Safety of Fecal Microbiota Transplantation in Children: A Single-Center Experience

Background/Objectives: Fecal microbiota transplantation (FMT) is an established treatment for recurrent or refractory Clostridioides difficile infection (rCDI) and is increasingly investigated for other gastrointestinal and neuropsychiatric conditions in children. However, pediatric safety data remains limited. This study aimed to evaluate the short-term safety of colonoscopy-delivered FMT in a heterogeneous pediatric cohort. Methods: We conducted a retrospective single-centre cohort study of all FMT procedures performed at Vilnius University Hospital Santaros Klinikos between 2017 and 2025. Fecal material from three screened healthy pediatric donors was delivered to the caecum via colonoscopy under general anesthesia. Adverse events (AEs) arising within 72 h of each procedure were identified from the clinical records and assigned a severity grade using version 5.0 of the Common Terminology Criteria for Adverse Events (CTCAE). AEs were analyzed on a per-procedure basis. Associations between sex, age, and underlying indication and AE occurrence were assessed using Fisher's exact test and the Mann-Whitney U test. Results: A total of 108 FMT procedures were performed on 73 children aged 2 to 17 years. The predominant indication was autism spectrum disorder (ASD) and other developmental disorders (80.6%). AEs were reported after 11 procedures (10.2%); all were classified as CTCAE grade 1 or 2, and no serious AEs occurred. The most common AEs were diarrhoea (5.6%) and vomiting (2.8%). In an exploratory comparison, procedures performed for ASD were associated with a lower AE rate compared with other indications combined (6.9% vs. 23.8%; OR 0.24, 95% CI 0.06-0.87). Conclusions: Colonoscopy-delivered FMT was well tolerated in pediatric patients, with a favorable short-term safety profile consistent with published pediatric FMT literature. As follow-up was largely passive and the cohort was predominantly children with ASD, the observed rate should be regarded as a lower bound, and these findings warrant confirmation in prospective studies.