Impact of a resuscitation protocol prioritizing airway clearance in the COVID-19 pandemic for patients with ventricular fibrillation: a retrospective observational study in Nagoya, Japan

Study design and setting

We retrospectively analyzed emergency medical transport data collected by the Nagoya City Fire Department in Nagoya, Japan. Eligible patients were individuals transported to emergency departments within Nagoya from April 2018 to March 2023 after witnessing CA with an initial waveform of VF/pulseless VT. Patients with cardiac arrest due to non-cardiac causes such as malignancy or poisoning were excluded. Additionally, patients who experienced CA after an emergency call was placed were excluded, since the interval from arrest onset to hospital arrival varied significantly. The pre-pandemic period was defined as April 1, 2018, to March 31, 2020, and the pandemic period was April 1, 2020, to March 31, 2023.

Nagoya has approximately 2.3 million inhabitants and covers an area of 326 km². Annually, the city receives about 130,000 emergency calls, managed through seven emergency centers [3]. A distinguishing feature of the Nagoya EMS system is the presence of a control system that secures a receiving medical facility upon receiving an emergency call indicating a patient is unconscious or not breathing, even before the EMS team’s arrival. EMS teams in Japan typically consist of three members, with at least one trained paramedic [4]. Permitted resuscitation procedures performed by EMS teams in Japan are limited to defibrillation using an automated external defibrillator, venous cannulation, adrenaline administration, and airway clearance, according to regional protocols. The initiation of venous cannulation, adrenaline administration, and airway management requires verbal authorization from a physician via telephone.

The conventional CA protocol for VF/VT involved defibrillation, venous access, and adrenaline administration within 10 minutes or less. To reduce COVID-19 transmission risks, a revised protocol was implemented, shortening field activity time to 6–8 minutes and prioritizing supraglottic airway management before establishing venous access and administering adrenaline. During the pandemic period, high-efficiency particulate air filters were attached to the bag-valve-mask (BVM) prior to patient contact. Resuscitation was performed in open or well-ventilated areas to reduce the risk of airborne transmission. Patient breathing was assessed solely through visual observation of chest movement, and close physical proximity between the patient and paramedics was minimized. The BVM mask was securely fitted over the patient's mouth and nose, with efforts made to maintain a tight seal during CPR to minimize air leakage. An automated chest compression device was utilized when available to optimize resuscitation quality and reduce direct physical contact.

Data collection

Patient characteristics (age, sex) and EMS process variables, including time to hospital determination, bystander CPR, bystander defibrillation, adrenaline administration timing, and airway management timing, were collected and analyzed for each study period.

Outcome

The primary outcome measures were prehospital ROSC, 1-month survival, and Cerebral Performance Category (CPC) score at 1 month.

Statistical analysis

Continuous variables were analyzed using the Mann-Whitney U test and presented as median values with interquartile ranges. Categorical variables were analyzed using the Fisher exact test and presented as actual counts and percentages. Multivariate logistic regression analysis was performed to identify factors influencing prehospital ROSC. Selected covariates included known influential factors in resuscitation outcomes: pandemic period, age, sex, bystander CPR, bystander defibrillation, and interval to EMS contact. Propensity scores were calculated using logistic regression, adjusting for age, sex (male), bystander CPR, bystander defibrillation, and time interval from witness to EMS contact (minutes). Nearest-neighbor matching (1:1) with a caliper of 0.2 was employed. Standardized mean differences (SMD) were computed to evaluate covariate balance before and after matching; an SMD of less than 0.1 was achieved for all variables, confirming adequate balance post-matching. Statistical significance was defined as a two-tailed P-value of <0.05. All analyses were performed using EZR statistical software (R statistical computing software).

Ethics

This study was approved by the Ethics Committee of Nagoya City University. Eligible patients were offered the opportunity to opt out and not participate in the study if their data were in the pre-processing stage.

Changes in the prehospital activity time during the COVID-19 pandemic

Regarding the total prehospital interval from emergency call to hospital arrival, Katasako et al. [5] analyzed nationwide OHCA registry data from Japan and reported a delay of approximately 1 minute, from 31 minutes in the pre-pandemic period to 32 minutes during the pandemic. In our Nagoya-based study, total prehospital activity time increased by 2 minutes, from 23 minutes in the pre-pandemic period to 25 minutes in the pandemic period. Even during the pandemic, Nagoya maintained a shorter overall prehospital activity duration, a distinctive feature attributed to the city’s advanced control system, which ensures hospital selection before EMS arrival. In a comparative study of EMS in other Asian cities, the interval to patient contact was reported as 5.7 minutes in Seoul, 8.2 minutes in Singapore, and 12.5 minutes in Bangkok, indicating Nagoya’s relatively rapid EMS response [6]. However, during the COVID-19 pandemic, the hospital selection interval increased significantly due to heightened demand for EMS services. This resulted in more frequent dispatches to remote scenes and longer transport distances to receiving hospitals, extending total prehospital activity time by approximately 2 minutes, although on-scene EMS duration remained unchanged. Multivariate analysis in our study indicated that bystander CPR and earlier EMS contact positively influenced prehospital ROSC outcomes. Given that rates of bystander CPR and defibrillation did not differ significantly between the pre-pandemic and pandemic periods, delayed EMS contact likely contributed to the decreased prehospital ROSC rates observed among CA patients with VF/VT.

EMS protocol changes and prognostic impact

To reduce COVID-19 transmission risks to EMS personnel, Nagoya modified its CPR protocol during the pandemic, prioritizing device-assisted airway clearance over adrenaline administration, along with implementing a policy to minimize on-scene activity duration. These changes resulted in increased utilization of closed esophageal airways and decreased intervals to airway clearance using these devices. However, the interval to adrenaline administration increased by approximately 3 minutes during the pandemic period. Although the COVID-19 protocol anticipated adrenaline administration timing similar to the conventional protocol, delays likely arose from the additional time needed for EMS personnel to don full personal protective equipment (PPE). Consequently, adherence to the established activity time limits of the new protocol resulted in reduced rates and delayed timing of adrenaline administration during the pandemic.

Among OHCA patients, a recent randomized controlled trial demonstrated non-inferiority of endotracheal intubation compared to bag-mask ventilation regarding 28-day neurologically favorable survival outcomes [7]. Differences in airway management methods thus may not significantly affect the prognosis of VF patients. In Australia, a significant increase in laryngeal mask airway use among EMS-witnessed OHCA patients occurred during the pandemic period (39.3% pre-pandemic vs. 61.1% pandemic, P<0.001), yet survival-to-discharge rates showed no significant differences (30.3% vs. 32.8%, P=0.16) [8]. Multiple reports similarly suggest that increased use of supraglottic airway devices during the COVID-19 pandemic did not appear to improve prognosis in OHCA patients.

Conversely, recent evidence suggests an association between delayed adrenaline administration and poorer clinical outcomes. In Canada, the interval from EMS arrival on scene to supraglottic airway device insertion decreased slightly during the pandemic period (11 minutes vs. 10 minutes, P<0.01), whereas the interval to adrenaline administration increased (11 minutes vs. 13 minutes, P<0.01). Furthermore, the rate of adrenaline administration also declined (69% vs. 65%, P<0.01) [9]. This resulted in a lower prehospital ROSC rate during the pandemic period (33.5% pre-pandemic vs. 29.7% pandemic), mirroring trends observed in our study.

Although adrenaline administration has traditionally been considered to increase the rates of ROSC and survival in CA patients without significantly improving neurological outcomes, recent studies have suggested that earlier administration of adrenaline might improve neurological outcomes. A propensity score-matched cohort study utilizing the Japanese registry for OHCA patients with all waveform types showed that delaying adrenaline administration beyond 20 minutes after an emergency call significantly reduced the proportion of patients with CPC scores of 1–2 at 1 month, compared to those receiving adrenaline within 20 minutes (0.8% vs. 2.2%; relative risk, 0.34; 95% CI, 0.30–0.39) [10]. Furthermore, among OHCA patients with shockable rhythms, a time-dependent propensity score analysis indicated that each additional minute of delay in adrenaline administration after EMS arrival decreased the 1-month survival rate by 5.5% (95% CI, 3.4%–7.5%; P<0.001 for interaction between adrenaline administration and matching time) [11]. In our study, the pandemic period was associated with a 3-minute delay in adrenaline administration, which might have contributed to the observed lower survival rate at 1 month (57.2% vs. 46.4%) and lower prehospital ROSC rate (54.3% vs. 43.3%) among CA patients with VF/VT.

The 2024 International Liaison Committee on Resuscitation (ILCOR) guidelines do not specifically reference COVID-19 or pandemics [12]. According to the 2023 ILCOR consensus, although PPE use during the pandemic increased rescuer fatigue, it did not negatively impact 30-day survival rates or ROSC in OHCA cases [13]. Consequently, the continued use of PPE is recommended for infection control. While the AHA issued temporary guidance during the early stages of the pandemic, no subsequent revisions have been made to the 2020 resuscitation algorithm. Our findings suggest that prioritizing airway management within the protocol might worsen patient outcomes, indicating the need for further research into this approach.

Limitations

Because this was an observational study, several risks of bias exist. Patient comorbidities were unknown, potentially introducing variability in patient characteristics. Additionally, we did not evaluate the quality of chest compressions or ventilation delivered during resuscitation. Ideally, to accurately examine prognostic factors such as the timing of adrenaline administration, a time-dependent propensity matching approach would have been preferable to control for biases related to resuscitation timing; however, the limited number of patients in our study precluded the use of this method. Information regarding the location of EMS dispatch or the identity of the individual who discovered the patient was unavailable, limiting detailed examination of life-saving interventions performed by bystanders. In-hospital treatment details were also unavailable, and thus, the data derived from prehospital activities alone were insufficient for thoroughly evaluating patient prognoses at 1 month.

In summary, we comprehensively analyzed EMS prehospital activities and outcomes for CA patients initially presenting with VF/VT in Nagoya. During the COVID-19 pandemic period, intervals from emergency call to EMS contact and hospital arrival were prolonged, and prehospital ROSC rates decreased. Delays in EMS contact and adrenaline administration resulting from changes to the resuscitation protocol prioritizing airway clearance likely influenced the lower prehospital ROSC rates observed among VF/VT patients during the pandemic.