A prospective study of blood pressure and risk of cataract in men

PURPOSE: Cataract is the leading cause of blindness worldwide. Blood pressu re has been identified as a risk factor in some, but not all, previous stud ies. We aimed to test prospectively the hypothesis that high blood pressure increases risk of age-related cataract. METHODS: Participants in the Physicians' Health Study of 22,071 men aged 40 tcr 84 years in 1982 completed annual questionnaires that provided medical history including self-reported blood pressure, treatment for hypertension , and cataract. Over 12 years, 1392 cataracts: were confirmed by medical re cord review among 17,762 physicians with complete data and no reported cata ract at baseline. We used proportional hazards regression models to examine relations of systolic blood pressure (SBP), diastolic blood pressure (DBP) , hypertension, as well as antihypertensive medications: with cataract, aft er control for potential confounding factors. RESULTS: In models adjusting for age and randomized treatment assignment, t here was a significant relationship of SEP (P = 0.01), but not DBP (p = 0.3 0), hypertension (p = 0.15), or antihypertensive medications teach p greate r than or equal to 0.23) with incident cataract. Estimates were attenuated after adjusting for multiple potential confounders, although the relationsh ip of SEP with incident cataract remained significant. The multivariate adj usted rate ratio (95% confidence interval) of cataract for SBP greater than or equal to 150 versus < 120 mmHg was 1.31 (1.04-1.66), P for trend = 0.04 . For DBP <greater than or equal to> 90 versus < 70 mmHg, the estimate was 1.11 (0.84-1.45), P fur trend = 0.33. CONCLUSIONS: Overall, these data suggest that the relationship of blood pre ssure with cataract is not strong, and is subject to confounding by other r isk factors. The modest magnitude of the association with SEP and lack of s ignificant relationships with DBP and hypertension may suggest a non-causal relationship of blood pressure with cataract. Ann Epidemiol 2001;11:104-11 0. (C) 2001 Elsevier Science Inc. All rights reserved.