ASSOCIATION OF CARDIOVASCULAR RISK AMONG DIFFERENT COMMON OFFICE BLOOD PRESSURE MEASUREMENT PROTOCOLS IN COMMUNITY-BASED CHINESE POPULATION

Abstract

Objective: The office blood pressure (BP) measurement protocols recommended by the guidelines is inconsistent. We aim to investigate the relationship between different office BP measurement protocols and cardiovascular risk in Chinese community population. Design and method: A total of 6098 participants without cardiovascular disease history were included in our study from the community atherosclerosis cohort conducted in Beijing, China, from December 2011 to April 2012. We conducted three measurements using the standard calibration method of Omron HEM-7117 electronic blood pressure monitor, with intervals between consecutive readings exceeding 1 minute. Four common office BP measurement protocols included: the average of the first two BP values; the average of the last two times; triple average value; and if BP fluctuation eceeds 10mmHg, take the average of the last two times; otherwise, take the average of the first two times. The primary endpoint is defined as the major adverse cardiovascular events (MACE) before December 31th, 2021, including acute myocardial infarction, stroke, and cardiovascular death. Cox proportional hazard regression models was used to examine the relationship between BP measurement protocols and MACE. Likelihood ratios and receiver operating characteristic curve (ROC) were also compared, and concordance of models assessed with Harrell C statistic. Results: Of all subjects, the mean age was 54.99 (8.41) years old, and 2126 (34.9%) were males. After a mean 9.86-year follow-up, a total of 486 (6.99%) participants experienced MACE. After adjusting for covariates, no difference of events predicted by different office BP measurement protocols was found in the COX analysis. The ROC curves for the different BP measurement protocols also did not show any significant differences (Figure1). There were no differences in the C-index among the different BP models (all P>0.05). These findings were consistent across subgroups stratified by age, gender, and hypertension grades (all P>0.05). Conclusions: The performance of all prediction models using four different BP measurement protocols is the same. All four blood pressure measurement protocols can accurately predict prognosis.