# The Association Between Perinatal Pharmacologic Treatments and Spontaneous Intestinal Perforation in Extremely Preterm Infants: A Propensity Score Matching Study

## TL;DR

This study finds that certain medications given to extremely low-birthweight preterm infants are linked to a higher risk of intestinal perforation.

## Contribution

The study is the first to use propensity score matching to identify pharmacologic risks for intestinal perforation in extremely preterm infants.

## Key findings

- Hydrocortisone use was significantly higher in infants with intestinal perforation compared to controls.
- Combined inotropic agents were also more commonly used in infants with intestinal perforation.
- Logistic regression confirmed hydrocortisone and combined inotropes as independent risk factors for intestinal perforation.

## Abstract

Background: The impact of perinatal pharmacologic agents on spontaneous intestinal perforation (SIP) in extremely low-birthweight (ELBW, <1000 g) preterm infants remains inconclusive based on findings from retrospective cohort or case–control studies. This study aims to address this uncertainty by using propensity score matching (PSM) to reduce bias. Methods: We retrospectively reviewed ELBW infants in our unit between 2014 and 2023 to identify SIP cases. Confirmed through medical notes, surgical consultation, and author review, each SIP case was matched at a 1:3 ratio using propensity scores on factors including the gestational age (GA), birthweight, gender, and birth year. Pharmacologic agents commonly given antenatally and postnatally were analyzed. Only medications that were started 24 h before the onset of SIP or the corresponding age (PSM-controls) were included. Results: A total of 858 ELBW infants were reviewed, 28 SIP cases (GA 25.3 ± 2.1 weeks, BW 735 ± 167 g) were identified, and 84 PSM-controls were matched. The SIP cases received hydrocortisone (25% (7/28) vs. 9.5% (8/84), p = 0.037) and combined inotropic agents (17.9% (5/28) vs. 2.4% (2/84), p = 0.020) to a significantly greater extent. No differences were observed in the use of other medications. In logistic regression, the use of hydrocortisone and combined inotropes remained independent risks for SIP, with ORs (95% CIs) of 3.4 (1.1–10.9) and 2.1 (1.2–3.8), respectively. Conclusions: This first PSM-based study supported postnatal hydrocortisone and combined inotrope use as independent risks for SIP in ELBW infants. Clinicians should be aware of these risks and remain vigilant for SIP when administering hydrocortisone and inotropes.