Results
Data from 19 eyes from 17 patients with 20 GDDs were included in this study. Six GDD exposure cases were included in our previous report at a shorter follow-up point. Baseline data is described in Table 1. The mean patient age (SD) was 59.1 (13.7) years, and 59% of patients were male. Eight of the subjects had primary open-angle glaucoma, 4 had chronic angle closure glaucoma, 1 had juvenile glaucoma, 1 had congenital glaucoma, and 3 had secondary angle closure glaucoma. The mean (SD) number of previous ocular surgeries in this population before the buccal membrane transplant was 4.8 (2.9).
Of the 20 GDDs, there were 19 Ahmed valves and 1 Molteno implant. The majority (10) of the GDDs were located in the superotemporal quadrant, 5 were positioned in the inferonasal quadrant, 4 in the inferotemporal quadrant, and 1 in the superonasal quadrant. Each of the Ahmed valves were originally covered using a corneal patch graft and superficial conjunctival coverage. Original coverage of the Molteno valve was done with a scleral graft and conjunctival reapproximation.
With respect to erosion with exposure, there were 16 cases of tube erosion, 2 cases of plate erosion, and 2 cases of pars plana clip erosion. The mean (SD) time from the original GDD procedure to the first erosion was 4.6 (3.7) years for cases of tube, plate, and clip erosion. However, for cases of tube erosion alone the mean (SD) interval from initial GDD implantation to erosion was 5.3 (3.8) years. Four cases (19%) underwent an erosion repair procedure before the buccal mucous membrane transplant procedure (Table 2). Three initial repairs involved a corneal patch graft with superficial conjunctival coverage, and 1 used a sclera graft with conjunctival advancement. The mean (range) length of survival of these initial repairs was 1.6 (0.1 to 3.3) years. In the other 16 cases included in this study, initial repair involving conjunctival reapproximation was not attempted because these subjects lacked sufficient viable or mobile conjunctiva necessary for such a procedure. The paucity of tissue was attributable to scarring around the exposed tube.
There were no instances of intraoperative complications. The mean (SD) length of follow-up from this procedure was 1.7 (1.2) years. Initial repair with buccal membrane transplant was considered a success in 70% of cases, which included 13 cases of tube erosion, and 1 case of plate erosion. An additional 3 cases (15%), consisting of 2 cases of tube erosion and 1 case of clip erosion, achieved coverage of the GDD after 2 repairs. There were no cases of qualified success reported. All patients who required minor additional procedures to maintain success progressed to reerosion requiring repeat buccal graft repair procedures. Outcome data is summarized in Table 3. Overall success rate after ≥1 buccal mucous membrane graft repairs was 85%. Three cases (15%), including 1 tube, 1 plate and 1 clip erosion, were ultimately failures despite 2 repair attempts.
Subgroup analysis by erosion site is summarized in Table 4. Tube erosions were successfully covered in 94% (15/16) of cases after ≥1 repairs. Cases of plate erosions were successful in one of 2 cases. Both cases of clip erosions were surgical failures after 1 repair, but 1 achieved success after a second repair.
The mean durations until initial exposure were 5.2 and 0.84 years after GDD implantation for the overall success and failure groups, respectively. The mean number of previous ocular surgeries was 5.3 and 2.0 for the success and failure groups, respectively. No other variables of interest, including age at the buccal mucous membrane repair procedure and number of ocular medications, proved to be statistically significant. Results are described in Table 5.