Risks of untreated and treated isolated systolic hypertension in the elderly: meta-analysis of outcome trials

Background Previous meta-analysis of outcome trials in hypertension have no t specifically focused on isolated systolic hypertension or they have expla ined treatment benefit mainly in function of the achieved diastolic blood p ressure reduction. We therefore undertook a quantitative overview of the tr ials to further evaluate the risks associated with systolic blood pressure in treated and untreated older patients with isolated systolic hypertension . Methods Patients were 60 years old or more. Systolic blood pressure was 160 mm Hg or greater and diastolic blood pressure was less than 95 mm Hg. We u sed non-parametric methods and Cox regression to model the risks associated with blood pressure and to correct for regression dilution bias. We calcul ated pooled effects of treatment from stratified 2 x 2 contingency tables a fter application of Zelen's test of heterogeneity. Findings In eight trials 15 693 patients with isolated systolic hypertensio n were followed up for 3.8 years (median). After correction for regression dilution bias, sex, age, and diastolic blood pressure, the relative hazard rates associated with a 10 mm Hg higher initial systolic blood pressure wer e 1.26 (p=0.0001) for total mortality, 1.22 (p=0.02) for stroke, but only 1 .07 (p=0.37) for coronary events. Independent of systolic blood pressure, d iastolic blood pressure was inversely correlated with total mortality, high lighting the role of pulse pressure as risk factor. Active treatment reduced total mortality by 13% (95% CI 2-22, p=0.02), card iovascular mortality by 18%, all cardiovascular complications by 26%, strok e by 30%, and coronary events by 23%. The number of patients to treat for 5 years to prevent one major cardiovascular event was lower in men (18 vs 38 ), at or above age 70 (19 vs 39), and in patients with previous cardiovascu lar complications (16 vs 37). Interpretation Drug treatment is justified in older patients with isolated systolic hypertension whose systolic blood pressure is 160 mm Hg or higher. Absolute benefit is larger in men, in patients aged 70 or more and in thos e with previous cardiovascular complications or wider pulse pressure. Treat ment prevented stroke more effectively than coronary events. However, the a bsence of a relation between coronary events and systolic blood pressure in untreated, patients suggests that the coronary protection may have been un derestimated.