A Comparison of Three Diagnostic Tests for Carpal Tunnel Syndrome Using Latent Class Analysis

Background: The current reference standard for carpal tunnel syndrome is under debate. Recent studies have demonstrated similar diagnostic accuracy between ultrasound and nerve conduction studies. The purpose of the present study was to determine the sensitivity and specificity of ultrasound, nerve conduction studies, and Carpal Tunnel Syndrome 6 (CTS-6) for the diagnosis of carpal tunnel syndrome using latent class analysis. Methods: Latent class analysis is a statistical technique that can be used to estimate the accuracy of diagnosis when there is no universally accepted reference standard. This type of analysis is useful in the setting of carpal tunnel syndrome as there remains substantial controversy with respect to the necessity of nerve conduction studies and other confirmatory testing. CTS-6 is a validated clinical diagnostic tool for the diagnosis of carpal tunnel syndrome that has been shown to have a high sensitivity and specificity. Data from a database on the cases of eighty-five consecutive patients who had had nerve conduction studies, CTS-6, and ultrasound were analyzed using classical latent class analysis, assuming that the three tests were imperfect and conditionally independent. Results: The sensitivities of ultrasound, CTS-6, and nerve conduction studies were 91% (95% confidence interval [CI], 81% to 98%), 95% (95% CI, 86% to 99%), and 91% (95% CI, 81% to 97%), respectively. The specificities of ultrasound, CTS-6, and nerve conduction studies were 94% (95% CI, 80% to 100%), 91% (95% CI, 74% to 99%), and 83% (95% CI, 66% to 95%), respectively. Conclusions: Ultrasound, nerve conduction studies, and CTS-6 have similar sensitivity and specificity for the diagnosis of carpal tunnel syndrome. The currently accepted reference standard (nerve conduction studies) had the lowest sensitivity and specificity of the three tests. These findings support previous studies that have suggested that CTS-6 and ultrasound are highly accurate in the diagnosis of carpal tunnel syndrome and that nerve conduction studies are not necessary in most cases. Level of Evidence: Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.