advertisement
A randomized, doubled blinded, controlled clinical trial by the European Glaucoma Prevention Study (EGPS), Miglior et al. (207), has raised more questions about the management of glaucoma than it has answered. Dorzolamide did not reduce the likelihood of developing primary open angle glaucoma (POAG) among ocular hypertensives compared with placebo. Why? Placebo treatment with the inactive vehicle, was not less effective in lowering intraocular pressure (IOP) than treatment with the drug. The magnitude of IOP lowering in the dorzolamide group, comparable to the 20% IOP reduction target goal of the Ocular Hypertension Treatment Study (OHTS)1-3 did not lessen the rate of progression compared to the placebo treatment group. What is different about the participants in the EGPS compared to those in the OHTS? Should important randomized controlled clinical trials, such as the Early Manifest Glaucoma Trial and the OHTS be repeated with placebo treatment to confirm published results, as the authors suggest? The EGPS eligibility criteria and patient follow-up likely explain the basis for these questions and differences in interpretation rather than the sound design of a double masked clinical trial. Compared to the OHTS baseline IOP eligibility of 24 mmHg, EGPS patients had a baseline IOP of 22 mmHg or higher as measured by an investigator who was not masked to the reading. Approximately 65% of EGPS patients would not have been eligible for OHTS.4 The mean baseline EGPS IOP was 23.5 mmHg, compared with 24.9 for OHTS. A regression to the mean likely contributed to some initial IOP lowering in both the dorzolamide and placebo groups. The proportion of patients lost to follow-up and the discontinuation rate was much higher than usual in EGPS. Patients in both the treatment and placebo treatment arms who dropped out of the study had much higher IOP than those who remained, thereby diminishing the IOP lowering difference between the treatment effect of dorzolamide and placebo. By recruiting patients with lower baseline IOP and losing those with higher follow-up IOP in both groups, no statistically significant treatment difference was noted between drug and placebo.
At first glance the fundamental question of whether IOP lowering prevents or delays the development of POAG among ocular hypertensive patients appears similar in the EGPS and OHTS; however, the apparently conflicting conclusions reflect substantial methodological differences rather than opposing statistically supported findings.