Interrater reliability of the needle examination in lumbosacral radiculopathy

Low back pain and lumbar radiculopathy are among the most common painful disorders affecting the adult population. This study hypothesizes that there is good correlation between the diagnostic impression of an unblinded electromyographer, using clinical and electromyographic information, and an independent electromyographer, who uses the needle examination only to assess for lumbar radiculopathy. This is a prospective, single-blinded, observational pilot study. The needle examination was electronically recorded, reproduced, and shown to a second examiner, blinded to all clinical data. Diagnostic impressions from both examiners were recorded and evaluated for agreement. Six recorded cases were reviewed by 66 blinded examiners. Overall diagnostic agreement was 46.9% (60.5% faculty level, 28.5% resident level). Logistic regression shows a strong association between training level and agreement on diagnostic impression (odds ratio, 1.9; 95% confidence interval, 1.12-3.22; P = 0.019). This study shows that there is fair interrater reliability between faculty-level examiners and poor reliability among resident-level examiners when the needle examination is used to evaluate patients with lumbar radiculopathy.