To determine whether filtering blebs resulting from adjunctive use of mitomycin C (MMC) leads to an increased risk of endophthalmitis. The authors retrospectively reviewed the records of 232 consecutive trabeculectomies performed at the W. K. Kellogg Eye Center with adjunctive use of MMC from May 1990 through June 1993. Data obtained from the records included patient age, sex, race, type of glaucoma, site of filtration surgery, concentration and duration of exposure to MMC, presence of early or late bleb leakage, and the occurrence of endophthalmitis. Three patients were lost to follow-up less than 1 month after surgery. A total of 229 eyes of 192 patients (11 women and 82 men) were included in the study. Mean follow-up of patients remaining free of infection was 18.5 +/- 10.8 months (range, 1-44 months). The overall incidence of bleb-related endophthalmitis was 2.6%. Endophthalmitis developed in 8% of patients (4 or 50) in whom an inferior approach was used and in 1.1% (2 or 179) in whom a superior approach was used (P = 0.02, Fisher's exact test). The estimated odds ratio for the development of endophthalmitis after trabeculectomy with adjunctive MMC for inferior versus superior filtration sites was 7.7. Short-term follow-up of trabeculectomies performed with adjunctive use of MMC demonstrates an overall incidence of endophthalmitis comparable to filtrationprocedures performed with 5-fluorouracil or without antifibrotic agents. However, inferior trabeculectomy performed with adjunctive MMC carries a significantly increased risk of bleb-related endophthalmitis compared with filters performed superiorly.