Eyecare Associates of East Texas, Tyler, Texas Presented, in part, at Nantucket Retina, Nantucket, Massachusetts, July 28, 2000. Reprint requests to David Weissgold, MD, University of Vermont, Department of Surgery, Division of Ophthalmology, 1 South Prospect St., Burlington, VT 05401. © 2001 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
Methods All patients diagnosed with exposed scleral buckle at our two centers during 1998 and 1999 were included in this retrospective case series. No patients with clinically infected scleral buckles were included. Informed consent, including the use of human cadaveric donor pericardium, was obtained from each subject before surgery. Each patient was followed postoperatively according to their clinical course and at the discretion of the treating surgeon. The operating surgeon in each case delivered the entirety of perioperative and longitudinal follow-up care. At surgery, the conjunctiva was undermined around the areas of buckle exposure. An attempt was made at anterior conjunctival mobilization, but the size of the exposed areas prohibited this. A piece of donor pericardium (Tutoplast; Innovative Ophthalmic Products, Inc., Costa Mesa, CA) was trimmed and inserted beneath the conjunctival edges posteriorly (Fig 1B). The anterior edge of the patch was sewn to the sclera a few millimeters posterior to the limbus with multiple interrupted No. 7-0 Vicryl sutures, and the conjunctival edges were approximated both to themselves and to the surface of the pericardial patch, also with No. 7-0 Vicryl suture (Fig 2B). The posterior edge of the pericardial patch was sewn onto the solid silicone buckle element in some eyes with multiple interrupted No. 7-0 Vicryl sutures and left unsutured in other cases. This left a well-covered buckle element, with a small area of exposed sclera just posterior to the limbus and uncovered donor pericardium anterior to the anterior conjunctival suture line.
Case Reports Case 1 A 62-year-old man was referred with an inferior RRD in his pseudophakic left eye. Snellen visual acuity was 20/25. The conjunctiva appeared healthy. The posterior segment examination ISSN 0161-6420/00/$–see front matter PII S0161-6420(00)00659-X
753
Ophthalmology Volume 108, Number 4, April 2001
Figure 2. Case 2, postoperative course. A, pericardial patch, in situ, 10 days postoperatively, without complete conjunctival epithelialization. B, complete healing of conjunctiva over grafted area, 10 months postoperatively.
Figure 1. Case 1, perioperative course. A, scleral buckle exposure at commencement of pericardial patch graft surgery. B, patch (arrow) in place, before conjunctival mobilization to cover most of its surface. C, complete healing of conjunctiva over grafted area, 5 weeks postoperatively.
revealed a macula-sparing, moderately bullous inferotemporal RRD. The detachment was treated with scleral buckling techniques with external drainage of subretinal fluid and transscleral cryoretinopexy in the areas of the breaks. A segmental No. 276 solid silicone exoplant buckle was used, and the eye was encircled with a silicone band. All No. 5-0 Mersilene anchoring suture knots were trimmed and rotated posteriorly, and the conjunctiva was closed
754
using several interrupted No. 6-0 plain gut sutures about the limbus. Ten weeks postoperatively, there was some epiphora, increasing conjunctival injection, and two broad areas of conjunctival dehiscence with resultant exposure of two anterior portions of the buckle. Owing to a minimum of conjunctival injection, a complete lack of purulent discharge, an absence of eyelid swelling, and no reported pain, we believed the scleral buckle was not infected. Because the buckle was placed relatively recently, treatment options aimed at buckle preservation were discussed with the patient. The subject consented to the use of donor pericardium as a patching material. Surgery was performed as outlined in the “Methods” section (Fig 1A, B). Postoperatively, the patient was maintained on topical ciprofloxacin drops every 3 hours. The conjunctival injection diminished, the surface of the exposed pericardial patch and then the anterior exposed sclera epithelialized slowly (Fig 1C), and complete healing was evident by 8 weeks postoperatively. The patient’s visual acuity remained 20/30.
Case 2 A 56-year-old monocular, pseudophakic woman was referred with a macula-sparing, bullous superotemporal RRD. Snellen visual
Weissgold et al 䡠 Rescue of Exposed Scleral Buckles acuity was 20/80. The conjunctiva appeared largely healthy and only minimally injected. Retinal breaks were treated with cryoretinopexy, external drainage of subretinal fluid was performed, and segmental scleral buckling and encircling were performed in a fashion similar to the procedure in case 1. Five weeks postoperatively, the patient was seen with a large area of temporal conjunctival dehiscence and scleral buckle exposure. There was no frank clinical evidence of buckle infection. Treatment options aimed at buckle preservation were discussed with the patient, and an attempt at patching with donor pericardium was chosen. Surgery was performed as outlined in the “Methods” section. The patient was placed on topical steroid drops postoperatively. The surrounding conjunctiva was moderately injected by 10 days postoperatively (Fig 2A). Re-epithelialization of the patch graft commenced slowly. Some loose Vicryl sutures were removed, and the previously exposed pericardial patch slowly epithelialized over the course of the ensuing 3 weeks. The conjunctiva and pericardial patch have remained stable since, without re-exposure. The visual acuity was 20/30 at the 10-month postoperative evaluation (Fig 2B).
Case 3 A 55-year-old woman was referred with hand motions vision and an RRD in her pseudophakic right eye. The conjunctiva and remainder of the anterior segment structures appeared unremarkable (except for the clear, well-positioned posterior chamber intraocular lens). Cryoretinopexy, external drainage of subretinal fluid, and segmental scleral buckling and encircling were performed in a fashion similar to the procedure in cases 1 and 2. Adjunctive pneumatic tamponade was used. Postoperatively, retinal reattachment was evident within days, and the conjunctiva healed normally within 2 weeks. However, the intermediate-term postoperative course was complicated by the development of a stage 4 idiopathic macular hole and several small peripheral inferior retinal tears. The tears were treated with argon laser retinopexy, but the patient declined surgery for the treatment of the macular hole. Eight weeks postoperatively there were some pigmented vitreous strands with increased vitreoretinal traction evident inferiorly, but the retina remained attached. At 10 weeks postoperatively, the patient required a second surgery. The retina had redetached completely and grade C3 proliferative vitreoretinopathy was evident. Standard three-port pars plana vitrectomy was performed. The sclerotomies were closed, and the conjunctiva was reapproximated with No. 7-0 Vicryl suture. The eye was maintained on topical tobramycin, steroid, and atropine drops for 10 weeks postoperatively. Thirteen months after the second surgery, an 8-mm ⫻ 8-mm area of conjunctival dehiscence was evident superotemporally, and the silicone buckle was exposed. Once again, there were no overt clinical signs or symptoms of buckle infection. Because of the history of proliferative vitreoretinopathy, buckle-preserving management strategies were discussed with the patient, and she opted for donor pericardial patching. Surgery was performed as outlined in the “Methods” section. Tobramycin drops were used four times daily postoperatively. The surface of the pericardial patch graft slowly epithelialized, and complete healing was evident by 3 weeks postoperatively. Eleven months after the patching surgery, the vision remained counting fingers at 6 feet (the macular hole never closed), the retina remained attached, and all of the conjunctival wounds were wellhealed.
Case 4 A 49-year-old man with a history of medically controlled chronic open-angle glaucoma was referred for management of an acute, macula-involving RRD in his phakic right eye. Grade B proliferative vitreoretinopathy was present. The conjunctiva and anterior segment structures appeared healthy in both eyes (despite chronic pilocarpine use on the right side). Snellen visual acuity was 20/ 100. Cryoretinopexy, external drainage of subretinal fluid, and segmental scleral buckling and encircling were performed in a fashion similar to the procedure in cases 1 to 3. The retina has remained completely reattached since the first postoperative day. Topical tobramycin, steroids, and scopolamine were used during the first postoperative month, in addition to timolol, brimonidine tartrate, and latanoprost drops. At no time in the postoperative course was pilocarpine used. The visual acuity recovered to 20/60. Nine months postoperatively, the patient was seen with diffuse conjunctival injection and a small bulbar conjunctival dehiscence temporally with exposure of the buckle, but no additional clinical evidence of buckle infection. Topical ciprofloxacin drops were commenced, and a pericardial patch graft was scheduled. Surgery was performed as outlined in the “Methods” section. A small anterior triangle of pericardium was left sutured down (to the episclera) but uncovered anteriorly, because wound tension prohibited conjunctival mobilization to the degree that would have been needed to completely cover the patch graft. Postoperatively, topical ciprofloxacin drops and pressure-lowering drops were continued. The anterior conjunctival wound edge retracted somewhat, the pericardial patch appeared to melt and, by 3 weeks postoperatively, the silicone buckle was exposed once again. A repeat pericardial patch was offered, but the patient declined, opting for removal of the scleral buckle. This was performed successfully, along with additional prophylactic laser retinopexy. It was possible to completely reapproximate the conjunctival edges anteriorly. Three months later, the conjunctiva was completely healed and the retina attached.
Results In three of four of our cases (75%), single surgeries to cover exposed scleral buckles with pericardial patches were successful, preventing the need for buckle removal for infection prophylaxis. In one case (25%), such surgery failed in the intermediate postoperative period. This patient’s scleral buckle was subsequently removed, and no retinal redetachment developed. In the three successfully grafted cases, reepithelialization of the pericardial surface took between 3 and 5 weeks. Postoperative conjunctival injection took between 3 and 8 weeks to abate.
Discussion Exposed implanted prosthetic materials, including scleral buckles, are at risk for becoming infected. Fortunately, scleral buckle infection is extremely uncommon.1–5 However, when it does occur, explantation of the prosthetic buckling material is generally required, because the response to local and/or systemic antibiotic therapy is usually poor. This may be related to the presence of a glycocalix biofilm secreted by bacteria surrounding the buckle element.6 Biofilm may protect the bacteria from host responses and may limit the penetration of antimicrobial drugs.
755
Ophthalmology Volume 108, Number 4, April 2001 Table 1. Cumulative Summary of Data Available in Literature Regarding the Risk of Recurrent Retinal Detachment after Removal of Exposed, but Clinically Uninfected, Scleral Buckles
Study Authors, YearReference No.
Number of Eyes with Recurrent Retinal Detachments after Removal of Scleral Buckles*
Number of Eyes with Exposure of Clinically Uninfected Scleral Buckles
Rate of Recurrent Retinal Detachment after Removal of Exposed Scleral Buckles (%)
Hadden, 19861 Ulrich and Burton, 19742 Smiddy et al, 19934 Shukla and Bhandari, 19957 Stratford, 19728 Yoshizumi, 19809 Wiznia, 198310 Lindsey et al, 198311 Hilton and Wallyn, 197812 Schwartz and Pruett, 197713 Deutsch et al, 199214 Total
5 8 9 2 12 1 0 1 1 2 5 46
14 24 31 26 46 16 3 9 23 23 61 276
36 33 29 8 26 6 0 11 4 9 8 17
*Eyes with recurrent retinal detachments that were present at the time of diagnosis of scleral buckle exposure are excluded from this analysis.
Although scleral buckle removal can often be accomplished without much surgical difficulty, at least in cases in which exoplants were used, retinal redetachment may result. Combining the results of a number of previously published series of scleral buckle outcomes, the overall retinal redetachment rate in this setting is approximately 17% (Table 1).1,2,4,7–14 A relatively short interval between initial RRD repair and the onset of scleral buckle exposure,7,11–13 persistent vitreoretinal traction evident at the time of detection of buckle extrusion,11 the presence of proliferative vitreoretinopathy at the time of initial RRD repair,11 and the lack of detection of retinal hole(s) at the time of the original RRD repair14 may all increase the risk of retinal redetachment at the time of scleral buckle explantation in a clinically uninfected eye. Strategies that might allow for preservation of preexisting buckle hardware in eyes with buckle exposure would be expected to greatly reduce the chance of retinal redetachment. Long-term prophylactic treatment of exposed scleral buckles with topical broad-spectrum antibiotic ophthalmic solutions is widely regarded by retinal surgeons as inadequate treatment in these cases.15,16 There is, however, one reported case of such treatment over an 11-year course, with no resultant clinically evident scleral buckle infection.17 We believe that topical antibiotic treatment alone is appropriate management for an exposed scleral buckle only in the short term while awaiting more definitive surgical management. Conjunctivoplasty is an uncommonly used management strategy in cases involving exposed scleral buckles.16 One must be certain that the exposed buckle is not infected, because a conjunctival flap will likely fail if it is. This technique is not useful when the area of exposure is large, because edge-to-edge conjunctival reapproximation will yield undesirable degrees of wound tension, and rotation of forniceal conjunctival flaps carries with it all of the problems inherent to conjunctival forniceal foreshortening. We agree with Kittredge and Conway16 that conjunctivoplasty is best viewed as a temporizing strategy in these cases rather than a permanent solution.
756
Autogenous fascial grafts have been successfully used as patches to cover areas of exposed scleral buckles.18 Temporalis fascia and fascia lata are two potential donor site materials that can be used in these cases. Although autogenous grafts carry lower risks of transmission of infection and immunogenic rejection than do heterogenous grafts, autogenous grafting procedures have the disadvantage of requiring a second surgical site for harvesting donor material. These second surgical sites are not generally anatomic regions where most ophthalmologists are comfortable operating, so additional surgical specialists may be needed. Donor scleral patching has been the most widely reported grafting technique for covering exposed scleral buckles.19 –21 Originally described in the mid-1970s,19 this technique generally uses preserved (sometimes irradiated) cadaveric human sclera. Full- or partial-thickness patches of sclera may be sutured to the recipient eye’s sclera anterior to the exposed buckle element. After being draped posteriorly over the exposed buckle element, the remaining edges of the scleral patch may be sutured or left unsutured posteriorly. Undermined, mobilized surrounding conjunctiva is then advanced anteriorly onto the exposed surface of the patch graft and sutured to it. In the intermediate postoperative period, viable conjunctiva will generally advance anteriorly to cover the entire surface of the scleral patch, ultimately preventing reexposure and infection, resulting in preservation of the original buckle. Watzke20 reported a 57% success rate in seven eyes treated with scleral patch grafts and followed for at least 6 months. Murdoch et al21 recently reported a 100% success rate in four cases treated with scleral patches overlaid with autogenous labial mucous membrane grafts and followed for 11 to 19 months postoperatively. Kittredge and Conway16 point out that conjunctival epithelium may grow around the edges of scleral patch grafts, undermining and potentially limiting success in some cases. Although there is not much discussion of this in the literature, we believe that gentle, but extensive, debridement of the surface of the exposed buckle and of the surrounding
Weissgold et al 䡠 Rescue of Exposed Scleral Buckles exposed sclera should be performed before any patching of these areas. Processed human cadaveric pericardium is widely available. According to its manufacturer,22 this tissue has been harvested from donors subjected to routine serologic testing (for human immunodeficiency virus-1 and -2, hepatitis B and C viruses, syphilis, human T-lymphocyte virus-1, and other transmissible pathogens). The pericardium is then reduced to its collagen and mineral components by passage through denaturation and inorganic solvents, preserved with organic solvents and, finally, sterilized with gamma-irradiation. The packaged final product, Tutoplast, is thin, pliable, easy to manipulate and trim, and has excellent tensile strength. In our experience, pericardium is much easier to work with than banked sclera, which is thicker and less pliable. Pericardium has been used to cover tube shunts in glaucoma surgeries,23 repair corneoscleral fistulas,24 repair necrotic sclera after pterygium surgery,25 promote reepithelialization of a penetrating keratoplasty donor button in a subject with ocular cicatricial pemphigoid,25 and close a leaking scleral tunnel wound after phacoemulsification and two earlier (nonpatching) surgical attempts at closure.25 In our cases 1, 2 and 3, short duration between the initial scleral buckling surgeries and the onset on buckle exposure was the primary reason for attempting to preserve the integrity of the buckles rather than remove them. In cases 1, 3, and 4, the patients had significant peripheral RRD and/or lattice retinal degeneration in their fellow eyes, and were felt to be at particularly high risk for redetachment with buckle removal. In case 2, the patient was monocular and we were quite concerned about the consequences of potential retinal redetachment. In case 1, our first attempt at this technique and the case whose success prompted attempts at pericardium use for patching in our subsequent three cases, the patient refused autogenous fascia grafting. Case 4 was our only patch failure. We postulate three possible explanations for this: 1. Lack of primary conjunctival coverage of the entire graft surface at the time of surgery. This was done, however, with complete success in cases 1 and 2. It would seem that this factor alone should not lead to graft failure. However, it is plausible that complete conjunctival coverage of pericardial patches may be necessary when other factors, such as lack of conjunctival health (see No. 2 below), tear film abnormalities, and topical drop toxicities, are at play in a particular case. 2. Conjunctival health. Although the conjunctiva in our patient appeared grossly healthy preoperatively, it is certainly possible that his long-term premorbid use of pressure-lowering medications may have damaged the conjunctival epithelium’s self-replicative capabilities. Pilocarpine, in particular, is well known to be toxic to the conjunctiva. Perhaps primary edge-toedge conjunctival reapproximation with coverage of the entire pericardial patch surface is required in eyes with less robustly healing conjunctiva.
3. Subclinical scleral buckle infection. This could have led to patch graft failure. We acknowledge the limitations of our study. This is a relatively small, nonrandomized, uncontrolled series. The duration of follow-up in case 1 is short, because the patient was lost to follow-up. Finally, we did not attempt to obtain cultures from the surfaces of the exposed buckles in our cases, as others have done,21 so we cannot be certain that our patients’ exposed buckles were not also infected. However, clinical signs of buckle infection other than exposure were absent in all four cases. Moreover, we believe that subclinical buckle infection would likely have biased our results toward patch failure, not the success that we are reporting. We do believe that it is advisable, however, to routinely culture exposed scleral buckle elements at the time of diagnosis and/or at the time of any surgical management. In conclusion, we believe that patch grafting of clinically uninfected exposed silicone scleral exoplants with processed cadaveric pericardium is relatively easy to perform and highly successful. We also believe that it is safe, although the theoretic risk of transmission of viral and other infectious agents always exists when heterogenous tissue is used. Pericardium may be preferable to banked sclera, because it is easier to use and has a similar, if not better, rate of long-term success. Although autogenous patch grafting techniques used with temporalis fascia and fascia lata completely eliminate the theoretic risk of transmission of infectious agents, they involve the added complexity and potential morbidity associated with second surgical sites. Processed cadaveric pericardium is readily available commercially. It is thin in comparison with banked sclera, which may facilitate conjunctival epithelialization over the grafts. On the basis of these factors, coupled with the 17% risk of retinal redetachment when removing a scleral buckle, we consider patching of exposed solid silicone exoplant buckles with donor pericardium to be an excellent management option for patients with clinically uninfected, exposed scleral buckles.
References 1. Hadden OB. Infection after retinal detachment surgery. Aust N Z J Ophthalmol 1986;14:69 –73. 2. Ulrich RA, Burton TC. Infections following scleral buckling procedures. Arch Ophthalmol 1974;92:213–5. 3. Flindall RJ, Norton EWD, Curtin VT, Gass JDM. Reduction of extrusion and infection following episcleral silicone implants and cryopexy in retinal detachment surgery. Am J Ophthalmol 1971;71:835–7. 4. Smiddy WE, Miller D, Flynn HW Jr. Scleral buckle removal following retinal reattachment surgery: clinical and microbiologic aspects. Ophthalmic Surg 1993;24:440 –5. 5. McMeel JW, Naegele DF, Pollalis S, et al. Acute and subacute infections following scleral buckling operations. Ophthalmology 1978;85:341–9. 6. Holland SP, Pulido JS, Miller D, et al. Biofilm and scleral buckle-associated infections. A mechanism for persistence. Ophthalmology 1991;98:933– 8. 7. Shukla M, Bhandari V. Management of extruded explants
757
Ophthalmology Volume 108, Number 4, April 2001
8. 9. 10. 11. 12. 13. 14. 15. 16.
after retinal detachment surgery. Afro-Asian Journal of Ophthalmology 1995;14:29 –32. Stratford TP. Fate of the reattached retina following removal of silicone elements. In: Pruett RC, Regan CJ, eds. Retina Congress, New York: Appleton-Century-Crofts, 1974; 623– 8. Yoshizumi MO. Exposure of intrascleral implants. Ophthalmology 1980;87:1150 – 4. Wiznia RA. Removal of solid silicone rubber exoplants after retinal detachment surgery. Am J Ophthalmol 1983;95:495–7. Lindsey PS, Pierce LH, Welch RB. Removal of scleral buckling elements. Causes and complications. Arch Ophthalmol 1983;101:570 –3. Hilton GF, Wallyn RH. The removal of scleral buckles. Arch Ophthalmol 1978;96:2061–3. Schwartz PL, Pruett RC. Factors influencing retinal redetachment after removal of buckling elements. Arch Ophthalmol 1977;95:804 –7. Deutsch J, Aggarwal RK, Eagling EM. Removal of scleral explant elements: a 10-year retrospective study. Eye 1992; 6(Pt6):570 –3. Wilkinson CP, Rice TA. Michels Retinal Detachment, 2nd ed. St. Louis: Mosby, 1997;1019,1054. Kittredge KL, Conway BP. Management of the exposed scleral explant. Semin Ophthalmol 1995;10:53– 60.
758
17. Schmidt CW, Cohen HB. Exposed scleral buckle: a case report in an eleven-year course. Ophthalmic Surg 1983;14: 238 –9. 18. Dresner SC, Boyer DS, Feinfield RE. Autogenous fascial grafts for exposed retinal buckles. Arch Ophthalmol 1991; 109:288 –9. 19. Wilson RS, Parker JC. Scleral patch for exposed silicone buckles. Ophthalmic Surg 1975;6:83–5. 20. Watzke RC. Scleral patch graft for exposed episcleral implants. Arch Ophthalmol 1984;102:114 –5. 21. Murdoch JR, Sampath R, Lavin MJ, Leatherbarrow B. Autogenous labial mucous membrane and banked scleral patch grafting for exposed retinal implants. Eye 1997;11(Pt 1):43– 6. 22. Tutoplast威 Processed Pericardium. [package insert]. Costa Mesa, CA: Innovative Ophthalmic Products, Inc. 23. Raviv T, Greenfield DS, Liebmann JM, et al. Pericardial patch grafts in glaucoma implant surgery. J Glaucoma 1998;7:27– 32. 24. Sibayan SAB, Latina MA. The use of processed pericardium in the repair of corneo-scleral fistulas. Ophthalmic Surg Lasers 1997;28:334 –5. 25. Schein OD. The use of processed pericardial tissue in anterior ocular segment reconstruction. Am J Ophthalmol 1998;125: 549 –52.