Moderate-to-late preterm birth, common respiratory pathogens, and severe lower respiratory tract infections across childhood

Technical Summary: Moderate-to-Late Preterm Birth, Common Respiratory Pathogens, and Severe Lower Respiratory Tract Infections Across Childhood

Problem Statement
Lower respiratory tract infections (LRTIs) remain a leading cause of pediatric hospitalization and mortality globally. While the respiratory vulnerability of extremely and very preterm infants (gestational age <32 weeks) is well-established, the long-term impact of moderate-to-late preterm (MLP) birth (gestational age 32 to <37 weeks) on LRTI severity across the entire pediatric lifespan is less understood. Existing literature predominantly focuses on the first few years of life, leaving a gap in knowledge regarding whether MLP-associated susceptibility to severe LRTIs persists into middle childhood and early adolescence. Furthermore, there is a lack of comprehensive data on how MLP status interacts with specific respiratory pathogens to influence disease severity, particularly given that MLP infants constitute the majority (80–85%) of the preterm population.

Methodology
This study employed a large-scale, 11-year retrospective design utilizing data from the Tongji Biobank at Tongji Hospital in Wuhan, China.

• Cohort: The study included 65,859 unique pediatric patients (aged 1 month to 14 years) hospitalized with LRTIs between January 1, 2015, and December 31, 2025. The cohort comprised 1,453 MLP children (2.21%) and 64,406 full-term children.
• Exclusions: Patients with primary/secondary immunodeficiency, congenital heart disease, bronchopulmonary dysplasia, cystic fibrosis, congenital airway/pulmonary malformations, neuromuscular disorders, or those admitted to surgery/PICU initially were excluded.
• Exposure and Outcomes: The primary exposure was MLP birth status. The primary outcome was a composite clinical severity measure defined by a physician diagnosis of severe pneumonia, extrapulmonary complications, oxygen administration, or PICU admission.
• Pathogen Analysis: Respiratory pathogens were identified via conventional microbiological tests (CMTs), multiplex real-time PCR (13-target panel), and targeted next-generation sequencing (tNGS). Serological IgM assays (9-plex) were also utilized.
• Statistical Analysis:
  • Multivariable logistic regression (sequentially adjusted for age, sex, delivery mode, birth order, calendar year, and season) was used to calculate odds ratios (ORs) for severity.
  • Stratified analyses were conducted across three developmental stages: infancy (1 month–<2 years), early childhood (2–<6 years), and middle childhood/early adolescence (6–14 years).
  • Propensity score matching (1:15 ratio) was used to balance covariates (age, sex, year, season) when comparing pathogen positivity rates between MLP and full-term groups.
  • Additive interactions between MLP status and 12 distinct pathogens were quantified using the Relative Excess Risk due to Interaction (RERI) and the Attributable Proportion (AP).

Key Results

1. Persistent Risk Across Age Groups: MLP birth was independently associated with an increased risk of composite severe LRTI outcomes across all age groups. The adjusted ORs were 1.16 (95% CI: 0.99, 1.34) for infants, 1.27 (95% CI: 1.03, 1.56) for early childhood, and 1.71 (95% CI: 1.21, 2.41) for middle childhood/early adolescence. The interaction between age and preterm status was not significant (p = 0.868), indicating the risk is age-independent.
2. Pathogen Susceptibility: After propensity score matching, the positivity rates for all 13 molecular targets and 9 serological targets were comparable between MLP and full-term children, with one minor exception (Influenza A virus in middle childhood showed a slightly higher rate in MLP). This suggests MLP birth does not alter the susceptibility to microbial acquisition.
3. Synergistic Interaction with RSV: A significant synergistic additive interaction was identified specifically between MLP status and Respiratory Syncytial Virus (RSV) infection.
   • RERI: 1.16 (95% CI: 0.30, 2.35).
   • Attributable Proportion (AP): 0.43 (95% CI: 0.11, 0.57).
   • This indicates that 43% of severe LRTI events in RSV-infected MLP children are attributable to the synergistic interplay of these two factors, beyond their independent effects. No such additive synergy was observed for the other 11 pathogens tested.
4. Specific Severity Components: MLP birth independently predicted higher risks for severe pneumonia diagnosis (OR = 1.63) and oxygen administration (OR = 1.51). Associations with extrapulmonary complications and PICU admission were less robust after full adjustment.

Significance and Claims
The authors claim that this study provides critical evidence that the respiratory vulnerability associated with MLP birth is not a transient neonatal phenomenon but a persistent predisposition extending well into middle childhood and early adolescence.

• Distinct Mechanism: The study distinguishes between susceptibility to infection and severity of disease. The findings suggest that while MLP children are not more likely to acquire common respiratory pathogens than their full-term peers, their underlying physiological characteristics (likely related to interrupted lung development) amplify the severity of the disease, particularly when co-occurring with RSV.
• Clinical Implications: The identification of a specific, synergistic additive interaction with RSV suggests that current preventive strategies, which often focus on the first year of life, may need to be re-evaluated. The authors propose that extending targeted RSV immunoprophylaxis and monitoring frameworks to include vulnerable MLP children in the 2-to-6-year age group could reduce the pediatric healthcare burden.
• Public Health Impact: Given that MLP infants represent the vast majority of the preterm population, even marginal increases in adverse outcomes impose a substantial economic and healthcare burden. These findings underscore the need for sustained clinical monitoring of MLP individuals throughout childhood.

Limitations Acknowledged
The authors note that as a single-center retrospective study, the cohort may be subject to selection bias toward more severe cases. Additionally, the lack of direct measurements for maternal smoking, environmental smoke exposure, and post-discharge lung function limits the ability to fully elucidate the physiological mechanisms. Finally, the sample size within the MLP cohort precluded granular stratification by specific weeks of gestation (32–36 weeks).