Feasibility of Basal Ganglia Microelectrode Recordings Under General Anesthesia with Combined Nitrous Oxide and Sevoflurane: A retrospective Single-Center Experience

Introduction Deep brain stimulation (DBS) is an established treatment for Parkinson’s disease (PD). The traditional method for accurate implantation is awake microelectrode recordings (MER) to map out the borders of the target nucleus. However, a significant portion of patients are unable to tolerate awake surgical procedures. Asleep MER techniques under different general anesthesia regimens have been described with variable effects on recording quality and required a lower inhaled sevoflurane level to obtain single unit recordings. Hence, a reliable method for asleep MER mapping is needed without compromising patient safety and comfort. We aimed to assess the feasibility and quality of basal ganglia MER under general anesthesia using inhalational agents including adding nitrous oxide as an adjunct to sevoflurane(N2O-GA). Methods This study retrospectively examined PD patients undergoing DBS implantation targeting either the subthalamic nucleus (STN) or the globus pallidus internus (GPi) at a single center. Anesthetic data on end-tidal (ET) sevoflurane and nitrous oxide, with the derived minimum alveolar concentration (MAC) were captured during the time of MER mapping. We evaluated the feasibility of identifying target nuclei borders, the quality of neuronal unit isolation, and the physiological dimensions of the targeted nuclei. We calculated the concordance between the nuclei sizes based on MER mapping and imaging. We also reported the firing characteristics of isolated units. Results We identified 18 patients (34 nuclei) who underwent STN (n=11) and GPi (n=7) DBS implantation. Background activity changes were reliable in all patients for border identification. The length of the tract identified by MER was highly concordant with the anatomical tract length identified by postoperative imaging (concordance correlation coefficient: 0.84, p <0.001). Firing in both nuclei showed higher bursting rates. Pallidal cells showed typical firing patterns with “pauser” cells in the GPe and continuous firing in the GPi. No complications were observed during follow-up.16 patients had MER data available for offline analysis. We identified 516 units (single/multi) across MER 28 tracts (STN=284, GP=232). In the 14 patients received the N2O-GA, anesthetic depth was maintained at 0.97 ± 0.06 MAC, compared to 0.525 ± 0.04 MAC in the sevoflurane-only cases. Conclusion MER under N2O-GA is feasible for DBS target nuclei identification for both STN and GPi and offers a safe and accurate surgical approach for PD patients unable to tolerate awake mapping.