Evisceration Techniques and Implant Extrusion Rates: A Retrospective Review of Two Series and a Survey of ASOPRS Surgeons. Don Liu, M.D., F.A.C.S.

Transcription

1 Ophthalmic Plastic and Reconstructive Surgery Vol. 23, No. 1, pp The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc. Evisceration Techniques and Implant Extrusion Rates: A Retrospective Review of Two Series and a Survey of ASOPRS Surgeons Don Liu, M.D., F.A.C.S. Department of Ophthalmology, University of Missouri-Columbia School of Medicine, Columbia, Missouri, U.S.A. Purpose: To compare implant extrusion rates in two retrospective series of patients, treated by different surgeons, and to determine if there are preferred evisceration and implant insertion techniques among oculoplastic surgeons. Methods: A retrospective review of cases of evisceration and implant insertion and a survey of oculoplastic surgeons. Results: The implant extrusion rate was zero in the author s series of 53 patients, compared with 27.1% (54 of 192 implants) in series treated by other surgeons. Statistically significant differences were found between the two series in surgical technique, wound care, and duration of antibiotic administration. The author routinely performed posterior sclerotomy, whereas the other surgeons did not. Average implant size was 18.8 mm in the author s series, compared with 13.7 mm in the other series. Antibiotic therapy was administered for 10 to 18 days in the author s series, compared with 28 to 46 days in patients treated by other surgeons. The response rate to the survey was 65% (223 of 343 surveys). Among the surgeons surveyed, immediate and secondary implant techniques were used in nearly equal frequency. Delayed implant technique is used least frequently. Reported implant extrusion rates varied greatly among surgeons, ranging from 0% to 20%. Only 3.9% of surgeons expressed a concern over sympathetic ophthalmia. Conclusions: Injudicious antibiotic use and smaller implants do not decrease the risk of implant extrusion. The survey revealed that immediate and secondary implant techniques are used with equal frequency. The theoretical risk of sympathetic ophthalmia does not appear to be a concern to most surgeons. At least three different surgical approaches are used for evisceration and implant insertion in patients with an infected eye. Many authors avoid placing an implant at the time of evisceration and believe implant placement should be performed only as a secondary procedure, if at all. 1 6 Variations in this approach include eradicating the infection before evisceration and implant insertion. 7 9 Other authors prefer to take the small risk of implant extrusion rather than not place an implant or subject the patient to a secondary implant procedure A few authors have reported good results with a third alternative, delayed Accepted for publication May 19, Supported in part by a nonrestricted departmental grant from Research to Prevent Blindness, Inc., New York, New York, U.S.A. The author has no financial interest in any of the surgical instruments, implants, suture materials, or medications used in this study. This manuscript is based in part on a thesis that was prepared in partial fulfillment of the requirements for membership in the American Ophthalmological Society. Address correspondence and reprint requests to Dr. Don Liu, Department of Ophthalmology, University of Missouri-Columbia School of Medicine, One Hospital Drive, Columbia, MO liud@health.missouri.edu DOI: /01.iop f3 primary implantation. With the use of this technique, primary wound closure is delayed after evisceration, and the scleral shell is packed with antibiotic-saturated gauze for a few days. Several days after evisceration, a second procedure is performed to insert the implant and then close the wound To date, no well-controlled, prospective, comparative studies have been performed to prove conclusively that one approach is preferred over another in patients with endophthalmitis; nor has there been a systematic review. How frequently each of these techniques is used and the associated implant extrusion rate are not known. Given the paucity of data, the current study includes a retrospective review of two large series of patients who underwent evisceration with implants, with or without an active eye infection. A survey of members of the American Society of Ophthalmic Plastic and Reconstructive Surgery (ASOPRS) was also conducted to examine current approaches in patients with endophthalmitis (evisceration versus enucleation, implantation techniques, and implant extrusion rates). The members were also asked specifically about their concern of the theoretical increased risk of sympathetic ophthalmia. 16

2 IMPLANT EXTRUSION AFTER EVISCERATION 17 METHODS Retrospective Review of Two Series of Patients Who Underwent Evisceration. A retrospective review was conducted of an uncontrolled series of 53 cases of evisceration performed by the author at two different teaching institutions from 1981 to 1997 and of a series of 192 cases performed from 1984 to 1998 by surgeons at the King Khaled Eye Specialist Hospital, Riyadh, Saudi Arabia. Particular attention was paid to the definition of eye infection, duration of antibiotic administration, surgical technique, suture materials, implant material and size, extrusion rate, and postoperative wound care regimen. A detailed description of the author s surgical technique of evisceration, suture material, postoperative wound care regimen, implant size, duration of antibiotic administration, and implant extrusion rate is published elsewhere. 16 The salient points include the following: Absolute alcohol was used to scrub the inside of the sclera; posterior sclerotomy was performed followed by instillation of gentamicin drops in the opened sclera shell; dirty instruments were removed from the field and surgeons were regloved; closure of sclera, Tenon, and conjunctiva was performed in layers, a conformer was inserted, and a pressure patch was placed over the socket for 5 days; there was minimal or no manipulation of the operative site until the patient was seen by an ocularist 6 to 8 weeks later, and antibiotics were generally administered for no more than 2 weeks. Survey of Oculoplastic Surgeons. A survey of ASOPRS members was conducted to examine concerns, preferences, and experiences of surgeons performing eviscerations and enucleation (Figure, Article Plus). The survey was mailed to 343 oculoplastic surgeons in the year Five questions were asked regarding evisceration techniques, implant extrusion rates, and concerns over sympathetic ophthalmia. RESULTS Review of the Author s Series. The author s series of 53 cases of evisceration included 11 patients with a clinical diagnosis of endophthalmitis. The duration of systemic antibiotic therapy in the 11 patients varied from 11 to 14 days (average, 13.6 days). The secondary implant technique was used in 7 of the 11 patients. The surgical procedure was performed from several days to many weeks after the infection had cleared. Different types of sutures and implants were used in these 7 patients. The other 4 patients with clinical endophthalmitis underwent immediate implantation, with evisceration and implant insertion performed within 72 hours of presentation. Only 2 of these 4 patients had positive intraoperative cultures that confirmed the diagnosis of endophthalmitis. Essentially, the same immediate implant technique was used in the 42 patients without endophthalmitis. In the overall group, implant size ranged from 16 to 22 mm (average, 18.8 mm). No implant extrusion or other complication occurred in these patients. Forty-two of the 53 patients received a hollow acrylic sphere; 4, a PMMA sphere; 4, a silicone sphere; and 3, a hydroxyapatite implant. Review of the Other Surgeons Series. From 1984 to 1997, a total of 192 patients underwent evisceration with PMMA implants at the King Khaled Eye Specialist Hospital, where the author completed a prospective study of evisceration techniques. 16 Review of this large series provided a unique opportunity to examine the author s prospective study and his own 53-patient series more critically in terms of surgical techniques and outcome by identifying factors that might influence the implant extrusion rate. All of these patients had an infected eye, and operations were performed by several different surgeons. Due to difficulties in coding and changes in computerization of the medical record, an infected eye did not necessarily mean either clinical or culture-proven endophthalmitis. Nonetheless, because of some type of eye infection, all of these patients had received various preoperative, intraoperative, and postoperative intravenous and topical antibiotics. Endophthalmitis was diagnosed by preevisceration anterior chamber or vitreous tap culture in 151 (78%) of the 192 patients. However, intraocular cultures in 37 patients at the time of evisceration, 2 to 17 days after the initial diagnosis, revealed only 4 positive results. Many of the 192 patients apparently had a blind, painful eye caused by corneal ulcer, bacterial keratitis, or hypopyon from unspecified causes. With vigorous intravenous and topical antibiotic therapy, the scleral cavity of most of these patients probably had become sterile by the time an implant was inserted 2 to 17 days later, as demonstrated by only 4 of 37 positive intraoperative cultures. Data on these 37 patients are summarized in the Table ( Article Plus). All of the surgeons at the eye center used essentially the same surgical technique, antibiotic regimen, and postoperative wound care regimen. Surgeons, however, used different suture materials: 5 0 or 6 0 Vicryl, 4 0 or 5 0 chromic, and 4 0 or 5 0 silk. Although an implant was inserted at the time of evisceration, the procedure was not performed until 2 to 17 days after admission. No surgeon made posterior sclerotomies or administered gentamicin during surgery. There was no documentation in the operative notes of clearing the dirty instruments from the field or changing gloves. Cleansing the scleral shell with absolute alcohol was documented in only 1 of 37 cases. All of the 37 patients received systemic antibiotics for 28 to 46 days (average, 34.2 days). The size of the implant ranged from 10 mm to 16 mm (average, 13.7 mm). After surgery, a light patch was placed over the eye. Ice cold compresses were not used. All 37 patients were examined by an in-house ocularist on the first postoperative day, with another appointment scheduled 1 to 4 weeks later. Follow-up examination with the surgeon was usually scheduled 4 to 6 weeks after surgery. In the meantime, patients were maintained on systemic antibiotics. They also continued to apply topical antibiotic ointment to the socket 4 to 6 times daily for 4 to 8 weeks. Once a patient had a prosthetic eye made, an appointment with the ophthalmologist would not be made unless a medical problem such as socket infection, implant exposure, or implant extrusion developed. Fifty-four (27.1%) of the 192 implants extruded, with 49 (90%) of the 54 implant extrusions occurring within the first 6 months. Implant extrusion probably was caused by infection, edema, hemorrhage, or faulty surgical technique. The extrusion rate of 27.1% is consistent with a previously unpublished rate of 21.6% in infected eyes at the same eye center from 1984 to 1990, using the same definition of infection. During that same period, implant extrusion occurred in 18 (11.4%) of 158

3 18 LIU uninfected eyes. In that unpublished review, it was concluded that suture materials did not affect the extrusion rate. In the subgroup of 37 patients with intraoperative cultures, the extrusion rate was 3 of 4 (or 75%) in patients with culture-proven endophthalmitis and 7 of 33 (21.2%) patients with sterile intraocular contents. All instances of implant extrusion occurred 1 week to 5 months after surgery. Statistically significant differences were found in the implant size, duration of antibiotic administration, and the implant extrusion rates between the author s retrospective series and prospective study and the retrospective series of patients operated on by other surgeons (p 0.000). Major differences existed in intraoperative techniques and postoperative wound care regimen. Survey of Oculoplastic Surgeons. A survey of ASOPRS members was conducted in 1999 to gather information on the following questions (Figure, Article Plus): (1) Was fear of sympathetic ophthalmia or fear of infection the reason a surgeon avoided evisceration? (2) What was the surgeon s preferred surgical technique for implant insertion (i.e., secondary implant, immediate implant, or delayed implant)? (3) What was the estimated implant extrusion rate? Of 343 questionnaires mailed, 223 were returned, constituting a response rate of 65%. Seven questionnaires were returned unopened because of change of address. Seventeen of the 223 returned surveys were subsequently lost, decreasing the response rate to 60%. Fifty-two (25.2%) of the 206 survey respondents stated that they never performed evisceration in an infected eye. Fear of spread of infection was the reason given by 32 surgeons (15.5%), whereas fear of sympathetic ophthalmia was the reason given by 8 surgeons (3.9%). The remaining 10 respondents who did not perform eviscerations gave a number of reasons, including lack of training in the technique. Survey respondents who indicated that they performed evisceration in an infected eye only sometimes gave a variety of reasons related to implant materials and extrusion rate. Clearly, fear of spread of infection rather than sympathetic ophthalmia was the main reason some surgeons avoided evisceration. When asked if an implant would be inserted in an infected eye as part of an evisceration procedure, 74 (35.9%) respondents replied never, and all believed that the implant probably would extrude. It was inferred that these surgeons prefer either the no implant technique or the secondary implant technique. The immediate implant technique was used in infected eyes by 95 (46.1%) of the respondents; 23 always used it, and 72 sometimes used it. Their estimates of extrusion rates over a wide range of years in practice varied a great deal. For example, several of these surgeons reported no implant extrusion in several procedures to up to 30 procedures. Some estimated 1 implant extrusion of 10 to 100 cases. Others estimated extrusion rates ranging from 2 of 20, 2 of 75, 2 of 250, and 5 of 400 operations. Although it was not clear if any of the infected cases was culture-proven endophthalmitis, it could be inferred that the implant extrusion rate in an infected eye was low and acceptable, using the immediate technique. The secondary implant technique was used by 89 (43.2%) of the 206 respondents. Twenty-one surgeons always used it, waiting for 1 to several weeks after the infection cleared, whereas 68 used it sometimes. A few surgeons reported no implant extrusion of several cases to up to 30 cases. One surgeon reported 1 extrusion out of several hundred procedures and another claimed a 15% extrusion rate. Fifteen surgeons always used the delayed primary wound closure technique and 40 surgeons sometimes used it, giving a total of 55 (26.7%) surgeons who reported use of the delayed implant technique. Estimates of implant extrusion rates with this technique varied from zero of 1 to 7 cases to 1 of 10 to 100 cases. Of the 206 respondents, 111 (53.9%) indicated that they never used the delayed implant technique. Forty (19.4%) surgeons never performed evisceration, even in uninfected eyes, preferring enucleation. Thirty-six of these surgeons stated that obtaining a surgical specimen was the reason and 4 surgeons feared sympathetic ophthalmia. Forty other surgeons always performed evisceration in patients without infection, whereas 115 surgeons did it sometimes. Estimates of implant extrusion rates after evisceration in uninfected eyes varied greatly. Some stated rates of zero out of a few cases, whereas others claimed zero of 100 to 500 cases. A few surgeons reported an estimate of 1 implant extrusion in 10 to 300 patients. A few other surgeons cited 2 implant extrusions of 2 to 400 cases. Still others estimated 3 extrusions in 20 to 300 cases. One surgeon each estimated 5 implant extrusions in 100, 200, and 450 cases, respectively. One surgeon reported 6 extrusions of 120 implants. One surgeon reported an extrusion rate of 15%. DISCUSSION Evisceration with implant insertion has been performed for more than 100 years in patients with endophthalmitis, yet, in contemporary ophthalmology, much is still unknown about preferred techniques or the timing of implant insertion for minimizing the risk of implant extrusion. No controlled studies or systematic reviews have been performed to compare the outcome of various surgical techniques or the timing of implant insertion. Careful analysis of the surgical techniques, duration of antibiotic administration, and postoperative wound care regimen in the two series of patients strongly suggests that these factors influence the outcome. Nevertheless, a limitation of this retrospective review of cases, however, is that it only allows aggregate analysis of the effects of the different techniques on implant extrusion rates. No conclusions can be drawn on whether a specific surgical technique, all by itself, used by the author and not used by the other surgeons, such as performing sclerotomy and cleansing the scleral shell with absolute alcohol, accounts for the author s low rate of implant extrusion. Similarly, although a statistically significant difference existed between the duration of antibiotic therapy used by the author (10 to 18 days) and that used by the other surgeons (28 to 46 days), no definitive conclusion can be drawn that a shorter duration of antibiotic therapy reduces the risk of implant extrusion. Use of systemic antibiotics for more than 10 to 14 days is deemed

4 IMPLANT EXTRUSION AFTER EVISCERATION 19 unnecessary, as injudicious use of antibiotics encourages the emergence of resistant bacterial strains Removal of contaminated instruments from the surgical field and the changing of gloves, as routinely practiced by the author, seem to be an obvious, logical approach to reduce the risk of infection ,32 34 This practice, however, apparently was not carried out or documented by other surgeons at the same hospital. Implant size also differed significantly in these two series. The author was able to accommodate larger implants by consistently using posterior sclerotomies All of the other surgeons at the same hospital never used posterior sclerotomies. They believed that the smaller the implant, the less the risk of extrusion. The role of intraoperative use of absolute alcohol is not clear; no controlled studies of this aspect have been performed. Although proponents of alcohol use believe it aids in destroying any remaining uveal tissues and microorganisms, opponents believe it irritates the tissues and causes prolonged inflammation, which might encourage early implant extrusion. Many authors advocate removal of the uveal tissues with various instruments or gauze. Berens and Breakey 38 appear to have been the first to use Metaphen to cleanse the scleral shell. Other authors have used absolute or 70% alcohol. 14,37,39 42 Based on personal communications with colleagues, many surgeons use alcohol routinely to denature uveal tissues and to take advantage of its antimicrobial properties. Conversely, just as many surgeons do not use alcohol, thinking that it may irritate the tissues and cause prolonged edema. Use of alcohol during the procedure was not a randomized, controlled aspect of the author s prospective study. Therefore, whether cleansing the scleral shell with alcohol contributes to implant retention is not known. Postoperative wound care is integral to total patient care but is often overlooked. Moderate to severe orbital and periorbital edema lasting several days is expected after evisceration with implant insertion. To prevent or subdue the edema, some surgeons prefer the application of ice-cold compresses 1,43,44 or a temporary tarsorrhaphy, 14,15,45 whereas others apply a pressure dressing over the operated socket for several days. 39,42,46 Unfortunately, too often a surgeon leaves the operating room and delegates the task of wound dressing to a resident or nurse, not knowing if the orders were carried out properly. In the author s prospective study, 16 the author made a point of personally placing the pressure dressing and inspecting it before removal on the fourth to sixth postoperative day. In the author s years of experience, only minimal residual edema occurs after evisceration and after enucleation in patients wearing a properly placed pressure dressing for several days. Compared with the wound care regimen used by the other surgeons, the wound care regimen used by the author seems to involve less tissue manipulation for a longer postoperative period than that used by the surgeons at the eye center. For example, the other surgeons initiated evaluation by an ocularist on the first postoperative day and reevaluation 1 to 4 weeks later, which appears to accomplish very little except irritating the already edematous tissues. 37,39,46 48 Similarly, it seems superfluous and a potential irritant to apply ointment to the operated socket 4 to 6 times daily for many weeks, in addition to continued and prolonged systemic antibiotic administration. Studies have shown that more experienced surgeons tend to have lower implant extrusion rates, suggesting that surgical technique and experience do make a difference in the outcome. 49 Zolli 49 showed in 1978 that the overall implant extrusion rate was 6% for evisceration and 1.8% for enucleation for the staff at the Wills Eye Hospital. However, these rates increased to 22% and 6%, respectively, if all others were included. No details were provided regarding surgical techniques and postoperative wound care. Of note, nearly half of the procedures in the author s prospective study 16 were performed by a resident or fellow under his supervision, demonstrating that the technique is reliable and the results are reproducible. In short, there were significant differences in surgical technique, implant size, duration of antibiotic therapy, and postoperative wound care regimen between the author and the other surgeons at the same eye center. It is possible that these differences may account for the different implant extrusion rates. Although each one of these factors in isolation may not have much impact on the outcome, together they seem to make a difference. The author s retrospective reviews, and the author s prospective study, 16 show that a smaller implant does not necessarily signify a lower risk of extrusion. Therefore, use of a smaller implant at the expense of good orbital volume replacement and possibly good cosmesis should be avoided. I strongly recommend performing posterior sclerotomy, which aids in the accommodation of larger implants. As another aspect of a good surgical technique, I recommend the intraoperative use of antibiotic drops and removal of dirty instruments from the surgical field. Suture and implant materials do not appear to affect implant extrusion rates. Since good postoperative wound care is essential in the total care of the patient, I recommend minimal manipulation of the surgical site during the immediate postoperative period, including application of topical antibiotics. In addition, evaluation by the ocularist should be delayed until 6 to 8 weeks after surgery. The findings also suggest that prolonged administration of antibiotics does not help prevent implant extrusion in patients with endophthalmitis. Antibiotics should be judiciously administered.

5 20 LIU Survey. Although data from the survey of ASOPRS members are not scientific, the findings nonetheless reveal certain relevant clinical information. First, the presumed theoretically increased risk of sympathetic ophthalmia does not appear to be a concern when performing evisceration. Only 8 (3.9%) surgeons surveyed gave sympathetic ophthalmia as a reason for not performing evisceration in an infected eye. Thirty-two (15.5%) surgeons stated they would not do so for fear of implant extrusion. In an eye without infection, only 4 (1.9%) surgeons would not perform evisceration for fear of sympathetic ophthalmia, and 36 (17.5%) surgeons would not for reason of inability to obtain an adequate surgical specimen. Ophthalmic surgeons with these beliefs constituted a relatively small fraction of the members surveyed. Interestingly, at least 10 respondents, who were ASOPRS members, never learned the evisceration technique during their residency and/or fellowship. Although the survey results suggest that sympathetic ophthalmia is not a major concern of most surgeons, answering the question of a possible increased risk of sympathetic ophthalmia after evisceration is beyond the scope of the retrospective review of the two series. Data were not collected on the occurrence of this complication and in the two series of cases. Although sympathetic ophthalmia has been a hotly debated topic over the decades, perhaps the lack of any well-documented, new case reports of an association with evisceration supports the notion that the connection is extremely weak, if indeed there is one. Second, the immediate implant and secondary implant techniques appear to have about an equal number of advocates, with 46.1% and 43.2%, respectively, of surgeons surveyed using them. Only 26.7% of the surgeons surveyed use the delayed implant technique. Many surgeons apparently use different techniques for various reasons, as they chose sometimes as their reply. They in fact constituted the largest group. Estimated implant extrusion rates of each technique varied widely, from zero to more than 20%. Although these figures are in general agreement with known reports, it must be emphasized that they are only rough estimates. In summary, the retrospective reviews show that use of smaller implants and prolonged antibiotic administration do not necessarily help prevent implant extrusion and that meticulous surgical technique and proper postoperative wound care are essential. The ASOPRS survey reveals that most surgeons are more concerned about the more realistic possibility of implant extrusion than the theoretical possibility of sympathetic ophthalmia. The survey results also show that controversies over evisceration and enucleation persist. About an equal number of surgeons use the immediate and the delayed implant techniques, and fewer surgeons use the secondary implant technique. Implant extrusion rates using these various techniques by different surgeons vary greatly, from 0% to 20% or more. REFERENCES 1. Fox SA. Ophthalmic Plastic Surgery. 5th ed. New York: Grune & Stratton; 1976: Hughes WL. Evisceration. Arch Ophthalmol 1960;63: Iliff CE, Iliff WJ, Iliff NT. Oculoplastic Surgery. Philadelphia: WB Saunders; 1979: Meltzer MA, Schaefer DP, Della Rocca RC. Evisceration. In: Smith BC, ed. Ophthalmic Plastic and Reconstructive Surgery. Vol 2. St. Louis: CV Mosby; 1987: Soll D. Enucleation and evisceration. In: Duane TD, ed. Clinical Ophthalmology. Vol. 5. Hagerstown, Md: Harper & Row; 1980: Witteman GJ, Scott R. Enucleation and evisceration. In: Peyman GA, Sanders DR, Goldberg MF, eds. Principles and Practice of Ophthalmology. Philadelphia: WB Saunders; 1987: Berens C, Rosa F. Evisceration with plastic intrascleral implants. Am J Ophthalmol 1953;36: Roper-Hall MJ. Evisceration. In: Stallard s Eye Surgery. 6th ed. Philadelphia: JB Lippincott; 1980: Kohn R. Textbook of Ophthalmic Plastic and Reconstructive Surgery. Philadelphia: Lea & Febiger; 1988: Baylis H, Shorr N, McCord CD, et al. Evisceration, enucleation and exenteration. In: McCord CD, Tanenbaum M, eds. Oculoplastic Surgery. 2nd ed. New York: Raven; 1987: Dresner SC, Karesh JW. Primary implant placement with evisceration in patients with endophthalmitis. Ophthalmology 2000;107: Holds J, Anderson R. Primary versus delayed implant in evisceration [letter]. Arch Ophthalmol 1989;107: Shore J, Dieckert P, Levine M. Delayed primary wound closure: use to prevent implant extrusion following evisceration for endophthalmitis. Arch Ophthalmol 1988;106: Levine MR, Older JJ. Enucleation, evisceration and exenteration and the extruding orbital implant. In: Waltman SR, Frueh BR, Keates RH, et al, eds. Surgery of the Eye. New York: Churchill Livingstone; 1988; Older JJ, Levine MR. Enucleation, evisceration, and exenteration. In: Stewart WB, ed. Ophthalmic Plastic and Reconstructive Surgery. San Francisco: American Academy of Ophthalmology; 1984: Liu D. A comparison of implant extrusion rates and postoperative pain after evisceration with immediate or delayed implants and after enucleation with implants. Trans Am Ophthalmol Soc 2005; 103: Mule PH. Evisceration of the globe with artificial vitreous. Trans Ophthalmol Soc UK 1885;5: Bickerton TH. Report of the committee of the Ophthalmological Society appointed in March 1896, to consider the relative value of simple excision of the eyeball, and the operations which have been substituted for it. Trans Ophthalmol Soc UK ;18: Gradle HS. Concerning removal of the eyeball: exenteration versus enucleation. Arch Ophthalmol 1915;44: Knapp H. A case of evisceration of the eyeball followed by orbital cellulitis (thrombosis). Arch Ophthalmol 1884;14: Hurley LD, Westfall CT, Shore JW. Prophylactic use of antibiotics in oculoplastic surgery. In: Shore JW, ed. International Ophthalmology Clinics: Orbital Disease. Vol 32, No. 3. Boston: Little, Brown; 1992: Postlethwait MD. Principles of operative surgery: antisepsis, technique, sutures, and drains. In: Sabiston DC Jr, ed. Textbook of Surgery. Philadelphia: WB Saunders; 1986: Vaudaux PE, Lew PD, Waldvogel FA. Host factors predisposing to foreign body infections. In: Bisno AL, Waldvogel FA, eds. Infections Associated With Indwelling Medical Devices. Washington, DC: American Society for Microbiology; 1989:3 26.

6 IMPLANT EXTRUSION AFTER EVISCERATION Waldvogel FA, Vaudaux PE, Pittet D, et al. Preoperative antibiotic prophylaxis of wound and foreign body infections: microbial factors affecting efficacy. Rev Infect Dis 1991;13(S10):S D Amico DJ, Noorily SW. Postoperative endophthalmitis. In: Albert DM, Jakobiec FA, eds. Principles and Practice of Ophthalmology. Vol 2. Philadelphia: WB Saunders; 1994: Forster RK, Abbott RL, Gelender H. Management of infectious endophthalmitis. Ophthalmology 1980;87: Olson JC, Flynn HW, Forster RK, et al. Results in the treatment of post-operative endophthalmitis. Ophthalmology 1983;90: Stern GA, Engel HM, Driebe WT. The treatment of postoperative endophthalmitis: results of differing approaches to treatment. Ophthalmology 1989;96: Jones DB. Emerging antibiotic resistance: real and relative. Arch Ophthalmol 1996;114: Kaiser AB. Antimicrobial prophylaxis in surgery. N Engl J Med 1986;315: Crossley K, Gardner LC. Antimicrobial prophylaxis in surgical patients. JAMA 1981;245: Morris J, Martin GH. Delayed primary closure of contaminated wounds: a preliminary report. Surgery 1949;26: Meissner K, Meiser G. Primary open wound management after emergency laparotomies for conditions associated with bacterial contamination. Am J Surg 1984;148: Verrier ED, Bossart JK, Heer FW. Reduction of infection rates in abdominal incisions by delayed wound closure techniques. Am J Surg 1979;138: Stephenson C. Evisceration of the eye with expansion sclerotomies. Ophthalmic Plast Reconstr Surg 1987;3: Jordan DR, Khouri LM. Evisceration with posterior sclerotomies. Can J Ophthalmol 2001;36: Massry GG, Holds JB. Evisceration with scleral modification. Ophthal Plast Reconstr Surg 2001;17: Berens C, Breakey AS. Evisceration utilizing an intrascleral implant. Br J Ophthalmol 1960;44: Forster RK, Abbott RL, Gelender H. Management of infectious endophthalmitis. Ophthalmology 1980;87: Strampelli B, Valvo A. Late results of a personal technique with 18 years experience. Am J Ophthalmol 1966;62: Iliff NT. Enucleation, evisceration, and exenteration. In: Iliff NT, ed. Complications in Ophthalmic Surgery. New York: Churchill Livingstone; 1983: Raflo GT. Enucleation and evisceration. In: Tasman W, Jaeger E, eds. Duane s Clinical Ophthalmology. Revised ed. Vol 5. Philadelphia: Lippincott-Raven; 1995: Burch F. Evisceration of the globe with scleral implant and preservation of the cornea. Trans Am Ophthalmol Soc 1939;37: Ruedemann AD. Modified Burch-type evisceration with scleral implant. Am J Ophthalmol 1960;49: Shields CL, Shields JA, De Potter P, et al. Lack of complications of the hydroxyapatite orbital implant in 250 consecutive cases. Trans Am Ophthalmol Soc 1993;91: Meltzer MA. Complications of enucleation and evisceration: prevention and treatment. Int Ophthalmol Clin 1992;32: Gougelmann HP. Evolution of the ocular motility implant. Int Ophthalmol Clin 1970;10: Soll D. The anophthalmic socket. Adv Ophthalmic Plast Reconstr Surg 1988;7: Zolli CL. Implant extrusion in eviscerations. Ann Ophthalmol 1988;20: