46 - Relationship between postnatal steroid exposure and growth, body composition, and early neurodevelopment in very low birth weight preterm infants at term and 4 months corrected age

Publication Number: 2043.46

Sara Jones, University of Minnesota Masonic Children's Hospital, Minneapolis, MN, United States; Scott Lunos, University of Minnesota, Minneapolis, MN, United States; Emily M. Nagel, University of Minnesota Department of Pediatrics, Minneapolis, MN, United States; Sara Gonia, University of Minnesota, Minneapolis, MN, United States; Megan Paulsen, Children's Hospitals and Clinics of Minnesota - St. Paul, Eden Prairie, MN, United States; Sara E. Ramel, University of Minnesota Masonic Children's Hospital, North Oaks, MN, United States

Background: Postnatal systemic steroids have been associated with adverse growth and neurodevelopmental outcomes in preterm infants. Event-related potential (ERP) measures auditory discrimination and can be administered in early infancy. Our group has shown strong associations between fat free mass and neurodevelopment via ERP and standard assessments. We hypothesized that systemic steroid exposure would be associated with increased adiposity, and suboptimal growth and auditory discrimination at term and 4 months.

Objective: To evaluate the relationship between postnatal systemic steroid exposure in preterm infants and growth, body composition, and ERP outcomes at term and 4 months corrected age.

Design/Methods: 79 preterm infants (born < 32 weeks' gestation) were enrolled in a clinical trial evaluating differences in standard vs enhanced parenteral nutrition the first week of life. Systemic steroid use (methylprednisolone, dexamethasone, and hydrocortisone) was converted to a total dexamethasone equivalent (mg/kg) throughout their NICU stay. Outcome measures: z-scores for anthropometrics and body composition measured by air displacement plethysmography, and ERP P3a component using auditory oddball task. Independent t-tests and linear regression analysis were used. Adjustments were made for gestational age, severe IVH, nutrition, trial study group, and sex.

Results: Mean postnatal steroid exposure was 4.26 (+/- 4.14) mg/kg dexamethasone equivalents, with n=43 (54.43%) exposed to systemic steroids. Systemic steroid exposed infants had lower gestational ages, antenatal steroid exposure, and nutrition, but more inhaled steroid use and NICU morbidities (Table 1). Systemic steroid exposed infants had lower weight (p=0.013), length (p < 0.001), OFC (p=0.001), and fat free mass (FFM) (p=0.003) z-scores at term; all resolved by 4 months. A difference in Deviant-Standard Latency Difference was seen at term and 4 months (p=0.049; p=0.014) (Table 2). There was a negative association between cumulative steroid exposure and z-scores for length at term and 4 months (β=-0.27(0.07): p< 0.001; -0.26(0.12); p=0.046), and weight (β=-0.21(0.07); p=0.004), OFC (β=-0.16(0.08); p=0.039) and FFM z-scores (β=-0.30(0.09); p=0.002), and Deviant-Standard Latency Difference (β=-14.75(6.74); p=0.03) at term only (Table 3).

Conclusion(s): Postnatal systemic steroid exposed preterm infants had lower fat free mass without higher adiposity, and suboptimal growth and auditory discrimination, potentially indicative of immaturity in neural discrimination, at term. At 4 months corrected age only suboptimal auditory discrimination persisted.

Table 1. Group Characteristics
Data shown as N (percent) or Median [IQR] unless otherwise indicated. BPD: Bronchopulmonary dysplasia, defined by oxygen requirement at 36 weeks post menstrual age. IVH: interventricular hemorrhage. NEC: necrotizing enterocolitis, defined by radiographic evidence and/or clinical suspicion. OFC: occipital-frontal circumference.

Table 2. Growth, Body Composition, and ERP Outcomes by Systemic Steroid Exposure at Term and 4 months Corrected Age
Data presented as Mean (SD). † P-values are from multiple linear regression/ANOVA models comparing outcomes by systemic steroid exposure while adjusting for gestational age, severe IVH, nutrition throughout the entire hospitalization, and randomization group. Additional adjustment for sex was done for ERP outcomes only. Z-scores account for corrected age and sex

Table 3. Associations Between Growth, Body Composition, and ERP and Total Dexamethasone Exposure at Term and 4 months Corrected Age
†Estimates (SE) and p-values are from multiple linear regression models testing the association between outcomes (growth, body composition, and ERP components) and total dexamethasone while adjusting for gestational age, severe IVH, nutrition throughout hospitalization, and randomization group. Additional adjustment for sex was included for ERP outcomes only. Z-scores correct for post-menstrual age and sex.