Idiopathic Vitritis: An Elusive Complication of Boston Keratoprosthesis Surgery

Abstract

Introduction: An idiopathic inflammation of the vitreous humor, “sterile vitritis,” has been observed in some keratoprosthesis recipients, typically months to years after implantation. Its etiology is unclear. Potential mechanisms include: 1) local tissue destruction and inflammation around the prosthesis, at the point of contact between cornea donor tissue and the device, 2) microbial cell wall or nucleic acid triggering inflammation, or 3) a systemic immune response with ocular manifestations. Purpose: 1) To present possible theories for vitreous inflammation, 2) revisit and redefine the clinical paradigm originally attributed to Boston keratoprosthesis recipients presenting with idiopathic culture-negative vitreous inflammation, and 3) investigate one potential mechanism of idiopathic vitritis via a case-control study, using a unique application of anterior-segment OCT imaging. Subjects: 346 adult patients with a keratoprosthesis, performed by three surgeons (JC, KAC, CHD) at Massachusetts Eye and Ear Infirmary from January 2000 to August 2013. Methods: A retrospective chart review was performed of the subject group at Massachusetts Eye and Ear Infirmary for demographic data, indication(s) for surgery, timing and presentation of vitreous inflammation, and best-corrected visual acuity at baseline, on presentation, and after resolution of vitritis. Of those in the subject group, 13 KPro eyes with a history of sterile vitritis and 34 KPro eyes without a history of sterile vitritis underwent anterior segment optical coherence tomography (AS-OCT) an average of 3.6 years postoperatively (range 1.9-14.2 yrs). Areas of corneal graft tissue loss (“gaps”) around the KPro stem were identified. Image analysis was performed by two masked observers. The difference in presence, number and size of gaps was compared between cases and controls. Results: 43 patients were found to have vitreous inflammation between 2 days and 8.5 years postoperatively. 23 cases (43 total episodes) showed no obvious cause for inflammation. 7/23 patients presented with signs and symptoms similar to infectious endophthalmitis but were culture-negative. The number of patients who fit the previous paradigm of sudden painless loss of vision without external signs of infectious endophthalmitis (“sterile vitritis”) was 8 of 23 cases. Vision decline was variable (median loss of 9 lines on Snellen Chart, range 0-24), and median time to recovery of best vision was 8.9 weeks (range 0.86-36.7). 12/23 patients had repeat bouts of vitritis. 10/43 episodes did not recover to baseline vision. 17/23 patients later developed retroprosthetic membranes (13), glaucoma (8), cystoid macular edema (3), and/or retinal detachment (2). Among sterile vitritis patients imaged using AS-OCT, thirteen patients (13 eyes) had a Type I keratoprosthesis. These patients were compared with 34 Type I controls. There was a significant difference in the proportion of patients presenting with at least 1 gap (11/13= 86%) on AS-OCT as compared to controls (11/34=33.3%, p<0.001). There was a significant difference in the number of gaps between cases and controls (2.6±1.6 vs. 0.5±0.8, p< 0.001), but no significant difference in gap size (0.056 ± 0.049 mm2 vs. 0.039 ± 0.025 mm2, p= 0.22). Conclusions: The paradigm for sterile vitritis after keratoprosthesis implantation includes sudden, painless loss of vision with full recovery of vision upon treatment with periocular corticosteroids. However, this existing paradigm does not capture the full spectrum of the disorder. Sterile vitritis after keratoprosthesis can also mimic infectious endophthalmitis, and vision may not recover to baseline despite treatment. A combination of factors likely contribute to each case of idiopathic vitreous inflammation. Specifically, tissue remodeling around the keratoprosthesis may create a dynamic space for inorganic or organic debris to enter the eye. A significantly higher proportion of KPro eyes with sterile vitritis had tissue loss around the KPro stem on AS-OCT than controls. Tissue loss may serve as an entry point for debris or bacterial components, triggering sterile vitritis. Our study underscores the importance of surveillance with AS-OCT imaging in the post-operative management of KPro patients.