How should peripheral neuropathy be assessed in people with diabetes in primary care? A population-based comparison of four measures

Aims To test the accuracy of four measures of peripheral diabetic neuropathy in a primary care population. Methods Type 2 diabetic (n = 544) and 544 non-diabetic participants aged 45-76 years were randomly selected from general practice registers. Neuropathy was assessed using vibration threshold (VT) and scores for light touch, thermal sense and modified Michigan Neuropathy Screening Instrument questionnaire. These measures were assessed for variation with diabetes status, age, diabetes duration, HbA(1c) , and presence of retinopathy and nephropathy. Light touch, thermal sense and questionnaire scores were assessed against VT using ROC curve analysis. Results Only VT and light touch were different between diabetic and non-diabetic groups (P = 0.02 and < 0.0001, respectively). All measures were significantly associated with diabetes duration and retinopathy, and all except questionnaire score (P = 0.14) with age. None was associated with nephropathy and only questionnaire score was associated with HbA(1c) (P = 0.033). VT varied as expected across scores of light touch (chi(2) = 41.65, P = 0.0001), thermal sense (chi(2) = 15.86, P = 0.015) and questionnaire (chi(2) = 21.22, P = 0.047). Area under the curve values for light touch, thermal and questionnaire scores were 0.72 (95% confidence interval (CI) 0.63, 0.82), 0.63 (95% CI 0.52, 0.73) and 0.64 (95% CI 0.53, 0.74), respectively. Conclusions All measures had associations with risk factors for neuropathy, but light touch score (monofilament) had the strongest association with vibration threshold (the chosen gold standard) and thus appeared the most appropriate tool for use in primary care, because of its validity and simplicity of use.