Respiratory mechanics modifies the impact of ventilator settings on clinical outcome in patients with acute brain injury.

Abstract

To investigate whether respiratory system elastance (E_RS) modifies the associations of tidal volume (V_T) and respiratory rate with clinical outcomes in mechanically ventilated patients with acute brain injury. In this post hoc analysis of the VENTIBRAIN study, E_RS was computed as the ratio of driving pressure (ΔP) to V_T/PBW. Effect modification by E_RS on outcome associations was assessed. In 1158 patients, V_T was similar across E_RS tertiles (median 7.9, 7.6 and 7.1 mL/kg), resulting in a marked ΔP gradient across tertiles (median 5, 9, and 12 cmH₂O). Higher V_T was associated with lower ICU mortality in patients with low E_RS (OR 0.52, 95%CI 0.36-0.74) but not in those with high E_RS (OR 1.16, 95% CI 0.89-1.51; P_int < 0.001). Higher respiratory rate was associated with worse outcomes, particularly at low E_RS, with the association attenuating at higher E_RS. The Bayesian optimal V_T/PBW decreased continuously with E_RS, from 11.4 mL/kg at E_RS 0.5 cmH₂O/(mL/kg) (≈ ΔP 6 cmH₂O) to 4.4 mL/kg at E_RS 2.5 cmH₂O/(mL/kg) (≈ ΔP 11 cmH₂O), while the optimal respiratory rate increased from 15 to 19 breaths/min. At constant alveolar ventilation, 1-cmH₂O ΔP increase had comparable prognostic impact to ≈ 3 breaths/min respiratory rate increases. These results were consistent irrespective of ARDS diagnosis. The impact of ventilator settings is modulated by E_RS. A physiological framework for V_T adjustment could consider ΔP along with the respiratory rate required to maintain isocapnia, with outcome effects governed by a trade-off of ≈ 3 breaths/min per 1-cmH₂O change in ΔP.