An Initial Experience of Completion Hemispherotomy via Magnetic Resonance-Guided Laser Interstitial Therapy.

INTRODUCTION: In carefully selected patients with medically refractory epilepsy, disconnective hemispherotomy can result in significant seizure freedom; however, incomplete disconnection can result in ongoing seizures and poses a significant challenge. Completion hemispherotomy provides an opportunity to finish the disconnection. We describe the use of magnetic resonance-guided laser interstitial thermal ablation (MRgLITT) for completion hemispherotomy. METHODS: Patients treated with completion hemispherotomy using MRgLITT at our institution were identified. Procedural and seizure outcomes were evaluated retrospectively. RESULTS: Five patients (3 males) underwent six MRgLITT procedures (one child treated twice) for completion hemispherotomy at a median age of 6 years (range 1.8-12.9). Two children had hemimegalencephaly, two had Rasmussen encephalitis, and one had polymicrogyria. All five children had persistent seizures likely secondary to incomplete disconnection after their functional hemispherotomy. The mean time from open hemispherotomy to MRgLITT was 569.5 ± 272.4 days (median 424, range 342-1,095). One patient underwent stereoelectroencephalography before MRgLITT. The mean number of ablation targets was 2.3 ± 0.47 (median 2, range 2-3). The mean length of the procedure was 373 min ± 68.9 (median 374, range 246-475). Four of the five patients were afforded improvement in their neurocognitive functioning and speech performance after ablation, with mean daily seizure frequency at 1 year of 1.03 ± 1.98 (median 0, range 0-5). Two patients achieved Engel Class I outcomes at 1 year after ablation, one was Engel Class III, and two were Engel Class IV. The mean follow-up time was 646.8 ± 179.5 days (median 634, range 384-918). No MRgLITT-related complications occurred. Delayed retreatment (>1 year) occurred in three patients: one child underwent redo ablation and two underwent anatomic hemispherectomy. CONCLUSION: We have demonstrated the feasibility of a minimally invasive approach for completion hemispherotomy using MRgLITT. Delayed retreatment was needed in three patients; thus, further study of this technique with comparison to other surgical techniques is warranted.

Acute Changes in Liver Function Tests During Initiation of Ketogenic Diet.

BACKGROUND: Ketogenic diet is an effective therapy for patients with medically refractory epilepsy. It is generally well tolerated, with the most common side effects being gastrointestinal. Hepatic toxicity has been described as an uncommon side effect of ketogenic diet, usually with long-term use. However, there are limited data to implicate ketogenic diet in acute liver toxicity. METHODS AND RESULTS: We analyzed all patients who underwent elective inpatient ketogenic diet initiation at our institution from June 2019 to June 2022. Of the 25 patients reviewed, we found 6 patients who showed acute, asymptomatic changes in liver function tests during initiation, in both hepatocellular and cholestatic patterns. Two patients stopped the ketogenic diet acutely and 3 patients continued ketogenic diet with changes in medications and/or addition of choline-all patients had improvement and normalization of liver function tests in the short term. One patient had acute normalization of chronically elevated liver function tests on ketogenic diet initiation. CONCLUSION: Ketogenic diet can cause acute changes in liver function tests during initiation of ketogenic diet, with both hepatocellular and cholestatic patterns, with and without the concurrent use of hepatotoxic medications. In most patients, ketogenic diet can be continued successfully by making changes to medications or addition of choline.

Inaccuracies in Parental Reporting of Treated Epileptic Spasms: Both Under- and Over-Reporting.

BACKGROUND: The purpose of this study was to evaluate the accuracy of parental reporting of epileptic spasms (ES) after 14 days of appropriate medical therapy for new-onset ES by comparison with extended video electroencephalography (vEEG) monitoring results. METHODS: Fifty-eight patients were identified from August 2019 to February 2021 with new-onset ES, confirmed on vEEG. Patients were initiated on appropriate treatment (high-dose steroids or vigabatrin). After two weeks of therapy, patients underwent overnight (18 to 24 hours) vEEG monitoring in the epilepsy monitoring unit. Parental reporting of presence or absence of ES on admission was compared with results of vEEG monitoring. RESULTS: The 58 patients ranged in age from three to 20 months (average 7.8 months). An underlying etiology was identified in 78%, whereas 22% patients had unknown etiology. The overall accuracy of parental reporting was 74% (43 of 58) when compared with results of vEEG within 14 to 18 days of starting therapy. Of these, 65% (28 of 43) reported ES resolution and 35% (15 of 43) reported continued ES. Of the 26% (15 of 58) families who were incorrect at two-week follow-up, 67% (10 of 15) reported resolution of ES. However, a minority of families, 33% (five of 15), who continued to report spasms clinically, were inaccurate. CONCLUSIONS: Although a majority of inaccurate parental reports at two weeks of treatment were due to unrecognized ES (a widely known phenomenon), a minority were conversely inaccurate due to persistent over-reporting of ES. This fact highlights the importance of correlating parental history with objective vEEG monitoring, to prevent inappropriate escalation of medication therapy.

Extent of EEG monitoring for detecting epileptic spasms.

INTRODUCTION: Epileptic spasms (ES) are the ictal manifestation of West syndrome. Due to poor interrater reliability in diagnosing hypsarrhythmia, identification of ES - ideally on video EEG monitoring - is important to start proper treatment. METHODS: We retrospectively analyzed inpatient video EEG recordings of 42 patients with new onset ES and calculated the time needed to capture the first ES, along with time to first sleep epoch, among other data. RESULTS: The average time to capture ES was 188.36 min with a 95 percent confidence interval from 116.95 to 259.76 min. The average onset of sleep occurred at 30.8 min with 95% Confidence interval between 19.9 and 41.9 min. There was no significant correlation between duration of symptoms and etiology and the time needed to detect first spasm on EEG. CONCLUSIONS: Routine or short duration EEGs are not sufficient to capture ES in most patients. 4-5-hour VEEG monitoring duration should be considered to capture ictal data in patients suspected of having ES.

Hyperperfusion on magnetic resonance imaging in acute chemotherapy-related leukoencephalopathy.

Acute chemotherapy-related leukoencephalopathy can present similar to acute stroke with symptoms including aphasia, dysarthria, and hemiplegia. Differentiation based on clinical appearance is challenging, and physicians must distinguish between the 2 conditions rapidly to institute appropriate therapies. An 8-year-old male with acute lymphoblastic leukemia receiving chemotherapy, including intrathecal methotrexate, presented to our emergency center with 2 hours of expressive aphasia and flaccid right hemiplegia. Emergent magnetic resonance imaging (MRI) was obtained, demonstrating diffusion restriction within bilateral corona radiata and centrum semiovale. Magnetic resonance perfusion revealed mildly increased perfusion, a finding inconsistent with ischemic stroke and previously unreported in acute chemotherapy-related leukoencephalopathy without necrosis. This increased perfusion conclusively eliminated stroke from the clinical differential. Magnetic resonance perfusion imaging proved valuable to rapidly distinguish acute chemotherapy-related leukoencephalopathy from ischemia, and the evaluation of perfusion alterations in this disorder may provide further insight into the pathophysiology of this entity.