Objective: To critically review the criteria used to diagnose nonarticular soft-tissue rheumatic disorders of the neck and upper limb.
Methods: An extensive search of the literature, including a search of Medline and EMBASE, authoritative recent reviews, and relevant textbooks, was completed. The diagnostic criteria used in epidemiologic studies were compared and the reliability and validity of these criteria were assessed.
Results: Altogether, the search identified 117 relevant research articles, among which 69 included a physical examination component, but few specified diagnostic criteria. Evidence supported respectable levels of between-observer repeatability regarding: symptom questionnaires (kappa, 0.52 to 0.79); measurement of shoulder range of motion with a goniometer (intraclass coefficients > 0.70); tests for carpal tunnel syndrome (Tinel's and Phalen's kappa, 0.53 to 0.80); and demonstration of neck tenderness (kappa = 0.43). The Katz hand diagram, and combinations of physical signs of carpal tunnel syndrome, show reasonable sensitivity and specificity for that diagnosis but only among patients referred to specialists with that putative diagnosis; no such validity has been shown among the general population. Only 1 diagnostic examination schedule has published data on both the reliability and the validity of its criteria and diagnoses. For the remaining soft-tissue upper-limb disorders, diagnostic criteria rely apparently on face and content validity and reliability data have not been published.
Conclusion: Classification of specific disorders of the neck and upper limb requires a back to basics approach. At present, the diagnosis of most of these conditions relies heavily on the clinical opinions of investigators and there are insufficient data to indicate that these criteria are repeatable, sensitive, or specific. Recent European initiatives offer scope to follow a more disciplined approach, but more work is urgently required.