Morphometric, biochemical, and Raman spectral analyses of muscle damage in patients with peripheral arterial disease: Providing objective criteria for diagnosis and treatment monitoring

Peripheral arterial disease (PAD), which affects approximately 10 million Americans, is characterized by atherosclerosis of the non-coronary arteries. PAD produces a progressive accumulation of ischemic injury to the limbs that is reflected in a gradual worsening in the myofiber morphology and oxidative damage in the gastrocnemius muscle. In this study, we evaluated the hypothesis that quantitative morphological and biochemical parameters of gastrocnemius myofibers change in a consistent manner during the progression of PAD, provide an objective grading of muscle degeneration in the ischemic limb, and correlate to clinical stage of PAD. Myofiber morphometrics were determined precisely with mathematical equations that incorporated multiple, objectively defined parameters including fiber density, roundness, minor and major axes, and solidity. Oxidatively altered proteins (hydrazide-reactive carbonyls) labeled with mechanisms-based molecular biomarkers were imaged with quantitative fluorescence microscopy and were used to measure oxidative damage in the muscle. A discriminant model was developed based on morphometric and oxidative stress parameters that accurately identified control (85.7% accuracy), claudicating (78.9% accuracy), and critical limb ischemia (85.0% accuracy) patients, with an overall classification accuracy of 83.3% using a cross-validation procedure. We have established that there are consistent and quantifiable changes in myofiber morphology and biochemistry that reflect the progression of PAD muscle degeneration and offer the possibility of an objective index for grading severity of disease within clinical classifications, i.e., Fontaine Stages. Further, we evaluated the use of Raman scattering spectral biomarkers to characterize altered muscle biochemistry in PAD patients and correlate patient blood flow (ABI) with key Raman spectral bands. A partial least squares regression model, based on the Raman spectra, was able to predict patient ABIs with a correlation coefficient of 0.85 using a full cross-validation. When using the first three partial least squares factor scores in combination with linear discriminant analysis, the discriminant model was able to correctly classify control, claudicating, and critical limb ischemia patients with 100% accuracy, using a full cross-validation procedure. In this study, we have demonstrated that Raman spectroscopy can be a powerful bioanalytical tool that is capable of detecting and correlating biochemical composition with patient ABIs and discriminating PAD clinical diagnosis