Authors :
Presenting Author: Alyssa Jobe, BS – Medical College of Wisconsin
Andrew Crow, BA – The Medical College of Wisconsin, Milwaukee
Pradeep Javarayee, MD MBA – The Medical College of Wisconsin, Milwaukee
Irene Kim, MD – Medical College of Wisconsin
Sean Lew, MD – Children's Wisconsin, Medical College of Wisconsin
Rationale:
Stereo-electroencephalography (SEEG) is used to localize epileptic foci in refractory epilepsy. While post-implantation CT is standard for detecting complications, its limitations may miss subtle hemorrhages. MRI offers greater sensitivity and anatomical detail but is less commonly used post-SEEG. This study evaluates whether 24-hour post-operative MRI improves detection of intracranial hemorrhage compared to CT in both adult and pediatric patients.
Methods: This retrospective study reviewed records of patients who underwent SEEG implantation at Froedtert Hospital and Children’s Wisconsin between January 2015 and September 2023. Data collected included clinical history, immediate post-operative CT and MRI findings (performed within 24 hours), SEEG details, and neurophysiological results. Statistical analyses included descriptive statistics, Wilcoxon rank sum and Fisher’s exact tests for group comparisons, and unweighted Cohen’s kappa to assess CT-MRI agreement.
Results: Descriptive statistics and clinical characteristics of the 111 patients (71 pediatric and 40 adult patients) with both MRI and CT scans are summarized in Table 1. Adults had a longer history of epilepsy and had been trialed on a higher number of antiseizure medications compared to pediatric patients. The groups also differed in epilepsy type, etiology, and insurance coverage, but were similar in gender, race, and ethnicity (Table 1). The mean number of sEEG electrodes and electrode contacts did not differ significantly between pediatric (13.79 electrodes, 125.96 contacts) and adult patients (14.38 electrodes, 138.05 contacts) (Table 1). With MRI considered as the gold standard, the agreement between CT and MRI in detecting intracranial hemorrhage and cerebral edema varied. For intracranial hemorrhage, CT demonstrated substantial agreement with MRI, correctly identifying 9 true positives and 96 true negatives. However, CT missed 3 minor hemorrhages < 5 mm in diameter detected by MRI and incorrectly identified 3 hemorrhages that were not present on MRI. In contrast, CT showed poor concordance with MRI in detecting cerebral edema. MRI identified six cases of cerebral edema that CT failed to detect, and CT identified two false positives, with no true positives identified (Table 2).
Conclusions:
This study highlights the added value of incorporating 24-hour post-operative MRI alongside routine CT imaging following sEEG implantation. CT failed to identify a few cases of hemorrhage and to reliably identify cerebral edema. These findings underscore the superior sensitivity of MRI for certain post-operative complications and support its use as a complementary modality to improve diagnostic accuracy in both pediatric and adult sEEG patients. Additionally, the findings support the safety of MRI with AdTech SEEG electrodes which received FDA approval mid-2023, building upon evidence from years of clinical practice.
Funding: There was no funding for this study.