This is a single-center, retrospective, observational, cohort study. The study was approved by the Clinical Research Ethics Committee of the Japanese Red Cross Society Kyoto Daini Hospital (Approval ID Sp2020–11). The Ethics Committee waived the requirement for informed consent because of the anonymous nature of the data. All procedures in this study were performed in accordance with relevant guidelines and regulations.
This study was performed at a tertiary critical care center in Kyoto City, Japan, which is an urban area with a population of approximately 1.5 million and about 90,000 ambulance calls annually . Our 672-bed hospital is one of four tertiary critical care medical centers in Kyoto City. Generally, tertiary critical care medical centers in Japan can accept emergency and critically ill patients transported by ambulance, including sepsis, acute coronary syndrome, cardiac arrest, severe trauma, and stroke patients, and can provide specialized treatment in an intensive care unit . In 2019, the emergency department cases had 7,610 patients who arrived by ambulance and 20,769 patients of “walk-in” status that arrived by other means .
This study included adult patients (age ≥18 years) with OHCA who were admitted to our emergency department while still in cardiac arrest at hospital arrival between January 1, 2019, and December 31, 2020. This study excluded OHCA patients who had a return of spontaneous circulation (ROSC) at hospital arrival.
COVID-19 cases in Kyoto City
In Kyoto City, the first case of COVID-19 infection was confirmed on January 30, 2020; by the end of March, 42 cases had been confirmed. After that, the number of infections increased through mid-May; by the end of May there were 248 cases. The number of infections decreased briefly; however, since late June, the number of cases continued to increase bimodally and reached 3,369 by December 31, 2020  (Supplementary Figure 1).
Emergency medical service resuscitation protocol in Kyoto City
During the COVID-19 infection-spreading period, the emergency medical service (EMS) in Kyoto city implemented a protocol to treat all the cardiac arrest patients as possibly having COVID-19. Bag-valve mask (BVM) ventilation and chest compression were performed with attention to the fact that virus-containing aerosols might be generated. In the EMS protocol, before entering the scene, all staff donned PPE that included N95 masks and eye protection, and a high-efficiency particulate air (HEPA) filter was attached securely to any manual or mechanical ventilation device in the path of exhaled gas. Breathing was assessed by observing chest wall movement in order to minimize the risk of infection. Chest compression was started after covering the mouth and nose of the patient with a BVM and holding it close to the patient's face. Chest compression was limited to as short a time as possible when the mouth of the patient was not covered with a mask or when advanced airway management such as laryngeal tube or tracheal intubation was not introduced. It was recommended that advanced airway management be introduced as early as possible. When administering positive pressure ventilation with a BVM, EMS staff held the BVM tightly against the patient's face to minimize air leakage. Since EMS personnel are generally not allowed to terminate resuscitation in a prehospital setting, all OHCA cases were transported to a hospital.
Revised resuscitation protocol during the COVID-19 period in our hospital
In the interest of controlling COVID-19 transmission, our hospital implemented a revised resuscitation protocol for OHCA patients on April 1, 2020. The details are described in the additional file. In brief, a restricted zone separated from other emergency beds by doors or plastic curtains was set up during resuscitation, and OHCA patients were admitted and treated only in this space (Supplementary Figure 2). All staff involved in the resuscitation procedures were required to wear PPE, including N95 masks (Supplementary Figure 3). An attending emergency physician was placed outside the isolated resuscitation area to direct the other team members who performed the resuscitation activities. The conventional cardiopulmonary resuscitation (CPR) period was defined as January 1, 2019 to March 31, 2020, and the COVID-19 safety protocol period was April 1, 2020 to December 31, 2020.
Prehospital resuscitation data and in-hospital data were obtained by electronic chart reviews by certified emergency physician. Prehospital resuscitation data included the presence of a witness, presence of a bystander who performed CPR, initial cardiac rhythm at the scene, prehospital epinephrine administration, prehospital advanced airway management, prehospital automated external defibrillator use, the call–hospital interval, and achievement of prehospital ROSC. The call–hospital interval was defined as the period from the incoming call to the time when the patient arrived at the hospital. In-hospital data included baseline characteristics of the patients (age and sex), treatments such as coronary angiography, and use of a mechanical circulatory device (extracorporeal membrane oxygen and/or intra-aortic balloon pumping). The cause of arrest was defined as having a cardiac (e.g., acute coronary syndrome, other heart disease, presumed cardiac cause), non-cardiac (e.g., cerebrovascular diseases, respiratory diseases, malignant tumors), or an external cause (including traffic injury, fall, hanging, drowning, asphyxia, drug overdose, or any other external cause) [15, 16]. The medical cause was defined as the cause of arrest other than an external cause. Patients were categorized by age as 18–64 years, 65–74 years, and ≥75 years. Patient outcomes were also collected.
The primary outcome of the study was survival of hospitalization. The secondary outcomes were ROSC after hospital admission and 1-month survival after OHCA occurrence.
Selection of variables
Based on previous studies [16-20], six potential confounding factors were selected: age, presence of a witness, presence of bystander CPR, initial cardiac rhythm at the scene, the call–hospital interval, and the first documented cardiac rhythm at hospital arrival.
Sample size estimation
It was estimated that at least 60–70 case outcomes would be required to account for the confounders using a logistic model, based on the generally accepted rule of 10 events per variable . Considering this, it was determined that including cases from January 2019 to March 2020, before implementation of the COVID-19 safety protocols, would result in an adequate sample size for analysis.
Data statistics for patient characteristics were calculated as a median with an interquartile range (IQR) for continuous variables and as a number with percent for categorical variables. The crude and adjusted odds ratios (AORs) of outcomes with 95% confidence intervals (CIs) were calculated using the multivariable logistic model including all potential confounders. Missing data were not replaced or estimated. Statistical analyses were performed using JMP Pro 14 software (SAS Institute, Cary, NC, USA).