A putative placebo comparison of the SCOPE and LIFE trials

The Losartan Intervention For Endpoint reduction in hypertension (LIFE) trial and the Study on Cognition and Prognosis in the Elderly (SCOPE) superficially produced comparable outcomes, with effects on stroke greater than those anticipated from blood pressure (BP) lowering alone. This, however, ignores important features of both studies. It ignores firstly the disparate comparator agents - atenolol in LIFE and predominantly hydrochlorthiazide in SCOPE, secondly the small, but potentially important BP differential between the treatment arms in SCOPE and finally the small, statistically non-significant increase in coronary heart disease (CHD) in both trials. This analysis compares the major cardiovascular outcomes in these trials with reference to placebo. Two alternative reference populations were employed to calculate the imputed placebo, firstly the MRC Trial in Elderly Hypertensives and secondly a meta-analysis of trials in the elderly, which included comparisons between diuretic- and b-blocker-based regimens. Overall, the choice of 'comparator placebo' did not substantially influence the derived results. Accounting for BP differences and based on the meta-analysis, both trials demonstrated statistically significant reductions in fatal/non-fatal stroke compared with placebo - relative risks (95% confidence intervals [CI]) of 0.53 (0.39, 0.73) and 0.56 (0.41, 0.76) for SCOPE and LIFE, respectively. For fatal/non-fatal MI, there were greater discrepancies between the studies, but with neither achieving statistical significance compared with placebo - relative risks of 0.85 (0.59, 1.24) and 1.08 (0.80, 1.46) for SCOPE and LIFE, respectively. This analysis clearly demonstrates that both candesartan in SCOPE and losartan in LIFE are associated with reductions in stroke events compared with placebo, greater than that observed in the well-established meta-analysis of placebo-controlled hypertensive trials. However, the CIs are such that it is not possible to suggest definitively that this is a benefit beyond BP reduction alone. Neither trial is sufficiently 'powered' to demonstrate a benefit in CHD outcomes, but with SCOPE there was a trend towards benefit with a point estimate compatible with the major meta-analysis.