Initial evaluation of the use of dynamic vascular optical imaging for predicting outcome of surgical intervention to improve blood perfusion in feet of diabetic patients

Abstract

Approximately 12 million people in the United States are affected by peripheral artery disease (PAD), characterized by an accumulation of plaque in the arteries of the lower extremities. In advanced stages, treating physicians often recommend a surgical intervention to improve blood flow to the feet. However, about 50% of patients require a second intervention within 12 months. Here we report on the potential of dynamic optical imaging (DOI) to predict the long-term outcome of such surgery. Our DOI system consists of four detection patches, each configured with two SI-detectors and four laser diodes at different wavelengths (678 nm, 780 nm, 808 nm and 850 nm). These patches are placed on four different angiosomes of the foot to record the dynamical responses to inflations and deflations of a thigh cuff. Inflating a cuff causes blood to accumulate in the foot, while deflating the cuff reduces the amount of blood. DOI measurements can be characterized by a response time to cuff inflation (rise time), and a plateau time between cuff inflation and deflation. For this study 16 patients with no previous history of interventions were enrolled, and DOI data was collected before and after the intervention. 4 of the 16 patients needed a second intervention within 6 months. We found a strong correlation between the changes in pre-and post-intervention rise time and the 6 months treatment outcome. A ROC analysis showed that it was possible to categorize outcomes correctly with an AUC (Area Under the Curve) of about 83%, and corresponding specificity of 100% and sensitivity of 75%.