The trusted source for
healthcare information and
By Philip R. Fischer, MD, DTM&H, and Roma Bhatia, BS,
Dr. Fisher is Professor of Pediatrics, Department of Pediatric and Adolescent Medicine, Mayo Clinic, Rochester, MN.
Ms. Bhatia is a medical student at the Mayo Clinic.
Dr. Fischer and Ms. Bhatia report no financial relationships in this field of study
SOURCE: Bailey LC, et al. Association of antibiotics in infancy with early childhood obesity. JAMA Pediatr published online September 29, 2014, doi:10.1001/jamapediatrics.2014.1539.
From 2001 to 2013, Bailey and colleagues studied 65,480 children in a primary care network affiliated with Children’s Hospital of Pennsylvania that covered urban and suburban parts of Pennsylvania, New Jersey, and Delaware. Premature and small for gestational age babies were excluded from the study, as were infants who did not have followup between 12-59 months and infants in whom BMI was unable to be assessed.
Medical records were analyzed and the use of antibacterial medications during the first two years of life was documented; antifungal and antiviral agents were not studied. Penicillin and amoxicillin were considered to be "narrow spectrum" agents while broad spectrum antibiotics consisted of other systemic antibacterial medications. Obesity was defined as a body mass index (BMI) of greater than the 95th percentile for age and sex. Potentially confounding factors were also evaluated, and careful multivariate analyses were undertaken.
Sixty-nine percent of children received antibiotic treatment during the first two years of life; 15% were obese at 4 years of age. Children were more likely to develop obesity during early childhood (24-59 months of age) if they received multiple courses of broad spectrum antibiotics (rate ratio, RR, 1.16 if four or more courses). Earlier broad-spectrum antibiotic use was also associated with more obesity (RR 1.11 for use at 0-5 months of age, 1.09 at 6-11 months). The use of narrow spectrum antibiotics (penicillin, amoxicillin) was not associated with later obesity.
Other factors such as male gender, Hispanic ethnicity, pubic insurance coverage, asthma, and steroid use were also significantly associated with development of obesity. The link between broad spectrum antibiotic use and obesity, however, was independent of these other factors.
Obesity is a problem of public health importance with 17% of US children and adolescents being overweight or obese.1 In Bailey’s large study described here, 69% of children received antibiotics during the first two years of life — an average of 2.3 antibiotic courses per child with 62% of children getting penicillin or amoxicillin at least once and 41% receiving a broad spectrum antibiotic at least once. It is likely that alterations in antibiotic use might decrease the risk of obesity.
During the first three years of life, the pattern of intestinal flora develops; by three years of age, the microbiome has assumed adult patterns.2 The establishment of the microbiome is likely fragile as it evolves during the early years of life. The resulting microbiome then affects energy metabolism and growth.3
In mice, early antibiotic use alters the microbiome and leads to alterations in short-chain fatty acids and affects lipid and cholesterol metabolism.4 In obese adults, antibiotic use alters the microbiome and changes insulin sensitivity.5 Thus, antibiotics can alter the intestinal microbiome at any age, but alterations in early life seem particularly important.
However, the etiologic relationships between the microbiome and obesity are not completely clear. In obese adults, microbiome patterns are associated with increased energy harvest.6 The new data provided by Bailey suggest that antibiotic-induced alterations in microbiome patterns at an early age might predispose to increased weight gain. Ten percent of Bailey’s subjects were already obese by two years of age, and it could be that obesity itself predisposes to risky microbiomes.
Of course, there are also other pediatric health-related reasons for avoiding unnecessary use of microbiome-altering broad spectrum antibiotics. Antibiotic use in the first year of life has been associated with a 2.9 fold increase in the risk of developing inflammatory bowel disease,7 and the risk appears to be twice that much when the antibiotics are effective against anaerobic organisms.8
As well noted by Bailey and colleagues, childhood obesity is multifactorial. Other illnesses and social factors contribute to obesity. Nonetheless, we now have clear data that practical interventions might moderate the risk of obesity — antibiotics should indeed be used judiciously during infancy, and narrower spectrum antibiotics should be used when broader coverage is not clearly necessary.