Continuous Blood Pressure Variability within Young, Healthy Adults: A Test-Retest Study to Assess Relative and Absolute Reliability

Variability in 24-hour and visit-to-visit blood pressure has been reported to be predictive of adverse cardiovascular outcomes. While recent studies have employed beat-to-beat blood pressure variability (BPV) as a predictor of cardiovascular risk, the reliability of beat-to-beat BPV remains equivocal. The purpose of this study was to examine intraindividual reliability of BPV, and to compare that reliability to more widely-documented heart rate variability (HRV) in young healthy adults. Continuous heart rate (HR, electrocardiogram), beat-to-beat blood pressure (BP, finger plethysmography), and respiratory rate (pneumobelt) were recorded during 10 minutes of rest in 77 participants (38 males, 39 females; age: 23±5 yr; BMI: 25±4 kg/m2) on two occasions separated by ≥2 days (25±19 days; 2-98 days). Reliability of beat-to-beat BPV, HRV (time- and frequency-domain), and additional cardiovascular (heart rate, blood pressure) assessments were quantified as relative and absolute reliability using the intraclass correlation coefficient (ICC) and the coefficient of variation (CoV), respectively. Beat-to-beat BPV demonstrated moderate relative and good absolute reliability (10-min duration: ICC=0.601-0.684, CoV=14-16%). In contrast, relative reliability of time-domain HRV was good-to-excellent, while absolute reliability was poor-to-good (10-min duration: ICC=0.844-0.904, CoV=13-34%). Frequency-domain HRV demonstrated moderate to excellent relative reliability and poor absolute reliability (10-min duration: ICC=0.626-0.903, CoV=34-37%). In summary, the rigor and reliability of continuous BPV and HRV are different when parsed out by relative vs. absolute reliability, with relative reliability being stronger with HRV and absolute reliability being stronger with BPV. These findings suggest both caution and specificity when employing continuous BPV or HRV for CVD risk stratification purposes.