The Effect of Intravenous Lidocaine on Intraoperative Somatosensory Evoked Potentials During Scoliosis Surgery

Abstract

Intraoperative somatosensory evoked potentials (SSEPs) are sometimes used to monitor the function of the spinal cord during scoliosis surgery. Various anesthetics can influence SSEPs, particularly the cortical responses, by attenuating amplitude or prolonging the latency [1-5].1 These alterations mimic the changes that can be seen during spinal cord ischemia and enjoin caution in the interpretation of the SSEPs. 1 Kimovec MA, Serpico LC, Sloan TB, et al. Effects of thiopental on cerebral envoked potentials [abstract]. Anesth Analg 1986;65:580. Lidocaine is used during general anesthesia in some specific situations. It is a rapid and easily titratable drug used to control ventricular arrhythmias. It can also prevent an increase in systemic blood pressure as well as increases in intracranial and intraocular pressure encountered during endotracheal intubation or suctioning [7-11]. One study evaluating the effects of intravenous (IV) lidocaine upon the intraoperative cortical SSEPs demonstrated an increase in the peak amplitudes and prolongation of the peak latencies of the cortical waveforms [12]. We report two cases in which IV lidocaine, given during scoliosis surgery, significantly reduced the amplitude and increased the latency of the cortical evoked potentials. Case Reports Two patients were admitted to our institution for surgical correction of their idiopathic scoliosis. Both underwent a posterior spinal fusion with an iliac crest bone graft and TSRH (Texas Scottish Right Hospital) instrumentation, under general anesthesia using fentanyl, midazolam, and nitrous oxide. Intraoperative SSEP monitoring and a wake-up test were performed. We evaluated the effects upon the SSEPs of IV lidocaine used as part of the anesthetic regimen to deepen the level of anesthesia. Heart rate, arterial blood pressure, arterial blood gases, and body temperature remained stable throughout the procedure. Patient 1 was a 12.5-yr-old girl diagnosed with idiopathic scoliosis 2 years earlier. Radiologic examination revealed right thoracic and left lumbar curvatures of 53 degrees and 49 degrees, respectively. Neurologic examination was normal. She underwent a posterior spinal fusion from T6 to L3 under general anesthesia. She was premedicated with IV midazolam (0.06 mg/kg). General anesthesia was induced with fentanyl (10 micro g/kg) and thiopental (5 mg/kg) IV. She also received glycopyrrolate (total dose 0.2 mg) during the induction period. Neuromuscular block was produced with IV vecuronium (0.16 mg/kg). Anesthesia was maintained with fentanyl (1-5 micro g [centered dot] kg-1 [centered dot] h-1), midazolam (total dose 2 mg), and 66% nitrous oxide. Vecuronium was repeated at a rate of 0.12 mg [centered dot] kg-1 [centered dot] h-1. A total of 90 mg IV labetalol was used to control blood pressure to help minimize bleeding. An IV bolus of lidocaine 1 mg/kg was injected 30 min before the wake up test, when the patient began to move prematurely. Patient 2 was a 15-yr-old female with idiopathic scoliosis characterized by a right thoracic curve of 47 degrees from T-4 to T-9 and a left thoracolumbar curve of 48 degrees from T-9 to L-1, as evidenced by her physical and radiological examinations. Neurological examination was normal. She underwent a posterior spinal fusion from T4 to L1 under general anesthesia. She was premedicated with IV midazolam (0.05 mg/kg). General anesthesia was induced with fentanyl (14 micro g/kg), midazolam (0.05 mg/kg), d-tubocurarine (0.5 mg/kg), and IV thiopental (1 mg/kg). She also received glycopyrrolate (total dose 0.2 mg) during the induction period. Lungs were ventilated with 66% nitrous oxide and 33% oxygen. Anesthesia was maintained with midazolam (total dose 5 mg), fentanyl (1.8-3.6 micro g [centered dot] kg-1 [centered dot] h-1), d-tubocurarine (total dose 15 mg), and 66% nitrous oxide. Arterial blood pressure was controlled with IV labetalol (total dose 15 mg). A 1-mg/kg bolus of lidocaine was injected IV 100 min after the induction to deepen the level of anesthesia. In both patients, the surgical procedure and the postoperative course were uncomplicated. SSEPs were performed according to the guidelines of the American Electroencephalographic Society for intraoperative monitoring [13]. After marking electrode sites in accordance with the International 10-20 System (C3 prime, C4 prime Cz prime, Fpz, Fz, M1, M2), the recording sites were prepared with a commercially available pumice-based solution. Gold cup electrodes were affixed using gauze and collodion. Additional gold cup electrodes were affixed to C-2 and C-3 of the cervical spine and the popliteal fossa bilaterally in the same manner. A ground electrode was placed on the left shoulder. Electrode impedances were maintained between 2000 and 5000 Omega. A monophasic stimulus of 100 micro s duration and 10-20 mA was delivered to the ankle with a rate of stimulation of 2 Hz. Posterior tibial nerve SSEP recordings were obtained from the popliteal fossa (PF) referenced to an electrode located 4 cm above the popliteal fossa, the upper cervical cord (CS) referenced to Cz prime and the ipsilateral ear, and the scalp (Cz prime) referenced to Fpz. Waveforms were recorded approximately every 15 min or more frequently during the critical period of instrumentation and distraction. Filter settings of 30-2000 Hz with a gain of 1000 were used. The analysis time was 120 ms, and responses identified after 350-500 repetitions were averaged using a digital averaging system (Nicolet CA-200 [Nicolet Biomedical Inc., Madison, WI] or Cadwell Excel [Cadwell Laboratories, Inc., Kennewick, WA]). Upgoing potentials were labeled "positive." Peak latency and amplitude for all waveforms were measured using the cursor on the averaging system. In Patient 1, 5 min after the lidocaine injection, the cortical SSEP disappeared but the cervical potential was preserved (Figure 1). The cortical waveform returned to its baseline value within 20 min.Figure 1: Intraoperative posterior tibial nerve somatosensory evoked potentials (SSEPs) from Patient 1. Trials 1, 2, and 3, respectively, are the trials obtained before, 5 min after, and 15 min after an intravenous bolus of 1 mg/kg lidocaine. Note the disappearance of the cortical evoked potential (EP, P37/N45) in Trial 2, whereas the cervical SSEP (CS) was preserved. In Trial 3, the cortical SSEP returned to baseline.In the second patient, there was a reduction in the amplitude of the cortical response 10 min after the injection (preinjection/postinjection lidocaine P37/N45 amplitude 1.34/0.6 micro V) and a prolongation of the latency (preinjection/postinjection lidocaine P37 latency 32.8/34.5), whereas the cervical responses remained unchanged (Figure 2). Of note, 10 min before the lidocaine injection, 0.25 mg/kg labetalol was given to control an increase in systemic blood pressure and resulted in an arterial blood pressure decrease from 130/80 to 105/70 mm Hg at the time the lidocaine was administered. However, 20 min after the lidocaine injection the blood pressure had decreased further (90/60 mm Hg), while the cortical SSEP had returned to its baseline value.Figure 2: Intraoperative posterior tibial nerve somatosensory evoked potentials (SSEPs) from Patient 2. Trials 1, 2, and 3 are, respectively, the trials before, 10 min after, and 20 min after an intravenous bolus of 1 mg/kg lidocaine. In Trial 2, the cortical SSEP amplitude (P37/N45) decreased by 50% and its latency (P37) increased by 1.7 ms compared with Trial 1, whereas the cervical SSEP remained unchanged. In Trial 3, the cortical SSEP returned to the preinjection values.Discussion Lidocaine acts on the nervous system, and on other excitable tissues, by preventing the development and the propagation of action potentials. This is probably the result of a direct interaction with specific receptors on voltage-dependent sodium channels in the neuronal membrane [7]. Lidocaine is rapidly distributed to the richly vascularized tissues, such as brain and heart, which accounts for the clinical manifestations of toxicity. The effect of IV lidocaine on the cortical SSEPs is probably a direct one upon the central nervous system (CNS) rather than an alteration of the electrical propagation in the peripheral or subcortical pathways. The preservation of the cervical evoked potentials after the IV injection of lidocaine in our two cases supports the suspected central action of the drug. The initial decrease in arterial blood pressure produced by labetalol, given 10 minutes before the lidocaine in Patient 2, probably did not contribute to the changes in the cortical SSEP. Indeed, 20 minutes after the lidocaine injection, when the blood pressure was even lower, the cortical SSEP had returned to baseline. Lidocaine has either an inhibitory [14,15] or excitatory influence [16] on synaptic transmission in the CNS in a dose-dependent fashion. A tetraphasic response of CNS electrical activity and behavior was observed in cats after an IV infusion of lidocaine at a subconvulsive constant dose (1 mg [centered dot] kg-1 [centered dot] min-1) [17]. An initial inhibitory phase was followed by a phase of excitation, a late inhibitory phase, and finally a stage of convulsions. When injected into the epidural space or when given IV, lidocaine has been shown to alter the cortical evoked response. Three studies evaluated the effect of epidural lidocaine upon the scalp-recorded evoked potentials to posterior tibial nerve [18,19] or median nerve stimulation [20]. Each showed a prolongation of the latency and a decrease in the amplitude of several components of the SSEP. These changes in the cortical responses are similar to what we have observed. Kasaba et al. [21] also evaluated the influence of epidural analgesia with lidocaine upon the scalp SSEP. In another study, they used an IV bolus followed by infusion of lidocaine during median nerve intraoperative SSEP monitoring [12] and, in contrast to the first studies, observed a significant increase in the peak amplitude (N1P2) and latencies of all waves. They suggested that the results found in the first three cases of epidural injections were due to the systemic effect of lidocaine on the CNS after its absorption from the epidural space. The amplitude enhancement of the cortical responses that Kasaba et al. observed might represent the excitation phase of the tetraphasic response to IV lidocaine described in cats [17], whereas our results after a bolus might represent the early inhibition phase. In conclusion, our observations in two patients undergoing scoliosis surgery show that an IV bolus of lidocaine can transiently depress, or even suppress, the intraoperative cortical SSEP, and may therefore mimic spinal cord ischemia. Although these results need to be confirmed in a larger number of patients, they warrant caution in the interpretation of the intraoperative SSEPs when IV lidocaine is used. Whenever possible, another drug with less marked influence on the scalp-recorded SSEP should be substituted for lidocaine.