Small for gestational age as an independent risk factor for long-term pediatric gastrointestinal morbidity of the offspring

Objective: The concept of neonatal programming has begun to emerge as an important component of adult health. Scarce data exist regarding perinatal risk factors for long-term gastrointestinal (GI) morbidity of the offspring. We aimed to evaluate the association between birthweight (BW) at term and long-term pediatric GI morbidity.

Study design: A population-based cohort analysis was performed, comparing the risk of long-term GI morbidity (up to the age of 18 years) in children delivered at term according to their BW. The study included all term deliveries occurring between 1991 and 2014 at a single regional tertiary medical center. Multiple gestations and fetuses with congenital malformations were excluded. BW was subdivided into: small for gestational age (small for gestational age (SGA) - BW ≤ 5th centile), appropriate for gestational age (AGA -5th centile < BW < 95th centile), and large for gestational age (LGA - BW ≥95th centile). Hospitalizations up to the age of 18 years involving GI morbidity were evaluated, using a predefined set of ICD-9 codes, as recorded in the hospital files. A Kaplan-Meier survival curve was used to compare cumulative GI morbidity incidence. A Cox proportional hazards model was constructed to control for confounders.

Results: During the study period, 225,600 term singleton deliveries met the inclusion criteria. Of them, 4.6% (n = 10,415) were SGA and 4.3% (n = 9796) were LGA. During the 18-years follow-up period, 11,791 (5.2%) children were hospitalized with GI morbidity. Hospitalizations were significantly more common in the SGA group, as compared with the AGA and LGA groups (6.6 versus 5.2 versus 4.5%, respectively, p < .001) Specifically, inflammatory bowel disease, celiac, hernia, hepatitis, and cholecystitis, were more common in the SGA group. The Kaplan-Meier survival curve demonstrated a significantly higher cumulative incidence of gastrointestinal morbidity in the SGA group (log rank p < .001). In the Cox proportional hazards model, controlled for relevant clinical confounders, SGA BW was found to be an independent risk factor for long-term GI morbidity (adjusted HR = 1.23, 95%CI 1.14-1.33, p < .001).

Conclusions: SGA offspring are at an increased and independent risk for long-term pediatric GI morbidity.