88 - Factors Associated with Prolonged Hospital Stay among Preterm Infants referred to Level IV Neonatal Intensive Care Unit (NICU)

Publication Number: 2084.88

Kristy Patel, Children's Hospital of Michigan, Troy, MI, United States; Girija Natarajan, Central Michigan University College of Medicine, Detroit, MI, United States; Abigail Cooley, Children's Hospital of Michigan, Ypsilanti, MI, United States

Background: Previous prediction models for length of stay (LoS) for preterm infants have been based mostly on perinatal risk factors. A small subset of preterm infants is referred to the highest (level IV) level NICUs for escalation of therapies, specialist services and surgeries. Data on the factors associated with prolonged LoS in this population are lacking.

Objective: Among preterm ( < 30 weeks gestational age-GA) infants referred to a level IV NICU, we examined the association between demographics, birth and clinical characteristics on admission, the initial week after transfer and at 36 weeks postmenstrual age (PMA), to predict LoS.

Design/Methods: This ongoing single-center retrospective study included preterm ( < 30 weeks GA) infants admitted to a Level IV NICU from 2011 to 2023 from a level III NICU until day 28 of age. We excluded infants with congenital malformations or genetic syndromes, who were admitted for palliative care or were transferred back or died. Infants were categorized as prolonged LoS (≥44 weeks PMA) vs. others. Demographics, clinical characteristics at admission, 1 week after admission and at 36 weeks PMA and discharge outcomes were compared using chi-square or t test, as appropriate. Linear regression was used to identify factors associated with LoS.

Results: In our cohort (n=131), 28 (21.4%) infants had a prolonged LoS and 103 (78.6%) did not. Infants with a prolonged LoS had lower mean (SD) birth weight, GA and higher surfactant doses (p < 0.05) (Table 1). At admission, mean (SD) weight was lower, and a higher proportion received pressors, ventilator support and inhaled nitric oxide. Mean (SD) FiO2 and mean airway pressure (MAP) were also significantly higher. Differences in proportions receiving invasive ventilator support persisted at 7 days after admission and at 36 weeks PMA. Comparing morbidities and outcomes at discharge in the two groups, infants with a prolonged LoS had higher mean (SD) days on oxygen, ventilator, parenteral nutrition (PN) and more surgeries (Table 2). Rates of laser surgery for retinopathy and ductal surgical or transcatheter closure were also significantly higher. This group was more likely to receive feeding tube or PN support at discharge home. On linear regression, PN duration was independently associated with LoS (Table 3).

Conclusion(s): Among preterm infants referred to and discharged from a level IV NICU, prolonged PN support was independently associated with LoS. Development of a prediction model for LoS in this uniquely vulnerable subset would allow better resource planning and parental counseling.

Table 1
Comparison of clinical characteristics at birth, admission, day 7 after admission to the level IV NICU and at 36 weeks PMA in groups with and without prolonged LoS

Table 2
Comparison of short-term outcomes in groups with and without prolonged LoS

Table 3
Linear regression analysis to examine association of factors with LoS