Assessment of test repeatability of arterial stiffness index

Background: Increased arterial stiffness is an early indicator of vascular disease. Several methods may be used to determine arterial stiffness. One method obtains an arterial stiffness index from the vascular dynamics of oscillometric-derived brachial artery pressure. Methods: To determine the test-retest repeatability of the CardioVision MS-2000, 47 healthy hospital employees had five consecutive measurements of arterial stiffness index measured after a 5-10 min period of rest and then repeated after an average of 146.8 days. Their mean age was 37 years and 71% were women. Results: The mean arterial stiffness index was 39.6 ± 9.7 and 37.2 ± 10.5 mmHg × 10 (P = 0.22) for the first and second time period, respectively. We computed an intraclass correlation coefficient of 0.31 and 0.33 for the first and second time periods, which is the measure of consistency or agreement of arterial stiffness index values within cases. The intraclass correlation coefficient for systolic blood pressure, diastolic blood pressure, heart rate and arterial stiffness index were 0.68 (P = 0.0001), 0.70 (P = 0.0001), 0.35 (P = 0.02) and 0.25 (P = 0.08), respectively. Conclusion: The results of this study suggest poor test-retest repeatability if consecutive measurements are used. The intraclass correlation coefficient, however, could be improved by eliminating the highest and lowest value from a set of measurements.