3066 — The Comparative Effectiveness of Chlorthalidone versus Hydrochlorothiazide
Lund BC, Center for Comprehensive Access & Delivery Research and Evaluation; Ernst ME, University of Iowa;
The preponderance of clinical trials supporting low-dose thiazide diuretics have used chlorthalidone, while hydrochlorothiazide remains more widely used. The drugs have not been compared in a head-to-head clinical trial, but indirect evidence suggests chlorthalidone may be superior to hydrochlorothiazide in lowering blood pressure and reducing cardiovascular events. This study compared the effectiveness of these chlorthalidone and hydrochlorothiazide in a real-world cohort.
Administrative VA data were used to compare clinical outcomes in patients initiating chlorthalidone (N = 2,257) or hydrochlorothiazide (N = 124,551), using a retrospective cohort design. Ineffectiveness was assessed during the year following thiazide initiation, and defined as (1) thiazide discontinuation, or (2) addition of a new antihypertensive medication.
At one year following initiation, discontinuation was significantly higher among patients initiating chlorthalidone (38.0%) compared to hydrochlorothiazide (27.6%) (p <0.001). In contrast, the addition of a new antihypertensive among persistent thiazide users, was significantly more common among patients initiating hydrochlorothiazide (29.9%) compared to chlorthalidone (23.6%) (p <0.001). Overall effectiveness, defined as persistent use without additional antihypertensive medication, was 47.4% for chlorthalidone and 50.7% for hydrochlorothiazide (p = 0.002). These relationships remained consistent after adjusting for demographics, concurrent antihypertensive medication, history of antihypertensive failure, and dose.
Our findings suggest that hydrochlorothiazide and chlorthalidone may not be therapeutically equivalent. Individuals who tolerated chlorthalidone were less likely to require additional antihypertensive treatment, perhaps reflecting greater efficacy for short-term blood pressure control. However, this advantage may be offset by a more substantial adverse event burden, as implied by a greater rate of discontinuation. In the absence of a randomized design, it is unclear whether our findings can be directly attributed to medication effects, or could be explained by unknown confounding factors, such as selection bias in the decision to prescribe chlorthalidone instead of hydrochlorothiazide.
More than one million Veterans currently take thiazide diuretics, of which more than 95% receive hydrochlorothiazide. There is sufficient evidence supporting the potential superiority of chlorthalidone to warrant a randomized head-to-head trial, which would have major implications for VA. The study design must employ appropriate thiazide dosing and monitoring strategies to ensure optimal tolerability.