Differences in blood pressure (BP) reading between multiple measurements, known as BP variability, is a strong predictor of cardiovascular and all-cause mortality. Lack of consensus on quantifying and reporting BP variability has limited its use in clinical practice. We aimed to determine the best possible way to quantify BP variability that would improve predictability of long-term mortality among elderly hypertensive population.Design and Method:
We used data from a subset of patients (n = 694) aged > 65 years from the Second Australian National Blood Pressure (ANBP2) study. To calculate Systolic BP (SBP) and Diastolic BP (DBP) variability, 24-hours ambulatory BP monitoring data at study entry and visit-to-visit clinic BP data over a median 4.1 years follow-up were used. BP variability was defined as within-individual standard deviations in BP across measurements for 24 hours, day-time, night-time and visit-to-visit. Cox-proportional hazard models were used to explore the relationship between outcome and BP variability.Results:
Over a median of 10.8 years, 161 patients died from any cause, of which 81 deaths were due to cardiovascular causes. We observed an increased risk of all cause and cardiovascular mortality with per mm increase in BP variability on univariate analyses irrespective of 24hrs, day-time, night-time and visit-to-visit BP variability, except DBP visit-to-visit BP variability. However, only day-time SBP and DBP variability from 24-hours ABPM were a strong predictor of long-term all-cause and cardiovascular mortality, when adjusted for age, gender and other possible risk factors (Table 1).Conclusions:
Day-time BP variability, both SBP and DBP, measured by 24-hours ambulatory BP monitoring is a better predictor of mortality over the long-term in elderly hypertensive patients than visit-to-visit clinic BP variability.