Research Article

Ventricular extrasystoles during thiazide treatment: substudy of MRC mild hypertension trial.

Br Med J (Clin Res Ed) 1983; 287 doi: (Published 29 October 1983) Cite this as: Br Med J (Clin Res Ed) 1983;287:1249


One short term and one long term study of the relation between ventricular extrasystoles and thiazide treatment were carried out during the Medical Research Council's mild hypertension trial. In the short term study 110 patients were randomly assigned to one of three treatment groups, bendrofluazide with or without potassium supplements, or placebo. They were studied before starting treatment and nine to 10 weeks later while still taking their randomly assigned drugs. No significant increase in the number of ventricular extrasystoles was associated with short term thiazide treatment, although serum potassium concentrations changed as expected. In the long term study 214 patients who had completed an average of two years' treatment with randomly assigned bendrofluazide or a placebo were studied while continuing to take their trial tablets; the 214 included 20 people who had been randomised at entry to the bendrofluazide group and who had a subsequent history of hypokalaemia. These 20 patients were studied before and after being further randomised to two groups, one continuing treatment without change and one continuing with bendrofluazide and also taking potassium supplements. Counts of ventricular extrasystoles were significantly higher (p = 0.025) in those receiving long term thiazide treatment than in their controls; however, there was no significant association between the number of ventricular extrasystoles and serum potassium concentrations in this group, although the correlation between number of extrasystoles and serum urate concentrations was significant (p = 0.035). Pooled data for both studies showed a highly significant correlation between number of ventricular extrasystoles, and serum potassium concentrations (r = -0.185; p = 0.003), but the correlation with serum urate concentrations was of similar strength (r = 0.178; p = 0.004). These biochemical changes may be acting merely as markers of thiazide intake, and the explanation of the association between thiazide treatment and ventricular extrasystolic activity therefore remains uncertain.