Value of dermatomal somatosensory evoked potentials in detecting acute nerve root injury: an experimental study with special emphasis on stimulus intensity.

STUDY DESIGN: Dermatomal somatosensory-evoked potentials (D-SSEPs) in rats were recorded at the spinal level after L2-, L4-, and L5-dermatome stimulation. Pre- and post-transection patterns and rates of change of corresponding nerve roots were compared to determine accuracy. OBJECTIVE: To investigate characteristics and normal values of D-SSEP elicited from lower limb dermatomes; to determine specificity, sensitivity, and utility of D-SSEP in detecting single-nerve root injury; and to determine optimal stimulation intensity. SUMMARY OF BACKGROUND DATA: D-SSEP allows assessment of single nerve root-specific pathways, electrodiagnosis of lumbosacral radiculopathy, and intraoperative neuromonitoring. Unacceptably low sensitivity and specificity make its value suspect. D-SSEP is insufficiently documented. METHODS: Eight rats were used to specify a standard D-SSEP waveform and its characteristics, evaluate stimulation sites and strengths, and determine appropriate stimulation and recording techniques. The L4 nerve root was transected in one group of 8 rats and the L5 in another. D-SSEPs were recorded at the thoracolumbar junction following submaximal and supramaximal stimulation at the L2, L4, and L5 dermatomal fields. Potentials recorded before transection, and immediately, 1 hour, and 1 week post-transection were compared. RESULTS: Reproducible spinal responses were obtained in all rats on all tests. Stimulation intensity, but not rates, affected amplitude. Relative amplitude reductions in transected-root D-SSEP were larger using submaximal than supramaximal intensity. D-SSEP elicited by submaximal than supramaximal intensity produced fewer false negatives and false positives. CONCLUSIONS: D-SSEP is valuable for detecting acute single nerve root injury. In clinical settings, submaximal dermatomal stimulation identifies conduction abnormalities more consistently and with fewer false negatives and false positives than does supramaximal stimulation. We recommend submaximal stimulation.