The comparison of multiple F-wave variable studies and magnetic resonance imaging examinations in the assessment of cervical radiculopathy.

Abstract

The aims of this study were to investigate the correlation of the findings of multiple median and ulnar F-wave variables and magnetic resonance imaging examinations in the prediction of cervical radiculopathy. The data of 68 patients who underwent both nerve conduction studies of the upper extremities and cervical spine magnetic resonance imaging within 3 mos of the nerve conduction studies were retrospectively reviewed and reinterpreted. The associations between multiple median and ulnar F-wave variables (including persistence, chronodispersion, and minimal, maximal, and mean latencies) and magnetic resonance imaging evidence of lower cervical spondylotic radiculopathy (i.e., C7, C8, and T1 radiculopathy) were investigated. Patients with lower cervical radiculopathy exhibited reduced right median F-wave persistence (P = 0.011), increased right ulnar F-wave chronodispersion (P = 0.041), and a trend toward increased left ulnar F-wave chronodispersion (P = 0.059); however, there were no other consistent significant differences in the F-wave variables between patients with and patients without magnetic resonance imaging evidence of lower cervical radiculopathy. In comparison with normal reference values established previously, the sensitivity and positive predictive value of F-wave variable abnormalities for predicting lower cervical radiculopathy were low. There was a low correlation between F-wave studies and magnetic resonance imaging examinations. The diagnostic utility of multiple F-wave variables in the prediction of cervical radiculopathy was not supported by this study.