Abstracts

Rationale: Hypothalamic hamartoma (HH) is a rare cause of epilepsy associated with medically refractory gelastic seizures, often associated with a progressive encephalopathy. While surgical intervention has led to seizure freedom in many patients, there are no broadly effective non-invasive treatments. We have performed 180 surgeries on 150 HH patients and have maintained a research database. We asked if those patients who had previously attempted the ketogenic diet (KD) noted clinical improvement. Furthermore, using fresh surgically-resected specimens, we asked whether ketone bodies might affect the spontaneous rhythmic firing of small HH neurons that have been implicated in a conceptual model of seizure genesis. Methods: Patients who had previously attempted the KD were identified by querying our proprietary HH database. Patients were contacted by phone or e-mail and administered a standard questionnaire documenting KD treatment course, efficacy, and adverse effects. Electrophysiological methods were performed on HH slices (300 ?m) from 4 patients in a recording chamber infused with physiological saline (34 C). Gramicidin (20 ?g/ml) perforated-patch microelectrode recordings were performed in small HH neurons under IR-DIC illumination. Results: Six patients (2.7%) were identified for inclusion (3 males, mean age 12 years). Table 1 summarizes the results. Of the four patients who responded to the KD (i.e., at least 50% reduction in total seizure frequency), two had a strong reduction in multiple seizure types (including complex partial, simple partial and atonic seizures). Two patients reportedly improved in school and/or behavioral functioning. Only minor side-effects (weight loss and hypercholesterolemia) were experienced. In electrophysiological recordings, a ketone cocktail consisting of 1mM ?-hydroxybutyric acid (BHB) and 1 mM acetoacetate (ACA) reduced the spontaneous firing in 5 out of 7 small HH neurons (see Figure 1). Slight suppression of rhythmic activity was seen in 3 out of 7 cells after application with the ketone cocktail, and full blockade of spontaneous firing was seen in 2 out of 7 cells. However, ketones did not affect the discharge pattern in 2 out of 7 cells. Conclusions: The efficacy of KD for epilepsy associated with HH is previously unreported. Our study demonstrates that while only a small percentage of patients have attempted the diet, two-thirds of subjects reported a reduction in seizures - with one patient having such a favorable response that they declined surgical intervention with 14 years of follow-up. Our preliminary cellular electrophysiological observations suggest that ketones directly modulate the intrinsic firing of small HH neurons in a manner that fits with our conceptual model of seizure genesis. The present study-despite the small number of patients-may support the use of the KD as a therapeutic option in those patients who do not have access to surgical intervention, who are particularly hesitant to pursue surgery, or who are incompletely controlled following surgery.