Abstracts

Rationale: The ketogenic diet (KD) is a medical treatment for refractory epilepsy. Neutropenia has been associated with the KD in adults, but the mechanism is not well understood. Potential causes of neutropenia in KD include folate, vitamin B12 and copper deficiencies as well as long-term caloric restriction known to inhibit neutrophil function and decrease neutrophil count. We describe the prevalence of neutropenia and associated factors in children treated with a KD. Methods: Children initiating a KD between September 2013 and June 2017 at the Hospital for Sick Children in Toronto, Canada were enrolled in a prospective observational study with parent consent. Dietary details, anthropometrics, and laboratory studies including neutrophil count, albumin, total protein, and folate were collected at diet initiation and 6, 12, and 24 months. Vitamin B12 and copper levels were not collected. Neutropenia was defined by age and sex lower end of reference interval ranging from 0.97 – 1.82 x 10^9/L. Neutrophil counts were correlated with diet type, change in height z-score, and dietary factors. Results: Eighty-one children were enrolled during the study period. Twenty-eight were excluded: 19 discontinued diet therapy prior to 6 months, 4 did not have baseline neutrophil data, 2 initiated the diet during status epilepticus, 2 died and 1 withdrew from the study.Of the included 53 children, 27 (50.9%) followed a classic KD, 19 (35.8%) a medium-chain triglyceride KD, and 7 (13.2%) a Modified Atkins diet. The median age at diet initiation was 4.86 years old (25^th,75^th percentile: 2.08, 8.27), and 27 (50.9%) were male.Neutropenia prevalence at baseline and 6 month follow-up was 20.8% and 22.9%. Of 11 children who were neutropenic at baseline, 3 (36.4%) had normal neutrophil counts at 6 months. Six of 42 (14%) children with normal neutrophil counts at baseline developed neutropenia at 6 months, which resolved in all 4 children with data available at subsequent time points. The prevalence of neutropenia at 12 (n=32) and 24 (n=15) months was 21.9%, and 26.7%.There were no significant association between neutrophil count and diet type, daily protein intake, serum albumin, total protein, folate level or height z-score at any time point. Conclusions: More than one-fifth of children initiating a KD are neutropenic at baseline and similar rates of neutropenia are seen during KD.   Most children with neutropenia during KD were neutropenic at baseline. Children who developed neutropenia at 6 months resolved in all cases with subsequent data available. No association with measured dietary factors was found. Further study is ongoing to investigate the role of medications and the clinical significance of neutropenia in this cohort. Funding: Ontario Brain Institute