Visit-to-visit blood pressure variability (VVV BPV) is a recognised risk factor for cardiovascular disease (CVD) that is underutilised in clinical practice. The reliability of electronic health record (EHR) data in estimating BPV and predicting CVD remains uncertain. This study compared BPV estimation methodologies using EHR versus non-EHR data and examined dose-response associations with CVD. A systematic review and meta analysis was conducted across five databases (MEDLINE, Scopus, EMBASE, CINAHL, and Web of Science) for studies published from January 2012 to August 2024. Studies assessing VVV BPV in adults and its association with CVD outcomes (e.g. myocardial infarction, stroke, heart failure, and cardiovascular mortality) were included. A dose-response meta-analysis (DRMA) evaluated BPV thresholds linked to increased CVD risk using standard deviation (SD) and coefficient of variation (CV). A total of 4,926 studies were screened, 49 of which met the inclusion criteria. No consensus has emerged on BPV estimation methodologies, although non-EHR studies have followed stricter protocols. The meta-analysis showed that VVV BPV predicted any CVD outcome. Effect sizes were comparable between EHR (HR: 1.17, 95% CI: 1.09-1.24) and non-EHR (HR: 1.14, 95% CI: 1.10-1.17) studies (P-value = 0.468). A BPV threshold of SD 6.72 mmHg or CV 9.05% was linked to a 10% higher CVD risk. The EHR data reliably estimate BPV, yielding effect sizes similar to those of non-EHR sources. A non-linear dose-response relationship suggests that a higher BPV increases CVD risk. VVV BPV needs to be incorporated into clinical practice, and further research is required to identify strategies to implement and scale up into routine workflow.
Lukitasari et al. (Thu,) studied this question.