Conference Coverage

Antibiotic-obesity link ‘clinically insignificant’ at age 10 years



– Antibiotic use in the first 2 years of life was associated with a small amount of weight gain by 10 years of age, but the amount is “likely clinically insignificant,” according to new data presented at Obesity Week.

Sheryl Rifas-Shiman Kari Oakes/MDedge News

Sheryl Rifas-Shiman

At 10 years of age, children without chronic health conditions who received any antibiotics had an odds ratio of 1.02 for being overweight or obese; for children with complex chronic conditions, the OR was 1.07 (95% confidence intervals, 0.97-1.07 and 0.96-1.19, respectively). The findings were based on data from almost 60,000 children studied in a large, multi-institutional national cohort.

“This is good news,” said first author Sheryl Rifas-Shiman, MPH, discussing the findings during a poster session. She noted that the 10-year data from the longitudinal study are consistent with findings at the 5-year mark that had previously been reported.

The study comes against the background of a recent meta-analysis finding that children given any antibiotics before age 24 months had a higher body mass index z-score (BMI-z) in childhood than children who didn’t take antibiotics. Disruptions that antibiotics can cause in the gut microbiome have been hypothesized to promote overweight and obesity in children.

The present study was conducted using electronic medical record data from 2009-2016 drawn from institutions participating in PCORnet, a national research collaboration and clearinghouse.

The analysis dichotomized the cohort into those who, by 24 months of age, had received any antibiotics and those who received none. Ms. Rifas-Shiman and her colleagues also looked at a categorical count of the number of antibiotic prescriptions, from 0 to 4 or more.

Finally, they broke the type of antibiotics into narrow- or broad-spectrum, she said in an interview during the poster session. In order for exposure to be considered narrow-spectrum only, the analysis was limited to participants who had no broad-spectrum antibiotic exposure during the same time frame.

The study’s multivariable analysis also took into account complex chronic conditions the children may have had.

Fifty-seven percent of children received antibiotics before the age of 24 months. Patients were overall just under half (48%) female, and about half (49%) were white. Black children made up 37% of the cohort, and Hispanic children constituted 12%.

By 10 years of age, 36% of participants were overweight or obese, with BMIs at or above the 85th percentile, according to 2000 Centers for Disease Control growth charts.

There was a suggestion of a dose-response relationship for both narrow- and broad-spectrum antibiotics because only at the higher antibiotic exposures did increased BMI-z reach statistical significance. However, broad-spectrum antibiotics were not more likely than narrow-spectrum antibiotics to be associated with increased BMI-z.

Other multivariable adjustment took into account site clustering and adjusted for sex, race, ethnicity, preterm birth, asthma or infectious disease diagnoses, and corticosteroid exposure, as well as health care visits in the first 2 years of life.

Limitations of the study included the fact that it was observational, raising the potential for undetected confounders. Also, since the study looked at prescription, rather than dispensing data, there could be some exposure misclassification by which children who were classified as having taken antibiotics did not actually take them or did not take the full amount prescribed.

“Antibiotics should always be used judiciously; however, the long-term risk of childhood obesity from antibiotics in infancy appears to be small and clinically insignificant,” wrote Ms. Rifas-Shiman, of the department of population medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, both in Boston.

The study was funded by an award from the Patient-Centered Outcomes Research Institute.

SOURCE: Rifas-Shiman et al. Obesity Week 2018, Poster TP-3155.

Next Article: