Pulse pressure is the best predictor of future left ventricular mass and change in left ventricular mass: 10 years of follow-up

Background Ambulatory blood pressure correlates more closely with left vent ricular mass (LVM) than casual blood pressure in cross-sectional studies, b ut prospective evidence is very limited. Objective To evaluate the best predictors of LVM and change in LVM during 1 0 years of follow-up, in a prospective study. Methods At baseline, blood pressure was recorded by casual measurements and 24 h intra-arterial ambulatory monitoring. The study participants were 97 healthy, untreated, 35-45-year-old men (34 normotensive, 29 borderline hype rtensive, and 34 mildly hypertensive). At 10-year follow-up, echocardiograp hy was performed in 86 (89%) of the men; echocardiographic data were availa ble both at baseline and at follow-up from 70 (72%) of them. Individuals wh o were not receiving antihypertensive medication (n = 66) were included in the prediction of LVM index (LVMI), which was analysed as a continuous vari able. Results The blood pressure variables that were best in predicting the LVMI were: 24 h pulse pressure (r = 0.308, P = 0.012), night-time pulse pressure (r = 0.291, P = 0.018), daytime pulse pressure (r = 0.253, P = 0.041), and casual systolic blood pressure (r = 0.212, P = 0.088). The LVMI was best p redicted by a model including 24 h pulse pressure, positive family history of hypertension, body mass index, and age (adjusted coefficients of determi nation (adj.R-2) = 0.197; that for the casual blood pressure model was adj. R-2 = 0.140). During the follow-up, LVMI increased by +7.5 g/m(2) and +23 g /m(2) in individuals receiving and not receiving antihypertensive medicatio n, respectively (P = 0.015). The change in LVMI was best predicted by the c hange in casual pulse pressure and use of anti hypertensive medication (adj .R-2 = 0.102). Conclusions Ambulatory blood pressure improved the prediction of future LVM I compared with that obtained from casual measurements. To our knowledge, t his is the longest prospective follow-up to show that pulse pressure is the most significant blood pressure parameter in predicting future LVMI and ch ange in LVMI. (C) 2001 Lippincott Williams & Wilkins.