Demographic, environmental, and genetic predictors of metabolic side effects of hydrochlorothiazide treatment in hypertensive subjects.

BACKGROUND Thiazide diuretics are recommended for first-line treatment of hypertension. Although considered safe and effective, their use is associated with dyslipidemia, hyperglycemia, and an increased risk of developing type 2 diabetes. The aim of this study was to characterize interindividual variation in glucose and lipid responses to hydrochlorothiazide and to identify demographic, environmental, and genetic predictors associated with this variation. METHODS A community-based sample of 585 adults with essential hypertension (291 African Americans [150 women and 141 men] and 294 non-Hispanic whites [126 women and 168 men]) underwent monotherapy with hydrochlorothiazide for 4 weeks. Linear regression was used to construct prediction models for the changes in plasma total cholesterol, triglyceride, and glucose concentrations. RESULTS The mean changes (+/- standard deviation) in response to hydrochlorothiazide were 6.13 +/- 22.8 mg/dL for total cholesterol, 17.21 +/- 70 mg/dL for triglycerides, and 3.5 +/- 9.5 mg/dL for plasma glucose. Ethnicity and baseline levels of the analytes were the only predictors that were significant for more than one response. Therefore, the mechanism for these metabolic effects are not entirely shared. CONCLUSIONS Taken together, demographic, environmental, and genetic factors accounted for only 13% of the total variation in total cholesterol response, 17% of the variation in triglyceride response, and 11% of the variation in glucose response to hydrochlorothiazide, and less than half of this predicted variation in response was explained by measured genotypes.