350 Surg Neurol 1992;37:350-5 Effect of Graded Hypoxia on Cortical and Spinal Somatosensory Evoked Potentials Siavash S. Haghighi, D.V.M., Ph.D., John J. Oro, M.D., Scott R. Gibbs, M.D., and Michael McFadden, B.S. Division of Neurosurgery, University of Missouri Hospital and Clinics, Columbia, Missouri Haghighi SS, Oro JJ, Gibbs SR, McFadden M. Effect of graded hypoxia on cortical and spinal somatosensory evoked potentials. Surg Neurol 1992;37:350-5. Cortical somatosensory evoked potential (CSEP), spinal somatosensory evoked potential (SSEP), and electroen- cephalogram were recorded in rats under pentobarbital anesthesia. After baseline recordings in room air (21% 02), animals were subjected to a graded hypoxia at 15.75%, 10.5%, and 5,25% oxygen levels for 10 minutes. Each level of hypoxia was followed by a 15-minute reoxy- genation period. With a moderate hypoxia (15.75% O2), measured latencies for the CSEP and the SSEP were not significantly different compared with baseline (p > 0.05). The CSEP amplitude showed a significant increase (p -- 0.02) during reoxygenation after the moderate hypoxia. Change in the latency or amplitude of SSEP at 15.75% hypoxia or during the reoxygenation period was not sig- nificant compared with the room air (p > 0.05). No change in the electroencephalogram was noticed with the moder- ate hypoxia. At severe hypoxia (10.5% 02), 80% of the animals lost CSEP within 2 minutes. The loss of CSEP was concomitant with significant attenuation of the electroen- cephalogram waves. The SSEP was resistant to the severe hypoxia and was present in all animals. We concluded that hypoxia affects CSEP with the tendency to increase the amplitude at moderate hypoxia (15.75%) and loss of the latency and amplitude with severe (10.5%) and extreme (5.25%) hypoxia. KEY WORDS: Cortical somatosensory evoked potential; Hypoxia; Spinal somatosensory evoked potential Somatosensory evoked potentials (SEPs) assess the func- tional state of the nervous system from peripheral nerve to cortex [2]. Clinical utility of SEP monitoring empha- sizes lesions of the nervous system, such as the brachial plexus [12], lumbosacral plexus [6,16], spinal cord [ 1,3,24], brain stem [ 13,14], and cerebral hemispheres Address reprint requests to: Siavash S. Haghighi, D.V.M., Ph.D., Division of Neurosurgery, University of Missouri Hospital and Clinics, Columbia, Missouri 65212. Received April 2, 1991; accepted October 24, 1991. [22]. The cortical somatosensory evoked potential (CSEP) and spinal somatosensory evoked potential (SSEP) in rats consist of a sequence of negative-posi- tive-negative waves that can be recorded in the first 70 ms following the onset of electrical stimulation to the posterior tibial nerve [23]. Previous studies have not systematically explored the relationship between oxygen delivery and changes in brain electrical activities [14]. Confirmation of the relationship of oxygen delivery and brain electrical activity yields an assessment of ade- quacy of oxygen availability when other factors, such as cerebral blood flow, cannot be measured. The aim of this project was to analyze the effects of moderate (15.75 % O2), severe (10.5 % Oz), and extreme hypoxic hypoxia (5.25% 02) on the CSEP, SSEP, and electroencephalogram (EEG) in rats. The EEG was cho- sen as an additional neurophysiological measure because it can be continuously monitored and quantified using power spectral frequency analysis [ 15 ]. Materials and Methods Animal Preparation Ten adult Sprague-Dawley rats of either sex, weighing 300-350 g, were utilized. Under pentobarbital anesthe- sia (Nembutal), each rat was intubated endotracheally and artificially ventilated with room air. The tidal volume was adjusted to approach normal ventilation (model 7088-600, Phillips and Bird, Richmond, VA). After tra- cheal cannulation of the animal, the head was rigidly fixed in a stereotactic head holder (model 1404, David Kopf Instruments, Tujunga, CA). Heart rate was moni- tored with an EKG machine (model 78304A, Hewlett- Packard Co., Palo Alto, CA). The left femoral artery was cannulated to monitor mean arterial blood pressure and sample a small amount of blood for analyzing pH, HCO;, PaO2, and PaCO2. The femoral venous line was established for the administration of fluids. The body temperature of the animals was monitored closely and maintained between 36.5 ° and 37.5°C using a heating blanket (Aquamatic K-model, Gorman Rupp Industries, Belteville, OH). © 1992 by Elsevier Science Publishing Co., Inc. 0090-3019/92/$5.00