Abstract P183: Visit to Visit Variability in Diastolic Blood Pressure Predicts Incident Cardiovascular Disease Events in Middle-aged Adults

Abstract

Introduction: visit-to-visit variability (VVV) in blood pressure (BP) relates to risk for cardiovascular disease (CVD) in clinical studies. It is unknown if VVV in systolic or in diastolic BP predicts incidents CVD in community-based middle-aged adults. Methods: The study sample comprised 1817 Framingham Heart Study (FHS) offspring cohort participants free of clinical CVD at the sixth examination (1995-1998) cycle who had attended examination cycles three, four and five at four year intervals. Exclusion criteria include medication treatment for elevated BP at any of the four examinations. We defined VVV in systolic and diastolic BP as the standard deviation of BP values across the four examinations. Incident CVD events were identified through physician panel review of medical records with follow-up through December 31, 2013. We used proportional hazards regression models to test associations of between VVV in systolic and diastolic BP and CVD risk adjusting for age, sex, body-mass-index, mean systolic and diastolic BP, total cholesterol, HDL cholesterol, medication treatment for hyperlipidemia, fasting blood sugar, diabetes, and cigarette smoking. Results: During follow up (average 16.2 years), 276 (15.2%) participants developed CVD. Mean and standard deviation values of VVV in systolic and diastolic BP were 8.09±4.48 and 5.38±2.5, respectively. Each standard deviation increase in diastolic BP VVV was associated with a hazards ratio (HR) of 1.08 (95% confidence interval [CI] 1.03-1.14) for incident CVD. The HR of systolic BP VVV was 1.03 (95% CI 1.00-1.06). Including both variables gave an HR of 1.01 (95% CI 0.98-1.04) for VVV in systolic BP and 1.08 (95% CI 1.03-1.13) for VVV in diastolic BP. There were significant linear trends in CVD risk according to quartiles of VVV in systolic and diastolic BP. Participants in the highest quartile of VVV in systolic BP had a HR of 1.68 (95% CI 1.13-2.50) for incident CVD; those with highest quartile of VVV in diastolic BP had a HR of 1.66 (95% CI 1.17--2.35). Conclusion: Visit-to-visit variability in BP predicts CVD risk in middle-aged adults.