Optimal cut-off value of the modified Toronto Clinical Neuropathy Score in the diagnosis of polyneuropathy

Abstract

Background and purpose: The modified Toronto Clinical Neuropathy Score (mTCNS) is a valid and reliable scale for the diagnosis and staging of diabetic sensorimotor polyneuropathy (DSP). The aim of this study was to determine the optimal diagnostic cut-off value of the mTCNS in diverse polyneuropathies (PNPs). Methods: Demographics and mTCNS values were retrospectively extracted from an electronic database of 190 patients with PNP and 20 normal controls. Sensitivity, specificity, and likelihood ratios and area under the receiver-operating characteristic (ROC) curve were determined for each diagnosis and different cut-off values of the mTCNS. Patients underwent clinical, electrophysiological and functional assessments of their PNP. Results: Forty-three percent of PNP was related to diabetes or impaired glucose tolerance. mTCNS was significantly higher in patients with PNP than in those without (15.27 ± 8 vs. 0.79 ± 1.4; p = 0.001). The cut-off value for diagnosing PNP was ≥3 (sensitivity 98.4%, specificity 85.7%, positive likelihood ratio 6.88). The area under the ROC curve was 0.987. Conclusion: A value of 3 or more on the mTCNS is recommended for the diagnosis of PNP.