Effects of pre-surgical administration of prostaglandin analogs on the outcome of trabeculectomy.

• Main Authors: Takako Miki, Tomoko Naito, Miyuki Fujiwara, Ryoichi Araki, Rieko Kiyoi, Yusuke Shiode, Atsushi Fujiwara, Yuki Morizane, Fumio Shiraga
• Format: Article
• Language: English
• Published: Public Library of Science (PLoS) 2017-01-01
• Series: PLoS ONE
• Online Access: http://europepmc.org/articles/PMC5519176?pdf=render

For primary open angle glaucoma (POAG), laser treatment or surgery is used when the target intraocular pressure (IOP) cannot be achieved by pharmacological agents, such as prostaglandin (PG) analogs; these drugs also have varied effects. We retrospectively reviewed the medical records of 74 POAG patients (74 eyes) whose IOP was inadequately controlled by PG analogs (bimatoprost [13 eyes], latanoprost [34 eyes], tafluprost [11 eyes], and travoprost [16 eyes]) and underwent primary trabeculectomy. The proportion of patients with no recurrent IOP elevation within 24 months post-trabeculectomy was significantly (P < 0.001) lower in the bimatoprost group (31.3%) than in the latanoprost (83.2%), tafluprost (45.5%), or travoprost groups (65.6%). Deepening of the upper eyelid sulcus (DUES) was observed before trabeculectomy in 18 of 74 eyes (24.3%) treated with bimatoprost (9 eyes; 50.0%), latanoprost (3 eyes; 16.7%), tafluprost (1 eye; 5.5%) and travoprost (5 eyes; 27.8%). The proportion of patients with no recurrent IOP elevation up to 24 months post-trabeculectomy was significantly (P < 0.0001) lower in the DUES(+) group (34.7%) than in the DUES(-) group (74.3%). Multivariate stepwise logistic regression analysis, with no recurrent IOP elevation used as dependent variable, and bimatoprost, latanoprost, travoprost, tafluprost, β-blocker, carbonic anhydrase inhibitor, brimonidine, gender, age, preoperative IOP, mean deviation, duration of PG analog use before surgery, and the number of ophthalmic solutions used as independent variables, identified only bimatoprost as a significant independent factor (P = 0.0368). Thus, the outcome of trabeculectomy varied depending on the PG analog used preoperatively, and bimatoprost use was associated with a high risk of recurrent IOP elevation up to 2 years post-trabeculectomy. This may indicate that the incidence of DUES differed with the PG analog used. Patients with glaucoma who are treated with bimatoprost should be monitored for DUES, and when these patients undergo trabeculectomy, the postoperative course of IOP should be followed carefully.