Vascular Ultrasound Data during Enhanced External Counterpulsation for Predicting Coronary Artery Disease

Acute exercise stimulus not only effectively highlights the response of cardiovascular function, but also plays an important role in the prediction of disease. This study aimed to investigate whether dynamic characteristics of peripheral artery function can predict coronary artery disease (CAD).

One-hundred subjects were enrolled to this study, including 50 patients with CAD and 50 matched volunteers as controls. All subjects received a single, 45-min session of enhanced external counterpulsation (EECP) therapy. Vascular ultrasound measurements in the carotid and brachial arteries were conducted before, during and immediately after EECP intervention. Blood flow spectrum curves in carotid and brachial artery analyzed by image processing are used to establish transfer function establishment based on 12-order ARX model. Two hundred and nineteen transfer function characteristics were calculated, including coefficient A, B, Zero(zx), Pole(px) and Pole(py) between CAD and control groups before EECP, during EECP and immediately after EECP. The classification was performed by Extreme Learning Machine combining with 10-fold cross-validation.

Averaged classification accuracy of transfer function characteristics before EECP, during EECP, immediately after EECP and EECP intervention (before, during and after EECP) were 89%, 93%, 94% and 90%, respectively. Additionally, Area Under Curve before EECP, during EECP, after EECP and EECP intervention were 0.87, 0.93, 0.95 and 0.89, respectively.

The findings suggest that transfer function characteristics of blood flow induced by EECP intervention may have the potential for the prediction of CAD. Moreover, data during EECP and immediately after EECP have a higher predictive value for CAD.Clinical Relevance-This evaluates the coronary artery disease by using noninvasive peripheral blood flow monitoring.