Abstract P175: Predictors of Long Term Blood Pressure Variability in Young Adulthood Through Middle Age: The Coronary Artery Risk Development in Young Adults Study

Abstract

Introduction: Long-term blood pressure variability (BPV) refers to fluctuation in BP that occur over weeks, months and years. BPV has been identified as a risk factor for the development of subclinical and clinical cardiovascular events independent of mean BP. However, little is known about the factors associated with long-term BPV. We hypothesized that long-term BPV will vary by demographic, behavioral, anthropometric, lab, and clinical factors. Methods: Using data from the CARDIA study – a longitudinal population based cohort study, we investigated whether baseline demographics (age, sex, race, education); behavioral factors (smoking, alcohol intake, physical activity); anthropometric measures (height, weight, body mass index - BMI); lab markers (total cholesterol, high density lipoprotein cholesterol - HDL-C, low density lipoprotein cholesterol – LDL-C, triglycerides, fasting blood glucose, glomerular filtration rate - GFR); and history of asthma are associated with different indictors of long-term BPV. Variability independent of the mean (VIM) and coefficient of variation (CV) of BP were calculated to quantify within-individual long-term BPV from baseline to visit 9 across 30 years. A least absolute shrinkage and selection operator (lasso) linear regression was used to identify variables that may be associated with long-term BPV and multivariate linear regression models were used to assess magnitude of association. Results: Participants were 3,095 individuals who were not taking antihypertensive medication (mean age 24.6 years, 45.5% male and 56.9% white). Mean VIM was 8.5 mmHg (SD=3.7) for systolic and 7.1 mmHg (SD=3.2) for diastolic BP. Age, sex, race, education, physical activity, alcohol intake, pack-years of smoking, height, weight, triglyceride, LDL-C and asthma were potential correlates of VIM or CV of diastolic BP by the lasso model. In addition to those variables, total cholesterol, HDL-C, fasting glucose, and GFR were potential correlates of VIM or CV of systolic BP. Variables that significantly associated with VIM of systolic BP were: age (years) (β=0.11, p<0.001), white race (β= -0.94, p<0.001), female sex (β=1.36, p<0.001), alcohol intake (drinks/wk) (β=0.01, p=0.001), height (cm) (β= -0.03, p=0.001), and history asthma (β=-0.47, p=0.02). Consistent findings were observed when the outcome was CV of systolic BP. VIM of diastolic BP was also significantly associated with age (β= -0.09, p<0.001) white race (β= -0.47, p=0.26), pack-years (β=0.06, p<0.001), height (cm) (β= -0.03, p<0.001), and history asthma (β=-0.47, p=0.01). The same variables were significantly associated with CV of diastolic BP. Conclusion: Identifying factors associated with long-term BPV can be useful to detect individuals who may be at a greater risk for future higher long-term BPV, which in turn is associated with greater cardiovascular risk.