Twenty-four-hour mean blood pressure and variability combined improve the prognostic value of ambulatory blood pressure

OBJECTIVE: To determine whether adding 24-h blood pressure (BP) variability to 24-h mean BP improves the BP predictive value for cardiovascular and all-cause fatal outcomes. METHODS: In 1993, among individuals of the Pressioni Arteriose Monitorate E Loro Associazioni (PAMELA) population, 24-h mean BP was calculated from the mean of three hourly measurements. BP variability was quantified as the weighted standard deviation of the 24-h mean BP, the 24-h average real variability and the 24-h erratic BP variations by Fourier analysis. Variability values were used after removal of the fraction explained by their linear relationship with 24-h mean BP. Data were adjusted for baseline covariates but because of within-subject comparisons unadjusted data were also used. RESULTS: During a 24.6 years follow-up, 250 cardiovascular and 798 all-cause deaths occurred. Adding any of the BP variability measures to 24-h mean BP significantly improved the goodness of fit of the data to the Cox prediction model used to estimate cardiovascular risk and all-cause mortality. This occurred also when BP variability was added to the net reclassification improvement (NRI) index. In subgroup analysis, a significant increase in the risk of cardiovascular and all-cause mortality was consistent in subjects with both 24-h mean and variability at least at their median value. CONCLUSION: Addition of 24-h BP variability improves the prognostic estimate of the risk of cardiovascular and all-cause mortality associated with 24-h mean BP. This supports an upgrading of information on BP variability in ambulatory BP monitoring reports to better exploit the potential of this approach as risk estimate.