Accuracy of diagnoses delivered by an automated hand-held nerve conduction device in comparison to standard electrophysiological testing in patients with unilateral leg symptoms.

INTRODUCTION: Automated hand-held nerve conduction study (NCS) devices are being marketed for use in the diagnosis of lumbosacral radiculopathy (LSR). In this study we compared the specificity and sensitivity of a hand-held NCS device for the detection of LSR with standard electrodiagnostic study (EDX). METHODS: Fifty patients referred to a tertiary referral electromyography (EMG) laboratory for testing of predominantly unilateral leg symptoms (weakness, sensory complaints, and/or pain) were included in the investigation. Twenty-five normal "control" subjects were later recruited to calculate the specificity of the automated protocol. All patients underwent standard EDX and automated testing. RESULTS: Raw NCS data were comparable for both techniques; however, computer-generated interpretations delivered by the automated device showed high sensitivity with low specificity (i.e., many false positives) in both symptomatic patients and normal controls. CONCLUSIONS: The automated device accurately recorded raw data, but the interpretations provided were overly sensitive and lacked the specificity necessary for a screening or diagnostic examination.