Mechanical Circulatory Support Following Out-of-Hospital Cardiac Arrest: Insights From the National Cardiogenic Shock Initiative

A locally optimised machine learning approach to early prognostication of long-term neurological outcomes after out-of-hospital cardiac arrest

Background

Out-of-hospital cardiac arrest (OHCA) represents a major burden for society and health care, with an average incidence in adults of 67 to 170 cases per 100,000 person-years in Europe and in-hospital survival rates of less than 10%. Patients and practitioners would benefit from a prognostication tool for long-term good neurological outcomes.

Objective

We aim to develop a machine learning (ML) pipeline on a local database to classify patients according to their neurological outcomes and identify prognostic features.

Methods

We collected clinical and biological data consecutively from 595 patients who presented OHCA and were routed to a single regional cardiac arrest centre in the south of France. We applied recursive feature elimination and ML analyses to identify the main features associated with a good neurological outcome, defined as a Cerebral Performance Category score less than or equal to 2 at six months post-OHCA.

Results

We identified 12 variables 24 h after admission, capable of predicting a six-month good neurological outcome. The best model (extreme gradient boosting) achieved an AUC of 0.96 and an accuracy of 0.92 in the test cohort.

Conclusion

We demonstrated that it is possible to build accurate, locally optimised prediction and prognostication scores using datasets of limited size and breadth. We proposed and shared a generic machine-learning pipeline which allows external teams to replicate the approach locally.

Evaluation of a Novel Left Ventricular Assist Device for Resuscitation in an Animal Model of Ventricular Fibrillation Cardiac Arrest

We evaluated an independently developed novel percutaneous implantable left ventricular assist device for resuscitation in a pig model of ventricular fibrillation cardiac arrest. The model was established in 10 domestic pigs by blocking the anterior descending coronary artery with a balloon after anesthesia. With ventilator-assisted ventilation, the independently developed percutaneous implantable left ventricular assist device was inserted via the femoral artery to assist circulation. According to whether effective circulatory support was achieved, the pigs were randomly divided into an experimental group and a control group. The experimental group was subjected to insertion of the assist device and received continuous circulatory support. The control group underwent insertion of the assist device; however, it did not start it within 15 minutes. For all animals, if successful rescue was achieved (sinus rhythm restoration within 15 minutes and maintenance for over 5 minutes), circulatory support was stopped, and the arterial blockage was removed. If sinus rhythm was not restored within 15 minutes, electric defibrillation, adrenaline injection, and removal of the arterial blockage were performed, and circulatory support was provided until sinus rhythm recovered. A determination of failed rescue was made when sinus rhythm was not restored after 1 hour. All successfully rescued animals were fed for 1 week. There were no significant differences in baseline data between the groups. All animals underwent successful novel left ventricular assist device implantation through the femoral artery. The rescue rate was significantly higher in the experimental group than in the control group (80% vs. 0%, [Formula: see text]). All successfully rescued animals survived after 1 week of feeding, and no eating or movement abnormalities were observed. We conclude that this independently developed percutaneous implantable left ventricular assist device can be conveniently and rapidly implanted through the femoral artery and can maintain basic circulatory perfusion during resuscitation in an animal model of cardiac arrest.

Unloading in Refractory Cardiogenic Shock After Out-Of-Hospital Cardiac Arrest Due to Acute Myocardial Infarction-A Propensity Score-Matched Analysis

Aims: Unclear neurological outcome often precludes severely compromised patients after out-of-hospital cardiac arrest (OHCA) from mechanical circulatory support (MCS), while it may be considered as rescue therapy for patients with refractory cardiogenic shock (rCS) in the absence of OHCA. This analysis sought to investigate the role of left ventricular (LV) unloading in patients with rCS related to acute myocardial infarction (AMI) after OHCA. Methods: Of 273 consecutive patients receiving microaxial pumps in the Hannover Cardiac Unloading Registry between January 2013 and August 2018, 47 presented with AMI-rCS following successful resuscitation. Subsequently, the patients were compared by propensity score matching to patients with OHCA AMI-rCS without MCS. The patient data for OHCA without LV unloading was available from 280 patients of the Hannover Cooling Registry for the same time period. Furthermore, the patients with OHCA without rCS were compared to the patients with OHCA AMI-rCS and LV unloading. Results: In total, 15 OHCA AMI-rCS patients without MCS were matched to patients with AMI-rCS and Impella. Patients without LV support had a higher proportion of a cardiac cause of death (n = 7 vs. n = 3; p = 0.024). LV unloading with Impella counteract rCS status and was associated with a preferable 30-day survival (66.7 vs. 20%, p = 0.01) and a favorable neurological outcome after 30 days (Cerebral Performance Category ≤2, 47 vs. 27%). Impella support is associated with a higher 30-day survival (odds ratio, 2.67; 95% confidence interval, 1.02-13.66). Conclusion: In patients after OHCA with AMI-rCS, Impella support incorporated in a strict standardized treatment algorithm results in a preferable 30-day survival and counteracts severe rCS status.