Digital Twins to Evaluate the Risk of Ventilator-Induced Lung Injury During Airway Pressure Release Ventilation...

To use digital twins constructed based on data from patients with acute respiratory distress syndrome (ARDS) to calculate all key indices of ventilator-induced lung injury (VILI) during airway pressure release ventilation (APRV), and to compare them with corresponding values obtained during pressure-controlled ventilation (PCV).

Digital twins were created by matching a high-fidelity cardiopulmonary simulation model to each patient's data.

Interdisciplinary Collaboration in Systems Medicine Research Network.

A dataset consisting of pairs of ventilator settings and arterial blood gases for 98 patients with ARDS receiving PCV.

VILI indices were calculated for each recorded PCV datapoint, and for typical APRV settings in fixed and time-controlled adaptive modes, in the same digital twins. Global optimization algorithms evaluated greater than 4.8 million changes to these settings to identify the lowest values of VILI indices that could be achieved in both modes while preserving adequate gas-exchange.

In digital twins, APRV settings of inspiratory pressure equals to 25 cm H 2 O, low-pressure setting equals to 0 cm H 2 O, inspiration time equals to 5 s, and expiration time set to achieve 75% of peak expiratory flow rate (mean 0.5 s), reduced mean mechanical power (MP) by 32% and mean tidal alveolar recruitment/de-recruitment by 34% compared with documented PCV settings, at the cost of moderate hypercapnia (mean PaC O2 58.5 mm Hg, pHa 7.32 vs. Pa CO2 45.6 mm Hg, pHa 7.37). Mean driving pressure, tidal volume, and lung stress/strain were similar in both modes. Computational optimization showed that these settings were close to optimal in terms of minimizing both mean MP and mean levels of tidal recruitment/de-recruitment during APRV.

Using digital twins we found possible lung-protective conditions and beneficial effects of APRV which need further evaluation in randomized clinical trials.