Closed-loop machine-controlled CPR system optimises haemodynamics during prolonged CPR

Abstract

Objectives: We evaluated the feasibility of optimising coronary perfusion pressure (CPP) during cardiopulmonary resuscitation (CPR) with a closed-loop, machine-controlled CPR system (MC-CPR) that sends real-time haemodynamic feedback to a set of machine learning and control algorithms which determine compression/decompression characteristics over time.

Background: American Heart Association CPR guidelines (AHA-CPR) and standard mechanical devices employ a "one-size-fits-all" approach to CPR that fails to adjust compressions over time or individualise therapy, thus leading to deterioration of CPR effectiveness as duration exceeds 15-20 ​min.

Methods: CPR was administered for 30 ​min in a validated porcine model of cardiac arrest. Intubated anaesthetised pigs were randomly assigned to receive MC-CPR (6), mechanical CPR conducted according to AHA-CPR (6), or human-controlled CPR (HC-CPR) (10). MC-CPR directly controlled the CPR piston's amplitude of compression and decompression to maximise CPP over time. In HC-CPR a physician controlled the piston amplitudes to maximise CPP without any algorithmic feedback, while AHA-CPR had one compression depth without adaptation.

Results: MC-CPR significantly improved CPP throughout the 30-min resuscitation period compared to both AHA-CPR and HC-CPR. CPP and carotid blood flow (CBF) remained stable or improved with MC-CPR but deteriorated with AHA-CPR. HC-CPR showed initial but transient improvement that dissipated over time.

Conclusion: Machine learning implemented in a closed-loop system successfully controlled CPR for 30 ​min in our preclinical model. MC-CPR significantly improved CPP and CBF compared to AHA-CPR and ameliorated the temporal haemodynamic deterioration that occurs with standard approaches.

Keywords: CPR; Cardiopulmonary resuscitation; Haemodynamics; Machine learning; Mechanical CPR; OHCA; Personalized medicine; Porcine; Refractory VF.

Fig. 1

Flow of Information During Closed-loop Control Waveforms of the piston’s distance from the chest and the CPP output from the animal were discretised to a cycle-by-cycle basis before being provided to the algorithms. The predictive algorithm (LINR) was used within the framework of the control algorithm (LQR), but they can be conceptualised as passing information in a stepwise manner as depicted. The user of the CPU sets the target CPP for an instant of time and gradually increases Q until the controller reaches the target, then increases the target itself. If the controller was overshooting, or if external forces were being applied to the animal, such as IV injections, R would be increased in order to minimise inappropriate reactions from the controller. CPP, coronary perfusion pressure; CPR, cardiopulmonary resuscitation; CPU, central processing unit; LQR, linear-quadratic regulator; LINR, linear regression; Q, quadratic characteristics of the controller; R, regulator characteristics of the controller.

Fig. 2

Amplitudes of Compression and Decompression Values are means with standard deviations represented by shading. MC-CPR showed the greatest variation in compression and decompression depths. AHA, American Heart Association recommendations; HC-CPR, human-controlled CPR; MC-CPR, machine-controlled CPR.

Fig. 3

Mean Coronary Perfusion Pressure for animals receiving CPR based on AHA Recommendations vs Human-Controlled CPR vs Machine-Controlled CPR CPP for each minute of CPR is plotted along with the linear curves used to quantify the rate of change of CPP per minute. AHA, American Heart Association; CPP, coronary perfusion pressure; CPR, cardiopulmonary resuscitation; HC-CPR, human-controlled CPR; MC-CPR, machine--controlled CPR.

Fig. 4

Mean Carotid Blood Flow of AHA vs Human-Controlled vs Machine-Controlled CPR CBF was measured every 5 ​min of CPR and expressed as a percentage of baseline. CBF was further analysed by linearly fitting the data. CBF, carotid blood flow; CPR, cardiopulmonary resuscitation.

Fig. 5

Closed-loop Feedback of Coronary Perfusion Pressure to the Prediction and Control Algorithms of Machine-Controlled CPR Closed-loop feedback resulted in forces and amplitudes of CPR which automatically oscillated with respirations. CO₂, carbon dioxide CPP, coronary perfusion pressure; RAP, right atrial pressure.