Background HF recruits alveoli during exhalation, improves gas exchange and prevents volutrauma. However, a positive-pressure first approach was recommended for RF associated with parenchymal lung injury (PLI). The aim was to quantify the effect size (ES) of HF preventing mortality in a broad range of PLI-mediated RF patients. We hypothesized patient age, sex, initial level of lung injury noted by PaO₂/FiO₂ and risk of death assessed by control group mortality (CGM) would predict the ES of HF.

Methods Prospero registered (CRD42022327304) study performed with random or mixed effects meta-analysis to quantify ES of HF on mortality. Meta-regression was used to assess the impact of listed variables on mortality. Bias was assessed with the RoB2 and ROBINS-I.

Results 54 studies compared HF (n=33724) with NIV (n=29209), HF mortality benefit OR 0.7 (95%CI 0.62-0.79), p<0.001, Prediction Interval (PI) 0.44-1.1, RCT OR 0.84 (0.71-0.98), p=0.03, Cohort OR 0.63 (0.53-0.75), p<0.001, Q=5.3, p=0.02. Comparing mixed NIV (OR 0.7, p=0.007) & biPAP alone(OR 0.67, p<0.01), HF was beneficial Q=0.07, p=0.9. Higher age, %Females (%F), and baseline PaO₂/FiO₂ had higher HF mortality. Age HF coefficient (CE) 0.02, p=0.04, %F HF CE 0.005, p=0.05, PaO₂/FiO₂ HF CE 0.002, p=0.02. HF mortality was inverse to CGM CE -0.009, p=0.01.

Conclusion HF had a significant mortality benefit for mixed populations of PLI-patients with RF. The PI highlighted that many patients are likely to benefit while others may not. Patients with RF that may benefit more from HF include younger, males, those with lower baseline PaO₂/FiO₂ or from undefined higher-risk-of-mortality populations.