Blood pressure-lowering drugs have different blood pressure-lowering profiles. We studied if differences in blood pressure mean and variability can explain the differences in risks of cardiovascular events and death among 15 245 high-risk hypertensive patients randomized to valsartan or amlodipine and followed for 4.2 years in the VALUE trial (Valsartan Antihypertensive Long-Term Use Evaluation). We selected patients with ≥3 visits and performed Cox regression analyses, defining mean blood pressure as a time-dependent covariate and visit-to-visit and within-visit blood pressure variability as the SD. Of 14 996 eligible patients, participants in the valsartan group had higher systolic mean blood pressure by 2.2 mm Hg, higher visit-to-visit systolic variability by 1.4 mm Hg, and higher within-visit systolic variability by 0.2 mm Hg (P values <0.0001). The higher risks of myocardial infarction and stroke in the valsartan group was attenuated after adjustment for mean and variability of systolic blood pressure, from HR 1.19 (95% CI, 1.02-1.39) to 1.11 (0.96-1.30) and from HR 1.13 (0.96-1.33) to 1.00 (0.85-1.18), respectively. The lower risk of congestive heart failure in the valsartan group was accentuated after adjustment, from HR 0.86 (0.74-1.00) to 0.76 (0.65-0.89). A smaller effect was seen on risk of death, from 1.01 (0.92-1.12) to 0.94 (0.85-1.04). In conclusion, the higher risks of myocardial infarction and stroke in patients randomized to valsartan versus amlodipine were related to the drugs' different blood pressure modulating profiles. The risk of congestive heart failure with valsartan was lower, independent of the less favorable blood pressure modulating profile.
Keywords: blood pressure variability; blood pressure- lowering drugs; heart failure; hypertension; myocardial infarction; stroke.