Background: The clinical prognostic value of visit-to-visit blood pressure (BP) variability (BPV) is debatable, and relative studies among patients receiving BP control to achieve lower BP targets are limited.
Methods: We analyzed a dataset from the STEP trail (Strategy of Blood Pressure Intervention in the Elderly Hypertensive Patients) to investigate the relationship between visit-to-visit BPV and cardiovascular events in patients with hypertensive aged 60 to 80 years. Visit-to-visit BPV was defined as the coefficient of variation, SD, delta, average real variability, and variability independent of the mean of BP measured at 6-, 9-, 12-, 15-, and 18-month follow-up visits. We computed hazard ratios for the risks associated with a 1-SD increase in BPV indexes in multivariable cox regression models.
Results: Among 7678 patients from the STEP trial, after adjustment for multiple confounders, diastolic BPV indexes were significantly associated with the primary composite end point (hazard ratios ≥1.21; P≤0.029) in the standard group, while there was no association between the clinical outcomes and systolic BPV (P≥0.091). In the intensive treatment group, either systolic or diastolic BPV was no association with clinical outcomes(P≥0.30). Sensitivity analyses using an alternative method to calculate BPV based on 7 BP records generated confirmatory results.
Conclusions: In older adults with hypertension, visit-to-visit diastolic BPV is an independent predictor of adverse health outcomes in the standard treatment group. However, BPV did not have prognostic value in the intensive treatment group.
Registration: URL: https://www.
Clinicaltrials: gov; Unique identifier: NCT03015311.
Keywords: blood pressure variability; cardiovascular outcomes; hypertension.