Long-term metabolic effects of spironolactone and thiazides combined with potassium-sparing agents for treatment of essential hypertension.

By using information prospectively collected in the computerized ARTEMIS databank, the long-term metabolic consequences of spironolactone, hydrochlorothiazide-amiloride combination and cyclothiazide-triamterene combination were evaluated in 100 patients for each group matched according to sex, age and blood pressure (BP). Spironolactone was prescribed at a mean dose of 98 mg, hydrochlorothiazide at 36 mg and cyclothiazide at 2 mg, during a mean follow-up of 20 months. Compared with the pretherapeutic values, BP decreased equally in both treatment groups (18/9 mm Hg on average). Creatinine increased significantly in the 3 groups (9, 8, 14 mumol/liter, p less than 0.001) as did uric acid (18, 31, 42 mumol, p less than 0.001). Plasma potassium increased with spironolactone (0.7 mmol/liter, p less than 0.001) and remained stable with the combinations of the thiazide-potassium-sparing agents. For the 3 groups, the slight and nonsignificant variations of fasting blood glucose and cholesterol were mainly the results of a phenomenon of regression to the mean. However, when both groups of thiazide-treated patients were considered, the reduction of plasma potassium was accompanied by a slight increase in glucose (0.1 mmol/liter) and cholesterol levels (0.2 mmol/liter) compared with when kalemia decreased (-0.1 and -0.1 mmol/liter, respectively, p less than 0.05 and p less than 0.05). It is concluded that in a clinical daily practice of a hypertension clinic low doses of spironolactone or of thiazides combined with potassium-sparing agents reduced BP without alteration in lipid or carbohydrate metabolism on long-term follow-up.