Blood pressure variability and organ damage in a general population: results from the PAMELA study (Pressioni Arteriose Monitorate E Loro Associazioni).

In hypertensive patients, 24-hour blood pressure (BP) variability (V) shows a positive relationship with organ damage, organ damage progression, and cardiovascular morbidity. The clinical relevance of BPV in the population has never been investigated. In a sample of 3200 individuals, randomly selected from the general population of Monza (Milan, Italy), we evaluated BP by an automatic oscillometric device every 20 minutes for 24 hours and left ventricular mass index (LVMI) by echocardiography. In each subject, individual systolic and diastolic BP readings were averaged to obtain a 24-hour mean. Systolic BPV was obtained by calculating (1) the standard deviation of the 24-hour mean, which was taken as the overall BPV, (2) the cyclic components (Fourier spectral analysis) that in the population as a whole explained >95% of the overall BPV, and (3) the fraction of the overall BPV that in each subject was not accounted for by the 2 cyclic components, termed individual residual BPV. A similar procedure was used for diastolic BP and heart rate. Participation rate was 64.1%. Patients receiving antihypertensive therapy (n=403) were excluded from the analysis, which was therefore limited to 1648 participants. In the population as a whole, LVMI significantly related to 24-hour systolic and diastolic BP mean (beta=0.40 and beta=0.37, respectively, P<0.001 for both) but not to the 2 cyclic components that accounted for most of the BPV. On the other hand, the individual residual BPV (which accounts on average for about 50% of overall BPV) showed a significant positive relationship with LVMI (beta =0.38 and beta=0.88 for systolic and diastolic BP, respectively, P<0.05 and P<0.01). No relationship was found between LVMI and heart rate values. These findings provide evidence that there is a relationship between LVMI and 24-hour average BP values in the population. They also provide the first demonstration that in the population there is also a positive independent association between LVMI and BPV. This association, however, can be exclusively seen with the BPV component that has an erratic rather than a cyclic nature.