The diagnostic accuracy of high-resolution ultrasound in screening for carpal tunnel syndrome and grading its severity is moderated by age

Abstract

ObjectiveTo assess the effect of age on the accuracy of high-resolution ultrasound (HRUS) in the diagnosis and grading of carpal tunnel syndrome (CTS). MethodsPatients with symptoms and signs of CTS (N = 527 wrists) were evaluated using electrodiagnostic studies (EDx) for CTS diagnosis and grading. Median nerve cross-sectional areas at carpal tunnel inlet (CSA) and at forearm level were measured by HRUS and the ratio of these values was calculated (WFR). Healthy controls underwent identical testing (N = 122 wrists). HRUS accuracy was assessed against the EDx standard by Receiver Operator Characteristic (ROC) curve analysis. ResultsIn patients >65 y with moderate and severe CTS, disease-related increases in CSA and WFR were negatively correlated with increasing age. Subjects were grouped by age into younger (<65 y) and older (≥65 y). The c-statistics for CSA and WFR respectively were: For CTS diagnosis, younger group: 0.94 and 0.96 (excellent); older group: 0.85 and 0.86 (satisfactory). For CTS grading, younger group: differentiating mild CTS from controls: 0.90 and 0.92 (excellent); mild from moderate: 0.79 and 0.74 (satisfactory); moderate from severe: 0.82 and 0.78 (satisfactory). For CTS grading, older group: differentiating mild CTS from controls: 0.83 and 0.83 (satisfactory); mild from moderate: 0.53 and 0.61 (poor); moderate from severe: 0.65 and 0.53 (poor). ConclusionsFor subjects aged <65 y, HRUS accuracy is excellent in CTS diagnosis and satisfactory in grading. For older subjects, accuracy is satisfactory in diagnosis but not in grading. SignificanceHRUS for CTS has diagnostic limitations selectively in older individuals.