Abstract
Background
Severe childhood infection has a dose-dependent association with adult cardiovascular events and with adverse cardiometabolic phenotypes. The relationship between cardiovascular outcomes and less severe childhood infections is unclear.
Aim
To investigate the relationship between common, non-hospitalised infections, antibiotic exposure, and preclinical vascular phenotypes in young children.
Design
A Dutch prospective population-derived birth cohort study.
Methods
Participants were from the Wheezing-Illnesses-Study-Leidsche-Rijn (WHISTLER) birth cohort. We collected data from birth to 5 years on antibiotic prescriptions, general practitioner (GP)-diagnosed infections, and monthly parent-reported febrile illnesses (0–1 years). At 5 years, carotid intima-media thickness (CIMT), carotid artery distensibility, and blood pressure (BP) were measured. General linear regression models were adjusted for age, sex, smoke exposure, birth weight z-score, body mass index, and socioeconomic status.
Results
Recent antibiotic exposure was associated with adverse cardiovascular phenotypes; each antibiotic prescription in the 3 and 6 months prior to vascular assessment was associated with an 18.1 μm (95% confidence interval, 4.5–31.6, p = 0.01) and 10.7 μm (0.8–20.5, p = 0.03) increase in CIMT, respectively. Each additional antibiotic prescription in the preceding 6 months was associated with an 8.3 mPa-1 decrease in carotid distensibility (-15.6– -1.1, p = 0.02). Any parent-reported febrile episode (compared to none) showed weak evidence of association with diastolic BP (1.6 mmHg increase, 0.04–3.1, p = 0.04). GP-diagnosed infections were not associated with vascular phenotypes.
Conclusions
Recent antibiotics are associated with adverse vascular phenotypes in early childhood. Mechanistic studies may differentiate antibiotic-related from infection-related effects and inform preventative strategies.
Funder
ZonMw
GlaxoSmithKline
National Health and Medical Research Council
Publisher
Public Library of Science (PLoS)