Compare the impact of initial extubation to positive airway pressure versus high-flow nasal cannula on postoperative outcomes in neonates and infants after congenital heart surgery.
Retrospective cohort study with propensity-matched analysis.
Cardiac ICU within a tertiary care children’s hospital.
Patients less than 6 months old initially extubated to either high-flow nasal cannula or positive airway pressure after cardiac surgery with cardiopulmonary bypass were included (July 2012 to December 2015).
Of 258 encounters, propensity matching identified 49 pairings of patients extubated to high-flow nasal cannula versus positive airway pressure. Extubation failure was 12% for all screened encounters. After matching, there was no difference in extubation failure rate between groups (positive airway pressure 16% vs high-flow nasal cannula 10%; p = 0.549). However, compared with high-flow nasal cannula, patients initially extubated to positive airway pressure experienced greater resource utilization: longer time to low-flow nasal cannula (83 vs 28 hr; p = 0.006); longer time to room air (159 vs 110 hr; p = 0.013); and longer postsurgical hospital length of stay (22 vs 14 d; p = 0.015).
In this pediatric cohort, primary extubation to positive airway pressure was not superior to high-flow nasal cannula with respect to prevention of extubation failure after congenital heart surgery. Compared with high-flow nasal cannula, use of positive airway pressure was associated with increased hospital resource utilization. Prospective initiatives aimed at establishing best clinical practice for postoperative noninvasive respiratory support are needed.
1Division of Critical Care Medicine, Department of Pediatrics, University of Alabama at Birmingham School of Medicine, Birmingham, AL.
2Department of Pediatrics, Section of Cardiology, University of Cincinnati College of Medicine, Cardiac Intensive Care Unit, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH.
3University of Alabama at Birmingham School of Medicine, Birmingham, AL.
4Department of Anesthesiology, Wake Forest Baptist Medical Center, Winston-Salem, NC.
5Department of Biostatistics, University of Alabama at Birmingham, Birmingham, AL.
6Division of Critical Care Medicine, Department of Pediatrics, University of Tennessee College of Medicine Chattanooga, Chattanooga, TN.
7Division of Cardiology, Department of Pediatrics, Section of Cardiac Critical Care Medicine, University of Alabama at Birmingham School of Medicine, Birmingham, AL.
*See also p. 195.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/pccmjournal).
Support for this project was provided, in part, through Division of Pediatric Cardiology funds.
Mr. King and Dr. Gans were supported in part by the National Institutes of Health/T35 grant (T35HL07473). The remaining authors have disclosed that they do not have any potential conflicts of interest.
Address requests for reprints to: Santiago Borasino, MD, MPH, Division of Cardiology, Department of Pediatrics, Section of Cardiac Critical Care Medicine, University of Alabama at Birmingham School of Medicine, Birmingham, AL 35233. E-mail: firstname.lastname@example.org