OBJECTIVE: To compare sonographic and surgical measured perimeters of the median nerve; to evaluate the diagnosis of carpal tunnel syndrome by median nerve cross-sectional area; to verify the association between cross-sectional area of the median nerve and carpal tunnel syndrome severity. MATERIALS AND METHODS: Thirty patients with established carpal tunnel syndrome were studied. Cross-sectional area and sonographic perimeter of the median nerve were measured. The correlation between clinical and sonographic findings and association with carpal tunnel syndrome severity were evaluated. Sonographic and surgical perimeters were compared. Clinical classification, surgical perimeter, cross-sectional area and sonographic perimeter of the median nerve were compared. Statistical analysis utilized paired samples t-test, Pearson's correlation, Bland-Altman's diagram, Kolmogorov-Smirnov's test, Welch's and Wilcoxon's tests. RESULTS:Five patients were excluded; 25 patients were studied; 60% of patients had moderated disease, and 60% presented cross-sectional area > 0.15 cm². Distribution of surgical perimeter was not normal (p = 0.5); the sonographic perimeter distribution was normal (p = 0). There was a statistically significant difference between perimeters (paired samples t-test, p < 0.0001, confidence interval = 95%). Pearson's correlation corresponded to 0.3913. Bland-Altman diagram demonstrated higher values for surgical perimeters. Median nerve cross-sectional area > 0.09 cm² was found in all the patients. CONCLUSION: No association was observed between median nerve sonographic and surgical perimeters. Median nerve cross-sectional area > 0.09 cm² was valid for diagnosis of carpal tunnel syndrome. No association was observed between median nerve cross-sectional area and carpal tunnel syndrome severity.

Keywords: Carpal tunnel syndrome; Ultrasonography; Diagnosis; Neuropathy.