Audit of trabeculectomy at a tertiary referral hospital in east Africa

Aim: To audit medium-term results of trabeculectomy at a tertiary referral unit in east Africa. Methods: A retrospective review of the files of all patients who underwent trabeculectomy surgery at our hospital during the year 2001. Results: One hundred fifty-seven patients were included. Mean age was 61 years (SD = 12). Thirty-one patients (20%) were women; 120 of 145 (83%) eyes had cup/disc ratios of 0.8 or greater preoperatively. Intraoperative 5-fluorouracil (5FU) was used in 57 cases (36%). Mean follow-up was 8 months (0-28, SD 7). Mean intraocular pressures (IOP) was 25 mm Hg (6-70, SD 10) preoperatively, 11 mm Hg (0-32, SD = 5) at 2 weeks postoperatively, and 13 mm Hg (0-42, SD = 6) at latest follow-up. One hundred fifteen patients (73%) had intraocular pressure of 15 mm Hg or less at latest follow-up and 141 (90%) had intraocular pressure of 21 mm Hg or less. Twenty (13%) had restarted antiglaucoma medication during the follow-up period. Thirty-seven of 148 (25%) lost 2 lines of Snellen acuity or equivalent between preoperative measurement and latest follow-up. Early postoperative complications (usually leaking bleb or shallow anterior chamber with hypotony) occurred in 18 patients (12%) and this was associated with postoperative loss of 2 or more lines of Snellen acuity (OR = 1.11 95% CI 1.02-1.22, P = 0.02). Patients having had 5FU had lower (t = 3.11, P = 0.02) mean intraocular pressure at 2-month follow-up (12 mm Hg, SD = 5) than those who did not receive it (16 mm Hg, SD = 6) but there was no significant difference at later follow-ups. Men (mean 14 mm Hg, SD = 6) had higher (t = 2.71, P = 0.01) intraocular pressures at latest follow-up than women (mean 11 mm Hg, SD = 3). Conclusions: The results of surgery compare well with other reported series. 5FU did not demonstrate significant benefit in terms of intraocular pressure lowering beyond 2 months postoperatively. Early postoperative hypotony should be avoided. The gender difference in intraocular pressure results has not previously been reported and deserves further investigation.