Abstract
Lack of a reliable reference standard for carpal tunnel syndrome (CTS) diagnosis could impact the diagnostic test characteristics. This systematic review sought to evaluate differences in the accuracy of CTS diagnostic modalities based on the reference standard used. METHODS: A systematic review was performed following PRISMA guidelines to investigate diagnostic modalities used in CTS. A literature search of Embase, PubMed, and Cochrane Reviews was conducted for the years of 2010-2021 for primary data, and 113 studies met final inclusion criteria. Studies were stratified based on the reference standard utilized and diagnostic modality assessed, and the weighted means of the sensitivities and specificities were calculated. RESULTS: Thirty-five studies used clinical diagnosis alone as a reference standard, and 78 studies used electrodiagnostic study (EDS). The specificity for MRI and ultrasound (US) were substantially lower when EDS was used as the reference standard. MRI was the test most affected by the reference standard used, showing increased sensitivity when using EDS as the reference compared to clinical diagnosis (77.1% versus 60.9%) and decreased specificity (87.6% versus 99.2%). Regardless of the reference standard used, all tests had anticipated false-positive and/or false-negative rates of at least 10%. CONCLUSIONS: Testing characteristics vary greatly based on the choice of reference standard, with the sensitivity of MRI most affected. Regardless of reference standard used, EDS, US, and MRI each had false-positive and/or false-negative rates too great to be appropriate for use as a screening examination.