K.I. Holden1, A.H. Ebanks1, G.B. Mychaliska2, R.A. Stark3, K.P. Lally1, N. Patel4, C.C. Miller5, M.T. Harting1 1University Of Texas Health Science Center At Houston, Pediatric Surgery, Houston, TX, USA 2University Of Michigan, Pediatric Surgery, Ann Arbor, MI, USA 3Seattle Children’s Hospital, Pediatric Surgery, Seattle, WA, USA 4Royal Hospital for Children, Neonatology, Glasgow, SCOTLAND, United Kingdom 5University Of Texas Health Science Center At Houston, Cardiovascular Surgery, Houston, TX, USA
Introduction:
Infants with congenital diaphragmatic hernia (CDH) may require extended postnatal hospitalizations for management of complex cardiopulmonary disease. Many remain inpatients for approximately 2 months on average, however, some will require stays beyond 3 months. Factors associated with a patient’s length of stay (LOS), along with outcomes of those with a prolonged LOS, remain unresolved. Our objective was to identify factors associated with, and outcomes among, CDH patients with an inpatient stay >3 months.
Methods:
Using the CDH study group (CDHSG) registry (2007-2022), patients were categorized by LOS (≤90 days as anticipated LOS (aLOS), >90 days as prolonged (pLOS)). pLOS cutoff was chosen based on clinical experience, data analysis, and prior publication. Demographics, CDHSG defect size, morbidity, and mortality were prospectively collected and retrospectively analyzed with non-parametric and multivariable tests to compare outcomes of length of stay and mortality. We developed two models: predicting pLOS and predicting outcome among pLOS patients.
Results:
Out of 9,212 patients, 7,830 (85.0%) had an aLOS, and 1,382 (15.0%) had a pLOS. In the pLOS group (vs aLOS), 78.9% had C/D defects (vs aLOS: 41.6% had C/D defects), 48.4% received extracorporeal life support (ECLS) (vs aLOS: 23.4%), with a median intubation time of 48 days (vs aLOS: 10 days), 15.10% had major cardiac defects (vs aLOS: 7.9%), and 12.9% had chromosomal anomalies (vs aLOS: 6.6%) (all p<0.001). Overall survival was 73.3%, with pLOS at 86.9% and aLOS at 70.9%. In multivariable analysis, pre-term delivery, low birthweight, presence of a major cardiac anomaly, prolonged intubation time, the use of ECLS, larger defect sizes, neurologic complications, and the need for surgical feeding access were all associated with pLOS (all p<0.01). These factors can be used to calculate the likelihood of a pLOS. Based on the devised scoring system, patients can be stratified into low (1-4 points), intermediate (3-7 points), and high (8-10 points) risk probability groups for pLOS. Once in the pLOS, the only factors significantly associated with mortality were a prolonged intubation time (p≤0.001) and presence of a major cardiac anomaly (p=0.04).
Conclusion:
This study identified factors associated with pLOS and developed a model which predicts pLOS and outcomes among pLOS patients. Factors like pre-term delivery, low birth weight, major cardiac anomalies, large CDH defect size, the use of ECLS, prolonged intubation, neurologic complications, and need for surgical feeding access can be used to predict LOS. These results provide evidence for family counseling and planning in this complex and highly variable patient population.