Longitudinal Changes in Mean and Pulse Pressure, and All-Cause Mortality: Data From 71,629 Untreated Normotensive Individuals.

Abstract

Blood pressure (BP) includes a steady (mean arterial pressure, MAP) and a pulsatile component that independently predict mortality. The association between longitudinal changes in central (c) pulse pressure (PP), brachial (b) PP, MAP, and incident mortality has never been investigated in this context. Brachial MAP and PP were measured at 2 routine checkups (1st visit 1992; mean interval, 5.8 ± 2.4 years) in 71,629 individuals, age 16-95 years, none on antihypertensive drugs. cPP was estimated with a validated algorithm. Each change (visit 2-1) in bPP, cPP, and MAP, expressed in mm Hg/year, was categorized as: increase, decrease, or no-change, with the latter representing the control-group. Follow-up data (6.9 ± 3.3 years) on all-cause mortality (2,033 deaths) were documented. All-cause mortality Cox regression models adjusted for confounders showed that compared to the subgroup with steady BP at both visits, the subgroup with: (i) increased bPP or cPP had an approximately 200% increase in relative risk (RR); (ii) decreased cPP and bPP had a 15% reduction in RR; (iii) increased MAP had a 68% increase in RR; (iv) decreased MAP had a 7% increase in RR of mortality. Interaction analysis stratified by gender showed that annual increases in PP, but not MAP, were greater in younger than older men and lower in younger than older women. Age cutoff value was 55 years. MAP and PP have distinct characteristics affecting all-cause mortality. PP integrates the effects of age and gender on all-cause mortality more notably than MAP, thus impacting significantly on cardiovascular risk.