A novel approach to the diagnosis and assessment of symptomatic diabetic neuropathy.

The McGill Pain Questionnaire has previously been shown to be useful in the differential diagnosis of painful conditions and has also been used to assess the efficacy of therapeutic intervention. We have applied this simple test to 42 patients with painful diabetic neuropathy and 49 control subjects with painful legs or feet of varying aetiologies. Analysis of responses by stepwise linear discriminant analysis confirmed a significant difference between the 2 groups (Wilks' lambda, P less than 0.001). Each questionnaire was given a single score as a result of the analysis, and this score correctly classified a total of 91% to either the neuropathic or control groups, and when applied prospectively to a further 25 ungrouped questionnaires a useful probability of their belonging to a diagnostic group was obtained. Use of the questionnaire might be a useful aid to the differential diagnosis of the painful diabetic leg.