Outcomes of Patients From All‐Japan Utstein Registry Managed by 2015 Cardiopulmonary Resuscitation Guidelines for Nonshockable Out‐of‐Hospital Cardiac Arrest

Abstract

BACKGROUND

The 2015 as well as the 2020 cardiopulmonary resuscitation guidelines have stressed that high‐quality cardiopulmonary resuscitation increases the survival rates of patients suffering from nonshockable cardiac arrest. However, whether the management of adult patients with out‐of‐hospital cardiac arrest (OHCA), performed in accordance with the International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations (CoSTR) 2015, is associated with favorable neurological outcomes remains unknown. The present study aimed to study the relationship between the implementation of the CoSTR 2015 clinical recommendations and outcomes in adult patients with OHCA, using data from the All‐Japan Utstein Registry, a prospective, nationwide, population‐based registry of OHCA.

METHODS

In this study, 73 874 adults from the All‐Japan Utstein Registry diagnosed with nonshockable OHCA that had been witnessed by bystanders or emergency medical service responders between 2011 and 2020 were included. The study outcomes were compared between 2012 and 2014 of the CoSTR 2010 era (2010 group) and 2017 and 2019 of the CoSTR 2015 era (2015 group).

RESULTS

The adjusted odds ratio for the 30‐day neurologically intact survival in the 2015 bystander‐witnessed group compared with the 2010 group was 1.175 (95% CI, 1.014–1.362; P=0.032). Similar results were observed in the emergency medical service responder‐witnessed group.

CONCLUSIONS

The cardiopulmonary resuscitation maneuver updates recommended by CoSTR 2015 for nonshockable OHCA were associated with higher survival rates without neurological complications compared with those recommended by CoSTR 2010 in Japan.

REGISTRATION

URL: http://www.umin.ac.jp/ctr/. Unique identifier: 000009918.

Nonstandard Abbreviations and Acronyms

BLS

basic life support

CoSTR

Consensus on CPR and Emergency Cardiovascular Care Science with Treatment Recommendations

ELST

emergency life‐saving technician

FDMA

Fire and Disaster Management Agency

OHCA

out‐of‐hospital cardiac arrest

ROSC

return of spontaneous circulation

Clinical Perspective.

What Is New?

• Compared with cardiopulmonary resuscitation maneuvers for out‐of‐hospital cardiac arrest with initial nonshockable rhythm recommended in the Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations (CoSTR) 2010, the updates recommended in CoSTR 2015 were associated with enhanced neurologically intact survival rates in Japan.

What Are the Clinical Implications?

• Early high‐quality cardiopulmonary resuscitation is crucial in ensuring survival of patients experiencing out‐of‐hospital cardiac arrest and initial nonshockable rhythm without neurological complications.

• After early high‐quality cardiopulmonary resuscitation, early advanced life support might improve outcomes of patients experiencing out‐of‐hospital cardiac arrest with an initial nonshockable rhythm.

Out‐of‐hospital cardiac arrest (OHCA) necessitating cardiopulmonary resuscitation (CPR) is a notable public health problem on a global level. ¹ , ² , ³ The International Liaison Committee on Resuscitation has published the guidelines for the International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations (CoSTR) about CPR and emergency cardiovascular care every 5 years from 2000 to 2016 and annually since 2017. ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ Despite the advanced management, neurological complications continue to affect nearly all OHCA survivors, particularly those with a nonshockable heart rhythm, including asystole or pulseless electrical activity. ¹⁰ , ¹¹ , ¹² The updates for nonshockable OHCA were minimal since the CoSTR 2010 (Figure S1). ⁴ , ⁵ , ⁶ , ⁷ , ⁹ However, it should be noted that while the CoSTR 2015, as well as the Guidelines 2020, emphasized the importance of high‐quality CPR as a part of the basic life support (BLS) performed on patients with nonshockable cardiac arrest since it can potentially improve the chances of surviving. ⁷ , ⁹

In Japan, using the Utstein‐style guideline, the Fire and Disaster Management Agency (FDMA) established the All‐Japan Utstein Registry on January 1, 2005. ¹³ , ¹⁴ , ¹⁵ In Japan, the emergency medical service (EMS) responders had modified their CPR technique every 5 years based on the CoSTR 2005, 2010, 2015, and Guidelines 2020. ⁵ , ⁶ , ⁷ , ⁹ The Japan Resuscitation Council guidelines were published every 5 years, as in the past, which led to the development of the 2020 guidelines based on CoSTR published by the International Liaison Committee on Resuscitation from 2017 to 2020. CPR efforts must continue until the return of spontaneous circulation (ROSC) is achieved or until they reach the hospital. ³ , ¹⁴ , ¹⁶ , ¹⁷

The comprehensive efficacy of the updated CPR maneuver guidelines for the management of nonshockable cardiac arrest has been investigated by some nationwide population‐based studies. No major changes in BLS and advanced life support were mentioned for patients with OHCA and nonshockable rhythm in CoSTR 2015 and Guidelines 2020. Thus, the primary objective of this study was to determine whether the pre‐hospital care based on CoSTR 2015 provided more promising outcomes in Japan, as compared with those based on CoSTR 2010.

METHODS

Data and study materials will not be made available to other researchers to reproduce the results or replicate the procedure.

Data Collection and Definitions

This study was a secondary analysis of the All‐Japan Utstein Registry, which is a prospective, nationwide, population‐based OHCA registry that has been developed following Japan’s ethical guidelines. ¹⁶ This registry has been described in detail in previous studies. ³ , ¹⁶ , ¹⁸ , ¹⁹ This registry involves the participation of all currently working fire stations with dispatch centers and all collaborating medical institutions. All EMS providers and physicians have been practicing and teaching CPR in accordance with the Japanese CPR guidelines, updated every 5 years based on international CPR recommendations. ⁴ , ⁵ , ⁶ , ⁷ , ⁹

This study was conducted by a Subcommittee of Resuscitation Science in the Japanese Circulation Society. The ethics committee at Nihon University Hospital approved this study on February 10, 2011 (Institutional Review Board number: 110202). This clinical trial was designed as a prospective and observational study and was published on a public website of the trial registry named “UMIN‐CTR,” with the URL http://www.umin.ac.jp/ctr/index/htm/ (UMIN‐CTR number: UMIN000009918). ³ , ¹⁶ The requirement for the written informed consent from recruited patients was waived. After complying with the prescribed governmental legal procedures, the Subcommittee of Resuscitation Science in the Japanese Circulation Society was provided with the trial registry data and the current study merely analyzed deidentified and anonymized data.

The EMS system in Japan has been previously described. ³ , ¹⁶ , ¹⁸ , ¹⁹ Briefly, nationwide there were 723 municipally governed fire stations with dispatch centers operating around a clock in 2022. Irrespective of rurality or population density, all EMS systems in Japan follow standardized CPR protocols. ¹⁴ , ¹⁶ Each ambulance was equipped with 3 EMS responders and included at least 1 emergency life‐saving technician (ELST), who was certified to insert intravenous lines and adjunct airways, but did not include any physicians. The number of ELSTs increased from 26 533 in 2011 to 40 043 in 2020. ²⁰ Since July 2004, specially trained ELSTs were permitted to insert endotracheal tubes and since April 2006, they could perform intravenous injections of epinephrine. In Japan, EMS responders are not permitted to terminate resuscitation in an emergency field. Thus, with the exception of patients with decapitation, incineration, decomposition, rigor mortis, or dependent cyanosis, most patients suffering from OHCA were transported to a hospital and recorded in this registry by the EMS responders. ¹⁸

Data elements collected prospectively in accordance with the Utstein guidelines depicted in the previously published papers have also been described. ¹³ , ¹⁴ , ¹⁶ Briefly, all event times were synchronized with the dispatch center clock. The information documented by the EMS responders included whether they or bystanders had witnessed the OHCA patient, the presence or absence of bystander CPR, and the types of techniques used by the bystanders to perform CPR. The initial cardiac arrest rhythm was classified either as a nonshockable, shockable, or an uncertain rhythm, based on the EMS report produced by analysis performed by an automated external defibrillator. Patients who received shocks from bystanders with public‐access defibrillators were classified as those having a shockable rhythm. The cause of arrest was clinically determined by the physicians‐in‐charge after arrival at the hospital. The detailed information on resuscitation and post‐cardiac arrest care was not collected; however, the hospital physicians collected the resuscitation outcome in collaboration with the EMS responders. A follow‐up interview determined the 30‐day neurological outcome for patients discharged alive. The data form was filled out by the EMS personnel with the physicians‐in‐charge. The obtained data which were subsequently integrated into the registry system on the FDMA database server was then logically revised by the computer system.

Study Patients

This study included adult patients from the All‐Japan Utstein Registry with OHCA, witnessed by bystanders or EMS responders and whose initial cardiac arrest was a nonshockable heart rhythm, between January 1, 2011 and December 31, 2020. The patients were divided into 2 groups based on the CoSTR protocol followed: the era of the CoSTR 2010 (2010 group) and that of the CoSTR 2015 (2015 group). Patients with the onset occurring from January 1, 2012 to December 31, 2014 belonged to the 2010 group, whereas those with the onset from January 1, 2017 to December 31, 2019 belonged to the 2015 group. Patients with disease onsets diagnosed in 2011, 2015, 2016, and 2020 were excluded from the analysis because of time lags between modifications to the resuscitation protocol in the EMS system in Japan and the publication of CoSTR.

The exclusions were as follows: patients aged <18 years or an unidentified age, patients with shockable OHCA or unidentified arrest rhythm, patients with unwitnessed OHCA or unidentified witnessed status, or patients for whom no resuscitation was attempted. Moreover, the enrolled patients were categorized into the 2 groups based on who witnessed them suffering from an OHCA: bystander‐witnessed (patients witnessed by bystanders) and EMS‐responder‐witnessed (patients with OHCA after the arrival of EMS responders) groups.

A subgroup of bystander‐witness patients who received bystander CPR was analyzed. Subgroups of unwitnessed patients and patients experiencing cardiac arrest of noncardiac cause were also analyzed. Noncardiac etiologies included cerebrovascular disease, respiratory disease, malignant tumor, trauma, poisoning, drowning, traffic accidents, hypothermia, and anaphylaxis.

Role of the Funding Source

The funders had no role in this study. The study protocol was designed by the implementation working group for the All‐Japan Utstein Registry of the FDMA. Although the FDMA collected and analyzed the data, it was not involved in revising or interpreting the data or writing this report. All authors have full access to the data from the All‐Japan Utstein Registry, which is a publicly accessible open database. The corresponding author has the ultimate responsibility for submitting the study for publication. The manuscript was prepared in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.

Endpoints

The primary endpoint was a favorable neurological outcome at 30 days following the onset of the OHCA, defined as a Cerebral Performance Category 1 (good performance) or 2 (moderate disability) on the 5‐category scale. ¹³ Unfavorable neurological outcomes were categorized as Cerebral Performance Category 3 (severe disability), 4 (vegetative state), or 5 (death). The secondary endpoints included the 30‐day survival (Cerebral Performance Category 1–4) after OHCA and pre‐hospital ROSC.

Statistical Analysis

Baseline characteristics and crude analysis of outcomes for binary and categorical variables were compared using the chi‐square test. Continuous variables were compared using the Mann–Whitney test. Cases with missing data were excluded. Multivariable logistic regression analyses were conducted for the independent predictors of resuscitation, including enrollment year as a primary exposure variable. Multivariable logistic regression analysis included potential confounders, which were chosen based on biological plausibility and previous studies. ³ , ¹⁴ , ¹⁸ , ¹⁹ The other covariates in the bystander‐witnessed group included factors such as age, sex, collapse‐to‐scene interval (minutes), airway management by EMS personnel (basic airway management alone or advanced after the former), and intravenous administration of epinephrine by EMS personnel (the number of times; 0, 1, 2, 3, 4, more than 5), as appropriate. The collapse‐to‐scene intervals were excluded from these covariates in the EMS‐responder‐witnessed group. Odds ratios, 95% CIs, and P values were calculated in the multivariable analysis. All hypothesis tests were 2‐sided, with a statistically significant P value set at <0.05. IBM SPSS Statistics version 24.0 (IBM Corp., Armonk, NY, USA) was used to perform all analyses.

RESULTS

Patient Population and Baseline Characteristics

Between January 1, 2011 and December 31, 2020, EMS responders performed CPR on 1 259 960 patients with OHCA and the patients were subsequently transported to a hospital (Figure 1). Among these patients, 653 262 were adult patients who had suffered OHCAs because of a nonshockable arrest as the initial rhythm from 2012 to 2014 and from 2017 to 2019. Among these patients, 264 782 were caused by noncardiac cause, and 253 805 were unwitnessed. In addition, 54 002 patients were witnessed and not received bystander CPR, 19872 were witnessed by EMS responders, and 57 738 were witnessed and received bystander CPR. In some patients, data on age (n=1), initial cardiac rhythm (n=27 019), witness status (n=1466), and bystander CPR performance (n=1597) were missing. Six patients did not have data on 30‐day neurological outcomes. These patients were excluded from the study. The baseline characteristics of the 2 groups in the bystander‐witnessed group significantly differed with respect to age, sex, time intervals, first monitored rhythm, advanced airway, administration of saline, and administration of epinephrine by EMS personnel (Table 1). The baseline characteristics of the 2 groups in the EMS‐responder‐witnessed group significantly differed with respect to age, as well as certain maneuvers conducted by the EMS personnel such as advanced airway, administration of saline, and administration of epinephrine (Table 2).

Figure 1. Study profile.

The patients who experienced an out‐of‐hospital cardiac arrest with the onset occurring from 2012 to 2014 belonged to the 2010 Group as the CoSTR 2010 era, and those with the onset occurring from 2017 to 2019 were assigned to the 2015 Group as the CoSTR 2015 era. International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations; CPR, cardiopulmonary resuscitation; and EMS, emergency medical service.

Table 1.

Baseline Characteristics of the Patients With Bystander‐Witnessed Nonshockable Cardiac Arrest (n=54.002)^*

Baseline characters—witnessed by bystanders and not received bystander CPR	2010 group (n=27 459)	2015 group (n=26 543)	P
Age, y median (IQR)	80 (71–87)	81 (72–87)	<0.001
Male sex, n (%)	16 314 (59.4)	16 100 (60.7)	0.003
Time interval from arrest to scene median (IQR)†	10 (6–15)	10 (7–15)	0.015
Time interval from arrest to epinephrine administration median (IQR)†	25 (19–32)	24 (19–31)	0.017
Time interval from arrest to hospital arrival median (IQR)†	36 (28–45)	35 (28–44)	<0.001
First monitored rhythm, n (%)			<0.001
Pulseless electrical activity	11 953 (43.5)	12 463 (47.0)
Asystole	15 506 (56.5)	14 080 (53.0)
Received shock after conversion to the shockable rhythm, n (%)	1916 (7.0)	1884 (8.6)	<0.001
Advanced airway by EMS person, n (%)	12 261 (44.7)	12 538 (53.8)	<0.001
Administrated saline by EMS person, n (%)	10 563 (38.5)	11 272 (42.5)	<0.001
Total dose of epinephrine by EMS person, n (%)			<0.001
No administration	16 932 (61.7)	17 316 (65.2)
1 mg	2918 (10.6)	3046 (11.5)
2 mg	2039 (7.4)	2538 (9.6)
3 mg	1491 (5.4)	2266 (8.5)
4 mg	704 (2.6)	736 (2.8)
More than 5 mg	3375 (12.3)	641 (2.4)

EMS indicates emergency medical service; and IQR, interquartile range.

^*Values are median (interquartile range) or numerator/total number (%).

^† Total number of patients calculating time intervals; 53 206 in the arrest‐to‐scene interval; 16 135 in the arrest‐to‐epinephrine‐administration interval; 53 664 in arrest‐to‐hospital‐arrival interval.

Table 2.

Baseline Characteristics of the Patients With EMS‐Responder‐Witnessed Nonshockable Cardiac Arrest (n=19 872)

Baseline characters—witnessed by EMS responders	2010 group (n=9398)	2015 group (n=10 474)	P
Age, y median (IQR)	79 (69–86)	80 (70–87)	<0.001
Male sex, n (%)	5296 (56.4)	5977 (57.1)	0.311
Time interval from arrest to scene median (IQR)	0 (0–0)	0 (0–0)	0.628
Time interval from arrest to epinephrine administration median (IQR)*	13 (9–18)	12 (9–17)	<0.001
Time interval from arrest to hospital arrival median (IQR)†	16 (10–25)	17 (10–24)	0.223
First monitored rhythm, n (%)			<0.001
Pulseless electrical activity	6944 (73.9)	8102 (77.4)
Asystole	2454 (26.1)	2372 (22.6)
Received shock after conversion to the shockable rhythm, n (%)	407 (4.3)	454 (5.3)	0.002
Advanced airway by EMS person, n (%)	2827 (30.1)	3421 (40.4)	<0.001
Administrated saline by EMS person, n (%)	2134 (22.7)	2903 (27.7)	<0.001
Total dose of epinephrine by EMS person, n (%)			<0.001
No administration	7235 (77.0)	8247 (78.7)
1 mg	725 (7.7)	911 (8.7)
2 mg	393 (4.2)	671 (6.4)
3 mg	219 (2.3)	436 (4.2)
4 mg	104 (1.1)	106 (1.0)
More than 5 mg	722 (7.2)	103 (1.0)

EMS indicates emergency medical service; and IQR, interquartile range.

^*Values are median (interquartile range) or numerator/total number (%).

^† Total number of patients calculating time intervals; 2122 in the arrest‐to‐scene interval; 3558 in the arrest‐to‐epinephrine‐administration interval; 19 804 in arrest‐to‐hospital‐arrival interval.

Outcomes

The value for each crude study outcome (favorable 30‐day neurological outcome, 30‐day survival, and pre‐hospital ROSC) was significantly higher in the 2015 group than in the 2010 group, both in the bystander‐witnessed group as well as the EMS‐responder‐witnessed group (Figure 2). The results of the multivariable logistic regression analyses for the independent predictors of resuscitation in the bystander‐witnessed group have been depicted in Table 3. The adjusted odds ratio for the 30‐day neurologically intact outcomes in the 2015 group compared with the 2010 group was 1.175 (95% CI, 1.014–1.362; P=0.032) as evident from Table 3. The adjusted odds ratio for survival at 30 days was 1.180 (95% CI, 1.076–1.294; P<0.001), while that for the rate of pre‐hospital ROSC was 1.131 (95% CI, 1.066–1.200; P<0.001). The adjusted odds ratio for the 30‐day neurologically intact outcomes in the 2015 group for the multivariate logistic regression analysis in the EMS‐responder‐witnessed group (Table 4), was 1.215 (95% CI, 1.064–1.386; P=0.004), while that for the 30‐day survival was 1.134 (95% CI, 1.027–1.254; P=0.005), and the adjusted odds ratio for the rate of pre‐hospital ROSC was 1.199 (95% CI, 1.098–1.308; P<0.001), compared with the 2010 group.

Figure 2. Crude study outcomes in patients with out‐of‐hospital nonshockable cardiac arrest stratified by witnessed status (A: bystander, B: EMS‐responder).

In both groups, each crude study outcome (prehospital ROSC, 30‐day survival, and favorable 30‐day neurological outcome) had significantly elevated values between the 2010 and 2015 groups. The Chi‐square test was performed in this analysis. ROSC indicates return of spontaneous circulation; and EMS, emergency medical service.

Table 3.

Adjusted Odds Ratios for a Favorable 30‐Day Neurological Outcome, 30‐day Survival, and Pre‐Hospital ROSC After Out‐of‐Hospital Nonshockable Cardiac Arrest in Bystander‐Witnessed Group Who Did Not Receive Bystander CPR

Variable	Favorable 30‐d neurological outcome	30‐d survival	Pre‐hospital ROSC
Age, y	0.967 (0.963–0.972), P<0.001	0.973 (0.970–0.976), P<0.001	0.997 (0.995–1.000), P=0.018
Male sex (reference female)	1.194 (1.014–1.362), P=0.026	1.057 (0.960–1.163), P=0.261	1.233 (1.159–1.313), P<0.001
Time interval from arrest to scene	0.925 (0.913–0.937), P<0.001	0.947 (0.940–0.955), P<0.001	0.968 (0.963–0.972), P<0.001
Advanced airway by EMS responders	0.428 (0.363–0.505), P<0.001	0.768 (0.699–0.845), P<0.001	0.848 (0.797–0.903), P<0.001
Administration of saline by EMS responders	1.017 (0.817–1.264), P=0.882	1.322 (1.160–1.508), P<0.001	1.628 (1.485–1.784), P<0.001
Total dose of epinephrine by EMS responders (reference no.)
1 mg	0.926 (0.700–1.233), P=0.587	1.402 (1.204–1.632), P<0.001	3.174 (2.873–3.507), P<0.001
2 mg	0.485 (0.329–0.715), P<0.001	0.784 (0.646–0.951), P=0.013	3.012 (2.702–3.357), P<0.001
3 mg	0.289 (0.174–0.479), P<0.001	0.380 (0.293–0.494), P<0.001	1.788 (1.581–2.021), P<0.001
4 mg	0.092 (0.023–0.374), P=0.001	0.270 (0.169–0.432), P<0.001	1.455 (1.216–1.743), P<0.001
More than 5 mg	0.231 (0.144–0.372), P<0.001	0.322 (0.245–0.424), P<0.001	0.987 (0.863–1.130), P=0.855
CoSTR 2015 (reference CoSTR 2010)	1.175 (1.014–1.362), P=0.032	1.180 (1.076–1.294), P<0.001	1.131 (1.066–1.200), P<0.001

CoSTR indicates International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations; EMS, emergency medical service; and ROSC, return of spontaneous circulation.

Table 4.

Adjusted Odds Ratios for a Favorable 30‐day Neurological Outcome, 30‐day Survival, and Pre‐Hospital ROSC After Out‐of‐Hospital Nonshockable Cardiac Arrest in EMS‐Responder‐Witnessed Group

Variable	Favorable 30‐d neurological outcome	30‐d survival	Pre‐hospital ROSC
Age, y	0.974 (0.970–0.978), P<0.001	0.976 (0.973–0.979), P<0.001	0.993 (0.990–0.997), P<0.001
Male sex (reference female)	0.981 (0.856–1.124), P=0.781	0.865 (0.781–0.985), P=0.005	1.042 (0.952–1.139), P=0.372
Advanced airway by EMS responders	0.427 (0.359–0.507), P<0.001	0.626 (0.557–0.704), P<0.001	0.565 (0.510–0.625), P<0.001
Administration of saline by EMS responders	1.115 (0.894–1.390), P=0.334	1.191 (1.015–1.397), P=0.032	1.601 (1.399–1.833), P<0.001
Total dose of epinephrine by EMS responders (reference no)
1 mg	0.704 (0.515–0.964), P=0.028	0.888 (0.718–1.100), P=0.277	1.473 (1.245–1.742), P<0.001
2 mg	0.472 (0.305–0.732), P=0.001	0.580 (0.435–0.775), P<0.001	1.644 (1.356–1.993), P<0.001
3 mg	0.087 (0.027–0.275), P<0.001	0.332 (0.215–0.512), P<0.001	1.121 (0.876–1.435), P=0.363
4 mg	0.079 (0.011–0.573), P=0.012	0.188 (0.076–0.462), P<0.001	0.825 (0.536–1.271), P=0.383
More than 5 mg	0.338 (0.210–0.545), P<0.001	0.445 (0.325–0.611), P<0.001	0.898 (0.716–1.126), P=0.351
CoSTR 2015 (reference CoSTR 2010)	1.215 (1.064–1.386), P=0.004	1.134 (1.027–1.254), P=0.005	1.199 (1.098–1.308), P<0.001

CoSTR indicates International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations; EMS, emergency medical service; and ROSC, return of spontaneous circulation.

Outcomes in the Subgroups

Table S1 shows the baseline characteristics of the 2010 and 2015 bystander‐witnessed groups who received bystander CPR, patients with cardiac arrest caused by a noncardiac etiological factor, and patients with unwitnessed cardiac arrest. Table 5 presents the favorable 30‐day neurological outcomes, 30‐day survival rates, and prehospital ROSC rate for these subgroups. These results were similar to those of the bystander‐witness patients who had received bystander CPR from no one, except for the 30‐day survival rates of patients with unwitnessed cardiac arrest.

Table 5.

The Primary and Secondary Outcomes in the Subgroups of Bystander‐Witness Patients Who Had Received Bystander CPR (A), Cardiac Arrest Patients Caused by Noncardiac Cause (B), and Unwitnessed Cardiac Arrest Patients (C)

Primary and secondary outcomes	Variable	Outcome (%)	Adjusted OR (95% CI)	P value
A. Subgroup analysis—witnessed by bystanders and received bystander CPR
Favorable 30‐d neurological outcome	2010 group	370/26943 (1.4)	(Reference)	0.001
	2015 group	499/30795 (1.6)	1.209 (1.102–1.478)
30‐d survival	2010 group	921/26943 (3.4)	(Reference)	0.001
	2015 group	1193/30795 (3.9)	1.171 (1.066–1.286)
Pre‐hospital ROSC	2010 group	2504/26943 (9.3)	(Reference)	<0.001
	2015 group	3428/30795 (11.1)	1.146 (1.081–1.216)
B. Subgroup analysis—Noncardiac etiology
Favorable 30‐d neurological outcome	2010 group	1067/135626 (0.8)	(Reference)	<0.001
	2015 group	1300/129154 (1.0)	1.278 (1.170–1.396)
30‐d survival	2010 group	4313/135625 (3.2)	(Reference)	0.043
	2015 group	4305/129154 (3.3)	1.049 (1.001–1.098)
Pre‐hospital ROSC	2010 group	10 876/135628 (8.0)	(Reference)	<0.001
	2015 group	12 544/129154 (9.7)	1.119 (1.086–1.254)
C. Subgroup analysis—Unwitnessed
Favorable 30‐d neurological outcome	2010 group	170/61611 (0.3)	(Reference)	0.001
	2015 group	238/60289 (0.4)	1.310 (1.110–1.546)
30‐d survival	2010 group	571/61781 (0.9)	(Reference)	0.346
	2015 group	636/60289 (1.1)	1.044 (0.954–1.142)
Pre‐hospital ROSC	2010 group	1460/61781 (2.4)	(Reference)	<0.001
	2015 group	1907/60289 (3.2)	1.112 (1.055–1.172)

EMS indicates emergency medical services; OR, odds ratio; and ROSC, return of spontaneous circulation.

The favorable 30‐day neurological outcomes, the 30‐day survival, and pre‐hospital ROSC in the subgroups of patients who had received advanced airway management have been shown in Table 6. The adjusted odds ratio for the 30‐day neurologically intact outcomes in the 2015 group in the multivariate logistic regression analysis of the bystander‐witnessed group was 1.240 (95% CI, 0.947–1.623; P=0.118), while that for the 30‐day survival was 1.129 (95% CI, 0.986–1.211; P=0.079). The adjusted odds ratio for the rate of pre‐hospital ROSC was 1.117 (95% CI, 1.030–1.211; P=0.007), compared with the 2010 group. However, the adjusted odds ratio for the 30‐day neurologically intact outcomes in the 2015 group for the multivariate logistic regression analysis in the EMS‐responder‐witnessed group was 1.560 (95% CI, 1.142–2.129; P=0.005) while that for 30‐day survival was 1.247 (95% CI, 1.027–1.514; P=0.026), and that the adjusted odds ratio for the rate of pre‐hospital ROSC was 1.326 (95% CI, 1.126–1.561; P=0.001), compared with the 2010 group.

Table 6.

The Primary and Secondary Outcomes in the Subgroups of Nonshockable Cardiac Arrest Patients Receiving Advanced Airway Management in Bystander‐Witnessed (A) and EMS‐Responder‐Witnessed (B) Groups

Primary and secondary outcomes	Variable	Outcome (%)	Adjusted OR (95% CI)	P value
A. Subgroup analysis—Advanced airway management in bystander‐witnessed group
Favorable 30‐d neurological outcome	2010 group	99/12261 (0.8)	(Reference)
	2015 group	120/12538 (1.0)	1.240 (0.947–1.623)	0.118
30‐d survival	2010 group	427/12261 (3.5)	(Reference)
	2015 group	478/12538 (3.8)	1.129 (0.986–1.292)	0.079
Pre‐hospital ROSC	2010 group	1329/12261 (10.8)	(Reference)
	2015 group	1603/12538 (12.8)	1.117 (1.030–1.211)	0.007
B. Subgroup analysis—Advanced airway management in EMS‐responder‐witnessed group
Favorable 30‐d neurological outcome	2010 group	64/2827 (2.3)	(Reference)
	2015 group	116/3421 (3.4)	1.560 (1.142–2.129)	0.005
30‐d survival	2010 group	189/2827 (6.7)	(Reference)
	2015 group	278/3421 (8.1)	1.247 (1.027–1.514)	0.026
Pre‐hospital ROSC	2010 group	272/2827 (9.6)	(Reference)
	2015 group	455/3421 (13.3)	1.326 (1.126–1.561)	0.001

EMS indicates emergency medical services; OR, odds ratio; and ROSC, return of spontaneous circulation.

DISCUSSION

This nationwide study compared the outcomes of CPR maneuvers in individuals who experienced nonshockable OHCA in Japan based on CoSTR 2010 and 2015 guidelines. Despite the persistently low rates of favorable neurological outcomes in patients with nonshockable cardiac arrest, neurologically intact survival rates differed between periods corresponding to the CoSTR 2010 and 2015 guidelines in Japan. Although these differences may reflect changes in guideline implementation, they could also be influenced by other time‐related factors or unmeasured confounders.

The outcomes of patients suffering from with OHCA and initial nonshockable rhythms are known to be worse compared with those with initial shockable rhythms. ²¹ , ²² , ²³ Therefore, compared with the OHCA patients with an initial shockable rhythm, those with the initial nonshockable rhythm tend to be older, have lower chances of a witnessed arrest and receiving less bystander CPR. ²⁴ This study included patients with witnessed arrests; however, the OHCA patients who received bystander CPR were excluded since we could not assess whether the CPR was performed in accordance with CoSTR 2010, 2015, or Guidelines 2020. However, disseminating BLS based on the latest guidelines to the general public could have contributed to the enhanced common practice of bystander CPR and could have led to improved neurologic outcomes. ¹⁶ Our previous study had demonstrated that the frequency of bystander CPR implementation had increased yearly in Japan. ¹⁶

Basic Life Support

The rates of neurologically intact survival were lower in the bystander‐witnessed group compared with the EMS responder‐witnessed group (Figure 2). In other words, the high‐quality CPR performed early in the onset by EMS personnel was conducive in increasing the chances of neurologically intact survival for patients suffering from OHCA. The time interval from a witnessed cardiac arrest to initiation of CPR was shorter in the EMS‐responder‐witnessed group compared with the bystander‐witnessed group. The quality of CPR affects ROSC rates and neurological outcomes; however, the data on CPR quality were lacking in this registry. The CoSTR 2015 was more strongly vocal in stressing that high‐quality CPR improves survival from cardiac arrest compared with the CoSTR 2010. ⁷ The changes in BLS maneuvers improved the neurological outcome in both groups, indicating that the changes were associated with the enhanced quality of CPR performed by EMS responders.

On the other hand, the rates of neurologically intact survival were similar between the group of patients who had received bystander CPR and the group of patients who had not. It was challenging to identify the cause of this similarity because the data on bystander CPR quality were also lacking in this registry. However, even among patients who received bystander CPR, the changes in BLS maneuvers improved neurological outcome. The dissemination of BLS guidelines based on the latest recommendations to the general public may increase the practice of bystander CPR and improve neurologic outcomes. ¹⁶

Although in the subgroup of cardiac arrest patients attributable to noncardiac cause or unwitnessed cardiac arrest patients, the rates of neurologically intact survivors were even lower, the changes in BLS maneuvers improved neurological outcome. Some previously conducted studies have demonstrated that the proportion of patients suffering from OHCA caused by cardiac causes was relatively low in patients with nonshockable OHCA. ²⁵ , ²⁶ In addition, patients with OHCA resulting from noncardiac causes tend to have poorer outcomes compared with those with cardiac causes. ²⁷ This registry included presumed cardiac causes in the cardiac cause when there is no apparent cause of the cardiac arrest. Therefore, the actual number of patients with cardiac cause might be lower.

Advanced Life Support

Advanced airway placement by EMS responders was associated with a decrease in 30‐day neurological intact, 30‐day survival, and pre‐hospital ROSC in both EMS‐responder‐witnessed and bystander‐witnessed groups (Tables 3 and 4). This finding suggested that advanced airway management might have been performed on patients with cardiac arrest who were in more critical conditions than those who achieved ROSC by BLS. However, multiple regression analysis demonstrated that in the EMS‐responder‐witnessed subgroups of patients undergoing advanced airway management, the update from the CoSTR 2010 to the CoSTR 2015 resulted in favorable 30‐day neurological outcomes, 30‐day survival, and pre‐hospital ROSC (Table 6). During the study period, the number of ELSTs increased from 26 533 to 40 043, ²⁰ which could indicate improved outcomes. In contrast, multiple logistic regression analysis for the bystander‐witnessed subgroups of patients subjected to advanced airway management, revealed that the update from the CoSTR 2010 to CoSTR 2015 was associated with pre‐hospital ROSC but not with favorable 30‐day neurological outcomes and survival (Table 6). These findings indicated that not only early advanced airway management but also implementation of high‐quality early chest compressions might be more critical for the prevention of neurological complications.

Saline administration by EMS responders was associated with the increase in 30‐day survival and pre‐hospital ROSC, but not neurologically intact survival rates in both EMS‐responder‐witnessed and bystander‐witnessed groups (Tables 3 and 4). Although saline administration has been related to metabolic acidosis and acute kidney injury, saline is the most commonly used fluid during CPR worldwide. ²⁸ Thus, whether saline is the optimal intravenous fluid remains unclear. In the future, studies on optimal fluid use are needed.

Epinephrine administration was not related to the increase in neurologically intact survival rates in both EMS‐responder‐witnessed and bystander‐witnessed groups (Tables 3 and 4). One‐time epinephrine administration was associated with improved 30‐day survival in the bystander‐witnessed group but not in the EMS responder‐witnessed group. Multiple epinephrine administration was indicative of improvements in ROSC in prehospital fields in both groups. Similar to the observations of the present study, some studies have demonstrated that epinephrine administration following a nonshockable cardiac arrest led to survival discharge and ROSC, but was not associated with neurologically intact survival rates. ²⁹ , ³⁰ The mechanisms by which epinephrine administration influences neurological outcomes in post‐cardiac arrest have not been elucidated to date. ³¹ In addition, some studies demonstrated that early administration of epinephrine was associated with higher survival rates in nonshockable cardiac arrest patients. ³² , ³³ Although, in this study, in the bystander‐witnessed group, which was associated with a longer time interval from a witnessed cardiac arrest to epinephrine administration, one‐time epinephrine administration was associated with higher 30‐day survival, similar results were not observed in the EMS‐responder‐witnessed group, which was associated with a shorter time interval from a witnessed cardiac arrest to epinephrine administration (Table 2). Moreover, the number of ELSTs increased during the study period in Japan, ²⁰ which might be related to more promising outcomes. In the future, further epinephrine administration studies using the All‐Japan Utstein Registry on patients with nonshockable OHCA are warranted.

Limitations

This study has several limitations. First, since this was not a randomized controlled trial, all conclusions are limited to the found associations, and the causality could not be determined. Second, although our plans included an incorporation of an analysis of the Guidelines 2020 era, owing to the constraints associated with the COVID‐19 pandemic, resuscitation algorithms were significantly modified to protect CPR providers, and the comparisons between the 2020 data with data from the previous years could not be performed. The COVID‐19 pandemic caused profound changes in resuscitation care worldwide. Third, the detailed information regarding ongoing CPR efforts after arrival at the hospital, along with the details on post‐cardiac arrest care, was lacking. Although the volume of tertiary centers increased from 244 to 294 during the study period, ³⁴ we cannot access the information on the quality of critical care provided. Advances in post‐cardiac arrest care and the quality of critical care might have improved the 30‐day neurological outcomes. Finally, neurological outcomes were measured 30 days after the detected OHCA; however, the recovery could have been more gradual in some patients. ³⁵

CONCLUSIONS

In conclusion, compared with CPR maneuvers for nonshockable OHCA recommended in CoSTR 2010, outcomes during the period corresponding to CoSTR 2015 differed with respect to neurologically intact survival rates in Japan. Although these differences could reflect other time‐related factors or unmeasured confounders, they may be related to changes in performing early high‐quality CPR with further timely advanced life support.

Sources of Funding

None.

Disclosures

None.

Supporting information

Figure S1

Table S1