Mark A. Greiner

Longer-Term Vision Outcomes and Complications with the Boston Type 1 Keratoprosthesis at the University of California, Davis

PURPOSE To evaluate retention of visual acuity and development of complications after Boston type 1 keratoprosthesis implantation over a longer follow-up period than previously reported. DESIGN Cohort study. PARTICIPANTS Forty eyes of 35 patients who underwent Boston type 1 keratoprosthesis surgery at the University of California, Davis, between 2004 and 2010. METHODS Preoperative, intraoperative, and postoperative parameters were collected and analyzed. MAIN OUTCOME MEASURES Best-corrected visual acuity (BCVA) and postoperative complications. RESULTS Preoperative visual acuity ranged from 20/150 to light perception and was ≤20/400 in 38 eyes (95%). Preoperative diagnoses included failed corneal transplants (19 eyes, 47.5%), chemical injury (10 eyes, 25%), and aniridia (5 eyes, 12.5%). Mean follow-up duration was 33.6 months (range, 5-72 months). Of 36 eyes followed for ≥1 year, 32 eyes (89%) achieved postoperative BCVA ≥20/200. Of eyes that achieved BCVA ≥20/200, at last follow-up, 19 of 32 eyes (59%) followed for ≥1 year retained BCVA ≥20/200; 16 of 27 eyes (59%) followed for ≥2 years retained BCVA ≥20/200; 7 of 14 eyes (50%) followed for ≥3 years retained BCVA ≥20/200; and 2 of 7 eyes (29%) followed for ≥4 years retained BCVA ≥20/200. End-stage glaucoma most commonly caused vision loss (7 of 13 eyes, 54%) when BCVA ≥20/200 was not retained (follow-up ≥1 year). Glaucoma was newly diagnosed in 11 eyes (27.5%); progression was noted in 9 eyes (22.5%). Glaucoma drainage device erosion occurred in 9 eyes (22.5%). Retroprosthetic membrane formed in 22 eyes (55%), 5 eyes (12.5%) developed endophthalmitis, 6 eyes (15%) developed corneal melt, 7 eyes (17.5%) underwent keratoprosthesis replacement, and 23 eyes (57.5%) required major surgery to treat postoperative complications. The initial keratoprosthesis was retained in 32 eyes (80%). CONCLUSIONS Keratoprosthesis implantation remains a viable option for salvaging vision. A significant number of patients lost vision over the postoperative course. Glaucoma and complications related to glaucoma surgery are significant challenges to maintaining good vision after keratoprosthesis surgery. Our study highlights the need for long-term follow-up and a team approach to management, and points to a more guarded long-term visual prognosis after surgery. FINANCIAL DISCLOSURE(S) The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Long-term Complications Associated with Glaucoma Drainage Devices and Boston Keratoprosthesis

PURPOSE To evaluate long-term complications related to glaucoma drainage devices in patients undergoing Boston type 1 keratoprosthesis surgery. DESIGN Retrospective case series. METHODS All patients who underwent Boston type 1 keratoprosthesis surgery at the University of California, Davis, between 2004 and 2010 were included. Preoperative and postoperative data were reviewed. Twenty-five eyes with glaucoma drainage devices were highlighted. Visual acuity and postoperative complications were tracked at postoperative months 1, 3, 6, 9, and 12 and at annual intervals thereafter. RESULTS Forty eyes of 35 patients were evaluated with an average follow-up of 33.6 months. Conjunctival breakdown occurred in association with 10 glaucoma drainage device implants in 9 eyes. Eleven eyes had glaucoma drainage devices placed before keratoprosthesis surgery, 3 eyes underwent glaucoma drainage device placement at the time of surgery, and 2 eyes had a glaucoma drainage device placed after surgery. All but one of the eroded glaucoma drainage devices were placed before surgery. Associated complications included endophthalmitis, hypotony, and keratoprosthesis extrusion, with 6 glaucoma drainage devices requiring removal. Long-term beset-corrected visual acuity was maintained better in eyes in which glaucoma drainage device erosions did not develop. CONCLUSIONS One of the main challenges with keratoprosthesis surgery is treating concurrent glaucoma. Glaucoma drainage devices have been advocated as a way to address this long-term complication, but this series suggests that glaucoma drainage device-related complications can cause significant vision loss.

Diabetes mellitus increases risk of unsuccessful graft preparation in Descemet membrane endothelial keratoplasty: a multicenter study.

Purpose: The aim of this study was to evaluate preparation outcomes of tissue prepared for Descemet membrane endothelial keratoplasty (DMEK) from diabetic and nondiabetic donors. Methods: In this nonrandomized, consecutive case series, DMEK grafts were prepared from diabetic and nondiabetic donors by experienced technicians in 2 eye banks using slightly different, modified submerged manual preparation techniques to achieve “prestripped” graft tissue. Graft preparation results were analyzed retrospectively. The main outcome measure was the rate of unsuccessful (failed) DMEK graft preparations, defined as tears through the graft area that prevent tissue use. Results: A total of 359 corneas prepared from 290 donors (114 diabetic and 245 nondiabetic) were included in the statistical analysis of graft preparation failure. There were no significant differences between diabetic and nondiabetic donor tissue characteristics with respect to donor age, death to preservation time, death to preparation time, endothelial cell density, percent hexagonality, or coefficient of variation. DMEK tissue preparation was unsuccessful in 19 (5.3%) cases. There was a significant difference in the site-adjusted rate of DMEK preparation failure between diabetic [15.3%; 95% confidence interval (CI), 9.0–25.0] and nondiabetic donors (1.9%; 95% CI, 0.8–4.8), and the corresponding site-adjusted odds ratio of DMEK graft preparation failure in diabetic donor tissue versus nondiabetic donor tissue was 9.20 (95% CI, 2.89–29.32; P = 0.001). Conclusions: Diabetes may be a risk factor for unsuccessful preparation of donor tissue for DMEK. We recommend caution in the use of diabetic tissue for DMEK graft preparation. Further study is needed to identify what subset of diabetic donors is at risk for unsuccessful DMEK graft preparation.

Fourier-Domain Optical Coherence Tomography and Adaptive Optics Reveal Nerve Fiber Layer Loss and Photoreceptor Changes in a Patient With Optic Nerve Drusen

Background: New technology allows more precise definition of structural alterations of all retinal layers although it has not been used previously in cases of optic disc drusen. Methods: Using Stratus and Fourier domain (FD) optical coherence tomography (OCT) and adaptive optics (AO) through a flood-illuminated fundus camera, we studied the retinas of a patient with long-standing optic disc drusen and acute visual loss at high altitude attributed to ischemic optic neuropathy. Results: Stratus OCT and FD-OCT confirmed severe thinning of the retinal nerve fiber layer (RNFL). FD-OCT revealed disturbances in the photoreceptor layer heretofore not described in optic disc drusen patients. AO confirmed the FD-OCT findings in the photoreceptor layer and also showed reduced cone density at retinal locations associated with reduced visual sensitivity. Conclusions: Based on this study, changes occur not only in the RNFL but also in the photoreceptor layer in optic nerve drusen complicated by ischemic optic neuropathy. This is the first reported application of FD-OCT and the AO to this condition. Such new imaging technology may in the future allow monitoring of disease progression more precisely and accurately.

Boston type 1 keratoprosthesis for chemical and thermal injury.

Purpose: To evaluate the outcome of the Boston type 1 keratoprosthesis (Kpro-1) in eyes with failed interventions for chemical and thermal injury. Methods: A retrospective review was performed of every eye with chemical or thermal injury that was treated with a Kpro-1 at a tertiary eye care center between January 1, 2008 and July 1, 2013. The main outcome measures were visual outcome, prosthesis retention, and postoperative complications. Results: Nine eyes met the inclusion criteria, including 7 eyes with alkali burns, 1 eye with an acid burn, and 1 eye with a thermal burn. After a mean follow-up of 40.7 months (range, 29–60 months), the median best-corrected visual acuity was 20/60 (range, 20/15 to no light perception). One eye was ≥20/20, 3 eyes were ≥20/40, and 6 eyes were ≥20/70. The initial Kpro-1 prosthesis was retained in 7 (77.7%) eyes and successfully replaced in the other 2 eyes. One or more serious complications occurred in 6 (66.7%) eyes. These included 2 cases of sterile corneal ulceration with prosthesis extrusion, 2 cases of microbial keratitis (1 bacterial and 1 fungal), 2 cases of bacterial endophthalmitis, and 2 cases of retinal detachments. These complications contributed to visual outcomes of hand motions in 2 eyes and no light perception in 1 eye. Conclusions: The Boston Kpro-1 is associated with highly satisfactory visual outcomes and prosthesis retention in most cases of severe chemical or thermal injury. Serious complications are common and may compromise the final outcome.

Graft survival versus glaucoma treatment after penetrating or Descemet stripping automated endothelial keratoplasty.

Purpose: The aim of this study was to assess and compare the association of glaucoma therapy with graft survival after performing penetrating keratoplasty (PKP) and Descemet stripping automated endothelial keratoplasty (DSAEK). Methods: A retrospective chart review was performed of cases: primary PKP from January 1, 2003, to December 31, 2005, or primary DSAEK from January 1, 2006, to December 31, 2008. Eyes with a surgical indication of pseudophakic corneal edema were included in the statistical analysis. Eyes were stratified by glaucoma treatment into those with (1) no glaucoma treatment, (2) medical therapy only, or (3) surgical intervention. The main outcome measure was graft survival. Results: Fifty-seven PKP-operated and 156 DSAEK-operated eyes met the inclusion criteria. After PKP and DSAEK, respectively, the 5-year Kaplan–Meier graft survival was 94.7% and 93.8% in eyes with no glaucoma treatment (P > 0.99), 93.8% and 96.3% in eyes with medical therapy only (P > 0.99), and 56.8% and 50% in eyes with surgical intervention (P > 0.99). After both procedures were performed, graft survival was significantly worse in eyes with surgical intervention compared with that in eyes with no glaucoma treatment (P < 0.0001) or in eyes with medical therapy alone (P < 0.0001). Conclusions: PKP and DSAEK have comparable graft survival in eyes without glaucoma management and in those with comparable glaucoma management.

Graft survival of diabetic versus nondiabetic donor tissue after initial keratoplasty.

Purpose: To compare corneal graft survival using tissue from diabetic and nondiabetic donors in patients undergoing initial Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty (PKP). Methods: A retrospective chart review of pseudophakic eyes that underwent DSAEK or PKP was performed. The primary outcome measure was graft failure. Cox proportional hazard regression and Kaplan–Meier survival analyses were used to compare diabetic versus nondiabetic donor tissue for all keratoplasty cases. Results: A total of 183 eyes (136 DSAEK, 47 PKP) were included in the statistical analysis. Among 24 procedures performed using diabetic donor tissue, there were 4 cases (16.7%) of graft failure (3 DSAEK, 1 PKP), and among 159 procedures performed using nondiabetic donor tissue, there were 18 cases (11.3%) of graft failure (12 DSAEK, 6 PKP). Cox proportional hazard ratio of graft failure for all cases comparing diabetic with nondiabetic donor tissue was 1.69, but this difference was not statistically significant (95% confidence interval, 0.56–5.06; P = 0.348). There were no significant differences in Kaplan–Meier curves comparing diabetic with nondiabetic donor tissue for all cases (P = 0.380). Statistical analysis of graft failure by donor diabetes status within each procedure type was not possible because of the small number of graft failure events involving diabetic tissue. Conclusions: We found similar rates of graft failure in all keratoplasty cases when comparing tissue from diabetic and nondiabetic donors, but further investigation is needed to determine whether diabetic donor tissue results in different graft failure rates after DSAEK compared with PKP.

Boston Type 1 Keratoprosthesis: Visual Outcomes, Device Retention, and Complications.

Purpose: To determine the visual outcomes, device retention, and complications after Boston type 1 keratoprosthesis (KPro-1) device implantation. Methods: Comprehensive review of every case of KPro-1 implantation at a tertiary eye care center. Results: The initial KPro-1 procedure, performed in 75 eyes without a previous keratoprosthesis, was included in the analysis. During the first 6 postoperative months, improvement occurred in the mean preoperative best-corrected visual acuity of 20/1265 to a mean best-obtained postoperative visual acuity of 20/97 (P < 0.001). After a mean follow-up period of 41.4 months (range, 0.8–82.8 months), the final mean best-corrected visual acuity was 20/428. Improved vision was recorded in 43 eyes (57.3%), ambulatory vision (≥20/400) in 47 eyes (62.7%), intermediate functional vision (≥20/80) in 23 eyes (30.7%), and full functional vision (≥20/40) in 11 eyes (14.7%). The initial device was retained in 64 eyes (85.3%), with a Kaplan–Meier retention probability of 96% at 6 months and 82% at 5 years. One or more sight-threatening complications occurred in 51 eyes (68%). These included device extrusion in 11 eyes (14.7%), ulcerative keratitis in 12 eyes (16%), endophthalmitis in 7 eyes (9.3%), sterile vitritis in 3 eyes (4%), retroprosthetic membranes in 25 eyes (33.3%), maculopathy in 26 eyes (34.7%), retinal detachment in 9 eyes (12%), and progressive optic neuropathy in 7 eyes (9.3%). Conclusions: Boston KPro-1 implantation is associated with satisfactory visual outcomes and excellent device retention in a majority of cases. However, serious postoperative complications are common and may compromise the final visual result.

Immediate endothelial cell loss after penetrating keratoplasty.

Purpose: To report the pattern and the extent of corneal endothelial cell loss immediately after a penetrating keratoplasty (PK) is performed. Methods: Ten donor corneal-scleral tissues with healthy endothelium were used to perform 10 PK surgeries on cadaver eyes. An 8.25-mm donor graft was placed into an 8.00-mm trephinated recipient bed of the cadaver eye. A 10-0 nylon suture was used with a 16-interrupted suture closure technique to secure each donor button. Viscoelastic was used to protect the endothelium in every case. The corneal donor buttons were removed and stained using trypan blue and alizarin red dyes. Digital high-definition photographs were obtained to record the pattern and quantity of stain resulting from endothelial damage and cell loss. Percent endothelial loss was calculated using the digital planimetry of our previously described Adobe Photoshop technique. Results: The immediate mean percent cell loss across the whole graft was 25.7% ± 7.5% (Range, 18–39). There was a distinct pattern of loss, with 58.3% of the loss in the peripheral 0.75 mm, and 12.4% of the loss in the central 3.00 mm. Conclusions: The immediate total endothelial cell loss after the PK was performed was about 25%. The immediate cell loss after the PK was consistent with areas of trephination and suture placement.

Boston keratoprosthesis type 1 for herpes simplex and herpes zoster keratopathy.

Purpose: The aim of this study was to evaluate and compare the outcomes of Boston keratoprosthesis type 1 (Kpro-1) in eyes with herpes simplex virus (HSV) and herpes zoster virus (HZV) keratopathy. Methods: A retrospective review was performed of the medical records of every patient treated with a Boston Kpro-1 at the University of Iowa Hospitals and Clinics between January 1, 2008 and July 1, 2012. Eyes with visual loss due to HSV or HZV keratopathy were included in the statistical analysis. The main outcome measures were graft retention, postoperative complications, and visual outcome. Results: Nine eyes met the inclusion criteria, including 5 eyes in the HSV group and 4 eyes in the HZV group. The graft retention rate was 100% in the HSV group after a mean follow-up of 48.4 months, compared with 25% in the HZV group after 50.5 months (P = 0.048). There were 3 cases of microbial keratitis, including 2 eyes that also developed endophthalmitis, in the HZV group, compared with no cases in the HSV group (P = 0.048). There was significantly better best-corrected visual acuity at the most recent examination in the HSV group than in the HZV group (P = 0.019). All 5 HSV eyes had improved best-corrected visual acuity compared with preoperative acuity, whereas only 1 HZV eye experienced a similar result (P = 0.048). Conclusions: Kpro-1 is associated with an excellent prognosis for graft retention, acceptably low prevalence of sight-threatening complications, and highly satisfactory visual improvement in eyes with HSV keratopathy, but not in eyes with HZV keratopathy.