Abstract Background inflammatory bowel disease (IBD), comprising Crohn’s disease (CD) and ulcerative colitis (UC), is a chronic intestinal disease driven by inflammatory T-cell responses thought to target intestinal bacteria. However, evidence for such microbiota-specific T-cell responses is scarce. Immunoglobulin (Ig) titers to bacterial flagellins are highly elevated in a subgroup of CD patients and associate with a more severe disease course, namely penetrating and stricturing disease. This prompted us to hypothesise that flagellin-specific humoral responses could help uncover the inflammatory T-cell responses underlying disease pathogenesis. Methods Plasma IgG and IgA reactivity for 12 flagellins from pathogenic and commensal human intestinal bacteria was assessed by antigen microarray in paediatric CD patients (n = 49), UC patients (n = 12), and age-matched paediatric healthy controls (n = 22) of the PIBD-SET Quality and TISKids cohorts before start of therapy and 10 weeks after the start of therapy. Positivity was defined as values above the mean + 2SD of the paediatric healthy control group. Peripheral blood CD4+ T cells of a subgroup of ‘flagellin high IgG responder’ (IgG positivity to 8 flagellins) and ‘flagellin low IgG responder’ (IgG positivity to 0 flagellin) therapy-naïve paediatric CD patients were cultured with recombinant flagellins. Frequency of flagellin-reactive T cells was assessed by measuring the expression of the activation marker CD154 by flow cytometry. Results As previously shown, IgG positivity to at least one flagellin was elevated in virtually all CD and UC patients (73% and 83%, respectively) compared with age-matched healthy controls. Crucially, a third of CD patients showed IgG reactivity to a high diversity of flagellins (8 or more) while UC patients showed IgG reactivity to a maximum of 3 flagellins. Notably, IgG levels specific for the Escherichia coli outer membrane porine (anti-OmpC IgG) were only increased in 8% of patients, demonstrating that a specific anti-flagellin response is mounted in IBD patients. Anti-flagellin IgA reactivity was not increased either in CD or UC patients compared with controls, suggesting that anti-flagellin CD4+ T-cell responses are required for these anti-flagellin humoral responses. In agreement, increased frequencies of CD4+CD154+ flagellin-reactive T cells were detected in ‘flagellin high IgG responder’ CD patients when compared with both controls and ‘flagellin low IgG responder’ CD patients. Conclusion Our study demonstrates that humoral responses to bacterial flagellins distinguish subgroups of paediatric CD patients. In particular, high diversity in anti-flagellin IgG reactivity is indicative of increased frequencies of anti-flagellin T cells.