Admission volume is associated with mortality of neonatal respiratory failure in emerging neonatal intensive care units

Abstract

Objective: The aim of this survey was to explore the relationship between admission volume and mortality of neonates with hypoxemic respiratory failure (NRF) in emerging neonatal intensive care units (NICUs).

Methods: NRF from 55 NICUs were retrospectively included with death risk as the major outcome. Perinatal comorbidities, underlying disease severity, respiratory support, facility utilization, and economic burden in the early postnatal period were compared among five NICU admission volume categories defined by NRF incidence, with score for neonatal acute physiology perinatal extension II (SNAPPE-II) also assessed as initial severity.

Results: Compared to NICUs with NRF < 50 cases/year, NRF incidence, NRF/NICU, NRF/NICU admissions, and magnitude of ventilator use were several times higher, and mortality rates 20–50% lower, in NICUs of 150–199 and ≥200 cases/year (p < .01), even after adjustment with SNAPPE-II in stratified ranges. Median SNAPPE-II values, which correlated with the death rate of NRF (r = .282, p < .001), were lower in NICUs of 150–199 and ≥200 than in <50, 50–99, and 100–149 categories (13 versus 18, p < .01). NRF mortalities were not correlated with the proportion of very low birth weight patients in each category.

Conclusions: Neonates in NICUs with smaller NRF admission volume and decreased magnitude of ventilator use had a higher risk of death as assessed by SNAPPE-II, which should be targeted in the quality improvement of newly established, resource-limited NICUs.

Keywords:

• Mechanical ventilation
• mortality
• neonatal intensive care unit
• respiratory failure
• score for neonatal acute physiology perinatal extension II

Acknowledgements

This study was based on the data file from the Chinese Collaborative Study Group for Neonatal Respiratory Diseases, and the contribution by all the coinvestigators and research staffs from participating centers are highly appreciated.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Additional information

Funding

This study was supported by grants from NSFC (No. 30611120518), “211” Project (Phase III) and Doctorate Research Program (No. 20090071110061), Ministry of Education, Shanghai Bureau of Health (No. LJ 06038), and China Medical Board of New York (No. 03-786).