Plantar Quad-Zone Analysis: A study of healthy feet

Abstract

The plantar aspect of the foot is a common site for pathology. Diabetic foot ulceration (DFU) and peripheral arterial disease (PAD) present the populations with the highest risk. Many people with diabetes experience foot neuropathy which can result in extensive skin damage, undetectable to the eye but which is potentially detectable using digital infra-red thermal imaging (DITI). This study aims to present a quad-zone approach to assess healthy feet for thermal symmetry, and multi-reader reliability, as a precursor to clinical trials. A FLIR T650 IR thermal camera (640x480) was used to image the soles of 30 healthy volunteers within a temperature-controlled environment. Five readers blind analysed the cases: the whole foot; and four regional zones (heel, arch, first ray, and lateral forefoot); and finally, the individual region of interest (ROI) analysis of the toes. Inter-operator reliability was assessed for the five readers. Mean values for each area of interest were assessed for contra-lateral symmetry. Healthy feet display a wide range of temperatures: left (min=22.60, max=34.00, mean=27.57, STDEV=2.23) and right (min=22.70, max=34.10, mean=27.63, STDEV=2.32). Differences between feet or quad-zones of the same individual were consistently <1 oC. Plantar toes can be significantly cooler than the rest of the foot and exhibit much greater differences (up to 2.6 oC), however always <1 oC between adjacent toes on the same healthy foot. No statistically significant differences between corresponding bilateral areas could be found. ICC=0.990 to 0.998 for all four quad-zones confirmed excellent inter-operator reliability in blinded assessment between five readers. The quad zone technique is easy and time efficient to perform, ideally suited to the clinical environment, with a very high level of reproducibility. This study of healthy feet provides the underpinning evidence for further research to determine differences in the diabetic and/or neuropathic foot and seek correlation with angiographic findings.