Diagnostic Accuracy of Neuromuscular Ultrasound vs. Electrodiagnostic Studies for Carpal Tunnel Syndrome: Systematic Review and Meta-analysis of Paired Accuracy Studies

Abstract

This systematic review with meta-analysis compared the diagnostic accuracy of neuromuscular ultrasound (NMUS) and electrodiagnostic studies (EDX) as confirmatory tests for carpal tunnel syndrome (CTS) diagnosis. We used bivariate random-effects models to estimate pooled sensitivity and specificity and generate hierarchical summary receiver-operating characteristic curves to assess diagnostic test accuracy. Nine paired accuracy studies were included, representing 1751 hands (743 clinically diagnosed CTS; 1008 without CTS) that underwent NMUS and EDX. Compared to the clinical diagnosis reference standard, the pooled sensitivity was 86.4% for NMUS and 91.6% for EDX. Pooled specificity was 79.3% for NMUS and 81.9% for EDX. The positive likelihood ratios were 4.2 and 5.1 for NMUS and EDX, respectively, and the negative likelihood ratios were 0.17 and 0.10, respectively. The diagnostic odds ratio was 24 for NMUS and 49 for EDX. No statistically significant differences were identified between NMUS and EDX for sensitivity, specificity, or overall diagnostic accuracy. Overall, the diagnostic accuracy of NMUS is similar to that of EDX for CTS diagnosis, with high sensitivity and moderate specificity for each. The choice between these confirmatory diagnostic tests should incorporate shared decision-making between patients and providers that weighs diagnostic accuracy as well as factors such as patient preferences, test availability, cost, and tolerability.