Meta-analysis shows optimal FiO2 for resuscitation of preterm infants remains equivocal

1. Systematic review with meta-analysis shows no statistically significant difference in morbidity and mortality outcomes when comparing use of low versus high FiO2 during initial resuscitation of preterm infants (<35 weeks gestational age).

2. Subgroup analyses of data on infants <32 weeks and ≤28 weeks also showed no difference in morbidity and mortality outcomes.

Evidence Rating Level: 1 (Excellent)

Study Rundown: Current literature suggests high FiO2 is particularly toxic to the incompletely developed organs of preterm infants, but no clear evidence exists regarding optimal FiO2 levels for resuscitation of these infants at birth. This systematic review with meta-analysis was achieved in collaboration with the International Liaison Committee on Resuscitation (ILCOR), which conducts a continuous literature review to determine the best evidence-based guidelines for resuscitation. This review included studies evaluating the morbidity and mortality effects of high versus low initial FiO2 for respiratory support of premature infants <35 weeks. Pooled study results showed no statistically significant difference in short-term mortality (STM), long-term mortality (LTM), long-term neurodevelopment impairment (NDI), or other major markers of morbidity when starting with low versus high FiO2 during resuscitation at birth. Subgroup analyses of very preterm (28 to 32 weeks gestational age) and extremely preterm (≤28 weeks gestational age) also showed no significant differences in outcomes based on level of FiO2 use at birth.

Of note, due to inconsistency, imprecision, and risk of bias, most study outcomes among studies included in this review received a “very low” GRADE (grading of recommendations) confidence score, which makes the certainty of point estimates low or very low. There was also high heterogeneity in several of the analyses, however this was attributed to results of one randomized controlled trial (RCT; the To2rpido study). Results from this systematic review and meta-analysis therefore suggest unclear evidence regarding the optimal starting FiO2 for resuscitation of preterm infants <35 weeks gestation. Updated ILCOR recommendations will be published in 2019.

Click to read the study, published today in Pediatrics

Relevant reading: Part 7: neonatal resuscitation: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations

In-depth [systematic review with meta-analysis]: This review included studies published between January 1980 and August 2018 (4 observational cohorts, 10 RCTs with 2 long-term follow-ups) comparing the effects of low (≤0.5) versus high (>0.5) FiO2 on major preterm morbidity (neurodevelopmental impairment, major intraventricular hemorrhage, necrotizing enterocolitis, etc), as well as short- and long-term mortality. Collectively, these 16 studies included data on 5697 patients (1007 from RCTs). Pooled results from the 10 RCTs reporting on showed no statistically significant difference in STM (RR 0.83, 95% CI 0.50 to 1.37, I² = 18%*) based on use of low versus high FiO2 use for resuscitation at birth. Similarly, there was no significant difference in LTM based on analysis of the 3 RCTs with data on this outcome (RR 1.05, 95% CI 0.32-3.39, I² = 79%). However, due to the low number of RCTs looking at this outcome and the high heterogeneity rating (79%), LTM data from 2 observational cohort studies was included in the analysis, after which results indicated a significantly beneficial effect of starting birth resuscitations with low vs high FiO2 at birth (RR 0.77, 95% CI 0.59-0.99, I²= 6%). Three RCTs and 2 observational studies reported data on long-term NDI. Analyses of data from the RCTs alone (389 patients) and RCTs combined with observational studies (1320 patients) showed no statistically significant difference in this outcomes when using low versus high FiO2 (RR 1.14, 95% CI 0.78-1.67, I² = 0% and RR 0.89, 95% CI 0.66-1.20, I² = 59% respectively). No statistically significant difference in significant morbidity outcomes using high versus low FiO2 were observed in the meta-analysis. Results from subgroup analyses on infants ≤32 and ≤28 weeks also showed no statistically significant difference in outcomes.

*I² is a measure of variance in a meta-analysis that is attributable to study heterogeneity.

Image: PD

©2018 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.