Quantitative somatosensory testing of warm and heat-pain thresholds: The effect of body region and testing method

Objectives: To study whether the sensitivity to noxious and innocuous heat varies across the body. Methods: Heat-pain threshold and warm sensation threshold were measured in 20 patients, using the Method of Limits (MLI) and the Method of Levels (MLE) in 5 regions: the chest, forearm, mid thigh, and the dorsal surface of the hand and foot. Results: With the Method of Limits, heat-pain threshold increased gradually from the lowest level in the chest (mean 42 degrees C) to peak level in the foot (44.5 degrees C, P < 0.001). With the Method of Levels, heat-pain threshold did not differ between body regions. Warm sensation thresholds measured with both Method of Limits and the Method of Levels was higher in the chest (36.2 degrees C) and foot (36.8 degrees C) compared to the other regions (mean of 35.3 degrees C, P < 0.01). The correlation between heat-pain threshold and warm sensation threshold was low to moderate, depending on the tested region. Differences in heat-pain threshold and warm sensation threshold between the methods were highest in the legs and smallest in the chest. The correlation between the Method of Limits and Method of Levels was moderate for heat-pain threshold (0.57) and good for warm sensation threshold (0.71). Discussion: The sensitivity to noxious heat is uniform across the body when measured with a reaction-time-free method (Method of Levels), but is greater in proximal than in distal regions, when measured with a reaction-time-dependent method (Method of Limits). Regardless of measuring method, the sensitivity to innocuous heat is not uniform across the body. It is concluded that the Method of Levels is preferred when heat-pain threshold is to be compared between body regions. For heat-pain threshold, within-patient comparisons can be made between each pair of regions tested. However, for warm sensation threshold, within-patient comparisons should be conducted between contralateral symmetrical regions.