Introduction: The study purpose is to determine the impact of etiology on ketogenic diet (KD) response in children with intractable epilepsy.

Methods: Children initiated on KD January 2009 through December 2012 were identified from the KD diet database at Mayo Clinic Rochester.  Charts of those with known etiology (genetic, structural, or both) were reviewed to determine seizure frequency (a) prior to KD onset, (b) 3 months after KD onset and (c) at final follow-up on KD.  Significant response was defined as >50% decrease in seizures.

Results: 49 children were identified (62% male).  Etiologies included genetic (33%), structural (49%), both (18%).  Median age of KD initiation was 30 months (range 3-194 months) and treatment duration was 21.5 months (range 3-120 months).   Structural or genetic etiologies responded equally, 46% and 44% significant response, respectively.  Only 22% of children with both structural and genetic etiology had significant response.   However, 2 of 4 children with tuberous sclerosis complex (TSC) had a significant response, whereas none of those with structural and genetic etiology not due to TSC (n=5) had significant response at final follow-up.  The non-TSC structural and genetic etiology patients were less likely to respond when compared to all others (p=0.05).

Conclusions: From etiology alone, it is difficult to predict which children will respond to KD.  Therefore, it is recommended that most children with intractable epilepsy be given a trial of KD.  However, those with both genetic and structural etiology for epilepsy, especially those without TSC, may be less likely to respond to KD.