Perinatal Antibiotic Exposure and Respiratory Outcomes in Children Born Preterm

Abstract

Importance: Animal models suggest a link between early antibiotic exposure and obstructive airway disease, but corresponding data for premature infants are lacking.

Objective: To investigate whether repeated perinatal antibiotic exposure in preterm neonates with very low birth weight (VLBW) is associated with obstructive airway disease at early school age.

Design, setting, and participants: In this population-based, multicenter cohort study, VLBW preterm neonates (22 weeks 0 days' to 36 weeks 6 days' gestation with birth weight <1500 g) were enrolled in 58 German Neonatal Network (GNN) centers from January 2009 to March 2017 and received a standardized follow-up at 5 to 7 years of age. To assess the sequential outcomes of antibiotic exposures, the post hoc analysis was restricted to participants born by cesarean delivery. Data were analyzed from May 2024 to February 2025.

Exposure: Perinatal antibiotic exposure, defined by an antibiotic risk score (ARS).

Main outcome and measures: The primary end point was the forced expiratory volume in 1 second (FEV1) z score at 5 to 7 years of age. The low-risk (ARS I) group was exclusively exposed to surgical antimicrobial prophylaxis (SAP) given to the mother before cesarean delivery. The intermediate-risk (ARS II) group was exposed to maternal SAP and postnatal antibiotic treatment of the neonate, while the high-risk (ARS III) group was additionally exposed to antenatal maternal treatment. Secondary outcomes included forced vital capacity (FVC) z score and childhood asthma episodes. Univariate and linear regression models were used to analyze outcome measures.

Results: Of 3820 VLBW preterm-born children with follow-up at age 5 to 7 years (median gestational age, 28.4 weeks [IQR, 26.6-30.3 weeks]; 1948 [51.0%] male; 1382 [36.2%] from a multiple birth), 3109 (81.4%) were born by cesarean delivery. Of these children, 292 (9.4%) were classified into ARS I, 1329 (42.7%) into ARS II, and 1488 (47.9%) into ARS III. Higher ARS levels were associated with lower FEV1 z scores at early school age (ARS II vs I: β, -0.31 [95% CI, -0.59 to -0.02]; P = .03; ARS III vs II: β, -0.27 [95% CI, -0.46 to -0.08]; P = .006). In the secondary analysis, a higher exposure level (ARS III vs II) was associated with impaired FVC z scores (β, -0.23; 95% CI, -0.43 to -0.03; P = .02) and an increased risk of early childhood asthma episodes (odds ratio, 1.91; 95% CI, 1.32-2.76; P = .001).

Conclusions and relevance: In this GNN cohort study, multiple episodes of perinatal antibiotic exposure were associated with impaired lung function in preterm-born children at early school age. Early identification of high-risk neonates may enable targeted strategies to support respiratory health and optimize long-term outcomes.

Figure 1.. Flowchart of the German Neonatal Network (GNN) Study Cohort of Preterm Individuals With Very Low Birth Weight Included in and Excluded From the Analysis

ARS indicates antibiotic risk score; BW, birth weight; GA, gestational age. ARS I indicates 1 antibiotic exposure (lowest risk); ARS II, 2 exposures (intermediate risk); and ARS III, 3 exposures (highest risk).

Figure 2.. z Scores for Forced Expiratory Volume in 1 Second (FEV₁) Across the 3 Risk Strata of the Antibiotic Risk Score (ARS)

Each dot represents an individual data point (z score); middle horizontal lines, means; upper and lower horizontal lines flanking the data clouds, 10th and 90th percentiles; whiskers, minimum and maximum values. Statistical comparison among groups was performed using the Kruskal-Wallis test. ARS I indicates 1 antibiotic exposure; II, 2 exposures; and III, 3 exposures.