Ambulatory Arterial Stiffness Index: Reproducibility of Different Definitions

BACKGROUND Ambulatory arterial stiffness index (AASI) has been proposed as a marker of arterial stiffness, which predicts cardiovascular mortality. This study compared the reproducibility of 24-h, daytime, night time, and symmetrical AASI. METHODS A total of 126 untreated hypertensives (mean age 48.2 +/- 10.7 (s.d.) years, 70 men) underwent 24-h ambulatory blood pressure (ABP) monitoring twice 2-4 weeks apart. The reproducibility of AASI was assessed using the following criteria: (i) repeatability coefficient (RC = 2 x s.d. of differences); (ii) RC expressed as a percentage of close to maximal variation (pMV = RC/(4 x s.d. of the mean of paired recordings)); (iii) coefficient of variation (CV); (iv) concordance correlation coefficient (CCC); (v) agreement (kappa) between the two AASI measurements to detect subjects at the top quartile of the respective AASI distributions. RESULTS There was no difference in average AASI values between the two assessments. For 24-h, daytime, night time, and symmetrical AASI, respectively, (i) RC values were 0.24, 0.38, 0.42, and 0.30; (ii) pMV 49.6, 68.8, 73.9, and 56; (iii) CV 40.3, 39.3, 62.9, and 116.3; (iv) CCC 0.60, 0.35, 0.28, and 0.52; (v) agreement 82.5% (kappa 0.54), 72.2% (0.28), 73% (0.22), and 81.7% (0.50). Differences in 24-h mean arterial ambulatory pressure (MAP) and in nocturnal MAP decline between the two assessments were significant determinants of the differences in 24-h and symmetrical AASI values. CONCLUSIONS Although no differences were found in average AASI values of the two ambulatory recordings, significant differences were observed in their reproducibility, with 24-h AASI being the most reproducible measure in terms of all the examined criteria.