Short-term blood pressure (BP) variability predicts cardiovascular complications in hypertension, but its association with large-artery stiffness is poorly understood and confounded by methodological issues related to the assessment of BP variations over 24h.

Carotid-femoral pulse wave velocity (cfPWV) assessment and 24-hour ambulatory BP monitoring were performed in 911 untreated, nondiabetic patients with uncomplicated hypertension (learning population) and in 2089 mostly treated hypertensive patients (test population). Short-term systolic BP (SBP) variability was calculated as the following: (1) SD of 24-hour, daytime, or nighttime SBP; (2) weighted SD of 24-hour SBP; and (3) average real variability (ARV), that is, the average of the absolute differences between consecutive SBP measurements over 24 hours. In the learning population, all of the measures of SBP variability showed a direct correlation with cfPWV (SD of 24-hour, daytime, and nighttime SBP, r=0.17/0.19/0.13; weighted SD of 24-hour SBP, r=0.21; ARV, r=0.26; all P<0.001). The relationship between cfPWV and ARV was stronger than that with 24-hour, daytime, or nighttime SBP (P<0.05) and similar to that with weighted SD of 24-hour SBP. In the test population, ARV and weighted SD of 24-hour SBP had stronger relationships with cfPWV than SD of 24-hour, daytime, or nighttime SBP. In both populations, SBP variability indices independently predicted cfPWV along with age, 24-hour SBP, and other factors.

We conclude that short-term variability of 24-hour SBP shows an independent relation to aortic stiffness in hypertension. This relationship is stronger with measures of BP variability focusing on short-term changes, such as ARV and weighted 24-hour SD.