Reproducibility of arterial stiffness and wave reflections in chronic obstructive pulmonary disease: the contribution of lung hyperinflation and a comparison of techniques.

Significant cardiovascular morbidity and mortality exists in chronic obstructive pulmonary disease (COPD). Arterial stiffness is raised in COPD and may be a mechanistic link. Non-invasive assessment of arterial stiffness has the potential to be a surrogate outcome measure, although no reproducibility data exists in COPD patients. Two studies (23 and 33 COPD patients) were undertaken to 1) assess the Vicorder reproducibility of carotid-femoral pulse wave velocity and Augmentation index in COPD; 2) compare it to SphygmoCor; and 3) assess the contribution of lung hyperinflation to measurement variability. There were excellent correlations and good agreement between repeat Vicorder measurements for carotid-femoral pulse wave velocity (r = 0.96 (p < 0.001); mean difference ±SD = -0.03 ± 0.36 m/s (p = 0.65); co-efficient of reproducibility = 4.02%; limits of agreement = -0.68-0.75 m/s). Augmentation index significantly correlated (r = 0.736 (p < 0.001); mean difference ±SD = 0.72 ± 4.86% (p = 0.48), however limits of agreement were only 10.42-9.02%, with co-efficient of reproducibility of 27.93%. Comparing devices, Vicorder values were lower but there was satisfactory agreement. There were no correlation between lung hyperinflation (as measured by residual volume percent predicted, total lung capacity percent predicted or the ratio of inspiratory capacity to residual volume) and variability of measurements in either study. In COPD, measurement of carotid-femoral pulse wave velocity is highly reproducible, not affected by lung hyperinflation and suitable as a surrogate endpoint in research studies. Day-to-day variation in augmentation index highlights the importance of such studies prior to the planning and undertaking of clinical COPD research.