Comparing test-retest reliability and magnitude of conditioned pain modulation using different combinations of test and conditioning stimuli

This study aimed to compare the reliability and magnitude of conditioned pain modulation (CPM) by applying different test stimuli (TS) and conditioning stimuli (CS). Twenty-six healthy male participants were recruited in the study of two identical sessions. In each session, four TS (electrical, heat, handheld, and cuff pressure algometry) were applied before and during CS (cold pressor test (CPT) or cuff algometry). The same procedure was repeated with 45-min intervals, but with the other CS. Five thresholds were measured including four pain detection thresholds from four TS and pain tolerance threshold from cuff TS (cuff PTT). Intraclass correlation coefficient (ICC (3,1)) and coefficient of variation (CV) were calculated as measures of reliability. The reliability of TS before and during CS was good for all combinations (ICC: 0.60-0.96, CV: 2.2-22.9%), but the reliability of the CPM effect varied (ICC: 0.04-0.53, CV: 63.6-503.9%). The most reliable combinations were considered to be the handheld pressure pain threshold with CPT (ICC: 0.49, CV: 63.6%) and the cuff pressure pain threshold with CPT (ICC: 0.44, CV: 107.6%). Significant CPM effects were found for all combinations, except the combinations of electrical and heat pain thresholds with cuff CS, which indicates the novel classification of the CPM mechanism. The combinations of handheld pressure and heat pain threshold with CPT would provide the minimum sample size to detect the significant CPM changes in further studies. It is beneficial to provide and compare both ICC and CV to design further clinical trials.