Cooling Detection Thresholds in the Assessment of Diabetic Sensory Polyneuropathy

OBJECTIVE: Cooling detection threshold testing may be an important quantitative method for assessing polyneuropathy, in that it has traditionally been viewed as a measure of small-fiber involvement. The present study sought to determine the agreement between two common testing devices and to determine whether these are concordant in their association with predictor variables for diabetic sensory polyneuropathy. RESEARCH DESIGN AND METHODS: A total of 83 patients with diabetes (10 patients with type 1 diabetes and 73 patients with type 2 diabetes) and a wide spectrum of diabetic sensory polyneuropathy severity underwent concurrent cooling detection threshold testing using the Medoc and CASE IV instruments. Common predictor variables for diabetic sensory polyneuropathy were measured on the same day. RESULTS: Measurements of cooling detection thresholds by both instruments were highly correlated (Spearman's correlation coefficient 0.81, P < 0.001) and demonstrated a high degree of agreement by the method of Bland and Altman (95% distribution critical values for the difference in cooling detection thresholds, +7.5 and -5.6 degrees C). Cooling detection thresholds by both instruments were strongly correlated with clinical indicators of large-fiber neuropathy but not with the symptoms of small-fiber neuropathy (pain). CONCLUSIONS: These two instruments available for assessment of cooling detection thresholds are interchangeable for research in diabetic sensory polyneuropathy. However, this modality is equivalent to other modalities of quantitative sensory threshold testing in its association with indicators of large-fiber neuropathy and does not seem to provide an advantage for the prediction of small-fiber involvement.