Does Childhood Antibiotic Use Cause IBD?

Question

Does childhood antibiotic exposure lead to inflammatory bowel disease later in life?

Response from William F. Balistreri, MD
Professor of Medicine, University of Cincinnati College of Medicine; Staff Physician, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio

Increasing Incidence of IBD

This inquiry likely was stimulated by the spate of recently published studies that strongly suggest that patients diagnosed with inflammatory bowel disease (IBD) in childhood are more likely to have been prescribed antibiotics early in life.^{[1,2,3,4,5,6,7]} The link may be a bit complex, however, because the pathogenesis of the IBDs is complex and multifactorial and may differ for Crohn disease (CD) and ulcerative colitis (UC). It is clear that the incidence of the IBDs has recently increased in pediatric populations, especially in high-resource countries.^{[8,9,10,11,12,13,14]} For example, a population-based study from Canada reported a 5%-7% annual increase in the incidence of IBD from 1995 to 2005 in children younger than 10 years.^[8,12] Another study found that the incidence of pediatric IBD increased by 6%-8% per year in Finland.^[13]

Pathogenesis of IBD

Traditional theories on the pathogenesis of IBD postulate interactions between a genetic predisposition and aberrant gastrointestinal immune responses to undefined environmental triggers and factors.^{[15,16,17,18]} Genome-wide association studies have identified genetic susceptibility to IBD; however, this accounts for fewer than half of the cases. The probability of disease development is modified by diet, lifestyle, and endogenous factors, including the gut microbiota.^[18] The recent increase in the incidence of pediatric IBD over such a short period of time, combined with the observation that individuals can carry high-risk genes without developing IBD, lends strong support to the involvement of other "drivers." These could include factors that may directly affect immune function or alter the intestinal bacterial community by encouraging or discouraging the growth of certain bacteria. In fact, some of these variant susceptibility genotypes have been linked to bacterial dysbiosis (imbalance) in patients with IBD.^{[18,19,20,21,22]}

A popular explanation is that the gut microbiota is altered in patients with IBD -- postulating either an increase in pathogenic bacteria or a reduction in protective bacteria, and then this dysbiotic imbalance triggers inflammation.^[19,20,23] Consistent with this hypothesis, multiple studies have found associations between gastrointestinal tract infections and the onset of childhood IBD. Patients with IBD have reduced diversity of intestinal microbiota (specifically, anaerobic bacteria) relative to healthy controls.^[19] Furthermore, the IBD phenotype can be induced or ameliorated in murine models of IBD through transfer of bacterial microflora.^{[1,2,18,19,24]}

Possible Role of Antibiotics

Broad-spectrum antibiotics will alter the composition of the human gut microbiome, with transiently reduced bacterial diversity. Thus, it is reasonable to hypothesize that antibiotic-induced alterations of the intestinal flora contribute to the etiology of IBD.

In a series of related studies, Shaw and colleagues^[1,2,3] aimed to determine whether early use of antibiotics was associated with the development of IBD in children. In a case-control analysis of the population-based University of Manitoba IBD Epidemiologic Database, a total of 36 children diagnosed between 1996 and 2008 were matched to 360 control patients.^[1] Antibiotic usage data were drawn from the Manitoba Drug Program Information Network, a comprehensive population-based database of all prescription drugs; the use of which reduced the issue of recall bias. The mean age at IBD diagnosis was 8.4 years; 58% had 1 or more antibiotic dispensations in their first year of life compared with 39% of controls. Those receiving 1 or more dispensations of antibiotics had an almost 3-fold greater risk for IBD. The researchers followed these studies with an assessment of 294 children whose IBD was diagnosed between 1989 and 2008, matching them to 2377 control patients.^[2]

The researchers hypothesized that a diagnosis of otitis media in early childhood, which serves as a proxy measure of antibiotic use, would be associated with the subsequent diagnosis of IBD. Approximately 5% of those with IBD and 12% of the controls did not previously have otitis media. In multivariate analyses, compared with cases and controls without otitis media, individuals with otitis media by age 5 years were 2.8-fold more likely to have IBD. This observation could be partially explained by the fact that susceptibility to otitis media in early childhood may be a link to other immune-related conditions, such as IBD, later in life.^[7] Repeated antibiotic use may also reflect a shared susceptibility to childhood infections and IBD or, alternatively, may trigger disease development.

Shaw and colleagues^[3] also determined that the use of antibiotics shortly before the diagnosis was associated with the development of IBD in adults. A total of 2234 adults diagnosed with IBD between 2001 and 2008 were matched to 22,346 control patients. The mean age at diagnosis of IBD was 43 years; 12% of cases had 3 or more antibiotic prescriptions 2 years before the case date, compared with 7% of control patients. This difference in antibiotic dispensations between cases and controls was fairly consistent at 3, 4, and 5 years before the diagnosis of IBD. Antibiotic dispensations were significantly associated with both CD and UC, and there was a dose-dependent relationship between the number of antibiotic dispensations and risk for disease.

Other studies have reported similar linkages. Virta and colleagues^[4] conducted a national, registry-based study and identified 595 children with IBD (233 with CD and 362 with UC) and 2380 matched control patients. The risk for pediatric CD, but not UC, increased with the number of antibiotic purchases from birth to the onset of IBD. The strength of the association with CD was related to the number of purchases and exposure periods but was not completely linear. A nationwide cohort study of all Danish children born between 1995 and 2003 found an association between filled antibiotic prescriptions and later development of IBD.^[5] The relative risk (RR) for IBD was approximately 2-fold higher for antibiotic users compared with nonusers. This association appeared to be an effect on CD alone (RR 3.4) and was strongest in the first 3 months following use (RR 4.4) and among children with more than 7 courses of antibiotics (RR 7.3).

Kronman and colleagues^[6] also found that certain antibiotic exposures throughout childhood were associated with an increased risk of developing IBD. This relationship declined with increasing age at exposure; those who received antibiotics in the first year of life had an adjusted RR of 5.5; the risk fell to 2.6 and 1.6 by 5 and 15 years, respectively. Each antibiotic course increased the risk for IBD by 6%, and a dose-response effect existed.

More Evidence Needed

Studies to date suggest that there could indeed be an association between antecedent antibiotic use (induced dysbiosis) and the ultimate development of IBD. They provide further evidence that manipulations in gut flora as a result of using antibiotics may be relevant in the pathogenesis of IBD. However, as in any observational study, causality cannot be inferred from the results. Confounding by indication -- in particular, prescribing antibiotics to children with intestinal symptoms of as yet undiagnosed IBD -- should also be weighed as a possible explanation. In addition, several of the studies are limited by recall bias, lack of controls, incomplete antibiotic capture, or including exposures between IBD symptom onset and diagnosis.

In summary, these studies provide impetus for further exploration of the gut microbiome in children who develop IBD. It will be instructive to use currently available methodology to investigate how the gut microbiome is altered by antibiotic use early in life and for how long it is altered. Because it is difficult to distinguish cause from effect, it will be important for future studies to clarify the relative roles of human and microbial genes, susceptibility, and environmental exposures (diet, lifestyle, smoking, and all medications) in the predisposition to IBD. The dysbiosis hypothesis, which posits that perturbations in the gut microbiome contribute to the development of IBD, may ultimately be validated.^[20] At present, we should heed the cautionary note and reinforce the importance of the judicious use of antibiotics.