Adaptive mechanical ventilation with automated minimization of mechanical power-a pilot randomized cross-over study

Adaptive mechanical ventilation automatically adjusts respiratory rate (RR) and tidal volume (V_T) to deliver the clinically desired minute ventilation, selecting RR and V_T based on Otis' equation on least work of breathing. However, the resulting V_T may be relatively high, especially in patients with more compliant lungs. Therefore, a new mode of adaptive ventilation (adaptive ventilation mode 2, AVM2) was developed which automatically minimizes inspiratory power with the aim of ensuring lung-protective combinations of V_T and RR. The aim of this study was to investigate whether AVM2 reduces V_T, mechanical power, and driving pressure (ΔP_stat) and provides similar gas exchange when compared to adaptive mechanical ventilation based on Otis' equation.

A prospective randomized cross-over study was performed in 20 critically ill patients on controlled mechanical ventilation, including 10 patients with acute respiratory distress syndrome (ARDS). Each patient underwent 1 h of mechanical ventilation with AVM2 and 1 h of adaptive mechanical ventilation according to Otis' equation (adaptive ventilation mode, AVM). At the end of each phase, we collected data on V_T, mechanical power, ΔP, PaO₂/FiO₂ ratio, PaCO₂, pH, and hemodynamics.

Comparing adaptive mechanical ventilation with AVM2 to the approach based on Otis' equation (AVM), we found a significant reduction in V_T both in the whole study population (7.2 ± 0.9 vs. 8.2 ± 0.6 ml/kg, p <  0.0001) and in the subgroup of patients with ARDS (6.6 ± 0.8 ml/kg with AVM2 vs. 7.9 ± 0.5 ml/kg with AVM, p <  0.0001). Similar reductions were observed for ΔP_stat (whole study population: 11.5 ± 1.6 cmH₂O with AVM2 vs. 12.6 ± 2.5 cmH₂O with AVM, p <  0.0001; patients with ARDS: 11.8 ± 1.7 cmH₂O with AVM2 and 13.3 ± 2.7 cmH₂O with AVM, p = 0.0044) and total mechanical power (16.8 ± 3.9 J/min with AVM2 vs. 18.6 ± 4.6 J/min with AVM, p = 0.0024; ARDS: 15.6 ± 3.2 J/min with AVM2 vs. 17.5 ± 4.1 J/min with AVM, p = 0.0023). There was a small decrease in PaO₂/FiO₂ (270 ± 98 vs. 291 ± 102 mmHg with AVM, p = 0.03; ARDS: 194 ± 55 vs. 218 ± 61 with AVM, p = 0.008) and no differences in PaCO₂, pH, and hemodynamics.

Adaptive mechanical ventilation with automated minimization of inspiratory power may lead to more lung-protective ventilator settings when compared with adaptive mechanical ventilation according to Otis' equation.

The study was registered at the German Clinical Trials Register ( DRKS00013540 ) on December 1, 2017, before including the first patient.