PURPOSE: To characterize long-term outcomes of tube shunt revision surgeries and identify factors associated with their failure. DESIGN: Retrospective chart review. PARTICIPANTS: One eye from each of 179 patients who underwent tube shunt revision surgery at the Wilmer Eye Institute between 2004 and 2015 with a minimum follow-up of 6 weeks. METHODS: Eligible eyes were identified from billing records and data related to their care were extracted from electronic medical records. Eyes were analyzed in aggregate and by indication for revision, including hypotony, high intraocular pressure (IOP), tube reposition, and tube exposure. MAIN OUTCOME MEASURES: Surgical failure, defined as a need for further tube shunt revision or other glaucoma surgery, unsatisfactory IOP at last follow-up, or both. Secondary outcomes included postoperative infection and functional visual impairment. RESULTS: With a median follow-up of 4.2 years, 126 failures occurred among 179 eyes. By Kaplan-Meier analysis, the cumulative rates of surgical failure at 1, 2, and 5 years after revision were 49%, 59%, and 74%, respectively. Most revision failures (105/126) were the result of the need for additional surgery, whereas 11 eyes showed IOP above target levels and 10 eyes showed dysfunctionally low IOP at the last follow-up. Factors associated with failure in a stepwise regression model were revision for hypotony (hazard ratio [HR], 6.79; P = 0.002), different surgeons performing the original and revision surgeries (HR, 2.80; P = 0.002), longer duration of symptoms before revision (P = 0.01), revision of a right eye (HR, 1.92; P = 0.03), and presumed preoperative infection (HR, 2.47; P = 0.04). In univariate analysis, success varied significantly by prior surgeon (P = 0.01), but not by revision surgeon. There was a 16% cumulative incidence of postoperative infection, with the highest risk in those with presumed preoperative infections (P = 0.01) and in persons of non-African, non-European derivation (P = 0.03). CONCLUSIONS: The estimated rate of failure of tube shunt revision is 75% by 5 years, most often because of a need for further surgery. The major potentially modifiable feature associated with success is tube shunt revision being performed by the original surgeon.
ASJC Scopus subject areas