Background: We aimed to investigate the association of the change in the ambulatory arterial stiffness index (AASI) with that in carotid-femoral pulse wave velocity (cfPWV) during treatment with antihypertensive medication.
Methods: We enrolled 207 hypertensive patients treated with olmesartan monotherapy for 12 weeks. Patients were randomly assigned to treatment with hydrochlorothiazide (HCTZ; n = 104) or azelnidipine (n = 103) for 24 weeks. The cfPWV and 24-h ambulatory blood pressure monitoring (ABPM) results were assessed at baseline and 24 weeks later. The AASI was defined as 1 minus the regression slope of diastolic blood pressure (DBP) on systolic BP (SBP), and was calculated by standard and symmetric regression.
Results: The changes in the AASI and symmetrical AASI were similar between the two groups, while cfPWV in the azelnidipine group decreased more than in the HCTZ group (P < 0.001). The change in AASI was not significantly correlated with change in cfPWV (r = 0.08, P = 0.26), whereas the change in symmetrical AASI was significantly but weakly correlated with change in cfPWV (r = 0.22, P < 0.001). The multivariable linear regression analysis revealed that the association of the change in symmetrical AASI with change in cfPWV remained significant even after adjustments for covariates derived from ABPM (regression coefficient (95% confidence interval): 1.33 (0.35-2.30), P = 0.01).
Conclusions: The present study demonstrated that neither AASI nor symmetrical AASI may be an unequivocal marker of arterial stiffness during antihypertensive treatment.