A multidomain 0D model for continuous positive airway pressure ventilation circuit design: Validation and applications

Abstract

This study focuses on optimizing a non-invasive ventilation (NIV) circuit for the treatment of hypoxemic respiratory failure using continuous positive airway pressure (CPAP). A multidomain 0D in silico approach was employed, creating a lumped circuit model of an innovative NIV-CPAP system in Mathworks® Simulink. The model relies on in vitro tests on commercial components characterizing pneumatic resistive behavior, and it exploits an extended resistance-inductance-capacitance model for the patient’s respiratory system, recurring to sigmoidal pressure–volume behavior characteristic of pathological conditions. The NIV-CPAP system was assembled in vitro and connected to a lung simulator to validate the model under healthy and pathological conditions (acute respiratory distress syndrome and chronic obstructive pulmonary disease). The study explored the impact of key features on the ventilation circuit, such as interface leakage, air volume within the circuit, and resistance induced by circuit components.

Validation of the 0D model through in vitro tests showed correlation coefficients between 0.9 and 1. Interface leakage caused reductions of up to 6% in delivered static pressure. Changes in air volume (mask or helmet interface, reservoirs adding) resulted in a maximum 8% decrease in pressure oscillations. Increased resistances from the starting ventilation circuit produced a tidal volume reduction of less than 1%. An optimized configuration that balanced resistances between limbs improved intrinsic positive end-expiratory pressure generation.

The proposed 0D model proved to be effective in guiding the design of the innovative device, providing computational efficiency and flexibility; it demonstrated its reliability as a tool to support the optimization of non-invasive ventilation circuits.