319 - Improvement of Extrauterine Growth Restriction Within a NICU: Quantitative Assessment Methods and Contributing Factors

Poster Number: 319
Publication Number: 319.129

Vanessa R. Eddy-Abrams, The Mount Sinai Kravis Children's Hospital, New York, NY, United States; Robert S.. Green, Icahn School of Medicine at Mount Sinai, New York, NY, United States; Diana S.. Lee, Icahn School of Medicine at Mount Sinai, New York, NY, United States; Emily Polidoro, The Mount Sinai Kravis Children's Hospital, New York, NY, United States; JING LIN, Icahn School of Medicine at Mount Sinai, New York, NY, United States

Background: Extrauterine growth restriction (EUGR) can have a significant impact on infants' later growth and developmental outcomes. Determining the rate of EUGR and contributing factors may lead to improved outcomes. EUGR is not universally defined. Two definitions are most used: discharge weight < 10^th percentile or change in z-score from birth to discharge < –1.0.

Objective: Determine incidence of EUGR based on two definitions in infants born < 33 weeks' gestation or < 1500g at time of discharge in order compare incidence between EUGR definitions, evaluate parameters for contribution to EUGR and compare incidence with historical cohort. 

Design/Methods: Data was collected from all infants < 33 weeks gestation or < 1500g at birth who were born at, admitted to, and discharged from Mount Sinai Hospital NICU between 1/2015-5/2019. Infants with congenital anomalies, cardiac disease, or who died or were transfer prior to discharge were excluded. Birth and discharge weights were available from a cohort with the same inclusion and exclusion criteria from 2006-2010. EUGR is defined as discharge weight < 10^th percentile or change in z-score from birth to discharge < –1.0. We compared the incidence of EUGR between the two definitions and to the historical cohort. Factors associated with EUGR were examined in the present cohort and analyzed with linear and logistic regression. Statistical significance was evaluated with t-tests, chi-square, and ANOVA where relevant. 

Results: Our sample included 604 neonates. The incidence of EUGR was 32.8% and 22% (for < 10^th percentile and change in z-score < -1, respectively). From 2015 to 2019 infants did have a statistical improvement in change in z-score (Fig. 1). Mean birthweight was comparable between past and current cohorts (1355 vs. 1360g respectively, p-value 0.8). There was no statistical difference in EUGR between the present and past cohort when using < 10th percentile, but a significant decrease in incidence of EUGR when using change in z-score < -1 (Fig. 2). Using a logistic regression model infants with EUGR (using both definitions) were statistically more likely to be female sex, have longer TPN use, and low z-scores at birth (Fig. 3).

Conclusion(s): EUGR continues to be a significant problem but has shown improvement over time when compared to previous data. Neonates who are female, with longer TPN use, and lower z-scores at birth are statically more likely to have EUGR.