Visit-to-Visit Blood Pressure Variability and Mortality and Cardiovascular Outcomes Among Older Adults: The Health, Aging, and Body Composition Study

Abstract

Background: Level of blood pressure (BP) is strongly associated with cardiovascular (CV) events and mortality. However, it is questionable whether mean BP can fully capture BP-related vascular risk. Increasing attention has been given to the value of visit-to-visit BP variability.

Methods: We examined the association of visit-to-visit BP variability with mortality, incident myocardial infarction (MI), and incident stroke among 1,877 well-functioning elders in the Health, Aging, and Body Composition Study. We defined visit-to-visit diastolic BP (DBP) and systolic BP (SBP) variability as the root-mean-square error of person-specific linear regression of BP as a function of time. Alternatively, we counted the number of considerable BP increases and decreases (separately; 10mm Hg for DBP and 20mm Hg for SBP) between consecutive visits for each individual.

Results: Over an average follow-up of 8.5 years, 623 deaths (207 from CV disease), 153 MIs, and 156 strokes occurred. The median visit-to-visit DBP and SBP variability was 4.96 mmHg and 8.53 mmHg, respectively. After multivariable adjustment, visit-to-visit DBP variability was related to higher all-cause (hazard ratio (HR) = 1.18 per 1 SD, 95% confidence interval (CI) = 1.01-1.37) and CV mortality (HR = 1.35, 95% CI = 1.05-1.73). Additionally, individuals having more considerable decreases of DBP (≥10mm Hg between 2 consecutive visits) had higher risk of all-cause (HR = 1.13, 95% CI = 0.99-1.28) and CV mortality (HR = 1.30, 95% CI = 1.05-1.61); considerable increases of SBP (≥20mm Hg) were associated with higher risk of all-cause (HR = 1.18, 95% CI = 1.03-1.36) and CV mortality (HR = 1.37, 95% CI = 1.08-1.74).

Conclusions: Visit-to-visit DBP variability and considerable changes in DBP and SBP were risk factors for mortality in the elderly.

Keywords: aged.; blood pressure; blood pressure variability; hypertension; mortality; myocardial infarction; stroke.

Figure 1.

Illustration of long-term visit-to-visit DBP variability: the SD of the residuals taken from person-specific linear regression of 5 measures of DBP (residual-mean-square error). Abbreviation: DBP, diastolic blood pressure.

Figure 2.

Association of considerable DBP and SBP changes (separately) with incident MI, incident stroke, all-cause mortality, and CV mortality. Estimates represent the hazard ratio per considerable change in blood pressure; a considerable change was an increase or decrease of at least 10mm Hg diastolic or 20mm Hg systolic between visits. Estimates adjusted for demographic measures (age, gender, race, education), clinical measures (fasting serum glucose, fasting plasma high-density lipoprotein cholesterol, fasting plasma low-density lipoprotein cholesterol, fasting plasma triglycerides), antihypertensive medication use (never, consistent, inconsistent), and health behaviors (current smoking status, former smoking status, body mass index), blood pressure mean, and blood pressure slope. Abbreviations: DBP, diastolic blood pressure; CV, cardiovascular; MI, myocardial infarction; SBP, systolic blood pressure.