a b s t r a c t

Introduction: We aimed to investigate the association between bystander cardiopulmonary resuscitation (CPR) with and without Automated external defibrillator use and neurological outcomes after out-of-hospital cardiac arrest (OHCA) in Korea.

Methods: This cross-sectional study used a nationwide Korean OHCA registry between 2015 and 2019. Patients were categorised into no bystander CPR and bystander CPR with and without AED use groups. The primary out- come was good neurological recovery at discharge. We also analysed the interaction effects of place of arrest, response time, and whether the OHCA was witnessed.

Results: In total, 93,623 patients were included. Among them, 35,486 (37.9%) were in the no bystander CPR group, 56,187 (60.0%) were in the bystander CPR without AED use group, and 1950 (2.1%) were in the bystander CPR with AED use group. Good neurological recovery was demonstrated in 1286 (3.6%), 3877 (6.9%), and 208 (10.7%) patients in the no CPR, bystander CPR without AED use, and bystander CPR with AED use groups, respec- tively. Compared to the no bystander CPR group, the adjusted odds ratio (95% confidence intervals) for good neu- rological recovery was 1.54 (1.45-1.65) and 1.37 (1.15-1.63) in the bystander CPR without and with AED use groups, respectively. The effect of bystander CPR with AED use was more apparent in OHCAs with witnessed arrest and prolonged response time (>=8 min).

Conclusion: Bystander CPR was associated with better neurological recovery compared to no bystander CPR; however, the benefits of AED use were not significant. Efforts to disseminate bystander AED availability and ensure proper utilisation are warranted.

(C) 2023

1. Introduction

Previous studies have shown that bystander automated external de- fibrillator (AED) use increases the survival outcomes of out-of-hospital cardiac arrest (OHCA) [1-4], and current guidelines recommend layper- sons to use AEDs, regardless of the patient’s characteristics or arrest cir- cumstances [5,6]. Patients with cardiac arrest usually present with a shockable rhythm, such as ventricular fibrillation, before transitioning to asystole [7,8]; many patients with OHCA could therefore be candi- dates for early AED use. AED use in non-Shockable rhythms can also

* Corresponding author at: Laboratory of Emergency Medical Services, Seoul National University Hospital Biomedical Research Institute, 101 Daehak-Ro, Jongno-Gu, Seoul 03080, Korea.

E-mail address: [email protected] (J.H. Park).

increase the frequency and quality of bystander cardiopulmonary resus- citation (CPR) via automatic guidance prompts [9].

Using AED in the field requires additional resources. Bringing an AED at an arrest site requires a dedicated person to retrieve it, which can take a significant amount of time [10,11]. Owing to the time-dependent availability of many AEDs, bystanders may not be able to use them when needed [12]. Even when a bystander AED is brought, there may be cases where bystanders have difficulty to apply it on the patient im- mediately [13,14]. Furthermore, CPR administration could be interrupted during the time spent retrieving and applying a bystander AED, resulting in an overall decrease in the Quality of CPR [15]. Those probable negative effects of bystander AED use could be intensified when patients present with a Non-shockable rhythm.

The effects of bystander AED use may be minimal in communities with predominant Non-shockable rhythms, and Asian countries are known to have such characteristics compared to Western countries [16].

https://doi.org/10.1016/j.ajem.2023.01.033

0735-6757/(C) 2023

In a previous study, Asian countries had about half the percentage of shockable rhythms compared to Western countries (6.5% in Japan, 7.9% in Taiwan, 13% in South Korea, and 16.2% in Singapore vs. 18.7% in Denmark, 20% in the United States, 20.0% in the United Kingdom, 25.0% in Norway, and 36.5% in Australia). The AED usage rate was also lower in Asian countries than in the United States and Nordic countries [16], which might have resulted from reduced training or inappropriate use. Therefore, in communities with a high proportion of non-shockable rhythm and a low proportion of bystander AED use, the overall additional benefit of bystander AED use in patients with OHCA, regardless of the ini- tial rhythm, is uncertain. In this study, we aimed to evaluate the effect of bystander CPR with and without AED use on Survival outcomes in individ- uals with OHCA in a community with predominantly non-shockable rhythm and low proportion of AED usage. We also evaluated whether the effect of bystander CPR with and without AED use differed in various arrest circumstances, including the location of arrest, witnessing status, and Ambulance response time. We hypothesised that the effect of by- stander CPR with AED use would not be significantly different from that of bystander CPR without AED use; however, the effect of bystander CPR with and without AED use could differ according to arrest circumstances.

1. Methods
   1. Study design and setting

This cross-sectional retrospective study used a nationwide population-based OHCA registry in South Korea. This study complied with the Declaration of Helsinki, and the protocol was approved by the Institutional Review Board (IRB) of the study institution (IRB No. 1103-153-357). The requirement for informed consent was waived.

The National Fire Agency (NFA) exclusively operates the emergency medical service (EMS) system in South Korea, with a total population of approximately 50 million in 100,210 km². The service level of EMS is basic-to-intermediate; EMS providers can administer CPR at the scene and during transport with automatic external defibrillation and ad- vanced airway management or intravenous fluid infusion under direct medical control. The NFA implemented a nationwide dispatcher- assisted CPR program in 2011 according to international guidelines [17]. All dispatch centres established a program for detecting OHCA, in- structing bystander CPR and AED use via telephone, and reporting the process.

The bystander AED program in South Korea began with the Emer- gence Medical Law in 2008. AEDs should be installed in specific loca- tions, including government buildings, airports, train stations, ports, and multi-user facilities. A Good Samaritan Law was established in 2008, and in 2012, AEDs were also to be installed in apartment buildings with >500 households after a law revision [18]. The total number of by- stander AEDs was 45,183 nationwide in 2019 (87 bystander AEDs per 100,000 persons).

• 1. Data sources

The Korea OHCA Registry (KOHCAR), which captures all incident cases of OHCA in the country, was retrieved from the following four sources: EMS run sheets for basic ambulance operation information, EMS cardiac arrest registry, dispatcher CPR registry for Utstein factors, and OHCA registry for hospital care and outcomes. The EMS providers recorded the EMS run sheets and EMS cardiac arrest registry for all EMS-assessed OHCAs, and the dispatchers entered the data in the dis- patcher CPR registry for all identified potential OHCA cases. All EMS reg- istries for each patient were linked using ambulance dispatch numbers in the electronic database server of the NFA and were integrated as a single episode. Medical record reviewers from the Korea Disease Con- trol and Prevention Agency abstracted the recorded information on hos- pital care and outcomes from approximately 700 hospitals. Professional medical record review experts were trained to use the Utstein

guidelines to conduct a medical record review of variables related to the etiology, risks, and outcomes of OHCA. To ensure the quality of the medical record Review process, a Quality management committee of emergency physicians, epidemiologists, statisticians, and medical re- cord review experts analysed the data every month, while providing feedback to each medical record reviewer. Detailed descriptions of the training and quality control of medical record reviewers have also been provided in previous studies [19-21].

• 1. Study population

Patients with OHCA of presumed cardiac etiology, aged 18 years or older, between January 2015 and December 2019, were included in the study. Patients were excluded from the analysis if the cardiac arrest was witnessed by an EMS provider or had missing information on bystander CPR or survival outcomes.

• 1. Main outcomes

The primary outcome was good Neurologic recovery at the time of discharge from the hospital, which was defined as a cerebral perfor- mance category score of 1 (good cerebral performance; ability to work) or 2 (moderate cerebral disability; ability to perform daily activ- ities independently). The secondary outcome was survival until hospital discharge.

• 1. Variables and measurements

Information on bystander CPR was obtained from the EMS cardiac arrest registry. Bystander CPR was categorised into the following three groups: no bystander CPR, bystander CPR without AED use, and by- stander CPR with AED use. Cases in which only bystander AED was used, and bystander CPR was not administered were classified as a no bystander CPR group.

We collected the patients’ demographic factors, including age, sex, urbanization level of residence (metropolitan and non-metropolitan), and medical history (diabetes mellitus, hypertension, heart disease, and stroke). We also collected community-related EMS factors, which included the location of arrest (public vs. non-public area, witness sta- tus, initial cardiac rhythm (shockable, pulseless electrical activity, and asystole), prehospital defibrillation by a bystander or an EMS provider, prehospital advanced airway management, EMS response time interval (time from EMS call to arrival at the scene), EMS scene time interval (time from EMS arrival to departure from the scene), EMS transport time interval [time from departure from the scene to arrival at the emergency department (ED)], and return of spontaneous circulation (ROSC) upon arrival at the ED.

• 1. Statistical analysis

The demographic findings and survival outcomes of the study popu- lation according to bystander CPR were described. Continuous variables were compared using the Wilcoxon rank-sum test, and categorical var- iables were compared using the chi-squared test.

Unadjusted odds ratios (ORs) with 95% confidence intervals (CIs) for the study outcomes were calculated using logistic regression analysis with the no bystander CPR group as the reference. The model was ad- justed for patient factors (age, sex, medical history, urbanization level of residence, location of arrest, and whether the OHCA was witnessed). We also analysed whether the effect of bystander CPR differed accord- ing to place of arrest, witness status, or EMS response time using an in- teraction analysis. Different interaction models were constructed for each characteristic. We put an interaction term of bystander CPR group and each characteristic to the multivariable logistic regression model to estimate the effect of bystander CPR group changes by each characteristic on the study outcome. If there is an interactive effect

between bystander CPR group and each characteristic, the adjusted OR of bystander CPR group on the study outcome would be different ac- cording to each characteristic. Sensitivity analysis was performed based on subgroups for patient with bystander CPR, and adjusted ORs with 95% CIs for study outcomes were calculated using logistic regres- sion analysis with the bystander CPR without AED group as the refer- ence. All statistical analyses were performed using the SAS version 9.4 software (SAS Institute Inc., Cary, NC). P-values were based on a two- sided significance level of 0.05.

1. Results
   1. Baseline characteristics

Among 140,152 EMS-treated OHCAs, 93,623 were included in the final analysis (Fig. 1). Among them, 35,486 (37.9%) were in the no by- stander CPR group, 56,187 (60.0%) were in the bystander CPR without AED group, and 1950 (2.1%) were in the bystander CPR with AED group. The median (interquartile range, IQR) age of patients was 73.8 (60.6-81.6) for no bystander CPR group, 73.4 (59.3-82.0) for bystander CPR without AED group, and 69.4 (56.1-80.0) for bystander CPR with AED group. The proportion of public places in the bystander CPR with AED group was higher than that in other groups (17.6% in the no by- stander CPR group, 19.3% in the bystander CPR without AED group, and 38.4% in the bystander CPR with AED group). There were 4631 (13.1%) cases of EMS Initial shockable rhythm in the no bystander CPR group, 10,878 (19.4%) in the bystander CPR without AED group, and 375 (19.3%) in the bystander CPR with AED group (p < 0.001). Survival to discharge and good neurological recovery were observed in 2321 (6.5%) and 1286 (3.6%) patients, 5507 (9.8%) and 3877 (6.9%) patients,

and 310 (15.9%) and 208 (10.7%) patients in the no bystander CPR group, bystander CPR without AED group, and bystander CPR with AED group, respectively (both p < 0.001) (Table 1).

• 1. Main analysis

Multivariate analysis showed that patients with OHCA who received bystander CPR with or without AED use were more likely to have good neurological recovery (adjusted OR 1.54 [95% CI 1.45 to 1.65] in by- stander CPR without AED group, and 1.37 [1.15-1.63] in bystander CPR with AED group), survival to discharge (adjusted OR 1.24 [95% CI

1.17 to 1.31] in bystander CPR without AED group, and 1.31 [1.13-1.52] in bystander CPR with AED group), and prehospital ROSC (adjusted OR 1.28 [95% CI 1.22 to 1.34] in bystander CPR without AED group, and 1.29 [1.13-1.47] in bystander CPR with AED group) than pa- tients with no bystander CPR. However, there was no difference in

outcomes between patients who received bystander CPR with or with- out AED (Table 2). In the sensitivity analysis for patients with bystander CPR, there was also no difference in outcomes of bystander CPR with and without AED group (adjusted OR for good neurological recovery,

1.00 [95% CI 0.88 to 1.14]; adjusted OR for survival to discharge 1.06

[95% CI 0.92 to 1.22]; adjusted OR for prehospital ROSC 1.00 [95% CIs

0.88-1.14]).

In the interaction model, the positive effect of bystander CPR with- out AED was more prominent in the non-public setting (p for interac- tion: 0.02). The effect of both bystander CPR with and without bystander AED was more prominent in witnessed arrest and with pro- longed (>=8 min) EMS response time (all p for interaction: <0.02) (Table 3).

1. Discussion

Using a nationwide population-based OHCA registry in South Korea, this study evaluated the effect of bystander CPR with and without AED use on survival outcomes. We identified a positive association between bystander CPR with and without AED use and survival outcomes of adult, non-EMS-witnessed OHCA with presumed cardiac etiology. Al- though survival outcomes were significantly better in the bystander CPR with AED use group in the univariate analysis, we found that there was no significant difference in survival outcomes between by- stander CPR with and without AED in the multivariate analysis (Table 2).

While many studies have reported the benefit of bystander AED use on survival outcomes [1-4,15,22], there was no additional benefit of by- stander AED use compared with bystander CPR alone on survival out- comes in our study. The positive effect of bystander AED use can only be seen in shockable rhythms [15]; since Asian individuals show a lower proportion of shockable rhythm compared to Western popula- tions which composed the sample of studies reporting the benefit of by- stander AED use, our findings are understandable [2,15]. Similar to our setting, Japan also reported a low proportion of shockable rhythms [16], but previous studies reported the benefit of bystander AED use in Japan [3,4]. However, these studies only included patients with shock- able rhythms. Considering the lack of definite benefit of bystander AED for non-shockable rhythms and the possible negative effect of re- trieving an AED on the quality of CPR at the scene [10-14], the benefit of bystander AED use might be different if it was analysed for all patients with OHCA regardless of the initial rhythm.

The low proportion of bystander AED use could also be a reason for the non-significant benefit of bystander AED use itself. South Korea has been reported to have the lowest level of bystander AED use among countries with national OHCA registries [16]; The number of installed

Fig. 1. Patient selection flowchart. EMS, emergency medical service; OHCA, out-of-hospital cardiac arrest; CPR, cardiopulmonary resuscitation.

Table 1

Demographic findings of study population according to the initial cardiac rhythm assessment and subsequent defibrillation.

Total No bystander CPR Bystander CPR without AED

Bystander CPR with AED

P-value

Variables	N	%		N	%		N	%		N	%
Total	93,623	100		35,486	100		56,187	100		1950	100
Gender, Male	59,287	63.3		23,216	65.4		34,791	61.9		1280	65.6	<0.001
Age, years												<0.001
18-49	10,374	11.1		3665	10.3		6408	11.4		301	15.4
50-74	39,887	42.6		15,234	42.9		23,771	42.3		882	45.2
75-	43,362	46.3		16,587	46.7		26,008	46.3		767	39.3
Median (IQR)	73.5 (59.7-81.8)			73.8 (60.6-81.6)			73.4 (59.3-82.0)			69.4 (56.1-80.0)		<0.001
Past medical history
Diabetes	20,755	22.2		7964	22.4		12,367	22.0		424	21.7	0.278
Hypertension	31,796	34.0		11,666	32.9		19,443	34.6		687	35.2	<0.001
Heart disease	16,120	17.2		6043	17.0		9740	17.3		337	17.3	<0.001
Stroke	8517	9.1		3131	8.8		5192	9.2		194	9.9	<0.001
Year of arrest												<0.001
2015	18,299	19.5		7725	21.8		10,219	18.2		355	18.2
2016	18,291	19.5		7015	19.8		10,877	19.4		399	20.5
2017	18,172	19.4		6723	18.9		11,064	19.7		385	19.7
2018	19,195	20.5		7046	19.9		11,769	20.9		380	19.5
2019	19,666	21.0		6977	19.7		12,258	21.8		431	22.1
Residential area, metropolis	37,909	40.5		15,250	43.0		21,858	38.9		801	41.1	<0.001
Public place	17,832	19.0		6243	17.6		10,820	19.3		769	39.4	<0.001
Witnessed	42,674	45.6		12,947	36.5		28,500	50.7		1227	62.9	<0.001
Bystander defibrillation	521	0.6		10	0		0	0.0		511	26.2	<0.001
Initial ECG rhythm of EMS												<0.001
Shockable	15,885	17.0		4631	13.1		10,878	19.4		376	19.3
PEA	14,914	15.9		6561	18.5		7976	14.2		377	19.3
Asystole	62,824	67.1		24,294	68.5		37,333	66.4		1197	61.4
EMS management
IV	40,644	43.4	14,052		39.6	25,783		45.9	809		41.5	<0.001
Advanced airway management	53,876	57.5	19,370		54.6	33,437		59.5	1069		54.8	<0.001
Defibrillation	22,633	24.2	7397		20.8	14,709		26.2	527		27.0	<0.001
EMS time interval, median (IQR), min
Response time	7 (5-9)		7 (5-9)			7 (5-9)			7 (5-10)			<0.001
Call to first EMS defibrillation Scene time	11 (8-17) 5 (9-13)		12 (9-18) 5 (9-13)			11 (8-16) 5 (9-13)			13 (9-19) 5 (10-12)			<0.001
Transport time	9 (13-10)		9 (13-10)			9 (13-10)			10 (12-9)			<0.001
Level of ED												<0.0001
Level 1	17,207	18.4	6287		17.7	10,503		18.7	417		21.4
Level 2	44,229	47.2	16,597		46.8	26,611		47.4	1021		52.4
Level 3	32,187	34.4	12,602		35.5	19,073		33.9	512		26.3
post-resuscitation care
Reperfusion	5649	6.0	1618		4.6	3818		6.8	213		10.9	<0.001
TTM	902	1.0	298		0.8	566		1.0	38		1.9	<0.001
ECMO	902	1.0	566		1.6	38		0.1	38		1.9	<0.001
Outcomes
Prehospital ROSC	10,883	11.6	3131		8.8	7381		13.1	371		19.0	<0.001
Survival to discharge	8138	8.7	2321		6.5	5507		9.8	310		15.9	<0.001
Good neurological recovery	5371	5.7	1286		3.6	3877		6.9	208		10.7	<0.001

AED, automated external defibrillator; IQR, interquartile range; CPR, cardiopulmonary resuscitation; EMS, emergency medical service; PEA, pulseless electrical activity; IV, intravenous; ED, emergency department; TTM, Targeted temperature management; ECMO, extracorporeal membrane oxygenation; ROSC, return-of-spontaneous circulation.

Table 2

Logistic regression analysis of survival outcomes according to the initial cardiac rhythm assessment.

	Total	Positive		Unadjusted	Adjusted +	Adjusted +
Outcomes	N	N	%	OR (95% CI)	OR (95% CI)	OR (95% CI)
Good neurological recovery
No bystander CPR	35,486	1286	3.6	Reference	Reference	0.65 (0.61-0.70)
Bystander CPR without AED	56,187	3877	6.9	1.97 (1.85-2.10)	1.54 (1.43-1.65)	Reference
Bystander CPR with AED	1950	208	10.7	3.18 (2.72-3.71)	1.37 (1.15-1.63)	0.89 (0.75-1.05)
Survival to discharge
No bystander CPR	35,486	2321	6.5	Reference	Reference	0.81 (0.77-0.86)
Bystander CPR without AED	56,187	5507	9.8	1.55 (1.48-1.63)	1.24 (1.17-1.31)	Reference
Bystander CPR with AED	1950	310	15.9	2.70 (2.38-3.07)	1.31 (1.13-1.52)	1.06 (0.92-1.22)
Prehospital ROSC No bystander CPR	35,486	3131	8.8	Reference	Reference	0.78 (0.75-0.82)
Bystander CPR without AED	56,187	7381	13.1	1.56 (1.50-1.63)	1.28 (1.22-1.34)	Reference
Bystander CPR with AED	1950	371	19.0	2.43 (2.16-2.74)	1.29 (1.13-1.47)	1.01 (0.89-1.15)

OR, odds ratio; CI, confidence interval; AED, automated external defibrillator; ROSC, return-of-spontaneous circulation.

*Adjusted for age, sex, past medical history (diabetes, hypertension, heart disease, stroke), year, urbanization level of residence, place of arrest, and witness status.

⁺ Same models with different references.

Table 3

Interaction analysis on good neurological recovery according to place of arrest, witness status, and response time.

Total	Positive		AOR	shock was protective against the negative impact of response time de-
N	N	%	(95% CI)?	lays on patient survival; our study also showed similar results [34].

AED use [32,33]. We also found that the effect of bystander CPR with and without bystander AED use was prominent in cases of delayed EMS response time. Clara et al. reported that bystander-delivered AED

Witness status Witnessed

No bystander CPR	29,243	641	2.2	Reference
Bystander CPR without AED	45,367	1939	4.3	1.66 (1.51-1.82)??
Bystander CPR with AED	1181	50	4.2	1.32 (0.97-1.80)

No bystander CPR	12,947	817	6.3	Reference	ate. Fourth, we analysed all arrest locations, including non-public areas.
Bystander CPR without AED	28,500	3244	11.4	1.78 (1.63-1.94)??	In South Korea, apartment buildings with >500 households are re-

Place of arrest Public place
No bystander CPR	6243	645	10.3	Reference ??
Bystander CPR without AED	10,820	1938	17.9	1.39 (1.26-1.55)
Bystander CPR with AED Non-public place	769	158	20.5	1.33 (1.07-1.64)

Our study has several strengths. First, this study analysed a nation- wide population-based OHCA registry in South Korea, and we could se- cure enough power for the analyses. Second, this is a leading study that analysed the effectiveness of bystander AED in a setting with a low pro- portion of shockable rhythms and bystander AED usage. Third, we analysed all cardiac arrest rhythms. Several studies have focused on shockable rhythms [1,7]; However, we thought that in a bystander CPR situation, it is difficult to identify the initial cardiac arrest rhythm; therefore, we thought this all-Rhythm analysis would be more appropri-

Bystander CPR with AED 1227 186 15.2 1.59 (1.31-1.92)??

Non-witnessed

No bystander CPR 22,539 469 2.1 Reference

Bystander CPR without AED	27,687	633	2.3	1.12 (0.99-1.27)??	areas are common targets of AED usage, the effectiveness of AED
Bystander CPR with AED	723	22	3.0	0.91 (0.58-1.42)??	usage for all OHCA cases needs to be analysed since the benefits and po-

quired by law to install bystander AEDs, with the aim of increasing AED use in residential areas. Although shockable rhythms and public

Response time

Response time < 8 min

No bystander CPR	14,650	297	2.0	Reference
Bystander CPR without AED	23,353	1190	5.1	1.49 (1.37-1.61)??
Bystander CPR with AED Response time >= 8 min	947	90	9.5	1.20 (0.96-1.51)??
No bystander CPR Bystander CPR without AED	20,836 32,834	989 2687	4.7 8.2	Reference 1.79 (1.56-2.05)?? ??
Bystander CPR with AED	1003	118	11.8	1.99 (1.51-2.61)

AOR, adjusted odds ratio; CI, confidence interval; CPR, cardiopulmonary resuscitation; AED, automated external defibrillator.

* Adjusted for age, sex, past medical history (diabetes, hypertension, heart disease,

stroke), year, urbanization level of residence, place of arrest, witness status, and interac- tion term (bystander CPR * place of arrest or bystander CPR * witness status).

?? p for interaction<0.05.

bystander AEDs was 87 per 100,000 persons in 2019, which is small compared to other countries (Denmark: 429 per 100,000 persons [11]; Japan: 474 per 100,000 persons) [23]. In addition, significant pro- portions of AED were reported as not ready for use and not having 24-h accessibility in a recent study [24,25]. Moreover, the mismatches be- tween the number of installed AEDs and the annual rate of AED use among places and provinces were also reported, which suggested that an AED redistribution strategy would be needed [18]. Public awareness is also low; according to a national survey conducted by the National emergency medical center in 2020, only 59.1% of lay public are confi- dent in performing CPR and 40.1% of lay public recognised AEDs, which is a small proportion compared to other countries (United States: 79% of lay public confident in performing CPR and 98% recognised AEDs) [26]. Poor accessibility and lack of education in the use of AED could increase the time and resources to retrieve an AED and would contribute to a longer CPR interruption, which may contrib- ute to lowering the effectiveness of bystander AED usage. Improving ac- cessibility to AEDs and strengthening public education is a priority for increasing the effectiveness of bystander AED use [14]. Dispatcher- assisted AED use using text message alarms or Smartphone applications could be an important solution for low AED usage status [27-29]. A pilot trial of text-message alarms was conducted in one city in South Korea, but it could not be expanded [30]. Continuing efforts are warranted to increase the effectiveness of bystander AED use in South Korea.

In the interaction analysis, the effect of bystander CPR without AED was more prominent in the non-public setting. Most bystander AEDs are installed in public areas that are difficult to access in the private sec- tor [31]. Therefore, the usability of bystander AED in private areas could be limited. We also found that in witnessed arrests, the effect of by- stander CPR with or without AED was greater than that of having no by- stander CPR. Since most patients with an unwitnessed status had a long No-flow time, the effect of bystander CPR may be reduced regardless of

tential harm differ according to the various OHCA circumstances. We also analysed the different effects of bystander CPR and AED in several circumstances using an interaction analysis.

This study had several limitations. First, bystander CPR and AED use were classified according to the EMS records obtained from bystanders at the scene by EMS providers. It is possible that bystander CPR and AED use rates were under- or overestimated. Second, detailed information on bystander CPR quality and AED use was not obtained due to a lack of information. Third, this study did not include a randomised controlled model. There is a significant potential of a confounding effect. Finally, this study was conducted in a single EMS system with an intermediate service level, which differs from that of countries with advanced Service levels. The generalisation of the study findings should be made with caution.

1. Conclusions

Bystander CPR with and without AED use in OHCA was associated with better neurologic outcomes than no bystander CPR; however, there was no significant difference between bystander CPR with or without AED. The low proportion of shockable rhythms and bystander AED use could contribute to our results. Since bystander AED use was not harmful to overall OHCA patients and bystander AED use has a strong potential to improve survival outcomes in shockable rhythm pa- tients, efforts to increase accessibility and strengthen public education are still needed. Therefore, prospective Intervention studies are war- ranted to increase the effectiveness of bystander AED use.

Funding

None.

CRediT authorship contribution statement

Sang Hun Kim: Writing - original draft, Formal analysis, Data curation. Jeong Ho Park: Writing - review & editing, Writing - original draft, Methodology, Formal analysis, Data curation, Conceptualization. Joo Jeong: Writing - review & editing, Validation, Supervision, Investi- gation. Young Sun Ro: Writing - review & editing, Methodology, Formal analysis, Conceptualization. Ki Jeong Hong: Writing - review & editing, Validation, Methodology. Kyoung Jun Song: Writing - review & editing, Project administration, Formal analysis, Data curation, Conceptualiza- tion. Sang Do Shin: Writing - review & editing, Validation, Investiga- tion, Formal analysis, Conceptualization.

Declaration of Competing Interest

None.

Acknowledgements

This study was supported by the National Fire Agency of Korea and the Korea Disease Control and Prevention Agency (KDCA).

References

1. Pollack RA, Brown SP, Rea T, Aufderheide T, Barbic D, Buick JE, et al. Impact of by- stander automated external defibrillator use on survival and functional outcomes in shockable observed public cardiac arrests. Circulation. 2018;137(20):2104-13. https://doi.org/10.1161/CIRCULATIONAHA.117.030700.
2. Blom MT, Beesems SG, Homma PC, Zijlstra JA, Hulleman M, van Hoeijen DA, et al. Improved survival after out-of-hospital cardiac arrest and use of automated external defibrillators. Circulation. 2014;130(21):1868-75. https://doi.org/10.1161/ CIRCULATIONAHA.114.010905.
3. Nakashima T, Noguchi T, Tahara Y, Nishimura K, Yasuda S, Onozuka D, et al. Public- access defibrillation and neurological outcomes in patients with out-of-hospital car- diac arrest in Japan: a population-based cohort study. Lancet. 2019;394(10216): 2255-62. https://doi.org/10.1016/S0140-6736(19)32488-2.
4. Kitamura T, Kiyohara K, Sakai T, Matsuyama T, Hatakeyama T, Shimamoto T, et al. Public-access defibrillation and out-of-hospital cardiac arrest in Japan. N Engl J Med. 2016;375(17):1649-59. https://doi.org/10.1056/NEJMsa1600011.
5. Olasveengen TM, Semeraro F, Ristagno G, Castren M, Handley A, Kuzovlev A, et al. European resuscitation council guidelines 2021: basic life support. Resuscitation. 2021;161:98-114. https://doi.org/10.1016/j.resuscitation.2021.02.009.
6. Panchal AR, Bartos JA, Cabanas JG, Donnino MW, Drennan IR, Hirsch KG, et al. Part 3: adult basic and advanced life support: 2020 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2020;142(16_suppl_2):S366-468. https://doi.org/10.1161/CIR.0000000000000916.
7. Kiyohara K, Nishiyama C, Kitamura T, Matsuyama T, Sado J, Shimamoto T, et al. The association between Public access defibrillation and outcome in witnessed out-of- hospital cardiac arrest with shockable rhythm. Resuscitation. 2019;140:93-7. https://doi.org/10.1016/j.resuscitation.2019.05.017.
8. Gilmore CM, Rea TD, Becker LJ, Eisenberg MS. Three-phase model of cardiac arrest: time-dependent benefit of bystander cardiopulmonary resuscitation. Am J Cardiol. 2006;98(4):497-9. https://doi.org/10.1016/j.amjcard.2006.02.055.
9. Pollack RA, Brown SP, May S, Rea T, Kudenchuk PJ, Weisfeldt ML. Bystander auto- mated external defibrillator application in non-shockable out-of-hospital cardiac ar- rest. Resuscitation. 2019;137:168-74. https://doi.org/10.1016/j.resuscitation.2019. 02.007.
10. Sondergaard KB, Hansen SM, Pallisgaard JL, Gerds TA, Wissenberg M, Karlsson L, et al. Out-of-hospital cardiac arrest: probability of bystander defibrillation relative to distance to nearest automated external defibrillator. Resuscitation. 2018;124: 138-44. https://doi.org/10.1016/j.resuscitation.2017.11.067.
11. Karlsson L, Malta Hansen C, Wissenberg M, Moller Hansen S, Lippert FK, Rajan S, et al. Automated external defibrillator accessibility is crucial for bystander defibrilla- tion and survival: a registry-based study. Resuscitation. 2019;136:30-7. https://doi. org/10.1016/j.resuscitation.2019.01.014.
12. Deakin CD, Anfield S, Hodgetts GA. Underutilisation of public access defibrillation is related to retrieval distance and time-dependent availability. Heart. 2018;104(16): 1339-43. https://doi.org/10.1136/heartjnl-2018-312998.
13. Andre AD, Jorgenson DB, Froman JA, Snyder DE, Poole JE. Automated external defi- brillator use by untrained bystanders: can the public-use model work? Prehosp Emerg Care. 2004;8(3):284-91. https://doi.org/10.1016/j.prehos.2004.02.004.
14. Delhomme C, Njeim M, Varlet E, Pechmajou L, Benameur N, Cassan P, et al. Auto- mated external defibrillator use in out-of-hospital cardiac arrest: current limitations and solutions. Arch Cardiovasc Dis. 2019;112(3):217-22. https://doi.org/10.1016/j. acvd.2018.11.001.
15. Holmberg MJ, Vognsen M, Andersen MS, Donnino MW, Andersen LW. Bystander au- tomated external defibrillator use and clinical outcomes after out-of-hospital cardiac arrest: a systematic review and meta-analysis. Resuscitation. 2017;120:77-87. https://doi.org/10.1016/j.resuscitation.2017.09.003.
16. Kiguchi T, Okubo M, Nishiyama C, Maconochie I, Ong MEH, Kern KB, et al. Out-of- hospital cardiac arrest across the world: first report from the International Liaison Committee on Resuscitation . Resuscitation. 2020;152:39-49. https://doi. org/10.1016/j.resuscitation.2020.02.044.
17. Travers AH, Rea TD, Bobrow BJ, Edelson DP, Berg RA, Sayre MR, et al. Part 4: CPR overview: 2010 American Heart Association guidelines for cardiopulmonary resusci- tation and emergency cardiovascular care. Circulation. 2010;122(18 Suppl 3): S676-84. https://doi.org/10.1161/CIRCULATIONAHA.110.970913.
18. Oh JH, Cho GC, Ryoo SM, Han SH, Woo SH, Jang YS, et al. Mismatches between the number of installed automated external defibrillators and the annual rate of auto- mated external defibrillator use among places. Prehosp Disaster Med. 2021;36(2): 183-8. https://doi.org/10.1017/S1049023X20001508.
19. Ro YS, Shin SD, Lee YJ, Lee SC, Song KJ, Ryoo HW, et al. Effect of Dispatcher-assisted cardiopulmonary resuscitation program and location of out-of-hospital cardiac ar- rest on survival and neurologic outcome. Ann Emerg Med. 2017;69(1). https://doi. org/10.1016/j.annemergmed.2016.07.028. 52-61 e1.
20. Ro YS, Shin SD, Song KJ, Lee EJ, Kim JY, Ahn KO, et al. A trend in epidemiology and outcomes of out-of-hospital cardiac arrest by urbanization level: a nationwide ob- servational study from 2006 to 2010 in South Korea. Resuscitation. 2013;84(5): 547-57. https://doi.org/10.1016/j.resuscitation.2012.12.020.
21. Park JH, Ro YS, Shin SD, Song KJ, Hong KJ, Kong SY. Dispatcher-assisted bystander cardiopulmonary resuscitation in rural and urban areas and survival outcomes after out-of-hospital cardiac arrest. Resuscitation. 2018;125:1-7. https://doi.org/10. 1016/j.resuscitation.2018.01.026.
22. Grubic N, Peng YP, Walker M, Brooks SC, Group CS. Bystander-initiated cardiopul- monary resuscitation and automated external defibrillator use after out-of- hospital cardiac arrest: uncovering disparities in care and survival across the urban-rural spectrum. Resuscitation. 2022;175:150-8. https://doi.org/10.1016/j.re- suscitation.2022.04.014.
23. Kiguchi T, Kiyohara K, Kitamura T, Nishiyama C, Kobayashi D, Okabayashi S, et al. Public-access defibrillation and survival of out-of-hospital cardiac arrest in public vs. residential locations in Japan. Circ J. 2019;83(8):1682-8. https://doi.org/10. 1253/circj.CJ-19-0065.
24. Kim TY, Jung YK, Yoon SH, Kim SJ, Cha KC, Jung WJ, et al. Trends in maintenance sta- tus and usability of public automated external defibrillators during a 5-year on-site inspection. Sci Rep. 2022;12(1):10738. https://doi.org/10.1038/s41598-022-14611-1.
25. Yoon CG, Jeong J, Kwon IH, Lee JH. Availability and use of public access defibrillators in Busan Metropolitan City, South Korea. Springerplus. 2016;5(1):1524. https://doi. org/10.1186/s40064-016-3201-6.
26. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, et al. Heart disease and stroke statistics-2014 update: a report from the American Heart Association. Circulation. 2014;129(3):e28-292. https://doi.org/10.1161/01.cir.0000441139. 02102.80.
27. Zijlstra JA, Stieglis R, Riedijk F, Smeekes M, van der Worp WE, Koster RW. Local lay rescuers with AEDs, alerted by text messages, contribute to Early defibrillation in a Dutch out-of-hospital cardiac arrest dispatch system. Resuscitation. 2014;85(11): 1444-9. https://doi.org/10.1016/j.resuscitation.2014.07.020.
28. Andelius L, Malta Hansen C, Lippert FK, Karlsson L, Torp-Pedersen C, Kjaer Ersboll A, et al. Smartphone activation of citizen responders to facilitate defibrillation in out- of-hospital cardiac arrest. J Am Coll Cardiol. 2020;76(1):43-53. https://doi.org/10. 1016/j.jacc.2020.04.073.
29. Ringh M, Rosenqvist M, Hollenberg J, Jonsson M, Fredman D, Nordberg P, et al. Mobile-phone dispatch of laypersons for CPR in out-of-hospital cardiac arrest. N Engl J Med. 2015;372(24):2316-25. https://doi.org/10.1056/NEJMoa1406038.
30. Lee SY, Shin SD, Lee YJ, Song KJ, Hong KJ, Ro YS, et al. Text message alert system and resuscitation outcomes after out-of-hospital cardiac arrest: a before-and-after population-based study. Resuscitation. 2019;138:198-207. https://doi.org/10.1016/ j.resuscitation.2019.01.045.
31. Sarkisian L, Mickley H, Schakow H, Gerke O, Starck SM, Jensen JJ, et al. Use and cov- erage of automated external defibrillators according to location in out-of-hospital cardiac arrest. Resuscitation. 2021;162:112-9. https://doi.org/10.1016/j.resuscita- tion.2021.01.040.
32. Chia MYC, Kwa TPW, Wah W, Yap S, Doctor NE, Ng YY, et al. Comparison of out- comes and characteristics of emergency medical services (EMS)-witnessed, bystander-witnessed, and unwitnessed out-of-hospital cardiac arrests in Singapore. Prehosp Emerg Care. 2019;23(6):847-54. https://doi.org/10.1080/ 10903127.2019.1587124.
33. Sasson C, Rogers MA, Dahl J, Kellermann AL. Predictors of survival from out-of- hospital cardiac arrest: a systematic review and meta-analysis. Circ Cardiovasc Qual Outcomes. 2010;3(1):63-81. https://doi.org/10.1161/CIRCOUTCOMES.109. 889576.
34. Stoesser CE, Boutilier JJ, Sun CLF, Brooks SC, Cheskes S, Dainty KN, et al. Moderating effects of out-of-hospital cardiac arrest characteristics on the association between EMS response time and survival. Resuscitation. 2021;169:31-8. https://doi.org/10. 1016/j.resuscitation.2021.10.014.