Abstract
Background Hypertension-associated cardiovascular events are particularly associated with elevated systolic blood pressure (SBP) in late life, yet long-term interactions between SBP, diastolic BP (DBP) and arterial stiffness in development of late-life hypertensive phenotypes remain unclear. Methods and Results In the UK Biobank, we determined associations between arterial stiffness index (ASI), SBP, DBP, and their progression, and transition from normotension (<140/90 mm Hg) to hypertension or elevated ASI (>10 m/s). Associations were determined by general linear and logistic regression, adjusted for cardiovascular risk factors and variability of measurements across follow-ups. Mean values of baseline SBP, DBP, and ASI were determined stratified by deciles of age, blood pressure, and ASI, with CIs determined by bootstrapping. In 169 742 participants at baseline, ASI was more strongly associated with DBP than SBP, before and after adjustment for risk factors (β: SBP, -0.01 [P<0.001]; DBP, 0.06 [P<0.001]), while DBP was more strongly associated with progression of ASI (n=13 761; β: SBP, 0.013 [P=0.01]; DBP, 0.038 [P<0.001]). Baseline ASI was associated with increasing SBP during follow-up (β=0.02, P<0.001) but not DBP (β=0.0004, P=0.39), but was associated with a younger age of transition from rising to falling DBP (highest versus lowest quartile: 51.2; 95% CI, 49.9-52.3 versus 60.4; 95% CI, 59.6-61.3 [P<0.001]). ASI predicted the development of isolated systolic hypertension (odds ratio, 1.30; 95% CI, 1.22-1.39), particularly after adjustment for measurement variability (odds ratio, 2.29). Conclusions Midlife DBP was the strongest predictor of progression of arterial stiffness, while arterial stiffness was associated with earlier transition to a falling DBP. Prevention of long-term harms associated with arterial stiffness may require more intensive control of midlife DBP.