Background
Multiple noninvasive respiratory support (NRS) modalities are used for postextubation support in preterm neonates. Seven NRS modalities were compared—constant flow continuous positive airway pressure (CPAP) (CF‐CPAP) (bubble CPAP; ventilator CPAP), variable flow CPAP (VF‐CPAP), high flow nasal cannula (HFNC), synchronized noninvasive positive pressure ventilation (S‐NIPPV), nonsynchronized NIPPV (NS‐NIPPV), bilevel CPAP (BiPAP), noninvasive high‐frequency oscillation ventilation (nHFOV).
Design
Systematic review and network meta‐analysis (NMA) using the Bayesian random‐effects approach. MEDLINE, EMBASE, CENTRAL, WHO‐ICTRP were searched.
Main Outcome Measure
Requirement of invasive mechanical ventilation within 7 days of extubation.
Results
A total of 33 studies with 4080 preterm neonates were included. S‐NIPPV, NS‐NIPPV, nHFOV, and VF‐CPAP were more efficacious in preventing reintubation than CF‐CPAP (risk ratio [RR] [95% credible intervals {CrI}]: 0.22 [0.12, 0.35]; 0.44 [0.27, 0.67]; 0.42 [0.18, 0.81]; 0.73 [0.52, 0.99]). Surface under the cumulative ranking curve (SUCRA) value ranked S‐NIPPV to be the best postextubation intervention (SUCRA: 0.98). S‐NIPPV was more effective than NS‐NIPPV, BiPAP, VF‐CPAP, and HFNC (RR [95% CrI]: 0.52 [0.24, 0.97]; 0.32 [0.14, 0.64]; 0.30 [0.16, 0.50]; 0.24 [0.12, 0.41]). NS‐NIPPV resulted in lesser reintubation compared to VF‐CPAP and HFNC (RR [95% CrI]: 0.61 [0.36, 0.97]; 0.49 [0.27, 0.80]). BiPAP, VF‐CPAP, and HFNC had comparable efficacies. The overall quality of evidence was very low to moderate.
Conclusion
Results of this NMA indicate that S‐NIPPV might be the most effective and CF‐CPAP the least effective NRS modality for preventing extubation failure.