Comparison of Surgical Outcome After Ahmed Valve Implantation for Patients With and Without Fluocinolone Intravitreal Implant (Retisert)
We appreciate the important questions raised by Drs Wan and Lam and the opportunity to further clarify our analysis in the manuscript titled “Comparison of surgical outcome after Ahmed valve implantation for patients with and without fluocinolone intravitreal implant (Retisert).”1
In this study, we performed anterior chamber tube placement of an Ahmed glaucoma valve (AGV) with adjuvant mitomycin-C (MMC) in all adult eyes and Retisert placement in a cohort of study eyes with a history of uveitis. Although we agree that patients with preexisting indications for Pars Plana vitrectomy may benefit from a Pars Plana tube placement with Retisert, the patients in our study did not have a separate indication for vitrectomy. A Pars Plana vitrectomy is not required for Retisert placement. Given that intravitreal steroids can achieve prolonged therapeutic levels in the anterior chamber2 and to avoid complications of vitrectomy, we believe Pars Plana tube placement is not necessary unless there is a separate indication.
Our reported intraocular pressure (IOP) analysis was a cumulative probability of IOP difference over all encounters during the first 12 months to avoid the challenges of multiple comparisons at each individual office visit or performing a Kaplan-Meier analysis in a retrospective study when the exact timing of an event is uncertain. In response to these questions, we have run individual visit IOP T test comparisons that show an increasing trend toward difference over follow-up that is not statistically significant until postoperative month 12. As indicated in the paper, all of the patients received MMC adjunctive treatment during AGV survey in both study and control eyes. Prior studies have shown that MMC significantly reduces early postoperative IOP3 and may reduce the incidence of a hypertensive phase to as low as 4%,4 though this effect may diminish with time.3 We speculate that the antimetabolite effect may have minimized early IOP differences between the Retisert and control groups, with later IOP differences becoming significant when the steroid depot remained but antimetabolite effect was reduced.
Drs Wan and Lam argued that the improved surgical outcome with AGV in study eyes may be due to better controlled uveitis activity with Retisert instead of the effect of corticosteroids on endplate modulation. We believe that Retisert may have effects on both. All control eyes either with or without uveitis had well-controlled inflammation before and after the AGV implantation. Therefore, controlling inflammation alone is unlikely to explain this result.
Interestingly, a similar effect of Retisert on AGV implantation has been reported by another group. In total, 60 eyes of 60 patients were followed for 2 years (22 patients with uveitic glaucoma receiving AGV combined with Retisert, 16 patients with uveitic glaucoma receiving AGV alone, and 22 patients with primary open-angle glaucoma receiving AGV alone). The authors found that patients with uveitis treated with Ahmed combined with Retisert had significantly greater success rates than primary open-angle glaucoma patients or uveitis patients treated with Ahmed alone.5
Lastly, uveitis is in many ways an exaggerated version of the normal ocular inflammation experienced by the combination of intraocular surgery and foreign body reaction that may lead to AGV failures in nonuveitic eyes. Although the results of this study cannot be directly extrapolated to nonuveitic patients, they provide an interesting insight from the available data of patients who have an intraocular steroid depot and Ahmed valve. The risks of long-term steroids are much higher in phakic patients and especially significant in children. However, pediatric patients were not included in this study. Future studies, ideally with lower cost, long-acting steroids especially in pseudophakic, adult patients are indicated to determine whether there is potential benefit for other patient populations.