Treatment outcomes in the neovascular glaucoma tube versus trabeculectomy study.

Abstract

To compare surgical outcomes between patients who underwent Baerveldt glaucoma implantation versus trabeculectomy (TLE) for the treatment of neovascular glaucoma. We performed a prospective, randomized, controlled clinical trial at Hiroshima University and Tsukazaki Hospital, Japan. Patients were randomized to Baerveldt glaucoma implantation (23 patients) or TLE (27 patients). The primary outcome measure was the rate of intraocular-pressure control. Secondary outcome measures included complications and interventions. We defined "failure" as intraocular pressure ≥ 22mmHg or < 20% reduction from baseline pressure, on two consecutive follow-up visits; need for additional glaucoma surgery; vision-threatening complications; or loss of light perception. Kaplan-Meier analysis was used to determine surgical success rates. Postoperative complications and interventions were compared between the two groups with Pearson Chi-square tests. There were no significant differences in demographic or ocular characteristics between the treatment groups at baseline. The postoperative follow-up was 26.6 ± 19.4months in the Baerveldt group and 27.3 ± 20.1months in the TLE group. There were no statistical differences (Mann-Whitney nonparametric tests with Bonferroni correction) in postoperative intraocular pressure measured at 6-month intervals. Success rates were 59.1 and 61.6% at 1-year after Baerveldt glaucoma implantation and TLE, respectively. Kaplan-Meier analysis showed no significant difference in success between groups (P = 0.71, log-rank test). Medication use, visual acuity, and interventions were similar between groups. Thirteen late postoperative complications developed in the Baerveldt glaucoma implantation group and four complications developed in the TLE group. Tube exposure was the most common late complication. These two procedures produced similar surgical success, intraocular-pressure reductions, visual acuity, and number of medications at the last visit. The TLE group showed smaller numbers of late complications and patients who lost more than two lines of visual acuity. Therefore, TLE might be a safer and better way to treat patients with neovascular glaucoma.