OBJECTIVE:Studies of the effects on lower-limb cortical somatosensory evoked potentials (CSEP) during total intravenous anesthesia are sparse for propofol and are lacking for midazolam. This study was designed to compare the effects of propofol and midazolam on CSEP under total intravenous anesthesia during intraoperative monitoring for surgical treatment of scoliosis.METHODS:CSEPs were recorded in two groups of 15 patients during posterior instrumentation for treatment of idiopathic scoliosis. The anesthesia used the combination of atracurium, alfentanil, and an hypnotic agent (propofol for Group I or midazolam for Group II). The main characteristics of the CSEPs (P40 latency and N34–P40 and P40–N50 amplitudes) were recorded using ankle posterior tibial nerve stimulation. The CSEPs were recorded before induction, 10, 70, 100, 130, and 160 minutes after induction, and before the wake-up test. The statistical analysis involved analysis of variance for repeated measures. Both groups were homogeneous before induction.RESULTS:Neither CSEP deterioration during risk-associated surgical procedures nor postoperative clinical abnormalities were observed. Both propofol and midazolam induced increases in P40 latencies, with the increases being greater and more regular for the propofol-treated group. The amplitude values changed with time for both groups, decreasing mainly after induction; in the midazolam-treated group, the amplitudes were smaller but more stable. Propofol modified the morphological characteristics of the response by decreasing the late P60 component amplitude; the W-shaped CSEP morphological pattern was maintained with midazolam.CONCLUSION:This study demonstrates the appropriate use of either propofol or midazolam in scoliosis monitoring. Preoperative small-amplitude CSEPs might favor the use of propofol anesthesia.