Arterial stiffness assessed by systemic versus regional indices: determinants and diagnostic agreement

Objective: Arterial stiffness is closely associated with aging and hypertension and may improve the stratification of cardiovascular risk in patients with hypertension. This study investigated the determinants and the diagnostic agreement of systemic (brachial-ankle pulse wave velocity [baPWV]) and regional (carotid distensibility coefficient [DC]) arterial stiffness indices. Design and method: Adults referred to a hypertension outpatient clinic were subjected to: (i) triplicate office blood pressure (BP) measurement using a validated professional oscillometric upper-arm cuff device Microlife WatchBP Office, (ii) baPWV measurements using an automated oscillometric device Microlife WatchBP Office Vascular, and (iii) DC and carotid intima-media thickness (cIMT) measurements with carotid ultrasound. Results: Data from 136 participants were analyzed (mean age 60±12 [SD] years, 71% males, body mass index 28±4 kg/m 2 , 83% hypertensives, 12% diabetics, 16% smokers, baPWV 14±2 m/sec, DC 24±9 10- 3 /kPa). There was a moderate negative association between baPWV and DC (r = -0.37, p<0.01). baPWV was positively correlated with age (r = 0.64, p<0.01), systolic office BP (r = 0.27, p<0.01) and cIMT (r = 0.29, p<0.01), while DC was inversely associated with age (r = -0.37, p<0.01), systolic office BP (r = -0.48, p<0.01), and cIMT (r = -0.29, p<0.01). In multivariable linear regression models (independent variables: age, sex, systolic office BP, cIMT, diabetes), age and systolic office BP appeared to be significant determinants of baPWV (beta = 0.12/0.05, respectively) and DC (beta = -0.32/-0.27, respectively). The agreement between the two methods in identifying individuals with increased arterial stiffness (1st quartile of DC distribution and 4th quartile of baPWV distribution) was fair (agreement 71%, kappa = 0.22, p<0.05). Conclusions: These results suggest that although these two methods for assessing arterial stiffness are correlated and determined by similar variables, their agreement in identifying individuals with increased arterial stiffness is fair, implying that they may provide complementary information.