Reliability of the clinical examination in the diagnosis of neurogenic versus vascular claudication.

Abstract

As research increasingly challenges the diagnostic accuracy of advanced imaging for lumbar spinal stenosis, the impression gleaned from the office evaluation becomes more important. Neurogenic claudication is a hallmark of lumbar spinal stenosis, but the reliability of clinical impression of claudication has not been studied. To determine the reliability of the clinical examination for neurogenic claudication in an idealized setting. Prospective masked controlled trial. Persons aged 55 to 90 years were recruited to form three groups: those offered surgery for spinal stenosis by academic spine surgeons, those who had peripheral vascular symptoms and positive ankle-brachial index (ABI), and those who were asymptomatic. All were extensively screened against confounding diseases. Forty-three neurogenic, 12 vascular, and 35 asymptomatic recruits were tested. Clinical impression of neurogenic claudication. A neurosurgeon and a vascular surgeon, masked to each other's findings, imaging, and recruitment status, performed a codified but unconstrained comprehensive spine and vascular history and physical examination for each subject. The surgeon's impression was recorded. Masked surgeons strongly agreed with the recruitment diagnosis (neurosurgeon kappa 0.761, vascular surgeon kappa 0.803, both p<.001) and with each other (kappa 0.717, p<.001). However, disagreements did occur between examiners and recruitment diagnosis (neurosurgeon n=13 cases, vascular surgeon n=10) and between examiners (n=14 cases). Pain level and marginally some measures of disability related to the agreement, but specific aspects of the physical examination, showed poor interrater reliability and did not contribute to the agreement. The clinical impression of neurogenic claudication is a reliable construct. The history, but not the poorly reproduced physical examination, contributes to reliability. The level of disagreement between experts in this simplified, yet severely involved, population raises concern about the risk of misdiagnosis in individual cases. Thus, surgical and other consequential decisions about diagnosis may require ancillary tests such as electromyography or ABI.