Objective.-To compare the sympathetic skin responses (SSRs) in patients with multiple sclerosis (MS), clinically isolated syndrome (CIS), and healthy controls. Methods.-SSR was recorded on both hands and feet in 30 patients and 20 healthy controls. SSR results (latency measurements) were compared in patients with normal or abnormal brainstem auditory evoked potentials (BAEPs), visual evoked potentials (VEPs) and somatosensory evoked potentials (SEPs). Results.-Twenty-three (76.6%) and sixteen patients (53.3%) with MS had abnormal SSR recordings based on 2-standard deviation (SD) or 3-SD (from the mean of the control group) abnormality criteria, respectively. Sixty-six percent and 40 percent of patients had abnormal (>2 SD) SSR in at least one hand and one foot, respectively. Patients with absent SSR had more severe disease and higher Expanded Disability Status Scale (EDSS) scores. Fourteen patients had an EDSS of zero, of whom nine had abnormal SSR and others had at least one abnormal EP study. Patients with abnormal SSR had significantly more abnormal BAEPs and SEPs than patients with normal SSR. SSR latencies were significantly correlated with EDSS and disease duration (P < 0.01). All patients had at least one abnormal electrophysiological study. ROC-curve analysis showed that a cut-off score of 7008 ms as the sum of all-4-limb SSR latencies had a 80% sensitivity and 95% specificity for differentiating MS patients from healthy controls. Conclusions.-This study suggests that SSR is a useful tool for assessment of autonomic function and can be complementary to EDSS and other electrophysiological studies in patients with MS and CIS. (C) 2011 Elsevier Masson SAS. All rights reserved.