Metabolic Risks of Diuretic Therapy

Abstract

Hypertension, like hypercholesterolemia, is a major risk factor for coronary heart disease (CHD). A characteristic such as hypertension attains risk factor status for CHD if its presence correlates with a high incidence of subsequent myocardial infarction. Although statistical correlation does not prove causation, the association suggests potentially fruitful avenues for research and possible prevention of the condition. Thus, one benefit of the treatment of hypertension was expected to be a reduction of CHD. Two sizable and lengthy placebo-controlled trials failed to demonstrate the expected reduction in CHD (1, 2). A trial of huge proportions was required to document the decrease in CHD and thereby substantiate the causal relationship between the two conditions (3). Does the difficulty in decreasing CHD mean that hypertension plays only a minor contributory role in the pathogenesis of CHD? Where can we now look for additional reductions in CHD? In an ongoing inquiry into these questions, we previously reported that diuretic drugs, when used in the treatment of hypertension, cause an increase in serum lipid concentrations (4). Two short-term studies have confirmed our observations (5, 6), but one large trial reported no change in serum lipids after 3 years of treatment with hydrochlorothiazide (7). To evaluate the durability of the effect of diuretics on serum lipids, we report herein our observations for intervals of treatment up to 18 months in duration, the longest periods for which we have treated sizable numbers of patients continuously with single agents. In addition, as part of a search for mechanisms of the effect, we report measurements of fasting serum insulin concentrations before and during treatment.