Fecal Transplants in Children with Autism Show Strong Safety Record in Study of 604 Procedures

Fecal microbiota transplantation, a therapy that transfers processed gut microbes from screened donors into recipients, has moved steadily from the fringes of gastroenterology into mainstream clinical research. Its best-established use remains the treatment of recurrent Clostridioides difficile infection, but a growing body of work has explored whether reshaping the gut microbiome might also help children with autism spectrum disorder, many of whom experience chronic gastrointestinal problems alongside the core features of the condition. What has been missing, critics have long argued, is large-scale, real-world safety data in pediatric populations. A new observational study from Shanghai Children’s Hospital, published in the journal Advances in Therapy, now offers one of the most detailed safety assessments to date, tracking 224 children with autism who collectively underwent 604 transplant procedures.

The research team, led by Youran Li and Ting Zhang of the Department of Gastroenterology, Hepatology and Nutrition at Shanghai Children’s Hospital, School of Medicine, Shanghai Jiao Tong University, set out to answer three questions that clinicians and families frequently ask: how often do adverse events occur after fecal microbiota transplantation in children with autism, how severe are those events, and do any patient or procedure characteristics predict who will experience them. The study was registered with the Chinese Clinical Trial Registry under identifier ChiCTR2200055943 and received ethical approval from the hospital’s review committee, with informed consent obtained from the guardians of all participants in accordance with the Declaration of Helsinki.

The children enrolled ranged in age from two to seventeen years, an age span that captures the early childhood period in which autism diagnoses are typically made as well as the school-age and adolescent years. Each child received transplants through one of three delivery routes: oral capsules containing lyophilized, freeze-dried donor microbiota; a nasojejunal tube, which passes through the nose into the jejunum, the middle section of the small intestine; or a transendoscopic enteral tube, which is positioned during an endoscopic procedure to deliver the transplant deeper into the digestive tract. These routes differ substantially in invasiveness, cost, and the degree of clinical support required, and the researchers were particularly interested in whether the method of delivery influenced tolerability.

To evaluate safety rigorously, the team graded every adverse event using the Common Terminology Criteria for Adverse Events, a standardized grading system widely used in clinical research in which Grade 1 denotes mild symptoms and Grade 5 denotes death. They then applied a multivariate generalized estimating equations model, a statistical approach well suited to repeated-measures data in which the same child contributes multiple procedures over time. This method allowed the investigators to adjust for correlated observations within individual patients and to isolate the independent effect of candidate risk factors, including delivery route, on the likelihood of an adverse event.

The headline finding is striking in its simplicity: across all 604 procedures, adverse events occurred in only 15, an overall incidence of 2.5 percent. Every single event was acute, meaning it began within 48 hours of the procedure, and every event was Grade 1, the mildest category on the severity scale. None of the events was severe, and none persisted. The median duration of symptoms was 28 hours, after which all events resolved spontaneously without long-term consequences. The most frequent symptoms reported were vomiting and irritability, both of which are recognizable, transient reactions in young children undergoing gastrointestinal interventions.

Long-term surveillance added a further layer of reassurance. The researchers followed the children longitudinally after their procedures and observed no delayed adverse outcomes, addressing one of the persistent concerns about microbiota-based therapies: that transferring living microbial communities might carry risks, such as the transmission of pathogens or the seeding of dysbiosis, that only become apparent months or years later. The field has grappled with this question seriously in recent years, particularly after documented transmissions of drug-resistant organisms in adult patients prompted regulatory tightening around donor screening and stool banking. The Shanghai team’s longitudinal data, while observational, suggest that in this pediatric population and under their screening and preparation protocols, delayed harms did not materialize.

Perhaps the most consequential finding concerns delivery route. The transendoscopic enteral tube approach carried an adverse event rate of 26.3 percent, dramatically higher than the other methods. Oral capsules produced adverse events in just 1.7 percent of procedures and the nasojejunal tube in 1.8 percent, both figures indicating favorable tolerability. When the multivariate model was run, transendoscopic delivery emerged as the sole independent risk factor, with an adjusted odds ratio of 21.90 and a P value below 0.001, meaning that after accounting for other variables, children receiving transplants via this route had roughly twenty-two times the odds of experiencing an adverse event compared with those receiving other routes.

The authors of the study conclude that fecal microbiota transplantation demonstrates favorable acute tolerability in children with autism spectrum disorder and that oral capsules, with their low adverse event rate and non-invasive nature, represent a preferred delivery route. This conclusion carries practical weight for clinics considering microbiota-based interventions in pediatric populations. Capsules avoid the discomfort and anxiety associated with tubes, do not require endoscopic facilities, and can be administered in outpatient settings, all of which matter when the recipients are young children, many of whom may have sensory sensitivities or difficulty tolerating medical procedures. The finding that the most invasive route was also the least well tolerated aligns with broader clinical intuition and with pediatric literature on procedure-related nausea and vomiting.

The study sits within a rapidly evolving research landscape. Previous work by some of the same investigators, published in BMC Microbiology in March 2025, examined the long-term safety of fecal microbiota transplantation in Chinese children from 2013 to 2023 across various pediatric indications, and the new report narrows the focus specifically to autism. Internationally, an open-label study published in Microbiome in 2017 reported that microbiota transfer therapy altered gut ecosystems and improved gastrointestinal and autism-related symptoms, and a follow-up in Scientific Reports in 2019 described benefits persisting two years after treatment. More recently, a randomized, double-blind, placebo-controlled trial published in Clinical Translation Medicine in 2024 tested oral fecal microbiota transplantation in children with autism, adding methodological rigor to a field long dominated by small and uncontrolled studies. Meanwhile, basic research published in Nature Microbiology in 2024 has identified multikingdom microbial markers associated with autism, and work in Cell Host and Microbe has shown that donor-recipient microbial interactions influence which microbial strains actually take hold after transplantation, a finding with implications for how donor-recipient matching might one day be optimized.

Caution remains warranted in interpreting the new findings. The study is observational rather than randomized, it reports safety rather than efficacy, and its single-center design means the results reflect one institution’s donor screening protocols, preparation methods, and clinical monitoring practices. The very low adverse event rate may partly reflect careful patient selection and experienced procedural teams, and other centers may not replicate identical figures. Nonetheless, the scale of the dataset, with 604 procedures across 224 children, the standardized adverse event grading, the statistical adjustment for repeated measures, and the longitudinal follow-up together make this one of the most informative real-world safety profiles available for pediatric fecal microbiota transplantation in autism. For families and clinicians weighing whether microbiota-based therapy has a place in the care of children with autism spectrum disorder, the message from Shanghai is measured but meaningful: in experienced hands, with appropriate donor screening and a preference for oral capsules, the acute and long-term safety picture appears reassuring, even as questions about therapeutic benefit continue to be tested in controlled trials.

Subject of Research: Safety and tolerability of fecal microbiota transplantation in children with autism spectrum disorder

Article Title: Acute Tolerability and Long-Term Surveillance of Fecal Microbiota Transplantation in Children with Autism: A Real-World Study of 604 Procedures

Article References: Li, Y., Yu, W., Liu, R., Xiao, P., Li, X., & Zhang, T. (2026). Acute Tolerability and Long-Term Surveillance of Fecal Microbiota Transplantation in Children with Autism: A Real-World Study of 604 Procedures. Advances in Therapy. https://doi.org/10.1007/s12325-026-03797-9

Image Credits: AI Generated

DOI: 10.1007/s12325-026-03797-9

Keywords: fecal microbiota transplantation, autism spectrum disorder, pediatrics, gut microbiome, adverse events, oral capsules, donor screening, long-term safety, Shanghai Children's Hospital, clinical trial, gastroenterology, neurodevelopmental disorders