Gastrostomy Tube Placement in Periviable Infants Surviving to Discharge

Background: There is increased survival in infants born between 22-25 weeks’ gestation. However, these infants experience significant morbidity. Need for gastrostomy tube placement prior to discharge is not a rare outcome for these infants. There currently is wide variation in practice. We sought to determine the percentage of periviable infants requiring gastrostomy tube placement prior to discharge at a single institution and identify associated clinical and demographic factors.

Methods: We conducted a single-center retrospective chart review of all live-born infants between 22-25 weeks’ completed gestation from 2013-2019. Infants not actively resuscitated and those with congenital and/or genetic anomalies were excluded from analysis. Chi-square test of independence and independent sample t-test were used to assess maternal and infant factors associated with need for gastrostomy tube. Factors associated with need for gastrostomy tube were incorporated into a multivariable logistic regression model.

Results: Two hundred and twenty-nine infants met criteria for inclusion and 144 (62.9%) survived until discharge. Of those surviving, 19% required gastrostomy tube placement. Self-identified maternal race/ethnicity (P = 0.01), periventricular leukomalacia (P= 0.03), intraventricular hemorrhage (P= 0.03), and chronic lung disease (P = 0.02) were associated with gastrostomy tube placement in this cohort. Maternal Black race remained significant in our multivariable logistic regression model, with an odds ratio of 5.0 [95% C.I. 1.8-13.6] (P < 0.01).

Conclusion: A high percentage of periviable infants surviving until discharge require gastrostomy tube placement at our institution. Identification of patient-specific factors may help predict which infants are at highest risk for this outcome and these data are useful when counselling families. The observed association between self-reported maternal race and gastrotomy tube placement requires further investigation.