DIABETES-MELLITUS IN TREATED HYPERTENSION - INCIDENCE, PREDICTIVE FACTORS AND THE IMPACT OF NONSELECTIVE BETA-BLOCKERS AND THIAZIDE DIURETICS DURING 15 YEARS TREATMENT OF MIDDLE-AGED HYPERTENSIVE MEN IN THE PRIMARY PREVENTION TRIAL GOTEBORG, SWEDEN

The objective of this study was to analyse predictive factors for the development of diabetes mellitus during long-term treatment of hyperte nsion and to compare the diabetogenic potential of thiazide diuretics and nonselective beta-adrenoceptor blockers. The study population comp rised 686 hypertensive men, aged 47-54 years, who were followed for 15 years. Patients were treated with either thiazide diuretics or beta-a drenoceptor blockers as monotherapy or in combination with one another or alternative other antihypertensive drugs. During the first part of the study, i.e. during the 1970s, only non-selective beta-adrenocepto r blockers were used. The average yearly incidence of the development of diabetes mellitus during follow-up was 1.3%. In univariate analysis body mass index, serum triglyceride level, radiographic heart enlarge ment and beta-blocker therapy were significantly associated with the d evelopment of diabetes mellitus. Predictors selected by stepwise Cox r egression were body mass index, radiological heart enlargement and bet a-blocker therapy. Two subgroups with patients treated with nonselecti ve beta-adrenoceptor blockers but not with thiazide diuretics during t he first five years of follow-up (the beta-blocker group; n = 93) or w ith thiazide diuretics but not with nonselective beta-adrenoceptor blo ckers during the first five years of follow-up (the thiazide-group; n = 96) were identified. The relative risk for developing diabetes melli tus was significantly higher in the beta-blocker group being 6.1 after 10 years and 3.5 after 15 years treatment in comparison with the thia zide group. In conclusion, the incidence of diabetes mellitus is high during long-term treatment of hypertension in middle-aged, male patien ts treated with thiazide diuretics and/or beta-adrenoceptor blockers. Factors related to insulin resistance are predictive for the developme nt of diabetes mellitus. Although the present study did not randomise to treatment groups, our results also indicate that nonselective beta- adrenoceptor blockers may be more diabetogenic than thiazide diuretics .