OBJECTIVE: To assess the factors that can predict the success of external needle revision of failing blebs. DESIGN: Retrospective, nonrandomized, comparative case series. PARTICIPANTS: Forty-nine eyes of 43 patients who underwent a needle revision of a filtration bleb with 5-fluourouracil (5-FU) by one physician (SVLB) between August 1993 and December 1994. INTERVENTION: Needle revision of a failing filtration bleb using 5-FU. MAIN OUTCOMES: The intraocular pressure, glaucoma medications, and any complications. RESULTS: The risk ratio of fornix-based trabeculectomies compared with limbus-based trabeculectomies was 3.781; P = 0.047. The risk ratios for gender, race, type of glaucoma, lens status, previous surgery, antimetabolite used for the initial trabeculectomy, and preoperative bleb characteristics were not statistically different for any one group. CONCLUSIONS: Fornix-based trabeculectomies were more likely to fail the needle revisions compared with limbus-based trabeculectomies. None of the five preneedling bleb characteristics showed a propensity toward success or failure. Other demographics such as gender, race, type of glaucoma, previous surgery, antimetabolite used for the initial trabeculectomy, and lens status were also not predictive for success.
OBJECTIVE: To assess the factors that can predict the success of external needle revision of failing blebs. DESIGN: Retrospective, nonrandomized, comparative case series. PARTICIPANTS: Forty-nine eyes of 43 patients who underwent a needle revision of a filtration bleb with 5-fluourouracil (5-FU) by one physician (SVLB) between August 1993 and December 1994. INTERVENTION: Needle revision of a failing filtration bleb using 5-FU. MAIN OUTCOMES: The intraocular pressure, glaucoma medications, and any complications. RESULTS: The risk ratio of fornix-based trabeculectomies compared with limbus-based trabeculectomies was 3.781; P = 0.047. The risk ratios for gender, race, type of glaucoma, lens status, previous surgery, antimetabolite used for the initial trabeculectomy, and preoperative bleb characteristics were not statistically different for any one group. CONCLUSIONS: Fornix-based trabeculectomies were more likely to fail the needle revisions compared with limbus-based trabeculectomies. None of the five preneedling bleb characteristics showed a propensity toward success or failure. Other demographics such as gender, race, type of glaucoma, previous surgery, antimetabolite used for the initial trabeculectomy, and lens status were also not predictive for success.