Abstracts

Rationale: Infantile Spasms (IS) is a devastating epileptic encephalopathy of infancy which is characterized by epileptic spasms, hypsarrhythmia, and profound neurodevelopmental deficits. Current first-line treatments are effective in only 50% of patients and can carry toxic side effects. The ketogenic diet (KD)—a high fat, low carbohydrate diet—has emerged as an effective alternative treatment for IS refractory to current treatments. Here we investigated the effectiveness of KDs in a neonatal rodent model of refractory and explored its mechanism of action.

Methods: Using the triple-hit paradigm, spasms were induced in rat pups on postnatal (P) day 4 via intracerebral injections of doxorubicin and lipopolysaccharide, and a subcutaneous injection of p-chlorophenylalanine at P5. The pups were then artificially reared using the “pup-in-cup” setup and put on either the KD (4:1 fats:carbohydrate+protein) or a control-milk diet (CM; 1.7:1). 31-phosphorus magnetic resonance spectroscopy (31P MRS) and head-out plethysmography were examined in conjunction with continuous video-EEG recordings to assess brain pH, respiration, and spasm frequency. Mitochondrial respirometry was conducted on liver and brain cortical tissue to examine metabolic changes induced by the KD. Ultrasonic vocalization (USV) and tracked open field activity (OFT) were conducted to examine the pups’ neurodevelopmental profile. This study had four experimental groups: IS pups on KD (KDL) or CM (CML) and control pups on KD (KDC) and CM (CMC).

Results: The KD resulted in a significant reduction in spasm frequency (P7: CML=6.68±0.85 vs KDL=3.55±0.20; P< 0.001), a decrease in the interictal spiking frequency (P12: CML=262.24±18.80 vs KDL=61.28±13.72; P< 0.001), and a reduction in lesion size (P12: CML=0.74±0.02 vs KDL=0.84±0.01; P< 0.001). The KD also improved the developmental profile in pups with spasms (USV, P7: CML=430.57>±20.44 vs KDL=541.67±17.15; P< 0.001. OFT, P11: CML=518.59±53.34 vs. KDL=1006.23±78.05 KDL; P< 0.001) and reduced mortality (CML=57% vs KDL=82%; P< 0.001). A lower brain pH was observed in pups with spasms fed the KD as measured with 31P MRS indicative of intracerebral acidosis (P7: CML=7.21±0.02 vs KDL=6.94±0.05; P< 0.001). Plethysmography also revealed KD treated pups with spasms took slower, deeper, longer breaths. This resulted in a decreased rate of expired CO2 (P7: CML=0.025±0.002 vs KDL=0.016±0.001; P< 0.001). Administration of the pH buffer, sodium bicarbonate, attenuated the spasm suppressing effect of the KD (P7: KDL=3.49±0.19 vs KDL-bicarb 5.41±0.52; P< 0.001). No differences were observed in the mitochondrial respiratory profiles between treatment groups. There were no sex-specific differences observed within the data.