84 aligned with the y-axis, which represents PC2 and therefore determines 17.9% of the outcome. CPT and WDT MLi diverge by a large angle (close to 180֯), which indicates a negative correlation. Figure 2 Principal component analysis of thermal threshold testing (TTT) parameters for the test foot. Sarcoidosis patients without SFN are displayed as blue dots and sarcoidosis patients with probable small fiber neuropathy (SFN) are displayed as yellow triangles. Overall, results overlapped between participants with probable SFN and those without SFN. When two vectors are close, with a small angle between them, the two variables represent a positive correlation. An example is CDT MLe and HPT. If they are at an angle of 90° to each other, they were not likely to be correlated. An example is PHS and WDT MLi. When they diverged and formed a large angle (close to 180°), they are negatively correlated. An example is CPT and WDT MLi. During TTT measurements, MLe could be time-consuming. In contrast with MLi, which always measured the CDT and WDT by a fixed number of 3 measurements, the number of levels required for MLe depended on the consistency of the participant’s performance. Measuring time varied between 1 min using 6 levels for successful measurements, up to 27 minutes and 25 levels for patients who seemed desensitized. Figure 3 visualizes the distribution of abnormalities measured with MLi or MLe. Most participants showed no abnormalities with either of the methods. By far the lowest proportion of participants showed corresponding abnormalities in both MLi and MLe. When comparing abnormalities measured with MLi only or MLe only, MLe found abnormal results more often than MLi (22% vs 17% for CDT and 11% vs 4% for WDT). 88 5
RkJQdWJsaXNoZXIy MjY0ODMw