This study compared the success rate of endoscopic cyclophotocoagulation (ECP) and the Ahmed drainage implant in the treatment of refractory glaucoma. Sixty-eight eyes of 68 patients were randomly assigned either ECP or Ahmed tube shunt implantation. Eyes that were included were pseudophakic with a history of at least one trabeculectomy with an antimetabolite, an intraocular pressure (IOP) equal to or above 35 mm Hg on maximum tolerated therapy, and a visual acuity of better than light perception. There were no statistically significant differences in the preoperative IOP of the Ahmed valve and ECP groups. Exclusion criteria included previous glaucoma drainage device implantation or a cyclodestructive procedure. All procedures were performed by a single surgeon. Success was defined as an IOP of more than 6 mm HG and less than 21 mm Hg, with or without topical anti-hypertensive therapy. After a mean follow-up was 19.82 ± 8.35 months and 21.29 ± 6.42 months for the Ahmed valve and ECP groups respectively (P = .04), there were no statistically significant differences in the mean IOP of the two groups (14.73 ± 6.44 mm vs 14.07 ± 7.21 mm, respectively). There were significant differences between pre- and postoperative IOP for both groups (P < .001). Kaplan-Meier survival curve analysis showed a probability of success at 24 months of 70.59% and 73.53% for the Ahmed and ECP groups, respectively (P = .7). Complications included choroidal detachment (Ahmed 17.64%, ECP 2.94%), shallow anterior chamber (Ahmed 17.64%, ECP 0%), and hyphema (Ahmed 14.7%, ECP 17.64%). The authors conclude that Ahmed shunt implant and ECP have similar success rate in the treatment of refractory glaucoma, but that the complication rate is lower with ECP.