Abstracts

Rationale: Independent component analysis (ICA) of resting state functional magnetic resonance imaging (rs-fMRI) can be used to identify normal and abnormal functional networks in patients with epilepsy that may impact cognition, seizure outcomes, and sensorimotor and language function. However, many pediatric patients may require sedation for this procedure. The effects of sedation on these networks in the form of functional connectivity is not well understood.  Methods: A review of all patients with refractory epilepsy who underwent epilepsy surgery between 2008 and 2017 at Nicklaus Children’s Hospital (formerly Miami Children’s Hospital) underwent fMRI with identification of resting state canonical and noncanonical components was performed. Demographics, seizure history, pre-operative neuropsychological data including verbal IQ, performance IQ, and full-scale IQ, etiology, and seizure outcomes were obtained for all patients.  Results: A total of 127 patients were reviewed, 43 of whom underwent a temporal lobe resection, 29 who underwent a frontal lobe resection, 10 with a parietal resection, 7 with an occipital resection, 34 of whom underwent resection of more than one lobe, and 4 who underwent other surgeries (callosotomy, hemispherectomy, hypothalamic hamartoma ablation). Of these, 97 (76%) patients underwent fMRI without sedation. When sedation was used, it either consisted of dexmedetomidine or Propofol, depending on the assessment of the anesthesiologist. There was a statistically significant difference between the mean number of non-canonical components in non-sedated patients versus sedated patients (2.6 versus 4.5, p=0.002). However, the difference between the mean number of canonical components or total components between the sedated and non-sedated groups was not statistically significant (p=0.4, p=0.2, respectively). There was no strong correlation found between the number of non-canonical components or the ratio of non-canonical components to total components and seizure outcome, region of epilepsy, or etiology of epilepsy.  Conclusions: This is one of the largest cohorts of pediatric epilepsy patients with resting state fMRI data described to date. As the use of rs-fMRI continues to grow and expand in its clinical application in the epilepsy domain, it will be important to interpret results of functional connectivity with an understanding of the confounding effects that sedation may have on the resting state networks, particularly the abnormal ones. Further studies with a larger cohort may better elucidate the relationship between the number and type of normal and abnormal networks and cognition and seizure outcomes.  Funding: No funding