Diagnostic Accuracy of Electrodiagnostic Comparative Latency Studies of Carpal Tunnel Syndrome: Single Test and Concordance Between Multiple Tests

<b>Background:</b> The optimal number of electrodiagnostic tests required to confirm carpal tunnel syndrome (CTS) has not been systematically evaluated. While single comparative latency study (COLS) is commonly used, it remains unclear whether diagnostic accuracy improves when concordance between multiple COLSs is required. <b>Methods:</b> We retrospectively reviewed the electrodiagnostic data of patients referred to our center with upper limb symptoms. Diagnostic accuracy was assessed for individual COLSs-median-ulnar mixed palmar latency difference (palmdiff), median-ulnar ring finger latency difference (ringdiff), and median-radial thumb latency difference (thumbdiff)-and for concordance between two COLSs. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated within diabetes mellitus (DM) and non-DM groups. <b>Results:</b> We included 538 patients, of whom 305 had CTS and 109 had DM. Among patients without DM, the PPV ranged from 87.6% to 92.6% for single COLS and 94.1% to 94.8% for concordance between two abnormal COLSs. When only patients with symptom durations of >6 months were considered, the PPV of concordance between two abnormal COLSs was consistently 100%. Among patients with DM who were younger than 60 years, the PPV for a single COLS was >89%, and that for concordance between two abnormal COLSs was >94%, whereas in those aged ≥60 years, PPVs dropped to 71%-83% for both strategies, and specificity remained high (>90%) only for concordance between two abnormal COLSs. <b>Conclusions:</b> Concordance between two abnormal COLSs enhances diagnostic precision for CTS, particularly in patients without DM and in patients with DM under 60 years of age. In patients with DM aged ≥60 years, the diagnostic accuracy of COLSs was low.