Early High-Frequency Oscillatory Ventilation in Pediatric Acute Respiratory Failure. A Propensity Score Analysis. Am J Respir Crit Care Med 2016 Mar 01;193(5):495-503
Date
10/23/2015Pubmed ID
26492410Pubmed Central ID
PMC4824923DOI
10.1164/rccm.201507-1381OCScopus ID
2-s2.0-84988876269 (requires institutional sign-in at Scopus site) 76 CitationsAbstract
RATIONALE: The use of high-frequency oscillatory ventilation (HFOV) for acute respiratory failure in children is prevalent despite the lack of efficacy data.
OBJECTIVES: To compare the outcomes of patients with acute respiratory failure managed with HFOV within 24-48 hours of endotracheal intubation with those receiving conventional mechanical ventilation (CMV) and/or late HFOV.
METHODS: This is a secondary analysis of data from the RESTORE (Randomized Evaluation of Sedation Titration for Respiratory Failure) study, a prospective cluster randomized clinical trial conducted between 2009 and 2013 in 31 U.S. pediatric intensive care units. Propensity score analysis, including degree of hypoxia in the model, compared the duration of mechanical ventilation and mortality of patients treated with early HFOV matched with those treated with CMV/late HFOV.
MEASUREMENTS AND MAIN RESULTS: Among 2,449 subjects enrolled in RESTORE, 353 patients (14%) were ever supported on HFOV, of which 210 (59%) had HFOV initiated within 24-48 hours of intubation. The propensity score model predicting the probability of receiving early HFOV included 1,064 patients (181 early HFOV vs. 883 CMV/late HFOV) with significant hypoxia (oxygenation index ≥ 8). The degree of hypoxia was the most significant contributor to the propensity score model. After adjusting for risk category, early HFOV use was associated with a longer duration of mechanical ventilation (hazard ratio, 0.75; 95% confidence interval, 0.64-0.89; P = 0.001) but not with mortality (odds ratio, 1.28; 95% confidence interval, 0.92-1.79; P = 0.15) compared with CMV/late HFOV.
CONCLUSIONS: In adjusted models including important oxygenation variables, early HFOV was associated with a longer duration of mechanical ventilation. These analyses make supporting the current approach to HFOV less convincing.
Author List
Bateman ST, Borasino S, Asaro LA, Cheifetz IM, Diane S, Wypij D, Curley MA, RESTORE Study InvestigatorsAuthor
Rainer G. Gedeit MD Associate Chief Medical Officer in the Children's Administration department at Children's WisconsinMESH terms used to index this publication - Major topics in bold
AdolescentChild
Child, Preschool
Female
High-Frequency Ventilation
Humans
Hypoxia
Infant
Infant, Newborn
Male
Odds Ratio
Propensity Score
Proportional Hazards Models
Prospective Studies
Randomized Controlled Trials as Topic
Respiration, Artificial
Respiratory Distress Syndrome, Newborn
Respiratory Insufficiency
Time Factors