Abstract
Gastrointestinal symptoms often persist in children with Hirschsprung's disease (HD) even after "corrective" pull-through surgery. Alteration of the gut microbiota ("dysbiosis") has emerged as a potential contributing factor. Animal studies show gut ecosystem changes that are both intrinsic to HD and caused by bowel resection itself, but human studies comparing the intestinal microbiota of children with HD and healthy children are limited. We collected food frequency dietary surveys, clinical and symptom data, and stool samples from 15 post-operative children with HD and 15 healthy controls (HCs). We performed 16S rRNA gene sequencing from the stool samples and quantified faecal calprotectin as a measure of gastrointestinal inflammation. Despite no global changes in the microbiota between HD and HC cohorts and no differences between individuals with and without a history of HD-associated enterocolitis (HAEC), we identified evidence of altered microbiota development and inflammatory trajectories in HD. In HCs, alpha diversity increased with age (r = 0.83, p < 0.001), while calprotectin levels declined (Spearman's ρ = -0.53, p = 0.04). These age-related patterns were absent in HD. Across the combined cohort, lower alpha diversity was associated with higher faecal calprotectin (Spearman's ρ = -0.47, p = 0.01). In HD, Fusobacteria abundance showed a strong positive correlation with calprotectin (Spearman's ρ = 0.76, adjusted p = 0.02). Pediatric Quality of Life (PedsQL) and gastrointestinal disease-specific symptom scores were lower in HD compared to HC but were not directly linked to microbial diversity or inflammation. Overall, we observed a divergence from healthy peers in the typical developmental trajectory of gut microbial communities and inflammation in children with HD that may involve Fusobacteria. Children with HD reported reduced health-related quality of life compared with HC, consistent with ongoing gastrointestinal symptoms. No microbiota differences were associated with HAEC history, though this may reflect limited sample size.