Evoked responses in the diagnosis of thoracic outlet syndrome.

The lack of an objective method for diagnosing thoracic outlet syndrome and monitoring the results of treatment has prevented a clear understanding of its natural history and the indications for surgery. Although one can detect vascular compression, the majority of patients have symptoms that are primarily neurogenic; and neither electromyograms nor studies of nerve conduction velocity has proved consistently helpful in their evaluation. We have studied evoked responses in 22 patients and believe that this technique allows selection of patients with significant neural compression for operative treatment. In this procedure, a bipolar stimulating electrode is placed on the median and ulnar nerves at the wrist; and bipolar recording is performed from active electrodes placed at Erb's point, over the spines of C6 and C2, and over the contralateral parietal scalp, all referenced to a midfrontal electrode. A minicomputer temporally summates the low-amplitude potentials elicited by repetitive electrical stimulation to increase the signal to noise ratio and therefore distinguish the potentials of interest from the background bioelectric activity. Normative values have been determined, allowing detection of changes in conduction of the electrical activity that occur between the sites of stimulation and the sites of recording. Of 19 new patients, 13 had abnormal evoked responses. Nine of these underwent surgical treatment; eight obtained good or excellent results. Abnormalities in evoked responses disappeared after operation in seven of the eight in whom it was measured. Four patients with abnormal responses but no surgery are being evaluated. Two of three long-term patients with recurrent symptoms had abnormal evoked responses that changed after treatment. Longer follow-up is needed to determine if this method is helpful in selecting patients for operative treatment or in monitoring its results.