The effect of blood pressure variability on the estimation of the systolic and diastolic pressures

Systolic (SBP) and diastolic (DBP) blood pressure are two parameters that are widely used in clinical and personal health monitoring and that are subject to significant physiological variability. The estimation of SBP and DBP is therefore subject to two sources of uncertainty: 1) measurement error, and 2) physiological variability. According to the ANSI/AAMI SP10 standard, the measurement error of an automated sphygmomanometer is limited to ±5 mmHg relative to simultaneous reference readings by at least two trained observers employing a calibrated manometer. In order to directly compare the contributions of physiological variability and measurement error to SBP and DBP estimation uncertainty, we analyzed continuous short recordings of arterial pulses (<; 5 min) and calculated the percentage of beats in which the SBP and DBP exceeds ±5 mmHg relative to the mean value in the analyzed interval. In a group of ICU patients, this percentage was 33.0% for SBP and 14.9% for DBP on average, while in a group of healthy individuals this percentage was 27.5% for SBP and 17.8% for DBP on average. True outliers occurred in 5.17% of the SBP and 12.2% of the DBP values in ICU patients, and 3.27% of the SBP and 3.63% of the DBP values in healthy individuals. These results indicate that blood pressure variability is an important contributor to SBP and DBP estimation uncertainty, and devices that intelligently determine the best time to initiate a recording and the frequency of recordings may alleviate this uncertainty.