308 - Accounting for Mortality as a Competing Outcome Suggests that Small for Gestational Age Birth Weight May Protect Preterm Infants from Severe Intraventricular Hemorrhage

Publication Number: 4302.308

Miles Bomback, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Samantha Pickell, Columbia University Vagelos College of Physicians and Surgeons, New York City, NY, United States; Aashna Banerjee, Middlebury College, Middlebury, VT, United States; Alex J. Lyford, Middlebury College, WEYBRIDGE, VT, United States; Veeral Tolia, Pediatrix, Dallas, TX, United States; Reese H.. Clark, Duke University School of Medicine, Marietta, SC, United States; Faith Kim, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Thomas Hays, Columbia University Irving Medical Center, New York, NY, United States

Background: Severe intraventricular hemorrhage is a devastating complication of preterm birth. Earlier gestational age increases risk and antenatal corticosteroids are protective. The role of low birth weight is poorly defined. Recent studies found lower risk in small for gestational age infants and following fetal growth restriction. However, conflicting reports exist regarding this protection, in part due to mortality as a competing risk.

Objective: To determine if low birth weight is associated with reduced risk of severe intraventricular hemorrhage when accounting for mortality and antenatal corticosteroid exposure.

Design/Methods: This retrospective observational study utilized the Pediatrix Clinical Data Warehouse. The study included 64,661 infants born from 23 and 0/7 to 33 and 6/7 weeks' gestation from 2000 to 2020. Competing risk analysis was used to account for mortality as a competing outcome. Infants were stratified by antenatal corticosteroid exposure.

Results: Competing risk analysis demonstrated a lower risk (95% confidence interval) of severe intraventricular hemorrhage in small for gestational age (3.6% (3.3%, 4.0%)) versus appropriate for gestational age (4.7% (4.6%, 4.9%)) infants. As normalized birth weight increased from the first decile, risk of severe intraventricular hemorrhage increased by 0.7% per decile (p = 1.8 × 10-5). Risk of severe intraventricular hemorrhage in first decile birth weight infants unexposed to antenatal corticosteroids (3.9% (2.7%, 5.1%)) approximated that of median birth weight infants exposed to antenatal corticosteroids (4.1% (3.4%, 4.7%)).

Conclusion(s): Small for gestational age birth weight may protect against severe intraventricular hemorrhage. This protection may be mediated by the same mechanism(s) as antenatal corticosteroids. These findings can inform improved risk stratification and counseling, and may lead to improved strategies to prevent severe intraventricular hemorrhage.

Table 1. Characteristics of Cohort


Figure 1. sIVH in a competing risk model accounting for mortality.
Figure 1. sIVH in a competing risk model accounting for mortality. Infants were grouped by birth weight according to Fenton criteria as small for gestational age or appropriate for gestational age. The risk of sIVH was determined for both groups using a competing risk model, accounting for mortality as a competing outcome, and plotted with 95% confidence intervals. Small for gestational age infants had a lower cumulative risk of sIVH when accounting for mortality as a competing outcome (A). Antenatal corticosteroid exposure was associated with lower cumulative sIVH risk for small for gestational age and appropriate for gestational age infants (B).

Figure 2. Cumulative risk of severe intraventricular hemorrhage by birth weight decile.
Figure 2. Cumulative risk of severe intraventricular hemorrhage by birth weight decile. The cumulative risk of severe intraventricular hemorrhage at 30 days was determined using a competing risk model, treating mortality as a competing outcome. This risk was determined for infants by birth weight decile (normalized for gestational age and sex). Infants large for gestational age and in the 10th decile were not included in the analysis. Infants were stratified on the basis of exposure to antenatal corticosteroids. Infants in the 1st decile birth weight exhibited the lowest cumulative risk of sIVH, which converged for infants exposed and unexposed to antenatal corticosteroids. Cumulative risk of sIVH increased with greater birth weight decile for infants exposed and unexposed to antenatal corticosteroids, remaining lower for those exposed.