216 - Non-invasive respiratory support in preterm infants as a primary mode of respiratory support: A systematic review and network meta-analysis

Poster Number: 216
Publication Number: 216.3163

Background: Various forms of non-invasive respiratory support are increasingly used to reduce the need for mechanical ventilation and associated lung injury in preterm neonates. However, their relative efficacy remains uncertain.

Objective: To assess the comparative efficacy of different forms of non-invasive respiratory support through a network meta-analysis of randomized controlled trials.

Design/Methods: A systematic search was conducted in multiple databases. Peer-reviewed publications of randomized trials comparing any 2 (or more) forms of non-invasive support in preterm neonates < 37 weeks’ GA were eligible. The studied modes included: continuous positive airway pressure (CPAP), nasal intermittent positive pressure ventilation (NIPPV), non-invasive neurally adjusted ventilatory assist (NIV-NAVA), non-invasive high frequency ventilation (NIHFV), high flow nasal cannula (HFNC), Biphasic CPAP (BiPAP), or High-CPAP (H-CPAP). Following data extraction, network meta-analyses were performed using a Bayesian approach, with CPAP as the reference mode. Primary outcomes were: (i) treatment failure; (ii) receipt of intubation; and (iii) moderate-severe bronchopulmonary dysplasia. Network risk ratios (nRR) with 95% credible intervals (CrI) were estimated using the random-effects model; relative rankings (with 95% CrI) were generated using surface under the cumulative ranking (SUCRA) values. Risk of bias was assessed using the Cochrane risk of bias tool; certainty of evidence (CoE) was assessed using the GRADE framework, which incorporated risk of bias, reporting bias, indirectness, imprecision, heterogeneity and incoherence.

Results: 47 studies comprising 6,225 participants were included. For treatment failure, NIPPV (nRR 0.63, 95% CrI 0.48-0.82, CoE: very low) and NIHFV (nRR 0.41, 95% CrI 0.23-0.70, CoE: moderate) showed a significant reduction compared to CPAP (Table 1). Similarly, for receipt of intubation, NIPPV (nRR 0.61, 95% CrI 0.50-0.76, CoE: low) and NIHFV (nRR 0.47, 95% CrI 0.29-0.73; CoE: moderate) showed a significant reduction compared to CPAP (Table 2). NIHFV was estimated to be the most effective mode for both outcomes. None of the non-invasive respiratory support modes showed any difference in moderate-severe bronchopulmonary dysplasia compared to CPAP (Table 3).

Conclusion(s): Available evidence suggests that NIPPV (very low to low certainty) and NIHFV (moderate certainty) may be superior to CPAP with respect to treatment failure and need for intubation. However, for the risk of moderate-severe BPD, there was no difference between any of the modes (low to moderate certainty) versus CPAP.