Contact Person

Arnhold Lohse

Project Description

In intensive care units, artificial ventilation is a life-saving therapy that provides gas exchange in patients with diseases such as Covid-19 or acute respiratory failure. During the initial ventilation phase in which the patients are sedated, the diaphragm remains inactive which may lead to diaphragm atrophy, called ventilator-induced diaphragmatic dysfunction (VIDD). VIDD can occur already after 18 hours of ventilation and prevents the weaning of patients from the mechanical ventilator. Patients who are difficult to wean occupy 40% of ICU beds.
To prevent VIDD, stimulation of the phrenic nerve is a promising solution. Electrical stimulation of the phrenic nerve allows the diaphragm to remain active despite patient sedation. In general, this procedure is established as an implant in patients with paraplegia, and its use is also conceivable in intensive care units.
In order to relieve clinical staff in intensive care units, the setting of stimulation and artificial ventilation parameters should be automated. During automation, stimulation and artificial ventilation must be synchronized and adapted to the patient's needs. The goals of automation are to maintain diaphragm activity for VIDD prevention and to maintain gas exchange while reacting to the changing effectiveness of stimulation.
The automation relies on control engineering and signal processing approaches. Pressure and volume flow values of the mechanical ventilator as well as voltages and currents of the stimulator are available as measurements. These measurements can be used to estimate internal states of the diaphragm and the respiratory system for control purposes. The control algorithms must have sufficient robustness to meet the safety requirements of the critical application.

Project Goals

• Prevention of VIDD by stimulation of the phrenic nerve
• Automation of stimulation and mechanical ventilation

Project Partner

Universitätsklinikum Aachen, Klinik für Operative Intensivmedizin und Intermediate Care

Publications

A. Lohse, M. Walter, M. Deininger, T. Seemann, D. Ziles, T. Breuer and S. Leonhardt, “Flexible pattern generation for phrenic nerve stimulation”, AUTOMED - Automation in Medical Engineering, Basel, 2021