The Boston Keratoprosthesis Type II: The Massachusetts
Eye and Ear Infirmary Experience
Siddharth Pujari, MS(Oph), MPH, Sana S. Siddique, MD, Claes H. Dohlman, MD, PhD,
and James Chodosh, MD, MPH
Purpose:To report the long-term outcomes of Boston keratopros-thesis type II implantation in the management of severe ocular surface disease and corneal blindness through a retrospective interventional case series.
Methods: This retrospective review included medical records of patients who underwent Boston keratoprosthesis type II implantation at the Massachusetts Eye and Ear Infirmary from January 1, 2000 through December 31, 2009. The main outcome measures analyzed were visual acuity, keratoprosthesis retention, and postoperative complications. Results: A total of 29 eyes of 26 patients received a Boston keratoprosthesis type II during the study period. Patients undergoing operation had corneal blindness because of mucous membrane pemphigoid (51.7%), Stevens–Johnson syndrome/toxic epidermal necrolysis (41.4%), or other ocular surface disease (6.9%). Visual acuity after surgery improved to 20/200 or better in 23 eyes (79.3%) and to 20/30 or better in 10 eyes (34.5%). In patients with at least 1 year of follow-up (n = 21), visual acuity of 20/200 or better was maintained in 12 eyes (57.1%). Of 13 eyes followed-up for more than 5 years, 6 eyes (46.2%) had visual acuity of 20/200 or better at the last follow-up examination. Eyes that did not improve to 20/200 or lost vision during the follow-up had end-stage glaucoma, previous retinal detachment, or age-related macular degeneration. Of the total of 29 eyes, 17 devices (58.6%) were retained without extrusion or replacement during a total follow-up time of 107.9 person-years. Conclusions:The Boston keratoprosthesis type II is a viable option for corneal blindness from severe autoimmune ocular surface diseases.
Key Words: Boston keratoprosthesis, mucous membrane pemphi-goid, Stevens–Johnson syndrome, long-term outcomes
Keratoprosthesis implantation is primarily reserved for patients with corneal blindness in whom standard corneal allograft surgery has a poor prognosis.1
The development of a keratoprosthesis was first proposed in 1789.2 The Boston keratoprosthesis, a collar button–shaped device composed of polymethylmethacrylate, was approved for marketing by the Food and Drug Administration in 1992 and is now the most commonly used keratoprosthesis in the world.3
The Boston keratoprosthesis type I is used in patients with repeated corneal allograft failure, corneal opacity with extensive neovascula-rization, and other indications in which conventional allograft surgery would likely fail, such as aniridia. In contrast, patients with severe autoimmune ocular surface diseases, such as Stevens–Johnson syndrome/toxic epidermal necrolysis (SJS/ TEN), mucous membrane pemphigoid (MMP), and end-stage keratoconjunctivitis sicca, and those after severe chemical burns require a modified device, the so-called type II keratoprosthesis, which is designed with an anterior extension to allow implantation through surgically closed eyelids. On practical grounds, the Boston keratoprosthesis type II (Fig. 1) is generally reserved for patients with significant symblephar-on or ankyloblepharsymblephar-on, ocular surface keratinizatisymblephar-on, and absence of normal lid function. Several studies in the literature report outcomes for the Boston keratoprosthesis type I,3–5
but, to our knowledge, no similar report exists regarding the less commonly implanted Boston keratoprosthesis type II. Here, we report outcomes of type II keratoprosthesis surgery performed at the Massachusetts Eye and Ear Infirmary over the past decade.
MATERIAL AND METHODS
The Human Studies Committee of the Massachusetts Eye and Ear Infirmary approved this study. The project did not fall under the Health Insurance Portability and Accountability Act requirements because no protected health information was recorded or linked by code to data.
We retrospectively reviewed the records of all patients who underwent implantation of the Boston keratoprosthesis type II at the Massachusetts Eye and Ear Infirmary from January 1, 2000 through December 31, 2009 (n = 26). The year 2000 was chosen as the start of our study because in 1999, the device was modified to include holes in the keratopros-thesis back plate6,7
and topical vancomycin was added to the postoperative regimen. These changes dramatically reduced the incidence of corneal melting7
Received for publication December 11, 2010; revision received January 16, 2011; accepted February 5, 2011.
From the Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, MA.
Supported in part by an unrestricted grant to the Department of Ophthalmology, Harvard Medical School from the Research to Prevent Blindness, New York, NY.
The authors state that they have no proprietary interest in the products named in this article.
Reprints: James Chodosh, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA, 02114 (e-mail: firstname.lastname@example.org). CopyrightÓ2011 by Lippincott Williams & Wilkins
respectively. Additionally, a previous study focused on SJS patients who received Boston keratoprosthesis implantation and showed a marked improvement in outcomes after 2000.9
We included only eyes in the study that had never previously had a type II keratoprosthesis.
The details of Boston keratoprosthesis type I implan-tation have been previously described.10
Implantation of the type II keratoprosthesis requires several additional steps. All ocular surface mucosa is removed, including tarsal and forniceal conjunctiva, before implantation of the device. Once the keratoprosthesis is in place, a notch is fashioned in the upper eyelid to accommodate the anterior extension of the keratoprosthesis and a complete tarsorrhaphy is performed, leaving only the anterior-most part of the keratoprosthesis exposed (Fig. 1). Topical fluoroquinolone and vancomycin drops are routinely used in patients after type II keratopros-thesis implantation to reduce bacterial colonization of the perioptic skin and to prevent infection.
Data were collected by 2 physicians with experience in ophthalmology, by retrospective chart review, and were recorded in a Microsoft Access database (Microsoft, Red-mond, WA). Nonidentifying patient demographics,
prekeratoprothesis and postkeratoprosthesis surgical proce-dures, and postoperative complications and outcomes were recorded.
Main Outcome Measures
The main outcome measures were postoperative visual acuity, device retention, and time to complications.
All analyses were performed using commercial statis-tical software (Stata version 9.1; Stata Corp, College Station, TX). Cumulative functional success, defined as the ratio of the cumulative years with visual acuity of 20/200 or better to total duration of time since implantation, was calculated. The Kaplan–Meier method was used to study maintenance of visual acuity of 20/200 or better if achieved after surgery and retention of the device. Multiple logistic regression was used to analyze continuous and categorical covariates potentially affecting retention.
A total of 26 patients (29 eyes) underwent type II Boston keratoprosthesis implantation. The demographics and clinical characteristics are summarized in Table 1. The mean age at
FIGURE 1. Before (L) and after (R) photographs of a patient with SJS who received implantation of a Boston keratoprosthesis type II.
TABLE 1.Pre–Boston Keratoprosthesis Type II Conditions
MMP SJS/TEN Other Total
No. patients (eyes) 13 (15) 11 (12) 2 (2) 26 (29) Female, n (%) 6 (46.2) 8 (72.7) 1 (50.0) 15 (57.7) Mean age (SD), yr 70.2 (12.7) 50.5 (18.2) 58.6 (29.9) 61 (18.4) Duration of inflammation (SD), yr 16.6 (20.3) 11.4 (9.8) 3.0 (0.9) 13 (15.3) Systemic immunosuppression, n (%) Prednisone 3 (11.5) 1 (3.9) — 4 (15.4) Antimetabolities 1 (3.9) 1 (3.9) — 2 (7.7) Alkylating agents 2 (7.7) 1 (3.9) — 3 (11.5) Intravenous immune globulin 1 (3.9) 0 (0) 0 (0) 1 (3.9) Total on immunosuppression 7 (6.9) 3 (11.5) 0 (0) 10 (38.5) Eye-specific characteristics, n (%)
Visual acuity 20/200 or worse 14 (48.3) 12 (41.4) 2 (6.9) 28 (96.6) Advanced glaucoma 8 (27.6) 1 (3.5) 1 (3.5) 10 (34.6) Previous glaucoma surgery 10 (34.5) 2 (6.9) 1 (3.5) 13 (44.8) Previous retinal detachment repair 1 (3.5) 0 (0) 0 (0) 1 (3.5) Previous Boston keratoprosthesis type 1 4 (13.8) 1 (3.5) 0 (0) 5 (17.3)
implantation was 61 years (SD, 18.4 years). Patient gender was approximately equally distributed (women, 57.7%). MMP was the most common diagnosis (15 eyes; 51.7%), followed by SJS/TEN (12 eyes; 41.4%) and other ocular surface disease (6.9%), with the latter including 1 patient with a chemical burn and 1 patient with severe radiation keratopathy after treatment for lymphoma. The patients with MMP were older than the patients with SJS/TEN. Seven patients (26.9%) with MMP and 3 patients (11.5%) with SJS/TEN were using immunosuppressive therapy at the time of implantation. Nearly all eyes (96.6%) had a visual acuity 20/200 or worse before Boston keratoprosthesis type II implantation (Table 1). Thirteen eyes (44.8%) had previous glaucoma surgery, and 10 (34.6%) were known to have advanced glaucoma at the time of keratoprosthesis surgery.
Procedures performed at the same time as Boston keratoprosthesis type II implantation are summarized in Table 2. Of the 29 eyes that underwent surgery, 11 eyes (37.9%) had lens extraction and 4 eyes (13.8%) additionally underwent anterior vitrectomy. Ten eyes (34.5%) had a tube shunt placed for glaucoma: 7 eyes (24.1%) with SJS/TEN that had not previously undergone glaucoma surgery and 3 eyes (10.4%) with MMP that had previous tube implants. Nine eyes (31.0%) had concurrent pars plana vitrectomy.
Visual Acuity Outcomes
Visual acuity after surgery improved to 20/200 or better in 23 eyes (79.3%) (Table 3). Ten eyes (34.5%) recovered a visual acuity of 20/30 or better. Cumulative functional success, defined as the ratio of the cumulative years with visual acuity of 20/200 or better to total duration of time since surgery (Table 4), was calculated for those patients who achieved at least 20/200 for each subgroup. Patients with SJS/TEN had a higher cumulative functional success (0.91) than those with MMP (0.64) or other ocular surface disease (0.45). In those patients with 1 year of follow-up (n = 21; overall mean follow-up time, 3.762.8 years), visual acuity of 20/200 or better was maintained in 12 eyes (57.1%). Thirteen eyes were followed-up for more than 5 years, and 6 (46.2%) had visual acuity of 20/200 or better at the last examination (Fig. 2). Visual acuity was maintained beyond the average
follow-up times (MMP 3.3 years and SJS/TEN 4.2 years) in 2 eyes (13.3%) with MMP and in 4 eyes (33.3%) with SJS/TEN. Six eyes (20.7%) of 29 did not improve to at least 20/200 because of preexisting conditions. In eyes with SJS/TEN (n = 4), end-stage glaucoma (n = 1), occult retinal detachment (n = 1), and age-related macular degeneration (n = 2) were the main reasons for failure to improve to 20/200, whereas occult retinal detachment was the cause in 1 eye with MMP. The cause could not be determined by chart review in 1 eye with MMP.
Retroprosthetic membranes were the most common complication after type II keratoprosthesis implantation. Fourteen eyes (48.3%) required treatment with the neodymium-doped Nd:YAG laser for a total of 17 procedures. One eye required surgical membranectomy. Eight eyes (27.6%) had retinal detach-ment develop; 3 eyes (10.3%) underwent retinal detachdetach-ment repair. The mean time (in years) to retinal detachment could be accurately assessed in 6 eyes and was longer in MMP (n = 2; 6.160.67 years) than in SJS/TEN (n = 4; 3.86 1.81 years). Two SJS/TEN patients had retinal detachment develop after replacement of the first keratoprosthesis or tectonic corneal patch graft (revision surgery). One patient had endophthalmitis develop on the fourth postoperative day after type II implantation and underwent pars plana vitrectomy and injection of intraocular antimicrobial medications, but that patient never regained useful vision. Details of other complications and procedures are listed in Tables 5 and 6, respectively.
Nine eyes (60%) with MMP and 6 eyes (50.0%) with SJS/TEN retained the device without requiring reimplantation or repair before their last follow-up (Fig. 3). Eight eyes (53.3%) with MMP required either reimplantation of the keratoprosthesis (n = 5; mean follow-up time, 3.561.7 years) or corneal patch graft repair (n = 2; mean follow-up time, 1.1 6 0.4 years). Similarly, 6 eyes (50.0%) with SJS/TEN required either reimplantation (n = 5; mean follow-up time, 4.061.4 years) or corneal patch graft repair (n = 1; follow-up time, 4 years). One eye (8.3%) with SJS/TEN required a second reimplantation of the keratoprosthesis. Of 6 eyes (MMP = 2, SJS/TEN = 4) that had 20/200 or better vision for more than 5 years, 3 required replacement or revision surgery. Ten of the 29 eyes (34.5%) had a wound leak develop after a mean of 3.46 2.25 years after primary implantation of the Boston keratoprosthesis type II. Eyes afflicted with MMP had wound leaks develop earlier (n = 4, 2.961.78 years) than in those with SJS/TEN (n = 6, 3.7 6 2.63 years), but there was no statistically significant difference between the mean times to wound leaks between the groups (P= 0.59). Eight of the 10 eyes that had wound leaks develop underwent subsequent surgery; the implant fully extruded in 1 eye, and all visual potential was lost before repair could be attempted. One eye (3.5%) had recurrent leak develop after the first repair. Three eyes (10.4%) presented with early extrusion of the implant, and it is presumed that they had wound leaks develop between follow-up visits. Of the total of 29 eyes, 12 eyes (41.4%) either underwent reimplantation of the keratoprosthesis or experienced partial or total extrusion of their keratoprosthesis during a total follow-up time of 107.9 years, corresponding to a hazard rate of 0.11 per person-year.
TABLE 2.Concomitant Procedures Performed With Boston Keratoprosthesis Type II Implantation
Concomitant Procedures MMP, n (%) SJS/TEN, n (%) Other, n (%) Total, n (%) Lensectomy 5 (33.3) 4 (33.3) 2 (100.0) 11 (37.9) Lensectomy/vitrectomy 1 (6.7) 3 (25.0) 0 (0) 4 (13.8) Pars plana vitrectomy 8 (58.3) 1 (8.3) 0 (0) 9 (31.0) IOL removal 2 (13.3) 0 (0) 0 (0) 2 (6.9) IOL insertion 1 (6.7) 2 (16.7) 1 (50.0) 4 (13.8) Iridectomy 2 (13.3) 4 (33.3) 1 (50.0) 7 (24.1) Tube shunt 3 (20.0) 7 (58.3) 0 (0) 10 (34.5) Other (blepharoplasty) 4 (26.7) 0 (0) 1 (50.0) 5 (17.2)
A search for factors affecting postoperative wound leak after Boston keratoprosthesis type II implantation using multiple regression analysis (including age at surgery, sex, type of ocular inflammation, preceding glaucoma surgery, and previous Boston keratoprosthesis type I surgery) did not show any statistically significant associations. One must interpret these results with caution given the small size and limited number of variables in this retrospective study.
This is the first reported study to our knowledge focused specifically on the long-term outcomes after implantation of the Boston keratoprosthesis type II. Reported success rates for implantation of a Boston keratoprosthesis (either type I or II) in high-risk patients with severe ocular surface diseases, such as MMP, SJS/TEN, and chemical burns, have been moderate. Yaghouti et al11
in 2000 reported success in maintaining at least 20/200 vision after 2 years of follow-up after Boston keratoprosthesis implantation in eyes with MMP, SJS/TEN, and chemical burns at 72%, 33%, and 64%, respectively; after 5 years of follow-up for SJS/TEN patients, the success rate was
essentially zero. In a subsequent study conducted between 2000 and 2005 in SJS/TEN eyes (n = 16) at our institution (some of these patients are included in our study) by Sayegh et al,9
visual acuity of 20/200 or better was maintained for a mean period of 2.56 2.0 years in 12 eyes (75%). In our study, 6 eyes (50.0%) with MMP and 5 eyes (62.5%) with SJS/TEN that achieved a vision of 20/200 or better maintained it for more than 2 years. The results of our study are difficult to compare with those previously published because patients in previous studies were not categorized by the type of Boston keratoprosthesis.
Glaucoma is an important problem after implantation of a Boston keratoprosthesis.11,12
Preexisting glaucoma may worsen after surgery, especially in eyes with chronic in-flammation. Furthermore, eyes with no history of glaucoma may have elevated pressures develop because of subsequent inflammation and progressive angle closure from peripheral anterior synechiae. A total of 20 eyes (70.0%) in our study had a glaucoma drainage device implant before or at the time of first implantation of a Boston keratoprosthesis type II. Despite these efforts, 2 eyes (9.1%) that achieved 20/200 or better did not maintain the improvement because of progressive glaucoma. Because of increased ocular rigidity after kerato-prosthesis implantation, currently available instrumentation does not accurately detect intraocular pressure, and digital palpation remains the best method in such eyes. Serial evaluations of the optic nerve and visual fields are important when possible.3
Oral medications may be the only effective way to treat elevated intraocular pressure in eyes immediately after Boston keratoprosthesis type II, followed by placement of an Ahmed valve or cyclophotocoagulation if medical treatment fails. Unfortunately, postoperative glaucoma procedures, such
TABLE 4.Eyes That Lost 20/200 Visual Acuity After Boston Keratoprosthesis Type II Implantation
No. Eyes Cumulative Years
No. That Failed to
Retain 20/200 or Better Acuity Postoperative
With Acuity of 20/200 or Better
MMP 13 9 46.4 27.3
SJS/TEN 8 3 32.8 29.7
Other 2 2 7.6 3.4
TABLE 3.Preoperative and Postoperative Best-Corrected Visual Acuity After Boston Keratoprosthesis Type II Implantation BCVA ($) Preoperative, n (%) Postoperative, n (%)* One Year Postoperative, n (%) At Last Examination, n (%) 20/20 — 8 (27.5) 3 (10.3) 1 (3.4) 20/25 — 1 (3.4) 3 (10.3) — 20/30 1 (3.4) 1 (3.4) 1 (3.5) 4 (13.8) 20/40 — — 2 (6.9) — 20/50 — 4 (13.8) — 2 (6.9) 20/60 — 1 (3.4) — — 20/70 — 1 (3.4) 2 (6.9) — 20/80 — 5 (17.2) — 1 (3.4) 20/100 — — — — 20/200 — 2 (6.9) 1 (3.4) 1 (3.4) 20/400 — 1 (3.4) 3 (10.3) 1 (3.4) CF 4 (13.8) 2 (6.9) — 4 (13.8) HM 9 (31) 1 (3.4) 1 (3.4) 2 (6.9) LP 15 (51.7) 2 (6.9) 5 (17.2) 13 (44.8)
The patient with preoperative visual acuity of 20/30 had a Boston keratoprosthesis type I implant but had complications necessitating replacement with a keratoprosthesis type II.
*BCVA at any time after surgery.
BCVA, best-corrected visual acuity; CF, count fingers; HM, hand motion; LP, light perception.
FIGURE 2.Kaplan–Meier survival curve for retention of 20/200 vision or better after Boston keratoprosthesis type II implan-tation, by preoperative diagnosis.
as valve implant or cyclophotocoagulation, necessitate opening the medial and lateral tarsorrhaphies to expose the globe, which can be difficult because of the loss of normal tissue planes after the union of ocular surface and eyelid submucosal tissues. Preplacement of a glaucoma drainage device during primary Boston keratoprosthesis type II may be justified given the risk of glaucoma in such cases, but existing drainage devices may not function well in the absence of conjunctiva.
Autoimmune diseases remain the worst prognostic category for Boston keratoprosthesis implantation because of ongoing ocular inflammation leading to tissue necrosis and melting, which can result in leakage, choroidal effusion, and even retinal detachment.13
Retroprosthetic membrane (48.3%) was the most common complication in patients in our study, and it can be managed conservatively with laser membranec-tomy in most cases. However, retinal and choroidal detach-ments (26.1%) together were the most common causes for the failure to maintain visual acuity of 20/200 or better. A previous study of outcomes after Boston keratoprosthesis type I reported a retinal detachment rate of 3.5% (n = 141).3
In an article reporting vitreoretinal complications after Boston keratoprosthesis types I and II, the retinal detachment rate was 12% (n = 110).13
Such differences may be related to the type of Boston keratoprosthesis or the specific indications for surgery or both. We speculate that chronic inflammation in patients with autoimmune disease contributes to vitreoretinal traction and subsequent retinal detachment.
Eyes with MMP and SJS/TEN are equally prone to development of tissue melts, potentially requiring revision or replacement. The introduction of vancomycin eyedrops to the postoperative keratoprosthesis regimen has significantly
reduced the risk of stromal infection by Gram-positive bacteria and also has reduced the rate of endopthalmitis,8
but the numbers of patients are too low to determine a positive effect of vancomycin in Boston keratoprosthesis type II patients. One eye in our study with SJS/TEN had severe culture-negative endophthalmitis develop, and the vision could not be salvaged, despite aggressive treatment; tissue melt was a possible predisposing factor. Regular examinations with aggressive management of tissue melts are particularly important for patients with a type II Boston keratoprosthesis. As the pathogenesis of MMP and SJS/TEN becomes better un-derstood, newer immunomodulating agents14–16
may be found to enhance retention. Various methods to achieve biointegra-tion of the keratoprosthesis in the future may also reduce tissue melts and increase retention rates.17
Our study comes with certain caveats. We have no data regarding which patients should receive a Boston keratopros-thesis type II versus a type I. Patients with cicatricial ocular surface disease and corneal blindness are generally considered for a Boston keratoprosthesis type II when a significant loss of conjunctival fornices, extensive symblepharon, severe aque-ous tear deficiency, and/or ocular surface keratinization make type I surgery seemingly unlikely to succeed. The inability to wear a contact lens because of forniceal foreshortening can be a reason for the failure of a type I keratoprosthesis and eventual replacement with the type II. The data in our modest patient series also were insufficient, and the patient population was too heterogeneous to determine the potential roles of total iridectomy, vitrectomy, and intentional aphakia in successful outcomes after keratoprosthesis implantation. Finally, com-parisons between the Boston keratoprosthesis type II and other types of keratoprosthesis currently in use around the world for those with severe cicatricial blindness18,19
are difficult because of differences in underlying patient populations and environ-ment and ongoing increenviron-mental improveenviron-ments over the past decade in the Boston keratoprosthesis device. Implantation of the Boston type II keratoprosthesis may be technically easier than other methods in such patients. Perhaps, in the future, diverse keratoprosthesis devices can be directly compared among similar patients in the same clinical center.
In conclusion, Boston keratoprosthesis type II implan-tation in selected patients can provide sustained recovery of vision. However, challenges clearly remain. Type II keratoprosthesis complications may be reduced by further advances in device design and materials, new approaches to
TABLE 5.Complications Leading to Loss of More Than 20/ 200 Visual Acuity After Boston Keratoprosthesis Type II Implantation
Type of Complication MMP SJS/TEN Other Total No. Eyes (%)
Endophthalmitis 0 1 0 1 (4.3) End-stage glaucoma 1 0 1 2 (8.7) Retinal detachment 3 1 0 4 (17.4) Choroidal detachment 2 0 0 2 (8.7) Unknown 3 1 1 5 (21.7)
Total number of eyes (n) = 23.
TABLE 6.Additional Procedures Required After Boston Keratoprosthesis Type II Implantation
Procedures MMP, Eyes (No. Procedures) SJS/TEN, Eyes (No. Procedures) Other, Eyes (No. Procedures) Total, Eyes (No. Procedures; %)*
Nd:YAG laser retroprosthetic membranectomy 6 (8) 6 (7) 2 (2) 14 (17; 48.3) Surgical retroprosthetic membranectomy 1 (1) 0 (0) 0 (0) 1 (1; 3.5) Tube shunt 0 (0) 2 (2) 0 (0) 2 (2; 6.9) Wound leak repair 4 (5) 5 (7) 0 (0) 9 (12; 31.0) Retinal detachment repair 1 (1) 2 (2) 0 (0) 3 (3; 10.3) Skin revision* 5 (7) 4 (9) 1 (1) 10 (17; 34.5)
immunomodulatory therapy, improvements in perioperative and postoperative care, and perhaps by a better appreciation for higher-risk subgroups within the larger heading of SJS/TEN and MMP patients.
1. Dohlman CH, Harissi-Dagher M, Khan BF, et al. Introduction to the use of the Boston keratoprosthesis.Expert Rev Ophthalmol. 2006;1:41–48. 2. de Quengsy P.Precis ou cours d’operations sur la chirurgie des yeux.
Paris, France: Didot; 1789.
3. Zerbe BL, Belin MW, Ciolino JB. Results from the multicenter Boston type 1 keratoprosthesis study.Ophthalmology. 2006;113:1779 e1–e7. 4. Bradley JC, Hernandez EG, Schwab IR, et al. Boston type 1
keratopros-thesis: the University of California Davis experience.Cornea. 2009;28: 321–327.
5. Chew HF, Ayres BD, Hammersmith KM, et al. Boston keratoprosthesis outcomes and complications.Cornea. 2009;28:989–996.
6. Khan B, Dudenhoefer EJ, Dohlman CH. Keratoprosthesis: an update.
Curr Opin Ophthalmol. 2001;12:282–287.
7. Khan BF, Harissi-Dagher M, Khan DM, et al. Advances in Boston keratoprosthesis: enhancing retention and prevention of infection and inflammation.Int Ophthalmol Clin. 2007;47:61–71.
8. Durand ML, Dohlman CH. Successful prevention of bacterial endoph-thalmitis in eyes with the Boston keratoprosthesis. Cornea. 2009;28: 896–901.
9. Sayegh RR, Ang LP, Foster CS, et al. The Boston keratoprosthesis in Stevens-Johnson syndrome.Am J Ophthalmol. 2008;145:438–444. 10. Dohlman CH, Waller SG, Netland PA. Keratoprosthesis surgery. In:
Lindquist TD, Lindstrom RL, eds.Ophthalmic Surgery Update. Chicago, IL: Mosby-Year Book; 1997:1–31.
11. Yaghouti F, Nouri M, Abad JC, et al. Keratoprosthesis: preoperative prognostic categories.Cornea. 2001;20:19–23.
12. Netland PA, Terada H, Dohlman CH. Glaucoma associated with keratoprosthesis.Ophthalmology. 1998;105:751–757.
13. Ray S, Khan BF, Dohlman CH, et al. Management of vitreoretinal complications in eyes with permanent keratoprosthesis.Arch Ophthalmol. 2002;120:559–566.
14. Koh MJ, Tay YK. An update on Stevens-Johnson syndrome and toxic epidermal necrolysis in children.Curr Opin Pediatr. 2009;21:505–510. 15. French LE, Trent JT, Kerdel FA. Use of intravenous immunoglobulin in
toxic epidermal necrolysis and Stevens-Johnson syndrome: our current understanding.Int Immunopharmacol. 2006;6:543–549.
16. Ahmed AR. Use of intravenous immunoglobulin therapy in autoimmune blistering diseases.Int Immunopharmacol. 2006;6:557–578.
17. Ciolino JB, Dohlman CH. Biologic keratoprosthesis materials. Int Ophthalmol Clin. 2009;49:1–9.
18. Iyer G, Pillai VS, Srinivasan B, et al. Modified osteo-odonto keratoprosthesis—the Indian experience—results of the first 50 cases.
19. Huang Y, Yu J, Liu L, et al. Moscow eye microsurgery complex in Russia keratoprosthesis in Beijing.Ophthalmology. 2011;118:41–46.
FIGURE 3. Kaplan–Meier survival curve for retention of first Boston keratoprosthesis type II implant, by preoperative diagnosis.