Abstract
ObjectiveWe aimed to find predictors for smartphone application-based ketogenic diet (KD) treatment effectiveness and safety.MethodsThe efficacy was evaluated according to the reduction in seizure frequency after the intervention of KD; safety was evaluated based on adverse effects. The ordinal logistic regression analysis was used to explore the influencing factors of efficacy.ResultsThe study sample included 116 males and 65 females with a median age of 2.27 years. The baseline frequency of seizure was more than five times/day in 123 children, 50.83% of them received three or more antiepileptic drugs (AEDs). Seventy-two patients' KD initiation mode was outpatient, and 73 completed the 12-month follow-up. A total of 88 (48.62%) patients had reported a reduction in seizure >= 50%. Compared with 12 months, those who had received KD therapy for only 3 (P = 0.009) and 6 months (P = 0.005) were more likely to show negative outcomes. Outpatient initiation had better outcomes (P = 0.029) than inpatient initiation. For the number of AEDs applied, patients on two AEDs were more likely to achieve better outcomes (P = 0.001). Adverse events had been noted among 77 patients; BMI Z-score at KD initiation was associated with adverse effects (P = 0.003).SignificanceOur study suggested that outpatient initiation and long-term treatment of KD should be encouraged.Plain Language SummaryOur research shows that the KD is a helpful treatment for children with refractory epilepsy, reducing seizures by more than 50% in nearly half of the cases, with some experiencing complete seizure freedom. We used a smartphone app to improve communication between patients and their healthcare teams, resulting in a high retention, and app usage was linked to reduced adverse effects. We recommend early consideration of KD treatment for patients failing two AED, encourage outpatient initiation, and advocate for longer-term KD use.