Abstract
Purpose :
In an observational study, hormone therapy (HT) containing estrogen, but not a combination of estrogen and progesterone, decreased the risk of primary open-angle glaucoma. As there has been no randomized trial designed to assess the effect of HT on OAG, we conducted a secondary analysis of a large, randomized, placebo-controlled trial to test if HT decreased the risk of open-angle glaucoma (OAG), and if the risk reduction varied by race.
Methods :
We analyzed a Medicare-linked database from 25,535 women in the Women’s Health Initiative Hormone Trial (1993 through 2014). The women without a uterus were randomized to receive either oral conjugated equine estrogens (CEE 0.625mg/day) or placebo; women with a uterus were randomized to receive oral CEE and medroxyprogesterone acetate (CEE 0.625mg/day+MPA 2.5mg/day) or placebo. We excluded women 1) who enrolled in Medicare after the trial concluded, 2) lacked eye examinations during a 4-year look-back period of continuous Medicare enrollment, or 3) had prevalent OAG (≥1 OAG-related codes during the look-back period). We used cox proportional hazards models to calculate hazard ratios (HR), considering possible confounders (age, race, age at menopause, diabetes, hypertension, alcohol intake, smoking, and body mass index). Effect modification by race was further tested.
Results :
Final analysis included 8,102 women (mean age=68.5±4.8years). The incidence of OAG was 10.7% (mean follow-up=13.2±5.3years; mean HT duration=6.0±1.8years). Women included in the analysis were older, more likely to be white or hypertensive, and less likely to be smokers, compared to those excluded. Increased age (p-trend=0.005) and African-American race (HR 2.46, 95%CI=2.00 to 3.02; white as a reference) were significant risk factors for incident OAG. We found no overall benefit of HT in reducing incident OAG (HR 1.00, 95%CI=0.81 to 1.23 in CEE trial, and HR 0.95, 95%CI=0.80 to 1.13 in CEE+MPA trial). However, a borderline risk reduction was demonstrated among African-American women treated with CEE (HR 0.61, 95% CI=0.37 to 1.01, p-interaction=0.06), but not CEE+MPA (HR 0.75, 95%CI=0.40 to 1.43, p-interaction=0.52), compared to placebo.
Conclusions :
Our analysis suggests that HT containing estrogen, but not a combination of estrogen and progesterone, may reduce the risk of incident OAG among African-American women. Further investigation is needed.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.