Optimal screening distance for ulnar neuropathy at the elbow.

Abstract

Focal demyelinating lesions typically occur within a 1-cm segment of a nerve. In electrodiagnostic studies, measurements over longer distances decrease the chance of detecting such lesions, but measurements over shorter distances result in greater experimental error. Our objective was therefore to determine the optimal screening distance for ulnar neuropathy at the elbow (UNE) incorporating previously derived experimental errors for calculating nerve conduction velocity (NCV). Using a lesion model wherein prolongation of 0.4 ms was added to the expected latency of a 1-cm nerve segment, new NCVs were derived for distances between 1 and 10 cm for nerves normally conducting between 40 and 65 m/s. Lesion detection, or sensitivity, was defined as the likelihood of calculating a decrease of 10 m/s from the normal NCV while including the experimental error. Specificity was related to the likelihood of an inadvertent calculation of such a decrease in NCV in a segment without a lesion. Sensitivity and specificity were derived at multiple distances with varying NCVs. The total percentage error was the sum of the false-negative and false-positive percentages. The least total percentage error occurred at 3-4 cm, 4-6 cm, and 6-8 cm for nerves normally conducting at 40-50 m/s, 50-60 m/s, and 60-65 m/s, respectively. We conclude that the optimal distance for screening UNE, considering both sensitivity and specificity, is significantly less than 10 cm, perhaps as low as 4-6 cm; considering in addition the likely locations of focal lesions, the best distance is 6-8 cm.