Estimating the risk of peripheral neuropathy in diabetes and chronic kidney disease: a new equation based on physical tests

AimDistal symmetric polyneuropathy (DSP) is a frequent complication of both diabetes mellitus (DM) and chronic kidney disease (CKD), but remains underdiagnosed.Diagnosis usually employs complex tools like the validated Michigan Neuropathy Screening Instrument (MNSI) that are however time consuming and hardly usable in routine practice.This study aimed to propose a predictive equation to assess the risk of DSP based on physical tests in order to provide a simple and reliable DSP screening. MethodsThe study included 59 participants with DM and/or stage 4-5 CKD not on dialysis.DSP was assessed using the MNSI.Participants completed gait, proprioception, postural control, and ankle/foot muscle strength tests.Multivariate regression models were developed using physical performances to predict MNSI outcomes, and area under the ROC curve (AUC) were calculated.Interval validity was considered using bootstrap inclusion frequency (BIF) and leaveone-out process. ResultsThe joint position sense (JPS) was the most discriminative in univariate analysis.In multivariable analysis, only the JPS was retained in all the models (BIF: 77.5% to 93.5%), while balance, step length and single-leg stance were also retained in different models after stepwise selection.The derived equations evidenced high AUC with respect to all the MNSI outcomes (AUC from 0.738 to 0.868), with satisfactory sensitivity and specificity. ConclusionThis study identifies a simple and feasible screening approach using functional tests to detect DSP in patients with DM or CKD.Early DSP detection by simple physical tests may enable timely referral to initiate strategies aimed at reducing long-term neuropathy complications.