Abstracts

Rationale:
Ketogenic diet therapy (KDT), a group of high-fat, low-carbohydrate, moderate-protein diets, are an effective treatment for drug-resistant epilepsy. There is conflicting evidence regarding impacts of KDT on growth. We aimed to determine if linear growth and weight were affected in children following KDT in the UK, and whether there were any associations with clinical, dietary or demographic factors.
Method:
A retrospective case note review was conducted, including children with epilepsy who followed KDT at two UK KD centres. Response was categorised as ≥ 50% seizure frequency in the 28 days prior to the 1-year appointment compared to before diet onset. Height and weight measurements taken as part of routine clinical management were recorded at 1, 2 and 3 years on diet, and 1-year post diet discontinuation. Anthropometric measurements were converted to z-scores and differences from baseline were analysed using Wilcoxon Signed Rank tests. Subgroup analyses of growth changes were performed to compare between feeding methods, ambulatory status, diet type, diet response, age at diet onset, blood beta-hydroxybutyrate level and protein prescription. Results144 patients were included, of which 43 had 1 year post diet data available. Median weight z-scores showed no significant change compared to baseline at any timepoint. Median height z-scores significantly decreased at 2 and 3 years on diet (p=0.000 and p=0.019). Tube-fed children had greater height z-score decrease at 2 years compared to orally fed children (p=0.031), as did non-ambulatory compared to ambulatory children (p=0.031). Height z-score at 2 years decreased less from baseline in individuals following modified KDs compared with those following a classical or medium-chain triglyceride KD at 2 years (p=0.030). A positive correlation was found between height z-score change at 2 years and age at diet onset (p=0.012), and also with beta-hydroxybutyrate level at 1 year (p=0.029). Responders had greater decrease in height z-scores compared to non-responders at 1 year (p=0.040). No significant changes in z-scores 1-year post diet discontinuation, compared to pre-diet, were found.
Conclusion:
In our cohort, linear growth was significantly decreased in children following KDT for 2 years, with no significant effect on weight. Tube feeding, non-ambulatory status, younger age, lower beta-hydroxybutyrate levels and ≥ 50% seizure reduction at 1 year were associated with greater decrease in height z-score, whilst modified KDs had less impact compared to other KD variants. However, height z-scores were no longer significantly different from baseline 1-year post diet. This may be due to the limited number of patients with post diet data available or the relatively short-term follow-up. These findings add to the literature regarding long-term side effects of KDT and help inform which groups may be at greater risk of impacted growth. Further investigation is warranted with prospective data collection and longer post diet follow-up.
Funding:
:No funding was received in support of this abstract.