Inter-rater agreement and accuracy of clinical tests used in diagnosis of Carpal Tunnel Syndrome

Abstract

A sample of 42 subjects (84 hands) with complaints of pain, numbness and/or tingling were used to study the inter-rater agreement and accuracy of seven clinical tests which can be used to assist in the diagnosis of carpal tunnel syndrome (CTS). A 'gold standard diagnosis', defined as a clinical diagnosis of CTS rendered by one of two hand surgeons and supported by electrodiagnostic evidence of CTS, was used to classify the 84 hands as to the presence of CTS. Two therapists were blinded to patient history, electrophysiology, diagnosis and all evaluations performed by the other therapist, until clinical testing was completed. Tests performed included: wrist flexion, wrist extension, Tinel's, tethered median nerve (TMN), pinch, vibration and Semmes-Weinstein monofilament (SWMF) tests. Substantial inter-rater agreement was observed between the two therapists for five of the clinical tests (k > 0.71) with SWMF and TMN having lower agreement. The most accurate test was Phalen's wrist flexion test. Good accuracy was demonstrated by pinch and vibration tests. Tinel's test was characterized by lower sensitivity, but false positives were rare. Wrist extension and TMN tests had poor sensitivity. SWMF testing was very sensitive, but a high number of false positives occurred when 'normal' was classified as 2.83. Reliability and accuracy of these tests supports their use as components of a clinical diagnosis of CTS.