Objective: To assess the feasibility of an automatic switch of a large number of patients with glaucoma or suspicion of glaucoma from latanoprost to bimatoprost, and to compare the efficacy of the 2 prostaglandin analogs before and after the switch.
Design: Retrospective nonrandomized comparison.
Participants: Forty-three thousand four hundred forty-one California patients and 538 patients at one Southern California clinical facility of a nationwide prepaid health maintenance organization (HMO) who were on either prostaglandin between March 2002 and December 2003 (21 months).
Methods: Beginning in April 2002, patients on latanoprost were systemically switched to bimatoprost by the HMO pharmacy service after obtaining approval from the entire ophthalmology staff. PART 1: computerized dispensing records of California patients were retrieved. PART 2: medical records of patients at one clinical facility were reviewed.
Main outcome measures: Rates of switching or switching back from one prostaglandin to another, intraocular pressure (IOP) control, and intolerability.
Results: PART 1: 17847 patients initially received latanoprost. Of them, 84.8% were switched from latanoprost to bimatoprost, and 13.0% were switched back to latanoprost. Twenty-five thousand five hundred ninety-four patients were started on bimatoprost without previous experience with latanoprost. Of them, 8.6% were later switched to latanoprost use instead. Patients who had previous experience with latanoprost had a statistically significantly higher rate of switching back to latanoprost after a period of bimatoprost use when compared with those who had no prior experience with latanoprost (13.0% vs. 8.6%, respectively; P<0.0001). PART 2: 309 patients were switched from latanoprost to bimatoprost. The mean IOP reduction of 0.51+/-2.77 mmHg (95% confidence interval, 0.20-0.82) after the switch was statistically significant (P = 0.001). Forty-one patients (13.3%) had a decrease of >3 mmHg of IOP. The statistical significance of the IOP reduction after the switch remains in the monotherapy group (P = 0.005) but not in the multitherapy group (P = 0.058). Thirty-three patients (10.7%) who were switched from latanoprost to bimatoprost discontinued bimatoprost and resumed latanoprost.
Conclusion: A systematic pharmacy-level switch from latanoprost to bimatoprost in a nationwide HMO achieved a high switch rate, with little switching back. There was a small but statistically significant reduction in mean IOP after the switch. An appreciable proportion of patients switched had a clinically significant reduction of IOP.