BACKGROUND: Conditioned pain modulation (CPM) is a set of psychophysical paradigms that is increasingly used clinically to evaluate descending pain modulation pathways. Impairment is common in chronic pain, suggesting CPM may serve as a mechanistic indicator. However, the lack of protocol standardization and reference data prevents clinical use in individual patients. METHODS: We compared two CPM protocols with different conditioning stimulus intensities, test stimulus types, and interaction timing. We assessed CPM effect size, test-retest reliability and sensitivity to detect loss of descending inhibition. RESULTS: Conditioning with 0°C water led to stronger inhibition of pressure pain threshold (PPT) than conditioning with 7°C water (Cohen's d = 0.52), when tested immediately after conditioning. When tested during conditioning, effects of 7°C water immersion on heat pain sensitivity had similar magnitude (D = 0.53) and test-retest reliability (ICC = 0.77) as those on PPT (D = 0.54, ICC = 0.73). For all outcomes assessed, 95% confidence intervals (CI) of CPM effect included some facilitation instead of inhibition. The maximum degree of facilitation compatible with normal CPM (upper cutoff of CI) indicates potential sensitivity to detect individual abnormality. This was most favourable for PPT assessed after conditioning with 0°C water (decrease by more than 75 kPa or 14% of baseline PPT). CONCLUSIONS: In conclusion, testing during conditioning stimulation yields medium to large effect sizes and good test-retest reliability. PPT testing immediately after ice water immersion has the narrowest 95% CI and hence offers the potential to generalize CPM assessments beyond group-level differences and compare inhibition among individuals in clinical practice. SIGNIFICANCE STATEMENT: Indicating the main aspects where this work adds significantly to existing knowledge in the field, and if appropriate to clinical practice. Simultaneous CPM protocols exhibit large effect sizes but are confounded by divided attention. We recommend a sequential protocol and provide model reference data for abnormal facilitation.
Lin et al. (Fri,) studied this question.