We read with appreciation the comments from Moussally and associates describing their experience with glaucoma drainage devices (GDD) in patients with Boston keratoprosthesis (KPro), in response to our recent article. We certainly concur with their conclusion that larger studies must be undertaken to investigate further the factors that may be involved in the erosion of GDD in patients with the Boston KPro.
Moussally and associates stated that they routinely place their KPro patients on a single antibiotic agent, a fourth-generation fluoroquinolone. In their patient population, there were no GDD erosions, and they have suggested that the use of fortified vancomycin may be more of a risk factor for GDD erosion than the use of a long-term soft contact lens.
Our speculation that the long-term soft contact lens used in Boston KPro patients may be contributing to GDD erosions has also been previously described in a study by Chew and associates. Soft contact lenses are routinely used in KPro patients to prevent evaporative damage to the ocular surface of these patients. We do not suggest discontinuing the use of long-term soft contact lenses in KPro patients. A multicenter study evaluating GDDs in KPro patients may provide us adequate numbers of patients to make more definitive conclusions regarding the risk factors leading to GDD erosions in KPro patients. The management of glaucoma continues to be one of the most vexing problems in keratoprosthesis surgery.
Regarding the use of fortified vancomycin in our patients, we have acknowledged in our article that these drops may cause some degree of toxicity to the ocular surface. This toxicity may accelerate breakdown of the ocular surface in these patients, many of whom have underlying ocular surface disease already. However, before any recommendations to discontinue this antibiotic drop are made, further study on the risks and benefits of fortified vancomycin drops in KPro patients must be conducted.
The routine use of fortified vancomycin eye drops in KPro patients was initially suggested in response to the high rate of endophthalmitis noted in an early Boston KPro study conducted between 1990 and 2000 by Nouri and associates. In that study, 13 patients (12%) developed bacterial endophthalmitis, most of which resulted in significant vision loss. All cases of endophthalmitis were caused by gram-positive cocci despite use of topical gentamicin, ofloxacin, or trimethoprim–polymyxin B for long-term prophylaxis. Since the routine use of topical vancomycin and fluoroquinolone drops at that institution, no cases of acute bacterial endophthalmitis have been reported.
Boston keratoprosthesis surgery continues to be an important procedure offering the hope of vision to patients with corneal blindness for whom traditional penetrating keratoplasty has a poor prognosis. More work is still to be done in order to understand how to best maximize the outcomes in these patients.