Blood pressure variability assessed by office, home, and ambulatory measurements before and during antihypertensive drug treatment: a sub-analysis of the REVERENT randomized trial.

Abstract

Blood pressure (BP) variability (BPV) can be assessed using office (OBP), home (HBP), or ambulatory BP (ABP) measurements. This analysis investigated the association and agreement between OBP, HBP, and ABP measurements for BPV assessment at baseline and 10 weeks after initiating antihypertensive drug therapy. Untreated hypertensive patients with elevated BPV were randomized to receive an angiotensin-converting enzyme inhibitor (ramipril) or a calcium channel blocker (nifedipine GITS) in a 10-week, open-label, blinded-end point study. BPV was assessed using standard deviation (SD) and coefficient of variation (CV) (reading-to-reading analyses). Data from 146 participants from three research centers (Athens/Greece; Milan/Italy; Shanghai/China) were analyzed [mean age 53 ± 10 (SD) years, male individuals 60%, baseline systolic OBP, HBP, and 24 h ABP 144 ± 9, 138 ± 10, and 143 ± 10 mmHg, respectively]. Post-treatment minus pre-treatment systolic CV difference was: OBP: 0.3%, P = 0.28; HBP: -0.2%, P = 0.20; 24 h ABP: 1.1%, P < 0.001. Home and ambulatory (not office) BPV indices presented weak-to-moderate correlation, both before and during treatment (range of coefficients 0.04-0.33). The correlation coefficient between systolic HBP and awake ABP CV was 0.21 and 0.28 before and during treatment, respectively ( P < 0.05/< 0.001, respectively). Home and ambulatory (not office) BPV indices presented slight-to-fair agreement (range 64-73%) in detecting participants with high systolic BPV (top quartile of respective distributions) both before and during treatment (kappa range 0.04-0.27). These data showed a weak-to-moderate association between out-of-office (but not office) BPV indices both before and during BP-lowering treatment, with reasonable agreement in detecting individuals with high BPV. Out-of-office BP measurements provide more similar and consistent BPV information than office measurements.