Ventilation of neonates.
Control of arterial CO₂ partial pressure.
The control of arterial carbon dioxide partial pressure is an important aspect, particularly for the mechanical ventilation of preterm babies and neonates. Fluctuations and relevant deviations from the physiological standard range can cause lasting damage to the immature brain.
Adjustment of ventilation at certain intervals.
Healthcare staff regularly adjust ventilation to the patient’s needs. Physicians and caregivers are normally responsible for several patients at a time and are unable to make adjustments continuously. Rather, they make them at certain intervals. This task is also challenging and time-consuming. User support could free up valuable resources for other patient care tasks. The problem here is that the existing methods of continuous measurement of the carbon dioxide partial pressure (paCO2), such as transcutaneous measurement, do not always provide reliable values and the end-tidal CO2 (etCO2) can often not be measured with the tiniest patients due to the dead volume of the measuring cuvettes.
Joint project between hospital and Löwenstein.
In a joint project with the Neonatology Section in the Department of Pediatric and Adolescent Medicine at RWTH Aachen University Hospital, the Chair of Computer Science 11 – Embedded Software at RWTH Aachen, and Löwenstein Medical, we set out to close this gap in care and develop an automated solution – as an add-on to the well-established CLAC® function of the oxygen equipment.
The aim of our work is to develop a control system that continuously adapts the ventilation so that the paCO2 lies in a target range specified by the medical staff and the patient is ventilated at the same time with the least possible strain on the lungs. The paCO2 control covers various modes in accordance with established mechanical ventilation practice. These modes map pressure- and volume-controlled ventilation.
Control is also envisaged that aims to achieve the ideal operating point of respiratory rate and tidal volume. The etCO2 measurement is used as input value for the control. Compared with commercially available measurement systems for preterm babies and neonates, we aim to improve the quality of the measurement. We expect that this high-quality etCO2 measurement can lead to a breakthrough in mechanical ventilation for preterm babies.
Control of the minute volume (MV) based on estimated paCO₂ value.
The control is based mainly on measuring the etCO₂ close to the tube connector. This value is used to estimate the paCO₂, which in practice is hampered by various technical and medical complications, such as a slipped tube, and requires further research. This estimate of the actual value is then compared with the paCO₂ target value so as to then define a minute volume (MV) for the ventilation. In a further algorithm step, this MV is translated into the specific therapy parameters of peak inspiratory pressure (PIP) and respiratory rate (RR).
Testing the control in an animal model.
To test the control in an animal model, LEONI plus ventilators were upgraded to be able to specify control commands for PIP and RR via a serial interface and to support capnometry in the respiratory gas. This setup enabled premature lambs to be ventilated from birth in SIMV therapy mode and with CLAC® support.
The first animal trials show the efficacy of the control. To this end, two ventilation situations were examined whose course is shown in Figure 1. In the first scenario, a hypercapnia was produced with hypoventilation and the controller then set to a paCO2 target value of 45 mmHg and a target range of ± 3 mmHg. Through blood gas analysis (BGA) a reference value of 59 mmHg at the start of control was measured. After 2.5 minutes the target value was reached and then maintained by automatic adjustments of the MV in the target range. A BGA 15 minutes after the start of the test confirmed a paCO2 value of 43.2 mmHg, which represents a minimal deviation from the estimated value of 44.5 mmHg. In the second part, the paCO2 estimated from the etCO2 was maintained constantly in the target range over a period of three hours. Reference measurements using BGA indicated only minor deviations of the actual paCO2 from 0.2 mmHg to 3.73 mmHg.
Thereby, we were able to show that paCO2 can be controlled on the basis of the measured etCO2 concentration and that acute hypocapnic and hypercapnic states can be compensated in under 15 minutes.
Automatic control of the paCO₂ for enhanced safety.
The automatic control of the paCO2 with ventilation in neonates therefore has the potential to increase safety for patients and reduce the workload of staff. Until the solution is deployed in practice, the estimate of the paCO2 in particular needs to be more robust so that regulation cannot cause incorrect ventilation. The focus of further development will also look at integrating control with the spontaneous breathing of the neonates.