Introduction Although manual therapists believe that they can diagnose symptomatic joints in the neck by manual examination, that conviction is based on only one study. That study claimed that manual examination of the neck had 100% sensitivity and 100% specificity for diagnosing painful zygapophysial joints. However, the study indicated that its results should be reproduced before they could be generalized. The present study was undertaken to answer the call for replication studies. The objective was to determine the sensitivity, specificity, and likelihood ratio of manual examination for the diagnosis of cervical zygapophysial joint pain. The study was conducted in a private practice located in a rural town. The practice specialised in musculoskeletal pain problems.
Methods Patients who exhibited the putatively diagnostic physical signs of cervical zygapophysial joint pain were referred to a radiologist who performed controlled, diagnostic blocks of the suspected joint, and other joints if indicated. The results of the blocks constituted the criterion standard, against which the clinical diagnosis was compared, by creating contingency tables. The validity of manual diagnosis was determined by calculating its sensitivity, specificity, and positive likelihood ratio.
Results The study sample was 173 patients with neck pain in whom cervical zygapophysial joint pain was suspected on clinical examination, and who were willing to undergo controlled diagnostic blocks of the suspected joint or joints. Manual examination had a high sensitivity for cervical zygapophysial joint pain, at the segmental levels commonly symptomatic, but its specificity was poor. Likelihood ratios barely greater than 1.0 indicated that manual examination lacked validity. Although the results obtained were less favourable than those of the previous study, paradoxically they were statistically not different.
Discussion The present study found manual examination of the cervical spine to lack validity for the diagnosis of cervical zygapophysial joint pain. It refutes the conclusion of the one previous study. The paradoxical lack of statistical difference between the two studies is accounted for by the small sample size of the previous study.
