Complications of subdural and depth electrodes in 269 patients undergoing 317 procedures for invasive monitoring in epilepsy.

Abstract

Intracranial monitoring is fundamental to epilepsy surgery, with reported complication rates of 3-17%. We aimed to assess the differences in complication rates between subdural and depth electrodes. We conducted a retrospective review of 317 electrode implantation procedures. All documented abnormal postoperative findings were recorded in our study. Those that resulted in a significant alteration of treatment course, including neurologic deficit, long-term medication use, reoperation, or hospital readmission, were deemed clinically significant. When possible, findings were attributed to a particular electrode type based on relative location to each electrode. Postoperative abnormalities were associated with SDE placement in 152 (47.9%) procedures and 40 (25.2%) DE placements (p < 0.001). Twenty-nine (9.1%) clinically significant complications were seen in the subdural electrode (SDE) group compared to 10 associated with DEs (6.3%, p = 0.37). SDEs were associated with increased rates of any postoperative hemorrhage (p < 0.001) or extraaxial collection (p = 0.007). Subdural grid placement was associated with an increased risk of any extraaxial collection (odds ratio [OR 2.42), as well as clinically significant collections (OR 9.47). Previous craniotomy was found to be associated with any abnormal postoperative finding (OR 1.71) as well as radiographic hemorrhage (OR 1.99). Concurrent resection is also associated with abnormal findings (OR 1.83) and extraaxial collections (OR 2.37). The overall complication rate was 9.1%, with 13 procedures (4.1%) resulting in neurologic deficit. However, only two patients (0.6%) had permanent neurologic sequelae resulting from lead placement. Subdural electrodes appear to have an increased rate of abnormal postoperative findings, including hemorrhage and extraaxial collections; however, there was no difference in clinically significant findings. Subdural grids also appear to be associated with symptomatic extraaxial collections, and previous craniotomy increases the risk of hemorrhage. Overall, intracranial monitoring remains a safe and effective procedure for localization of operative seizure foci. Patient selection and risk education for various modalities is an essential aspect of preoperative evaluation.