The Effect of General Anesthesia on the Microelectrode Recordings From Pallidal Neurons in Patients With Dystonia

The most common anesthetic technique for patients undergoing insertion of deep brain stimulators (DBS) is local anesthesia with or without conscious sedation as this facilitates intraoperative microelectrode recordings (MERs) for target localization. However, general anesthesia (GA) may be needed in some of the patients especially those with dystonia. The purpose of our study was to determine the effects of GA on MERs from pallidal neurons in patients with dystonia undergoing DBS implantation surgery.

After IRB approval, we retrospectively reviewed the medical records of all patients who had insertion of DBS from January 2009 to December 2013. Data collected and analyzed included demographics, indications for DBS, targets of insertion, MER, and anesthetic management. From the records we identified patients with dystonia who received GA for DBS insertion. We then compared the MER data under GA with the data from patients who had surgery under local anesthesia only during the same time period. Because of the small sample size, the effects of various anesthetic regiments on MER and localization of target nuclei were compared qualitatively.

Of the 435 patients who underwent DBS insertion during the study period, 20 (4.3%) patients had GA for the procedure. Dystonia was the most common indication for GA (16/20 patients, 80%). Good-quality MER data obtained from 10 patients with dystonia under GA was compared with 8 patients who had no sedation for the procedure. Administration of GA made target localization difficult due to suppression of both spontaneous and evoked neuronal discharges from internal globus pallidus. Although not studied systematically, propofol (>100 mcg/kg/min) seemed to suppress pallidal discharges more than GA with a lower dose of propofol (<75 mcg/kg/min), remifentanil, and 0.2% to 0.4% end-tidal sevoflurane or desflurane.

Our retrospective review suggests that there was a difference in spontaneous and evoked neuronal discharges with MER performed under GA compared with no sedation. MER recordings during GA appeared most robust during a combination of anesthetics including low-dose propofol infusion, remifentanil, and a low concentration of either sevoflurane or desflurane. Our findings can inform a power analysis to determine the sample size that would be required to prospectively test the hypothesis that there is a difference in spontaneous and evoked neuronal discharges with MER performed under GA compared with no sedation.