Elaine Wirrell, Susan Eckert, Lily Wong-Kisiel, Eric Payne and Katherine Nickels. Ketogenic Diet Therapy in Infants: Efficacy and Tolerability. Pediatric Neurology. In press.
The aim of this study was to evaluate tolerability and efficacy of the ketogenic diet in infants less than 12 months of age.
Infants less than 12 months of age, commencing the ketogenic diet between 09/2007 and 07/2016 were identified. Records were reviewed for epilepsy details, diet initiation details, efficacy and tolerability.
27 infants commenced the ketogenic diet (56% male, median age 7 months). Median age at seizure onset was 1.9 months and 92% had daily seizures. An epilepsy syndrome was noted in 19 (West-11, Epilepsy in Infancy with Migrating Focal Seizures-5, Early Myoclonic Encephalopathy-1, Ohtahara-1, Dravet-1). Infants were on a median of 2 and had failed a median of 1 antiepileptic drugs for lack of efficacy. All initiated a traditional ketogenic diet at full calories without fasting, and all but one started the diet in hospital.
Significant hypoglycemia during initiation was seen in two - both had emesis +/- decreased oral intake. 88% developed urinary ketosis by 48 hours and all were successfully discharged on the diet (median ratio 3:1).
Of those continuing the ketogenic diet, responder rates at 1, 6 and 12 months were 68%, 82% and 91%, with 20%, 29% and 27% achieving seizure freedom. By 12 months, 2 stopped the diet for serious adverse effects (1-markedly increased triglycerides, 1-dehydration and severe ketoacidosis), 5 discontinued for lack of efficacy, 6 were lost to follow-up and 2 died of unrelated causes.
The ketogenic diet is an effective and well-tolerated treatment for infants with intractable epilepsy. In-hospital initiation is strongly recommended due to risk of hypoglycemia with emesis or reduced intake.
From the article
One 10 month old with known acquired structural etiology had outpatient dietary teaching and then initiated the ketogenic at home, starting at a 1:1 ratio, and increasing by a ratio of 0.5 per day, with close telephone follow-up with an experienced ketogenic dietician. This child achieved moderate to large ketosis within 6 days and had no adverse effects with dietary initiation…
The majority of infants in our study had very severe epilepsy with daily seizures and etiologies and syndromes highly correlated with medical intractability. In addition, most also had significant developmental delay, a common comorbidity of early-life epilepsy. Seizure types predominantly included both spasms and focal seizures, which are typical of what are seen in refractory infantile epilepsies. Despite the young age at which dietary therapy was started (median of 7 months), these children were clearly medically refractory.
With the exception of one case, all infants were initiated on the diet in hospital, consistent with the consensus statement recently published on ketogenic diet initiation in infants. All patients in our study were started on a traditional ketogenic diet. A recent Korean study found that a traditional ketogenic diet was significantly more likely to result in seizure freedom in children younger than two years of age than a modified Atkins diet. Infants were commenced on a 2:1 ratio, which is higher than the 1:1 ratio recently recommended in the consensus statement.
No infant was fasted, and all started the diet on full calories, to minimize the risk of hypoglycemia. Younger children are potentially at greater risk of hypoglycemia due to more limited glycogen stores, and the consensus statement also recommended dietary initiation without fasting in infants. Most, but not all infants in our study underwent routine glucose checks every 6-12 hours until ketosis was well-established. However, only two infants developed hypoglycemia during dietary initiation, and both of these were associated with reduced oral intake and/or emesis. In both cases, additional glucose checks were performed based on our protocol, which mandates glucose testing if infants take less than two thirds of their allotted meal, or with any emesis…
Despite lack of fasting, the majority of infants rapidly attained ketosis, with nearly all showing urine ketones by 24-48 hours. The infant brain may be uniquely programmed to utilize ketones. Work in animal models has shown that as a consequence of high fat content in maternal milk, the brain metabolism of suckling animals represents a model of naturally occurring ketosis with an increased rate of uptake and metabolism of ketone bodies. This unique ability to metabolize ketones may also explain the high efficacy rates in our study, as well as that of Dressler et al…
We found that the diet was well-tolerated in most infants. However, two developed serious adverse effects over the course of follow-up, emphasizing the importance of careful education of families and periodic, routine laboratory studies to screen for side effects. Furthermore, growth parameters must be carefully followed and caloric adjustments made.