-National registry study demonstrates clear association; causal direction uncertain To understand the consequences of early-life antibiotic exposure on the future development of celiac disease, Stine Dydensborg Sander, MD, PhD, of Hans Christian Andersen Children's Hospital in Odense, Denmark, and colleagues, used two independent nationwide cohorts, one from Denmark (n=1,168,656) and one from Norway (n=537,457) for an observational, register-based cohort study. In general, food allergies are a public health concern in the U.S., affecting up to 6% of children. Among the common food allergies, celiac disease has been described as an immune-mediated reaction to dietary gluten (storage protein for wheat, barley, and rye), which mainly affects the small intestine in genetically predisposed individuals and usually resolves by following a gluten-free diet. However, small bowel bacterial overgrowth, microscopic colitis, exocrine pancreatic insufficiency, irritable bowel syndrome, and lactose or fructose intolerance have also been implicated in nonresponsive celiac disease. Global estimates for celiac disease range from 0.7% to 1.4%. Prevalence and diagnostic rates for the autoimmune enteropathy rose substantially in recent years, according to the American College of Gastroenterology. What causes the autoimmune reaction to gluten is uncertain. Genetically predisposed individuals have mutations in an immunity-related gene, HLA-DQ, with one or more variants present in virtually every person with celiac disease. But the haplotypes are also found in 30% of the general population and only 3% of people with those haplotypes eventually develop celiac disease. Numerous environmental factors could interact with HLA-DQ and non-HLA-DQ genes to trigger celiac disease. In the respective Danish and Norwegian cohorts, the median ages at the end of follow-up were 11.6 and 5.4 years. Mothers filled in questionnaires related to infectious diseases in the child, and infant feeding, at ages 6 and 18 months. The main exposure was systemic antibiotics dispensed in the first year of life. Medical information was collected across both cohorts for more than 1.7 million children, of whom 3,346 were diagnosed with celiac disease. About 43.6% of children with celiac disease had been exposed to systemic antibiotics during the first years of their lives in the Danish cohort versus 38.7% of children without celiac disease. More Norwegian children with celiac disease (20.3%) had also been exposed to early-life systemic antibiotics versus children without celiac disease (18.4%). Overall, early-life systemic antibiotics was positively associated with diagnosed celiac disease in both cohorts (pooled odds ratio 1.26, 95% CI 1.16-1.36). A dose-response relationship was also seen between an increasing number of antibiotics and risk of celiac disease (pooled OR 1.08 for each additional antibiotic, 95% CI 1.05-1.11). Adjustment for the number of maternally reported infections in the child in two large sub-cohorts reduced the association slightly (pooled OR 1.18, 95% CI 0.98-1.39). Neither a specific type of antibiotic nor age at exposure were prominent factors in celiac disease, suggesting there is no particularly vulnerable age and no differing effect among antibiotic classes. The association was at least as strong for exposure from 0 to 24 months as for 0 to 12 months, Sander's group reported. Study limitations included the difficulty of disentangling the effect of infections and antibiotics in an observational study lacking details of the infections, and indications for antibiotic use. Since maternally reported infectious episodes at 6 and 18 months may be an imprecise measure for the actual incidence, the authors also raised the possibility of residual confounding in their analysis. Source Reference: Gastroenterology, 2019; DOI:10.1053/j.gastro.2019.02.039 Study Highlights: Explanation of Findings Systemic antibiotics dispensed in the first year of life was consistently associated with diagnosed celiac disease (n=3,346) among more than 1.7 million children enrolled in independent Danish and Norwegian study cohorts. Researchers observed a dose-dependent relationship with an increasing number of dispensed antibiotics. Taken together, the findings suggested that early-life, systemic antibiotic exposure was associated with a later diagnosis of celiac disease. Adjustments for potential confounders did not change current study outcomes. Results were analyzed against a backdrop of growing knowledge that early-life exposure to antibiotics, either prenatally or postnatally, is common and mainly due to the use of intrapartum prophylaxis, or to antibiotic usage during delivery by cesarean section. Concurrently with the rise in individuals exposed to antibiotics in early infancy, several researchers noticed an increased risk in non-communicable diseases, including allergies. The allergic response to gluten -- seen as a typical hallmark of celiac disease -- could possibly be elicited by distinct intestinal microbiota in celiac disease with proinflammatory properties. Some, but not all studies, have described associations between antibiotic use and the later onset of celiac disease. Consistent with current study findings, Italian researchers reported an incidence rate ratio of 1.31 (95% CI, 1.10–1.56) for celiac disease confirmed by pathology reports (510 cases). By contrast, The Environmental Determinants of Diabetes in the Young (TEDDY) study found no association between parentally reported antibiotic exposure and persistently positive celiac disease-specific antibodies at age 4 years (783 cases of celiac disease autoimmunity). Experts concluded that the TEDDY study provided some reassurance "that appropriate use of antibiotics will not have unintended consequences and cause autoimmune diseases." On the other hand, a Swedish register-based study showed that antibiotic exposure up to 3 years prior to the diagnosis was associated with celiac disease at all ages. Further sub-analysis of this underpowered study showed that exposure in the first 6 months of life was associated with celiac disease, in accordance with current study findings. Sander and colleagues could not exclude the possibility of reverse causality. This phenomenon refers to possible scenarios where the symptoms of celiac disease could mimic an infection, exaggerate symptoms of infections, or increase the risk of infectious diseases, thereby increasing the propensity for prescriptions for antibiotics. But the association between antibiotics in the first year of life and celiac disease was comparable across the strata of age at diagnosis, so the authors noted that it was less likely that the findings were due to reverse causality. "Future studies should attempt to separate the effect of infections and antibiotics by using more detailed information on indication for use of antibiotics and type of infections preferably by using biomarkers; and to elaborate on types and ages of exposure and interaction between risk factors and if the effect of antibiotic differs between risk groups," noted the authors. Jocelyn A. Silvester, MD, of Boston Children's Hospital and Harvard Medical School, pointed out that figuring out the potential ties between antibiotics and celiac disease poses a challenge. "This is a very difficult question to answer, even though this is one of the largest datasets we have to look at," Silvester, who was not involved in the current study, said in an interview. It's difficult to tease out the true relationship because of potential confounding factors, she added, noting that that not all antibiotic types have the same effect on the microbiota and not all the infections treated with antibiotics were bacterial, but may have included viral and fungal. Underlying infection rather than antibiotics may have been driving the celiac risk, she stated. "But having large well-done studies that try to answer difficult questions is always a step in the right direction," Silvester said. Source
