SOMATOSENSORY-EVOKED POTENTIALS RECORDED FROM THE POSTERIOR PHARYNX TO STIMULATION OF THE MEDIAN NERVE AND CAUDA-EQUINA

Somatosensory evoked potentials (ppSEPs) in response to stimulation of the median nerve at the wrist and the cauda equina at the epidural sp ace (the L4 level) were recorded from the posterior wall of the pharyn x in 15 patients who underwent spinal surgery under general anesthesia , using disc electrodes attached to the endotracheal tube, and compare d with segmental spinal cord potentials (seg-SCPs) that were recorded simultaneously from the posterior epidural space (PES). ppSEPs consist ed of the initially positive spike (P9) followed by slow positive (P13 ) and negative (N22) waves. The P13 and N22 of ppSEPs had phase revers al relationship with the P2 and N2 recorded from the PES, respectively . The peak latencies of P9 (9.40 +/- 0.7 ms) (mean +/- SD), P13 (13.1 +/- 0.9 ms), and N22 (22.0 +/- 2.1 ms) of ppSEPs coincided with those of P1, N1 and P2 of seg-SCPs, respectively. ppSEPs were recorded more clearly with a reference electrode on the dorsal surface of the neck t han with the reference electrode at the earlobe or back of the hand. T he threshold and maximal stimulus intensities were also similar betwee n the ppSEPs and seg-SCPs. Thus, the P9, P13, and N22 components of pp SEPs were thought to have the same origin as the P1, N1 and P2 of seg- SCPs, respectively. Therefore, the P9, P13 and N22 of ppSEPs may refle ct incoming volleys through the root, synchronized activities of the i nterneurons and primary afferent depolarizations (PAD), respectively. ppSEPs in response to cauda equina stimulation showed that the latenci es of the two initial components (4.6 +/- 0.4 and 6.4 +/- 0.6 ms) corr esponded to those of the SCPs recorded from the PES (4.6 +/- 0.3 and 6 .3 +/- 0.5 ms), suggesting that these potentials reflect impulses cond ucting through the spinal cord, similar to epidurally recorded SCPs.