Model-based Patient Matching for in-parallel Multiplexing Mechanical Ventilation Support.

Area of Science:

• Biomedical Engineering
• Respiratory Medicine
• Critical Care

Background:

• COVID-19 surges can cause mechanical ventilator shortages, necessitating care rationing.
• Multiplexing mechanical ventilators (co-MV) is a potential solution, but lacks guidelines for patient matching and settings.
• Existing methods do not address parallel ventilation requirements for co-MV.

Purpose of the Study:

• To propose a method for patient matching and mechanical ventilator settings for two-patient co-MV.
• To enable safe and effective mechanical ventilation sharing during critical care shortages.
• To provide clinicians with a tool for in silico analysis before clinical implementation.

Main Methods:

• Utilized patient-specific, clinically validated respiratory mechanics models.
• Developed a method to simulate and estimate tidal volume for co-ventilated patients.
• Determined mechanical ventilator settings for co-MV under pressure control mode, ensuring patient-specific constraints are met.

Main Results:

• The proposed method successfully matches patients and determines appropriate mechanical ventilator settings for two-patient co-MV.
• Simulations estimate resultant tidal volumes for various patient combinations under shared ventilation.
• The approach ensures both patients meet ventilation requirements with similar settings.

Conclusions:

• This research provides a novel, model-based method for implementing two-patient co-MV.
• The in silico analysis tool aids clinicians in safely managing mechanical ventilator sharing.
• The findings support the clinical implementation of co-MV as a temporary solution for ventilator shortages.