Editorial: Antibiotics and risk of IBD—Can we close the books?

While the exact pathogenesis of inflammatory bowel diseases (IBD) is not fully understood, it is widely held that it is determined by a complex interplay between genetic, immunological and environmental factors. Globally, diverging temporal incidence patterns can be observed with sharply rising incidence rates in newly industrialised countries in Asia and South America, as opposed to stabilising rates in Western countries.1– 3 These differences may to a large extent be explained by changes in lifestyle factors such as improved hygiene, introduction of Western dietary habits and more frequent antibiotic use.1 In fact, there is a growing body of evidence linking exposure to certain antibiotics, that are known to perturb the gut microbiota, with a higher risk of developing IBD. So far, the largest studies on this topic come from Scandinavian countries suggesting an association between exposure to antibiotics and later development of IBD. Among those, a nationwide Swedish case– control study4 (2007– 2016, n = 23,982 cases and 117,827 controls) and a Danish longitudinal cohort study5 (2000– 2018, n = 52,898 cases in 6,104,245 individuals) demonstrated a clear dosedependent association of cumulative antibiotic use with adultonset Crohn's disease (CD) and ulcerative colitis (UC). Furthermore, in another case– control study, early life exposure to antibiotics has been linked to an increased risk of childonset IBD.6 However, these studies represent Western countries with predominantly Caucasian populations, yet studies from other parts of the world are scarce. Recently, Oh and colleagues report findings from a nationwide case– control study from Korea involving 68,633 individuals with IBD and 343,165 controls (2004– 2018).7 The authors found an increased risk of IBD (adjusted OR = 1.24; 95% CI = 1.21– 1.27) following exposure to antibiotics with similar estimates for both CD and UC. The risk increased with cumulative antibiotic use and seemed to persist even up to 9 years after exposure. Interestingly, consistent with the Scandinavian studies, the highest risk was seen after use of broadspectrum antibiotics (i.e. fluoroquinolones) which supports the hypothesis of antibioticinduced perturbations of the gut microbiota.8 A strength of their study is the adjustment for previous gastrointestinal infections, since it has been a matter of debate whether symptoms of gastroenteritis prompting antibiotic treatment could as well have been the first signs and symptoms of IBD. Nevertheless, residual confounding is still an issue due to lack of data on several clinical covariates and other gut microbiotaaltering medications. However, of note, when previous Scandinavian studies either adjusted for healthcare utilisation or for other gut microbiotaaltering medications, estimates largely remained unchanged. Overall, the current study elegantly confirms and extends existing evidence, so the question arises whether the books on this subject can now be closed? Perhaps not, because unanswered questions on this important topic remain: Does antibiotic treatment act as a trigger or accelerator of an already developing disease? What are the molecular mechanisms and does previous exposure to antibiotics affect the clinical course of IBD? Finally, studies reporting absolute risk estimates and familial/twin studies to disentangle the complex interplay with genetics are needed to bridge the current knowledge gaps before we can close the books.