a b s t r a c t

Introduction: The best location for safe and timely implementation of extracorporeal cardiopulmonary resuscita- tion (ECPR) is currently uncertain. We aimed to evaluate the association between the location of ECPR and sur- vival outcomes in out-of-hospital cardiac arrest (OHCA) patients. We also evaluated whether the effects of ECPR location on survival differed between patients who underwent Coronary angiography and those who did not.

Methods: We used data collected between 2013 and 2020 from a nationwide OHCA database. Adult OHCA pa- tients with presumed cardiac etiology who underwent ECPR were included in the study. The primary outcome was survival to discharge. The main exposure was the ECPR location (emergency department [ED] or cardiac catheterization laboratory [Cath lab]). We compared primary outcomes of ECPR between the ED and Cath lab using multivariable logistic regression. The interaction between ECPR location and CAG was also evaluated.

Results: Of 564 ECPR patients, 448 (79.4%) and 116 (20.6%) underwent ECPR in the ED and Cath lab, respectively. CAG was observed in 52.5% and 72.4% of the patients in the ED and Cath lab groups, respectively. There were no significant differences in survival to discharge between the ED and Cath lab groups (14.1% vs. 12.9%, p = 0.75, adjusted odds ratio [AOR] [95% confidence interval] 1.87 [0.85-4.11]). AOR of interaction analysis (95% CI) for survival to discharge of the ED group was 2.34 (1.02-5.40) in patients with CAG and 0.28 (0.04-1.84) in patients without CAG (p for interaction was 0.04).

Conclusion: In adult OHCA patients who underwent ECPR and CAG, ECPR in the ED shortened time to ECMO pump-on time and increased survival to discharge compared to ECPR in the Cath lab.

(C) 2022

1. Introduction

Out-of-hospital cardiac arrest (OHCA) constitutes a major public health burden worldwide [1]. Despite advances in resuscitation tech- niques and systematic post-cardiac arrest treatment, few patients can be successfully resuscitated. Extracorporeal cardiopulmonary resuscita- tion (ECPR) is the application of extracorporeal membrane oxygenation on patients who have failed to achieve sustained spontaneous

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

E-mail addresses: [email protected] (Y. Kim), [email protected] (J.H. Park), [email protected] (S.Y. Lee), [email protected] (Y.S. Ro), [email protected] (K.J. Hong), [email protected] (K.J. Song), [email protected] (S.D. Shin).

circulation recovery (ROSC) with conventional cardiopulmonary resus- citation (CPR) measures [2]. ECPR maintains cerebral perfusion and per- fusion of other vital organs until Reversible causes of OHCA can be identified and treated [3]. Although the evidence to recommend routine use of ECPR in cardiac arrest patients is insufficient [4,5], some guide- lines suggest that selected cardiac arrest patients be considered for ECPR if there are potential reversible causes [6,7].

Clear guidelines on target patient groups, appropriate timing, and appropriate location for ECPR are currently lacking despite results from several observational studies [2,8-10]. With respect to target pa- tients and timing, the Extracorporeal life support Organization (ELSO) recommends that ECPR be considered for patients with good prognosis, such as young age (< 70 years), witnessed arrest, bystander CPR, or ini- tial shockable rhythm patients; the goal of ECPR is to enable appropriate application of ECMO within 60 min of cardiac arrest [2,11,12]. ECPR can

https://doi.org/10.1016/j.ajem.2022.11.013

0735-6757/(C) 2022

be conducted at various hospital locations, but is most often conducted in the emergency department (ED) or cardiac catheterization laboratory (Cath lab) [5]. The best location for safe and timely implementation of ECPR is currently uncertain.

Acute coronary syndrome is a major reversible cause of OHCA in adults with presumed cardiac etiology; emergency coronary angiogra- phy (CAG) identifies culprit coronary arteries and enables immediate treatment through percutaneous coronary intervention (PCI) in resusci- tated OHCA patients [13,14]. Current Resuscitation guidelines recom- mend performing emergency CAG in OHCA patients with suspicious Cardiac origin and ST elevation on electrocardiogram immediately after ROSC [6,7]. ELSO recommends emergency coronary angiography for all ECPR patients without an overt non-cardiac cause, regardless of age and post-resuscitation rhythm [2]. For better survival outcomes, it is important to minimize low flow time by rapidly applying ECMO to OHCA patients who require ECPR. In addition, it is important to re- duce the time to reperfusion because the longer the time to coronary reperfusion, the more severe myocardial damage. ECPR in the ED can reduce low flow time by shortening ECMO application time but pro- long the coronary reperfusion time by transporting to the Cath lab with ECMO devices. Conversely, ECPR in the Cath lab can decrease coronary reperfusion time but increase low flow time. Although there are risks involved in transportation of OHCA patients to the Cath lab while performing resuscitation and a possibility that CAG may not be performed after ECMO insertion, the Cath lab may be bet- ter prepared for ECMO implementation than the ED, and the benefits of immediate CAG could be maximised if ST-segment elevation was seen on electrocardiograms following ROSC. A recent study reported a Survival benefit for early ECMO-facilitated resuscitation in the Cath lab among OHCA patients [15].

There is no clear association between ECPR location and Survival outcomes in OHCA patients who received ECPR. As ECPR locations may have differing effects on the time to ECMO application and time to reperfusion, the association between ECPR locations and survival out- comes might be different if CAG was performed. This study aimed to evaluate the association between ECPR locations and survival outcomes in OHCA patients who received ECPR. We also evaluated whether the association between ECPR and survival outcomes differed according to CAG. We hypothesised that if CAG was not performed, the survival out- comes of patients may be poor regardless of ECPR locations. Because minimising low flow time might be more effective than reducing time to reperfusion in OHCA patients [16], we also hypothesised that if both ECPR and CAG were performed, ECPR in the ED might have a sur- vival benefit compared to ECPR in the Cath lab.

1. Methods
   1. Study setting

In Korea, emergency medical services (EMS) is provided by the National Fire Agency, a government-based system which offers basic- to-intermediate level ambulance services from fire departments and serves approximately 50 million people over a 100,210 km² area. All OHCA patients are transported to the nearest ED of a PCI-capable hospi- tal by protocol; however, this is not mandatory. The national govern- ment has designated the following three levels of ED: Level 1 (n = 36) and Level 2 (n = 119) EDs, which have abundant resources and fa- cilities and an emergency medical doctor on staff 24 h per day, 7 days per week; and Level 3 EDs (n = 261), which may be staffed by general physicians. All OHCA patients are admitted to the ED and proceed with ACLS. Decisions regarding ECPR, location of procedure, method of vascu- lar access, and device type are determined by the attending physicians at each centre. If ECMO application is necessary, ECPR is performed in the emergency room or moved to the Cath lab depending on the situa- tions of medical resources including hospital facilities and medical staff that can insert the ECMO. In 2020, 74 EDs conducted at least one ECLS

for OHCA patients, and the median case volume for ECLS in OHCA pa- tients was 4 (interquartile range [IQR] 2-6). Age, comorbidities, and ini- tial rhythm are used as inclusion criteria for ECPR, and specialised teams for ECPR have been established at some centres [17,18]. General acute care and post-resuscitation care are provided according to international standard guidelines by Attending emergency physicians. In most level 1 and 2 EDs, CAG is performed by a cardiologist in the Cath lab [19]. Emergency physicians are responsible for making critical decisions on patient care.

• 1. Data sources

We obtained data from the nationwide Korea Out-of-Hospital Car- diac Arrest Registry (KOHCAR), which captures all cases of OHCA occur- ring nationwide from four sources: the EMS run sheets for basic ambulance operation information, EMS CPR registries, dispatcher CPR registries, and review of hospital medical records on hospital care and outcomes from approximately 700 hospitals. To ensure the quality of the medical records, Quality management committees (comprising emergency physicians, epidemiologists, statistical experts, and medical record review experts) analyze monthly data and send feedback to each reviewer. To ensure the quality of the medical records, quality management committees (comprising emergency physicians, epidemi- ologists, statistical experts, and medical record review experts) analyze monthly data and send feedback to each reviewer. We used data for which quality control was completed by the quality management com- mittee in this study.

• 1. Study participants

Data from January 2013 to December 2020 were included for OHCA adults (>=18 years) with presumed cardiac etiology who underwent ECPR in the ED or Cath lab. The presumed cardiac etiology was reviewed

by examining patient medical records and excluding trauma, drowning, poisoning, burns, asphyxia, and hanging. Patients were excluded if they had missing ECMO location information, ECMO initiation (pump-on) time, or CPR termination time. As 99% of patients received ECPR in the ED or Cath lab, patients who received ECPR elsewhere (Operating rooms or intensive care unit) were also excluded.

• 1. Outcome measures

The primary outcome was survival until hospital discharge. The sec- ondary outcome was good neurological status, defined as a cerebral per- formance category (CPC) score of 1 (good recovery) or 2 (mild disability).

• 1. Variables

The main exposure was the location where ECPR was performed. ECPR was defined as successful veno-arterial ECMO implantation and pump-on during cardiac massage. The location of ECPR was determined by ECMO cannulation and classified as ED or Cath lab.

We collected information on age, sex, medical history (diabetes mellitus, hypertension, heart disease, and stroke), place of arrest (public or others), and bystander CPR (yes or no). We also collected information on the type of initial cardiac rhythm (shockable or non-Shockable rhythms), EMS resuscitation (defibrillation, advanced airway manage- ment, and epinephrine use by EMS providers), response time (call to ambulance arrival at the scene), scene time (arrival to departure from the scene), transport time (scene departure to ED arrival), level of ED (Level 1, 2 or 3), Targeted temperature management, coronary angiogra- phy, percutaneous coronary intervention, and total ECMO duration (time from ECMO pump-on to ECMO turn-off time).

• 1. Statistical analysis

Demographic findings and outcomes according to ECPR location and CAG implementation are provided. Categorical variables were compared using the chi-square test and continuous variables were compared using the Wilcoxon rank-sum test. Multivariable logistic regression anal- ysis was conducted to estimate the effects of ECPR location on survival outcomes for all study patients and to calculate adjusted odds ratios (AOR) and 95% confidence intervals (CI) after adjusting for age, sex, loca- tion of arrest, witnessed status, bystander cardiopulmonary resuscitation, metropolitan area, pre-hospital advanced airway management, pre- hospital epinephrine use, and EMS defibrillation. We added an interaction term of ECPR location and coronary angiography implementation to the multivariable logistic regression model to estimate the effects of ECPR lo- cation on survival outcomes based on CAG implementation. Because some patients could not undergo PCI after CAG, our sensitivity analysis also included an interaction term between ECPR location and PCI.

A p value <0.05 was considered the level of significance for all anal-

yses. We tested for multicollinearity between the covariables in the models using the variance inflation factor (VIF), and there was no mul- ticollinearity (VIF < 10 in all variables). All statistical analyses were per- formed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

• 1. Ethical statements

The study was conducted in accordance with the principles of the Declaration of Helsinki. This study was approved by the Institutional Review Board (IRB) of study institution, and the requirement for informed consent was waived owing to the retrospective nature of the study (IRB No. SNUH-1103-153-357).

1. Results
   1. Demographics

Of 222,554 EMS-treated OHCA patients, 564 underwent ECPR in the ED or Cath lab and were included in the analysis (Fig. 1).

Demographic characteristics according to ECPR locations are shown in Table 1. Sex, age, witnessed status, proportion of bystander CPR, and proportion of Initial shockable rhythm presence were similar between the ED and Cath lab groups. The median (interquartile range [IQR]) from ED arrival to ECMO pump-on time was 43.0 (34.0-59.0) minutes in the ED group and 57.0 (43.0-78.0) minutes in the Cath lab group (p < 0.01). There was no significant difference in the total time that pa- tients were on ECMO between the ED and Cath lab groups (p = 0.66). The ED group had similar outcomes as the Cath lab group: 14.1% vs. 12.9% for survival to discharge and 9.4% vs. 11.2% for good neurological recovery (both p > 0.05) (Table 1).

Demographic findings of the coronary angiography patients are shown in Table 2. The non-CAG group had poorer outcomes than the CAG group (4.5% vs. 21.0% for survival to discharge and 2.9% vs. 15.0% for good neurological recovery, both p < 0.01) (Table 2).

The time from ED arrival to ECMO pump-on was shorter in the ED than in the Cath lab group regardless of whether CAG was conducted (median, ED: 49 min in the non-CAG group and 41 min in the CAG group vs. Cath lab: 63.5 min in the non-CAG group and 54 min in the CAG group) (Table 3). In the CAG group, there was a significant difference in the ED arrival to CAG start time and ECMO pump-on to CAG start time between the ED and Cath lab groups. The median (IQR) of ECMO pump- on to CAG start time was 96.0 (61.0-131.5) minutes in the ED group and 12.0 (0.0-24.0) minutes in the Cath lab group (Table 3).

• 1. Main analysis

There was no significant difference in survival outcomes between the ED and Cath lab groups after adjusting for other covariates (AOR

EMS-treated OHCAs (2013-2020) N=222,554
			Paediatric OHCAs (<18 years of age) n=4,107
Adult OHCAs N=218,447
			Non-medical etiology n=54,476
OHCAs with presumed cardiac etiology N=163,971
			ECMO not applied n=162,281
ECMO applied N=1,690
			ECMO applied after sustained ROSC n=1,116
ECPR cases N=574
			Ineligible ECMO or stop CPR time (n=2) ECMO cannulation in other places (n=8)
Eligible ECPR cases in ED or Cath lab N=564

Fig. 1. Patient flow. EMS, emergency medical service; OHCA, out-of-hospital cardiac arrest; ECMO, extracorporeal membrane oxygenation; ECPR, extracorporeal cardiopulmonary re- suscitation; ROSC, return-of-spontaneous circulation; CPR, cardiopulmonary resuscita- tion; ED, emergency department; Cath lab, cardiac catheterization laboratory.

[95% CI] for survival to discharge was 1.87 [0.85-4.11] and AOR for good neurological recovery was 1.10 [0.48-2.50]) (Table 4).

The interaction model showed significant interactions between ECPR location and CAG in the model for survival to discharge and the model for good neurological recovery (p for interaction: 0.044 in the model for survival to discharge and 0.023 in the model for good neuro- logical recovery). For patients who underwent CAG, ECPR in the ED was associated with better survival outcomes than in the Cath lab (AOR [95% CI] 2.34 [1.02-5.40]) for survival to discharge). For patients who did not undergo CAG, there was no significant association between ECPR loca- tions and survival outcomes (Table 5). The sensitivity analysis also found a significant interaction between ECPR location and PCI, and ECPR in the ED was associated with better survival than in the Cath lab for patients who underwent PCI (p for interaction: 0.032).

1. Discussion

This study found no difference in survival outcomes between ECPR in the ED and Cath lab, but the positive effects of ECPR in the ED in- creased in patients receiving CAG, emphasising the benefits of ECPR in the ED compared with the Cath lab for OHCA patients with a presumed cardiac etiology who underwent both ECPR and CAG.

ECPR is a complex and time-sensitive intervention that requires teamwork, clearly defined roles, and well-trained healthcare providers [20]. Therefore, various factors, including cost effectiveness, should be considered when performing ECPR in OHCA patients [5,21-23]. Since it is important to apply ECPR as soon as possible, it is also important to try to reduce the time to screen patients and to decide where and how to conduct ECPR. According to one systematic review of 27 studies, 44% of ECPRs were performed in the Cath lab, 26% in the ED, and 22% in multiple sites. [5] Hospital survival in that study did not differ between the Cath lab and ED groups, and were 31% and 32%, respectively. The lo- cation of ECPR was not the main exposure for each analysis, and

Demographics of study population by extracorporeal cardiopulmonary resuscitation location.

				ECPR location
	Total		ED		Cath Lab
Variables	N (%)		N (%)		N (%)	P-value
All	564 (100)		448 (100)		116 (100)
Gender						0.74
Female	101 (17.9)		79 (17.6)		22 (19.0)
Male	463 (82.1)		369 (82.4)		94 (81.0)
Age, Median (IQR) (years)	56.4 (46-64.5)		56.3 (45.7-64.3)		56.7 (48-66)	0.52
Age group (years)						0.48
18-49	193 (34.2)		158 (35.3)		35 (30.2)
50-74	327 (58.0)		254 (56.7)		73 (62.9)
75-	44 (7.8)		36 (8.0)		8 (6.9)
Past medical history
Diabetes	128 (22.7)	101 (22.5)		27 (23.3)		0.87
Hypertension	202 (35.8)	160 (35.7)		42 (36.2)		0.92
Heart disease	109 (19.3)	78 (17.4)		31 (26.7)		0.02
Stroke	16 (2.8)	11 (2.5)		5 (4.3)		0.28
Year						0.22
2013-2015	140 (24.8)	115 (25.7)		25 (21.6)
2016-2018	218 (38.7)	165 (36.8)		53 (45.7)
2019-2020	206 (36.5)	168 (37.5)		38 (32.8)
Residential area, Metropolitan	344 (61.0)	293 (65.4)		51 (44.0)		<0.01
Public location	338 (59.9)	266 (59.4)		72 (62.1)		0.60
Witnessed by bystanders	408 (72.3)	325 (72.5)		83 (71.6)		0.83
Bystander CPR	302 (53.5)	243 (54.2)		59 (50.9)		0.52
Initial shockable rhythm at the scene	330 (58.5)	263 (58.7)		67 (57.8)		0.98
Elapsed EMS time interval, Median (IQR) (min)
Response time	7 (5-9)	7 (5-9)		7 (5-9)		0.62
Scene time	14 (9-19)	14 (9-19)		13 (8-19)		0.35
Transport time	7 (4-9)	7 (5-9)		6 (4-10)		0.50
EMS management
Defibrillation	374 (66.3)	297 (66.3)		77 (66.4)		0.99
Advanced airway management	385 (68.3)	309 (69.0)		76 (65.5)		0.48
Epinephrine use	102 (18.1)	79 (17.6)		23 (19.8)		0.58
Level of ED						0.99
Level 1	238 (42.2)	189 (42.2)		49 (42.2)
Level 2 or 3	326 (57.8)	259 (57.8)		67 (57.8)
Post-resuscitation care
Coronary Angiography	319 (56.6)	235 (52.5)		84 (72.4)		<0.01
Percutaneous Coronary Intervention	211 (37.4)	162 (36.2)		49 (42.2)		0.23
Targeted Temperature Management	90 (16.0)	71 (15.8)		19 (16.4)		0.89
Elapsed ECPR time interval, Median(IQR), (min)
Total prehospital time	28 (22-34)	28 (22-34)			29.5 (22-35)	0.27
ED arrival to ECMO pump on	46 (35-63)	43 (34-59)			57 (43-78)	<0.01
Total duration of ECMO	1474 (275.5-4562.5)	1468 (239.5-4586.5)			1602 (367-4322)	0.66
Survival outcomes
Survival to discharge	78 (13.8)	63 (14.1)		15 (12.9)		0.75
Good neurological recovery	55 (9.8)	42 (9.4)		13 (11.2)		0.55

ECPR, extracorporeal cardiopulmonary resuscitation; ED, emergency department; Cath lab, cardiac catheterisation laboratory; IQR, interquartile range; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; ECMO, extracorporeal membrane oxygenation.

unadjusted analysis was conducted with a limited sample size, so cau- tion is needed when interpreting these results.

ECPR in the ED or Cath lab has both advantages and disadvantages. In patients with OHCA, refractory ventricular fibrillation/pulseless ven- tricular tachycardia is the main target for ECMO application, and those patients have a high rate of coronary artery disease [24]. For these pa- tients, performing CAG and reducing the time to reperfusion may im- prove survival rates [14]. ECPR in the Cath lab may shorten time to reperfusion because CAG can be directly conducted at the same location as ECPR. However, ECPR in the Cath lab carries an additional risk of mov- ing intra-arrest patients with chest compressions. Additionally, if ECPR is conducted in the ED, the duration from collapse to ECMO pump-on could be shortened compared to ECPR conducted in the Cath lab.

Observational studies show that the shorter the interval between cardiac arrest and ECPR initiation, the better the survival rate [21,25- 27]. Previous studies recommend time from arrest to ECPR to be <60 min [11]. In our study, three-quarters of the ED group had ECMO cannu- lation inserted within 60 min, and approximately half of the Cath lab

group had ECMO cannulation inserted within 60 min; the time from ED arrival to ECMO insertion was shortened by 14 min compared to that of the Cath lab. We found that 57% of patients who received ECPR underwent CAG. The time from ECMO pump-on to the start of CAG dif- fered by 84 min, with a median value of 12 min for Cath lab and 96 min for ED (Table 1). There were conflicting results in time to ECPR initiation and time to CAG according to ECPR location, but survival rates were higher in the ED group than in the Cath lab group, suggesting that re- ducing low flow time is more effective for survival than reducing the time to reperfusion in OHCA patients. Reducing the low flow time may be of greater importance because it applies to all ECPR patients and precedes time to reperfusion [26,28]. In contrast, reducing time to reperfusion is only effective for patients who underwent CAG (57% in our study), and the benefits of early CAG may be insignificant in patients without ST-segment elevation after return of spontaneous circulation [29,30]. Although the presence or absence of ST-segment elevation can- not be determined before ROSC, the benefits of early CAG is only limited to a fraction of patients undergoing ECPR.

Demographics of study population by coronary angiography.
			Coronary	angiography
	Total		No	Yes
Variables	N (%)		N (%)	N (%)	P-value
All	564 (100.0)		245 (100.0)	319 (100.0)
ECMO insertion place					<0.01
Emergency Department	448 (79.4)		213 (86.9)	235 (73.7)
Cardiac Catheterization Lab.	116 (20.6)		32 (13.1)	84 (26.3)
Gender					<0.01
Female	101 (17.9)		60 (24.5)	41 (12.9)
Male	463 (82.1)		185 (75.5)	278 (87.1)
Age, Median (IQR) (years)	56.4 (46-64.5)		54.7 (43.1-63.9)	57.1 (48.3-64.9)	0.02
Age group (years)					0.25
18-49	193 (34.2)		93 (38.0)	100 (31.3)
50-74	327 (58.0)		133 (54.3)	194 (60.8)
75-	44 (7.8)		19 (7.8)	25 (7.8)
Past medical history	(0.0)		(0.0)	(0.0)
Diabetes	128 (22.7)		40 (16.3)	88 (27.6)	<0.01
Hypertension	202 (35.8)		79 (32.2)	123 (38.6)	0.12
Heart disease	109 (19.3)		50 (20.4)	59 (18.5)	0.57
Stroke	16 (2.8)		4 (1.6)	12 (3.8)	0.13
Year					<0.01
2013-2015	140 (24.8)		105 (42.9)	35 (11.0)
2016-2018	218 (38.7)		74 (30.2)	144 (45.1)
2019-2020	206 (36.5)		66 (26.9)	140 (43.9)
Residential area, Metropolitan	344 (61.0)		152 (62.0)	192 (60.2)	0.65
Public location	338 (59.9)		157 (64.1)	181 (56.7)	0.08
Witnessed by bystanders	408 (72.3)		173 (70.6)	235 (73.7)	0.42
Bystander CPR	302 (53.5)		138 (56.3)	164 (51.4)	0.25
Initial shockable rhythm at the scene	330 (58.5)		117 (47.8)	213 (66.8)	<0.01
Elapsed EMS time interval, Median(IQR) (min)
Response time	7 (5-9)		7 (5-9)	7 (5-9)	0.68
Scene time	14 (9-19)		14 (9.5-19)	14 (9-19)	0.45
Transport time	7 (4-9)		7 (4-9)	7 (5-9)	0.93
EMS management
Defibrillation	374 (66.3)		146 (59.6)	228 (71.5)	<0.01
Advanced airway management	385 (68.3)		147 (60.0)	238 (74.6)	<0.01
Epinephrine use	102 (18.1)		30 (12.2)	72 (22.6)	<0.01
Level of ED					0.01
Level 1	238 (42.2)		89 (36.3)	149 (46.7)
Level 2 or 3	326 (57.8)		156 (63.7)	170 (53.3)
Post-resuscitation care
Coronary Angiography	319 (56.6)	.		319 (100.0)
Percutaneous Coronary Intervention	211 (37.4)	.		211 (66.1)
Targeted Temperature Management	90 (16.0)	26 (10.6)		64 (20.1)	<0.01
Elapsed ECPR time interval, Median(IQR), (min)
Total prehospital time	28 (22-34)	27 (22-33)		29 (22-35)	0.12
ED arrival to ECMO pump on	46 (35-63)	50 (36-65)		44 (34-59)	0.02
Total duration of ECMO	1474 (275.5-4562.5)	461 (90-3100)		2543 (509-5398)	<0.01
Survival outcomes
Survival to discharge	78 (13.8)	11 (4.5)		67 (21.0)	<0.01
Good neurological recovery	55 (9.8)	7 (2.9)		48 (15.0)	<0.01

ECPR, extracorporeal cardiopulmonary resuscitation; ED, emergency department; Cath lab, cardiac catheterisation laboratory; IQR, interquartile range; CPR, cardiopulmonary resuscita- tion; EMS, emergency medical services; ECMO, extracorporeal membrane oxygenation.

As ECPR location is closely associated with time to ECPR, it is impor- tant to reduce the time to ECPR through adequate decision of ECPR loca- tion. One approach is to mobilise an ECMO-equipped emergency response unit to the scene to initiate on-site ECMO. Pre-hospital ECPR has been suggested in a specific setting to reduce low-flow time and im- prove outcomes [10,31,32]. However, implementing pre-hospital ECPR is difficult in most EMS systems, especially when physicians are not also dispatched. Rapid EMS mobilisation and transportation to an ECPR-capable centre may also reduce time to ECPR. In this approach, pre-hospital selection of patients, distinguishing hospitals according to ECPR capacity and volume, and EMS hospital designation protocol for those patients would be of high importance [33]. Adequate preparation and optimisation of the protocol, including ECPR location is another approach to reduce time to ECPR. multidisciplinary collaboration can reduce time to ECPR regardless of where ECPR is conducted [34,35]. In Japan, a hybrid resuscitation room can be created in emergency rooms

so that ECMO can be applied more quickly and safely [36-38], radio- graphic images can be checked through the C-arm, and CAG and PCI are available immediately, removing problems that may occur when moving patients to the Cath lab. The feasibility of implementing a Hybrid resuscitation room may vary depending on the hospital, but it is still im- portant to reduce the time to ECPR.

This study had some limitations. First, ECPR protocols vary among hospitals. Although most hospitals in Korea carry out ECPR in accor- dance with international recommendations, specific ECPR protocols may differ depending on patient conditions and preparations in the emergency room. Therefore, we were unable to obtain information on the exact indications, specific procedures, and barriers to ECPR from each centre, which could create selection bias. Second, we used a South Korean nationwide registry encompassing data for 8 years, and the population of the final analysis was above 500 persons, but the sam- ple size was too small to conduct a comprehensive analysis. In the Cath

Table 3 Extracorporeal cardiopulmonary resuscitation and coronary angiography time characteristics of study population by the location of extracorporeal cardiopulmonary resuscitation and cor- onary angiography.

Coronary angiography (-) Coronary angiography (+)

Variables	ED	Cath lab		ED	Cath lab
ECPR time characteristics Total pre-hospital time, Median (IQR) (min) ED arrival to ECMO pump-on, Median (IQR) (min) Total duration of ECMO, Median (IQR) (hr)	213 27.0 (22.0-33.0) 49.0 (36.0-64.0) 6.1 (1.2-47.7)	32 30.0 (22.5-34.0) 63.5 (44.0-72.5) 29.1 (6.3-78.3)		235 29.0 (22.0-34.0) 41.0 (33.0-53.0) 49.4 (10.4-95.7)	84 29.0 (22.0-37.5) 54.0 (41.5-83.0) 25.4 (6.0-72.0)
CAG time characteristics ED arrival to CAG start, Median (IQR) (min)	- - 142.5 (105.5-184.5)				77.0 (58.0-98.0)
ECMO pump-on to CAG start, Median (IQR) (min)	- - 96.0 (61.0-131.5)				12.0 (0.0-24.0)

ED, emergency department; Cath lab, cardiac catheterisation laboratory; IQR, interquartile range; ECPR, extracorporeal cardiopulmonary resuscitation, CAG, coronary angiography.

Table 4 Effect of the location of extracorporeal cardiopulmonary resuscitation and coronary inter- vention on survival outcomes.

Finally, this was an observational study and not a randomised controlled trial. There may have been significant potential biases that were not controlled.

Outcomes, n/N (%)

Primary outcome: survival to discharge Total 78/564 (13.8)

ECPR location

Unadjusted OR (95% CI)

Adjusted OR (95% CI)?

1. Conclusions

In adult OHCA patients who underwent ECPR and CAG, ECPR in the ED shortened low-flow time and increased survival to discharge com-

ED 63/448 (14.1) 1.10 (0.60-2.02) 1.87 (0.85-4.11)

Cath lab 15/116 (12.9) Reference Reference Coronary angiography

Yes 67/319 (21.0) 5.66 (2.92-10.97) 4.40 (1.74-11.12)

No 11/245 (4.5) Reference Reference Secondary outcome: good neurological recovery

Total 55/564 (9.8)

ECPR location

ED	42/448 (9.4)	0.82 (0.42-1.58)	1.10 (0.48-2.50)
Cath lab	13/116 (11.2)	Reference	Reference
Coronary angiography
Yes	48/319 (15.1)	6.02 (2.67-13.56)	4.76 (1.49-15.18)
No	7/245 (2.9)	Reference	Reference

OR, odds ratio; ECPR, extracorporeal cardiopulmonary resuscitation; ED, emergency department; Cath lab, cardiac catheterisation laboratory.

* Adjusted for age, sex, year, place of arrest, witnessed status, bystander cardiopulmo-

nary resuscitation, metropolitan, pre-hospital advanced airway management, pre-hospital epinephrine use, and EMS defibrillation.

lab group, only 2 of the 32 patients showed good Neurological prognosis when CAG was not performed. Therefore, we could not conduct an in- teraction analysis for good neurological recovery. Third, we were unable to collect data on complications associated with ECPR such as bleeding, leg ischaemia, and Vascular injury. These factors might be associated with the location of ECPR cannulation, but we could not conduct an analysis due to lack of data. [39] Fourth, this study was conducted in Korea. Therefore, the generalisability of the results to other countries is limited. Depending on the environment, the effect of the ECPR loca- tion on the time to ECMO pump-on may be different, and in this case, the effect of the ECPR location on the patient’s survival may be different.

pared to ECPR in the Cath lab. Hospitals that perform ECPR should con- sider ways to minimize time to ECMO pump-on. Further prospective studies are warranted to evaluate the effects of specific criteria, proto- cols, equipment, or facilities to reduce time to ECPR and improve OHCA survival outcomes.

Author contributions

Drs. Y Kim and JH Park had full access to all the data in the study and take responsibility for the integrity of the data as well as the accuracy of the data analysis.

Study concept and design: Drs. SD Shin, KJ Song, and JH Park. Acquisition, analysis, and interpretation of data: Drs. JH Park, SY Lee,

SD Shin, and KJ Hong.

Drafting of the manuscript: Drs. Y Kim, SY Lee, KJ Hong, and JH Park. Critical revision of the manuscript for important intellectual content:

Drs. KJ Hong, SY Lee, and YS Ro.

Statistical analysis: Drs. Y Kim, JH Park and SY Lee Obtained funding: Dr. SD Shin.

Administrative, technical, or material support: Drs. SY Lee, KJ Hong, YS Ro, and KJ Song.

Study supervision: Drs. SD Shin and KJ Song. Manuscript approval: All authors.

Source of funding

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

Table 5

Effect of the location of extracorporeal cardiopulmonary resuscitation in the interaction model with coronary angiography.

Survival to discharge Good neurological recovery

	Outcomes, n/N (%)	Adjusted OR (95% CI)?		Outcomes, n/N (%)	Adjusted OR (95% CI)?
Coronary angiography (+)
ED	54/235 (23.0)	2.34 (1.02-5.40)		37/235 (15.7)	1.76 (0.72-4.31)
Cath lab	13/84 (15.5)	Reference		11/84 (13.1)	Reference
Coronary angiography (-)
ED	9/213 (4.2)	0.28 (0.04-1.84)	5/213 (2.4)		0.11 (0.01-1.00)
Cath lab	2/32 (6.3)	Reference	2/32 (6.3)		Reference

OR, odds ratio; ED, emergency department; Cath lab, cardiac catheterisation laboratory.

The p-value of the interaction term (ECPR location x coronary angiography) was 0.044 in the model for survival-to-discharge and 0.023 in the model for good neurological recovery.

* Adjusted for age, sex, year, place of arrest, witnessed status, bystander cardiopulmonary resuscitation, metropolitan area, advanced airway, epinephrine use, and EMS defibrillation.

CRediT authorship contribution statement

Yoonjic Kim: Writing - original draft, Data curation, Formal anlaysis. Sun Young Lee: Data curation, Project administration, Writing - review & editing. Jeong Ho Park: Methodology, Formal analysis, Conceptuali- zation, Data curation, Writing - review & editing. Ki Jeong Hong: Data curation, Project administration, Writing - review & editing. Young Sun Ro: Project administration, Writing - review & editing. Kyoung Jun Song: Supervision, Conceptualization, Project administration. Sang Do Shin: Supervision, Conceptualization, Data curation, Funding acquisition.

Declaration of Competing Interest

There are no conflicts of interest for all authors in this study. This re- search did not receive any specific grant from funding agencies in the public, commercial, ornot-for-profit sectors.

Acknowledgement

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

References

1. Berdowski J, Berg RA, Tijssen JG, Koster RW. Global incidences of out-of-hospital car- diac arrest and survival rates: systematic review of 67 prospective studies. Resusci- tation. 2010;81(11):1479-87. https://doi.org/10.1016/j.resuscitation.2010.08.006.
2. Richardson ASC, Tonna JE, Nanjayya V, Nixon P, Abrams DC, Raman L, et al. Extracor- poreal cardiopulmonary resuscitation in adults. Interim guideline consensus state- ment from the extracorporeal life support organization. ASAIO J. 2021;67(3): 221-8. https://doi.org/10.1097/MAT.0000000000001344.
3. Halenarova K, Belliato M, Lunz D, Peluso L, Broman LM, Malfertheiner MV, et al. Pre- dictors of poor outcome after extra-corporeal membrane oxygenation for refractory cardiac arrest (ECPR): a post hoc analysis of a multicenter database. Resuscitation. 2022;170:71-8. https://doi.org/10.1016/j.resuscitation.2021.11.015.
4. Alfalasi R, Downing J, Cardona S, Lowie BJ, Fairchild M, Chan C, et al. A comparison between conventional and extracorporeal cardiopulmonary resuscitation in out- of-hospital cardiac arrest: a systematic review and Meta-analysis. Healthcare (Basel). 2022;10(3). https://doi.org/10.3390/healthcare10030591.
5. Downing J, Al Falasi R, Cardona S, Fairchild M, Lowie B, Chan C, et al. How effective is extracorporeal cardiopulmonary resuscitation for out-of-hospital cardiac ar- rest? A systematic review and meta-analysis. Am J Emerg Med. 2022;51:127-38. https://doi.org/10.1016/j.ajem.2021.08.072.
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. Soar J, Bottiger BW, Carli P, Couper K, Deakin CD, Djarv T, et al. European resuscita- tion council guidelines 2021: adult advanced life support. Resuscitation. 2021;161: 115-51. https://doi.org/10.1016/j.resuscitation.2021.02.010.
8. Choi DS, Kim T, Ro YS, Ahn KO, Lee EJ, Hwang SS, et al. Extracorporeal life support and survival after out-of-hospital cardiac arrest in a nationwide registry: a propen- sity score-matched analysis. Resuscitation. 2016;99:26-32. https://doi.org/10. 1016/j.resuscitation.2015.11.013.
9. Kumar KM. ECPR-extracorporeal cardiopulmonary resuscitation. Indian J Thorac Cardiovasc Surg. 2021;37(Suppl. 2):294-302. https://doi.org/10.1007/s12055-020- 01072-2.
10. Singer B, Reynolds JC, Lockey DJ, O’Brien B. Pre-hospital extra-corporeal cardiopul- monary resuscitation. Scand J Trauma Resusc Emerg Med. 2018;26(1):21. https:// doi.org/10.1186/s13049-018-0489-y.
11. Fagnoul D, Combes A, De Backer D. Extracorporeal cardiopulmonary resuscita- tion. Curr Opin Crit Care. 2014;20(3):259-65. https://doi.org/10.1097/MCC. 0000000000000098.
12. Reynolds JC, Grunau BE, Elmer J, Rittenberger JC, Sawyer KN, Kurz MC, et al. Preva- lence, natural history, and time-dependent outcomes of a multi-center north American cohort of out-of-hospital cardiac arrest extracorporeal CPR candidates. Re- suscitation. 2017;117:24-31. https://doi.org/10.1016/j.resuscitation.2017.05.024.
13. 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.
14. Camuglia AC, Randhawa VK, Lavi S, Walters DL. Cardiac catheterization is associated with superior outcomes for survivors of out of hospital cardiac arrest: review and meta-analysis. Resuscitation. 2014;85(11):1533-40. https://doi.org/10.1016/j.resus- citation.2014.08.025.
15. Yannopoulos D, Bartos J, Raveendran G, Walser E, Connett J, Murray TA, et al. Ad- vanced Reperfusion strategies for patients with out-of-hospital cardiac arrest and re- fractory ventricular fibrillation (ARREST): a Phase 2, single Centre, open-label, randomised controlled trial. Lancet. 2020;396(10265):1807-16. https://doi.org/10. 1016/S0140-6736(20)32338-2.
16. Deakin CD. The Chain of survival: not all links are equal. Resuscitation. 2018;126: 80-2. https://doi.org/10.1016/j.resuscitation.2018.02.012.
17. Cho YS, Song KH, Lee BK, Jeung KW, Jung YH, Lee DH, et al. Five-year experience of extracorporeal life support in emergency physicians. Korean J Crit Care Med. 2017; 32(1):52-9. https://doi.org/10.4266/kjccm.2016.00885.
18. Ryu JA, Cho YH, Sung K, Choi SH, Yang JH, Choi JH, et al. Predictors of neurological outcomes after successful extracorporeal cardiopulmonary resuscitation. BMC Anesthesiol. 2015;15:26. https://doi.org/10.1186/s12871-015-0002-3.
19. Kim YM, Jeung KW, Kim WY, Park YS, Oh JS, You YH, et al. Korean guidelines for car- diopulmonary resuscitation. Part 5. post-cardiac arrest care. Clin Exp Emerg Med 2021. 2020;8(S):S41-64. https://doi.org/10.15441/ceem.21.025.
20. Soar J, Maconochie I, Wyckoff MH, Olasveengen TM, Singletary EM, Greif R, et al. 2019 international consensus on cardiopulmonary resuscitation and emergency car- diovascular care science with treatment recommendations: summary from the basic life support; advanced life support; pediatric life support; neonatal life support; ed- ucation, implementation, and teams; and first aid task forces. Circulation. 2019;140

(24). https://doi.org/10.1161/CIR.0000000000000734. e826-e80.

1. Debaty G, Babaz V, Durand M, Gaide-Chevronnay L, Fournel E, Blancher M, et al. Prognostic factors for extracorporeal cardiopulmonary resuscitation recipients fol- lowing out-of-hospital refractory cardiac arrest. Syst Rev Meta-anal Resuscitation. 2017;112:1-10. https://doi.org/10.1016/j.resuscitation.2016.12.011.
2. Holmberg MJ, Geri G, Wiberg S, Guerguerian AM, Donnino MW, Nolan JP, et al. Ex- tracorporeal cardiopulmonary resuscitation for cardiac arrest: a systematic review. Resuscitation. 2018;131:91-100. https://doi.org/10.1016/j.resuscitation.2018.07.

029.

1. Beyea MM, Tillmann BW, Iansavichene AE, Randhawa VK, Van Aarsen K, Nagpal AD. Neurologic outcomes after extracorporeal membrane oxygenation assisted CPR for resuscitation of out-of-hospital cardiac arrest patients: a systematic review. Resusci- tation. 2018;130:146-58. https://doi.org/10.1016/j.resuscitation.2018.07.012.
2. Yannopoulos D, Bartos JA, Aufderheide TP, Callaway CW, Deo R, Garcia S, et al. The evolving role of the cardiac catheterization Laboratory in the Management of pa- tients with out-of-hospital cardiac arrest: a scientific statement from the American Heart Association. Circulation. 2019;139(12). https://doi.org/10.1161/CIR. 0000000000000630. e530-e52.
3. D’Arrigo S, Cacciola S, Dennis M, Jung C, Kagawa E, Antonelli M, et al. Predictors of favourable outcome after in-hospital cardiac arrest treated with extracorporeal car- diopulmonary resuscitation: a systematic review and meta-analysis. Resuscitation. 2017;121:62-70. https://doi.org/10.1016/j.resuscitation.2017.10.005.
4. Wengenmayer T, Rombach S, Ramshorn F, Biever P, Bode C, Duerschmied D, et al. In- fluence of low-flow time on survival after extracorporeal cardiopulmonary resusci- tation (eCPR). Crit Care. 2017;21(1):157. https://doi.org/10.1186/s13054-017- 1744-8.
5. Kawakami S, Tahara Y, Koga H, Noguchi T, Inoue S, Yasuda S. The association be- tween time to extracorporeal cardiopulmonary resuscitation and outcome in pa- tients with out-of-hospital cardiac arrest. Eur Heart J Acute Cardiovasc Care. 2022. https://doi.org/10.1093/ehjacc/zuac010.
6. Otani T, Sawano H, Natsukawa T, Nakashima T, Oku H, Gon C, et al. Low-flow time is associated with a favorable neurological outcome in out-of-hospital cardiac arrest patients resuscitated with extracorporeal cardiopulmonary resuscitation. J Crit Care. 2018;48:15-20. https://doi.org/10.1016/j.jcrc.2018.08.006.
7. Desch S, Freund A, Akin I, Behnes M, Preusch MR, Zelniker TA, et al. Angiography after out-of-hospital cardiac arrest without ST-segment elevation. N Engl J Med. 2021;385(27):2544-53. https://doi.org/10.1056/NEJMoa2101909.
8. Lemkes JS, Janssens GN, van der Hoeven NW, Jewbali LSD, Dubois EA, Meuwissen M, et al. Coronary angiography after cardiac arrest without ST-segment elevation. N Engl J Med. 2019;380(15):1397-407. https://doi.org/10.1056/NEJMoa1816897.
9. Lamhaut L, Hutin A, Puymirat E, Jouan J, Raphalen JH, Jouffroy R, et al. A pre-hospital extracorporeal cardio pulmonary resuscitation (ECPR) strategy for treatment of re- fractory out hospital cardiac arrest: an observational study and propensity analysis. Resuscitation. 2017;117:109-17. https://doi.org/10.1016/j.resuscitation.2017.04.

014.

1. Pozzi M, Dubien PY, Cesareo E, Pinero D, Obadia JF, Richard JC. Pre-hospital extracor- poreal cardiopulmonary resuscitation. J Cardiothorac Vasc Anesth. 2020;34(2): 571-2. https://doi.org/10.1053/j.jvca.2019.08.005.
2. De Charriere A, Assouline B, Scheen M, Mentha N, Banfi C, Bendjelid K, et al. ECMO in cardiac arrest: a narrative review of the literature. J Clin Med. 2021;10(3). https:// doi.org/10.3390/jcm10030534.
3. Tonna JE, Selzman CH, Mallin MP, Smith BR, Youngquist ST, Koliopoulou A, et al. De- velopment and implementation of a comprehensive, multidisciplinary emergency department extracorporeal membrane oxygenation program. Ann Emerg Med. 2017;70(1):32-40. https://doi.org/10.1016/j.annemergmed.2016.10.001.
4. Yannopoulos D, Bartos JA, Martin C, Raveendran G, Missov E, Conterato M, et al. Min- nesota resuscitation Consortium’s advanced perfusion and reperfusion cardiac life support strategy for out-of-hospital refractory ventricular fibrillation. J Am Heart Assoc. 2016;5(6). https://doi.org/10.1161/JAHA.116.003732.
5. Hifumi T, Inoue A, Takiguchi T, Watanabe K, Ogura T, Okazaki T, et al. Variability of extracorporeal cardiopulmonary resuscitation practice in patients with out-of-

hospital cardiac arrest from the emergency department to intensive care unit in Japan. Acute Med Surg. 2021;8(1):e647. https://doi.org/10.1002/ams2.647.

1. Nishi R, Nakahara S, Miyake Y, Sakamoto T. Extracorporeal cardiopulmonary resus- citation in a hybrid resuscitation room. Am J Emerg Med. 2020;38(12):2748-9. https://doi.org/10.1016/j.ajem.2020.04.067.
2. Hayashida K, Kinoshita T, Yamakawa K, Miyara SJ, Becker LB, Fujimi S. Potential im- pacts of a novel integrated extracorporeal-CPR workflow using an interventional

radiology and immediate whole-body computed tomography system in the emer- gency department. BMC Cardiovasc Disord. 2020;20(1):23. https://doi.org/10. 1186/s12872-020-01332-4.

1. Inoue A, Hifumi T, Sakamoto T, Kuroda Y. Extracorporeal cardiopulmonary resuscita- tion for out-of-hospital cardiac arrest in adult patients. J Am Heart Assoc. 2020;9(7): e015291. https://doi.org/10.1161/JAHA.119.015291.