Dynamic Thermal Imaging in the Assessment of Chronic Limb-Threatening Ischemia

Chronic limb-threatening ischemia (CLTI) is a serious medical complication, where the blood perfusion and tissue oxygen delivery are substantially impaired for the affected lower limb. The condition is typically formed as an advanced stage of un- treated peripheral arterial disease and usually affects one or more of the main lower limb arteries or main artery branches. The lower limb can be divided into distinct perfusion areas according to the arterial tree anatomy, known as the angiosome concept, where each anatomical area is sourced mainly by a single source artery. This coarse, yet useful division provides means for non-invasive assessment of the source artery state, corresponding to spatially segmented skin locations. This work is based on four publications, Studies I-IV, which investigate the utility of thermal imaging in the diagnostics of vascular conditions and especially, for CLTI. The work focuses mainly on spatiotemporal analysis during thermal and hydrostatic modulation. In this work, the method is referred to as dynamic thermal imaging, an extension of the traditional steady-state measurements.

Dynamic thermal imaging is a relatively new technique in clinical research, thus currently lacking standardized research tools or methodology. The first and second parts of this work focus on developing and testing potential tools and such method- ologies for dynamic thermal imaging, establishing a connection between thermal modulation tests and current clinical parameters for CLTI diagnostics. Study III proceeds with the methodology and dynamic parameters, introduced in Studies I and II, and applies them to healthy individuals to find the dependency between the thermal recovery time constant and subject-specific confounding factors. Such factors are age, blood pressure, body-mass index, and sex. Study IV presents a blood flow-thermal multiphysics simulation model for the lower limb, based on clinical imaging data, aiming toward individualized simulation models. The study also suggests a method- ology for converting the clinical imaging data to a simulation-compatible form.

The clinical pilot in Study II found a moderate to high linear correlation between the recovery time constant, and ankle-brachial and toe-brachial indices, currently used in the clinical assessment. The thermal recovery time constant was found to be relatively independent from most of the confounding factors, excluding patient age with respect to the observed recovery, especially in the medial angiosomes. The data also suggests that absolute temperatures, frequently used as an indication for PAD or CLTI, might be a poor diagnostic metric due to the altering thermal expression through the development of the disease and high inter-individual variation. The spectral analysis results indicated decreased endothelial and neurogenic spectral band powers for CLTI and claudication patients when compared to healthy individuals. Although, the statistical significance cannot be properly assessed from the pilot data. Additionally, the hydrostatic modulation test did not induce detectable thermal changes, nor showed any correlation with the clinical reference measurements nor with the clinical status in any test parameters, including the spectral band power analysis. The observed inter-individual thermal variation depends on various factors, for which at least age and extensive clinical background should be accounted for in future research. Additionally, environmental factors such as ambient temperature and humidity should be strictly controlled. The true usefulness of thermal and dynamic thermal imaging for diagnostic purposes remains uncertain and requires further research and validation.