[Effectiveness and tolerability of ketogenic diet therapy in children with genetic developmental and epileptic encephalopathy onset within 2 years of age].

Objective: To investigate the effectiveness and tolerability of ketogenic diet therapy (KDT) in children with developmental and epileptic encephalopathy (DEE) caused by genetic variants onset within 2 years of age. Methods: A retrospective cohort study was conducted. Data were reviewed from 117 children with DEE who started KDT in the Department of Neurology of Beijing Children's Hospital Affiliated to Capital Medical University, Jinan Children's Hospital, Henan Children's Hospital and Urumqi Children's Hospital from January 2016 to June 2023. Data included demographic information, genetic testing results, KDT regimens, and treatment outcomes. Based on the effectiveness of KDT, the children were classified into 4 grades, Grade 1: seizure-free for at least 6 months; Grade 2: 75%-<100% seizure reduction; Grade 3: 50%-<75% seizure reduction; and Grade 4:<50% seizure reduction. The seizure-free group was defined as Grade 1, and the uncontrolled group as Grade 2+3+4; the significantly effective group was defined as Grade 1+2, and the non-responsive group as Grade 3+4. Inter-group comparisons were performed using the χ2 test, the independent sample t-test, or the Mann-Whitney U test. Results: A total of 90 children were enrolled, including 44 males and 46 females, with the age at seizure onset of 2.0 (0.7, 5.3) months. Thirty genes were involved, with ion channels being the most common neurobiological mechanism (49 children (54%)), followed by synaptic function (7 children (8%)). The age at KDT initiation was 11.0 (5.8, 21.8) months, and the time interval between seizure onset and KDT initiation was 8.0 (4.0, 18.3) months. Following KDT, 31 children (34%) achieved Grade 1 (seizure freedom), 19 children (21%) achieved Grade 2, 13 children (14%) achieved Grade 3, and 27 children (30%) achieved Grade 4. There were 24, 9, 9, 6, 5, and 5 children with SCN1A, CDKL5, SCN2A, KCNQ2, STXBP1 and KCNT1 genes variants, respectively; 7, 4, 2, 2, 4, and 1 of these children, respectively, achieved seizure freedom. Among the seizure-free and uncontrolled groups (31 and 59 children), the maintenance lipid-to-nonlipid ratio was (2.2±0.7) and (2.5±0.7):1, 19 children (61%) and 22 children (37%) had non-ion channel gene variants, respectively, both represented statistically difference (both P<0.05). Among the significantly effective and non-responsive groups (50 and 40 children), there were 31 females (62%) and 15 females (38%), the interval between seizure onset and KDT initiation was 6.0 (3.8, 12.3) and 10.0 (5.3, 27.5) months, both with statistically difference (both P<0.05). Adverse events occurred in 59 children (66%), with gastrointestinal symptoms being the most common (53 children). The main reasons for discontinuing KDT in 73 children were dissatisfaction with efficacy (29 children (40%)) and adverse reactions (14 children (19%)); among them, 2 were due to intolerance resulting from a rapid increase in the lipid-to-nonlipid ratio. Conclusions: KDT is effective in most children with genetic DEE onset within 2 years of age. Earlier initiation yields better results, and seizure freedom is not associated with higher lipid-to-nonlipid ratios. Children with variants in genes encoding non-ion channel proteins, such as STXBP1, respond well to KDT. Most children tolerate KDT well, and a cautious, standardized ketogenic regimen helps to prevent discontinuation due to adverse reactions.