Non-invasive respiratory support in preterm infants as primary mode: a network meta-analysis.

Summary

Across 61 trials (7554 neonates), NIPPV and NIHFV may reduce treatment failure and need for intubation compared with CPAP or HFNC, though certainty is low to very low; effects on moderate-severe CLD appear minimal. Evidence is limited for infants <28 weeks’ GA, and many trials did not match mean airway pressures across modes.

Key Findings

• NIPPV reduced treatment failure versus CPAP (network RR 0.63, 95% CrI 0.48–0.82; very low certainty).
• NIHFV reduced treatment failure versus CPAP (network RR 0.41, 95% CrI 0.23–0.69; low certainty).
• Little to no effect on moderate-severe chronic lung disease across modes; evidence generally very uncertain.
• Most studies did not compare modes at equivalent mean airway pressures; infants <28 weeks’ GA were under-represented.

Clinical Implications

Consider NIPPV or NIHFV as first-line non-invasive support over CPAP/HFNC to reduce early failure/intubation where expertise and equipment are available, while recognizing low-certainty evidence and ensuring equivalent mean airway pressure settings across modes.

Limitations

• Low to very low certainty due to within-study bias, imprecision, and incoherence.
• Non-equivalent mean airway pressures across modes in most trials; paucity of data for GA <28 weeks.

Future Directions

Head-to-head RCTs matching mean airway pressures are needed, especially in infants <28 weeks’ GA, with standardized failure criteria and long-term outcomes.