Incident open-angle glaucoma and intraocular pressure

Abstract

Purpose: To evaluate the role of baseline intraocular pressure (b-IOP) as a risk factor for incident open-angle glaucoma (OAG) in participants of African origin from the Barbados Eye Studies.

Design: Population-based 9-year cohort study.

Participants: Three thousand two hundred twenty-two persons examined during the study period who were free of glaucoma at baseline and at risk of developing OAG during the 9-year follow-up.

Methods: Study protocols were standardized and included ophthalmic and other measurements, automated perimetry, applanation tonometry, fundus photography, and comprehensive ophthalmologic examination for those referred. The product-limit approach was used to estimate incidence. Relationships between b-IOP and incidence were evaluated by adjusted relative risk ratios (RRs) with 95% confidence intervals (CIs), based on Cox regression models.

Main outcome measure: The 9-year incidence of OAG was based on both visual field and optic disc abnormalities, with ophthalmologic evaluations to exclude other possible causes.

Results: The overall 9-year incidence of OAG was 4.4% (95% CI, 3.7%-5.2%), and the mean (standard deviation) b-IOP among persons at risk was 18.0 mmHg (4.1). Among the 125 incident OAG cases, the mean b-IOP was 21.9 mmHg and 46% had b-IOP of >21 mmHg. In contrast, the nonincident group had a mean b-IOP of 17.8 mmHg and only 12% had b-IOP of >21 mmHg. Overall, OAG risk increased by 12% with each 1-mmHg increase in IOP (RR, 1.12; 95% CI, 1.08-1.16). Incidence steadily increased from 1.8% (95% CI, 1.2%-2.7%) for persons with b-IOP of < or =17 mmHg (referent group) to 22.3% (95% CI, 15.8%-31.1%) for those with b-IOP > 25 mmHg, resulting in an adjusted RR of 13.1 (95% CI, 7.1-24.1) among the latter group. The attributable risk for IOP of >25 mmHg was 19%. Using 21 mmHg as a cutoff, the RR was 7.9 (95% CI, 3.8-16.2) and the attributable risk was 37%.

Conclusions: After 9 years' follow-up, the risk of OAG was positively related to IOP levels at baseline. Although persons with b-IOP of >25 mmHg had a 13-fold RR of developing OAG, most cases arose with lower b-IOP. This study thus confirms the role of IOP as an influential risk factor, yet at the same time underscores its limitations in predicting OAG.