Resuscitation 83 (2012) 1427 1433
Contents lists available at SciVerse ScienceDirect
Resuscitation
jo u rn al hom epage : www.elsevier.com/locate/resuscitation
Clinical paper
Acute coronary angiography in patients resuscitated from out-of-hospital cardiac
arrest A systematic review and meta-analysis
Jacob Moesgaard Larsen", Jan Ravkilde
Department of Cardiology and Centre for Cardiovascular Research, Aalborg University Hospital, Hobrovej 18-22, 9000 Aalborg, Denmark
a r t i c l e i n f o a b s t r a c t
Article history:
Introduction: Out-of-hospital cardiac arrest has a poor prognosis. The main aetiology is ischaemic heart
Received 22 June 2012
disease.
Received in revised form 30 August 2012
Aim: To make a systematic review addressing the question:  In patients with return of spontaneous
Accepted 30 August 2012
circulation following out-of-hospital cardiac arrest, does acute coronary angiography with coronary
intervention improve survival compared to conventional treatment?
Methods: Peer reviewed articles written in English with relevant prognostic data were included. Compar-
Keywords:
ison studies on patients with and without acute coronary angiography were pooled in a meta-analysis.
Coronary angiography
Results: Thirty-two non-randomised studies were included of which 22 were case-series without patients
Heart arrest
with conservative treatment. Seven studies with specific efforts to control confounding had statistical
Outcome
evidence to support the use of acute coronary angiography following resuscitation from out-of-hospital
cardiac arrest. The remaining 25 studies were considered neutral. Following acute coronary angiography,
the survival to hospital discharge, 30 days or six months ranged from 23% to 86%. In patients without an
obvious non-cardiac aetiology, the prevalence of significant coronary artery disease ranged from 59% to
71%. Electrocardiographic findings were unreliable for identifying angiographic findings of acute coronary
syndrome. Ten comparison studies demonstrated a pooled unadjusted odds ratio for survival of 2.78 (1.89;
4.10) favouring acute coronary angiography.
Conclusion: No randomised studies exist on acute coronary angiography following out-of-hospital cardiac
arrest. An increasing number of observational studies support feasibility and a possible survival benefit
of an early invasive approach. In patients without an obvious non-cardiac aetiology, acute coronary
angiography should be strongly considered irrespective of electrocardiographic findings due to a high
prevalence of coronary artery disease.
© 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction return of spontaneous circulation (ROSC) after OHCA is less clear,
especially in comatose survivors. The topic was evaluated in the
Out-of-hospital cardiac arrest (OHCA) has a poor prognosis and 2010 International Consensus on Cardiopulmonary Resuscitation
is a leading cause of death. The incidence of OHCA treated by the and Emergency Cardiovascular Care Science with Treatment Rec-
emergency medical service in Europe has been estimated to be ommendations (2010 CoSTR).4 The recommendation was: acute
approximately 275,000 persons per year with a survival of 10.7% for CAG should be considered in STEMI or clinical suspicion of coro-
all rhythms and 21.2% for ventricular fibrillation arrest.1 The most nary ischaemia as a likely cause of the arrest, and that it may be
frequent cause of OHCA is ischaemic heart disease.2 Acute coro- reasonable to be included in a systematic standardised post cardiac
nary angiography (CAG) with percutaneous coronary intervention arrest protocol. Several new studies have emerged. The aim of this
(PCI) is the treatment of choice in patients with acute coronary syn- study was to make an updated systematic review of the evidence
drome (ACS) with ST-segment elevation (STEMI) or new left bundle on performing acute CAG following ROSC after OHCA.
branch block (LBBB) in the electrocardiogram (ECG) without pre-
ceding cardiac arrest.3 The prognostic value of acute CAG following
2. Methods
The study was conducted in accordance with the principles
stated by the Meta-analysis Of Observational Studies in Epi-

A Spanish translated version of the abstract of this article appears as Appendix
demiology (MOOSE) group and the Preferred Reporting Items
in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2012.08.337.
" for Systematic Reviews and Meta-analysis (PRISMA) group.5,6 In
Corresponding author.
E-mail address: [e-mail] (J.M. Larsen). short, we defined a structured question describing the Population,
0300-9572/$  see front matter © 2012 Elsevier Ireland Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.resuscitation.2012.08.337
1428 J.M. Larsen, J. Ravkilde / Resuscitation 83 (2012) 1427 1433
Intervention, Comparison and Outcome (PICO). This was followed 3. Results
by literature search and critical appraisal of the evidence. The eli-
gible studies were summarised in tables, and the outcome was 3.1. Eligible studies
evaluated in a meta-analysis.
The literature search is illustrated in the flow diagram in Fig. 1.
Thirty-two studies met the criteria for inclusion in the review.
2.1. PICO question
Ten were included in the meta-analysis. Seven studies were clas-
sified as supporting acute CAG following ROSC after OHCA, and
 In patients with ROSC following OHCA (P), does acute CAG with
the remaining 25 studies were neutral. Twelve studies were not
coronary intervention (I), compared to conventional treatment (C),
considered in the 2010 CoSTR, primarily due to publication after
improve survival (O)?
completion of the 2010 CoSTR evaluation process. Table 1 sum-
marises the LOE and design of the included studies: LOE 2 (one
2.2. Literature search study), LOE 4 (22 studies) and LOE 5 (nine studies). In all cases, LOE
5 was due to inclusion of patients with in-hospital cardiac arrest or
The literature search was performed on May 1st, 2012, in collab- cardiac arrest without specification of location. Seventeen studies
oration with experienced research librarians. PubMed search terms were retrospective.
were:  Heart arrest [Mesh] AND ( Coronary Angiography [Mesh]
OR  Angioplasty, Balloon, Coronary [Mesh]). Embase search terms
3.2. Studies on acute coronary angiography in ST-segment
were:  exp heart arrest AND ( exp angiocardiography OR  exp
elevation myocardial infarction
transluminal coronary angioplasty ). SveMed+ search terms were:
 exp Heart-Arrest AND ( exp Coronary-Angiography OR  exp
Table 2 summarises the characteristics of the 15 studies on
Angioplasty,-Transluminal, Percutaneous Coronary ). The identi-
acute CAG in patients with STEMI following ROSC.8 22 The survival
fied records were managed using reference management software
ranged from 41% to 92%. Common characteristics were male sex,
(RefWorks 2.0, ProQuest LLC, USA). Duplicates were identified and
witnessed cardiac arrest, OHCA and shockable rhythm. The inclu-
deleted. Screening of the records was done by one author (Larsen
sion periods were generally before or in the early era of therapeutic
JM). Reviews, case reports, editorials, letters, comments, conference
hypothermia (TH), and the use of TH was therefore low despite
abstracts, records with clearly no relevance to the PICO question,
a high prevalence of comatose survivors on arrival to hospital.
and articles not written in English were excluded. Full text arti-
The largest study is a retrospective case series of 186 consecutive
cles were evaluated for eligibility by both authors. Articles without
patients undergoing acute CAG due to ST-segment elevation or
prognostic data at hospital discharge, 30 days or six months for
patients with acute CAG or with double publication of prognostic
Table 1
data were excluded. Other literature sources were screening of the
Evidence level and design of the studies included in the systematic review.
reference lists of the included articles and 2010 CoSTR and the peer
Study design LOE 1 LOE 2 LOE 3 LOE 4 LOE 5
review process.
Studies supporting acute CAG following OHCA
Prospective    Croniera 
2.3. Evidence appraisal
Dumas
Gräsnera
Tömtea
The level of evidence (LOE) was evaluated by both authors: LOE
Spaulding
1 randomised controlled trials or meta-analyses of randomised
Retrospective  Strotea   Merchant
controlled trials; LOE 2 studies using concurrent controls without
Studies neutral to acute CAG following OHCA
randomisation for comparison; LOE 3 studies using retrospective
Prospective    Bendz Quintero-Moran
controls for comparison; LOE 4 studies without a control group
Kahn Szymanskia
for comparison; and LOE 5 studies not directly related to the spe-
Lettieri
cific population.7 Comparison studies without matched concurrent
Mooneya
Möllmanna
controls were classified as LOE 4. The studies were categorised as
Nielsen
prospective or retrospective as a simple evaluation of quality. Stud-
Peels
ies favouring acute CAG in a propensity score analysis or reporting
Pleskot
a significant adjusted odds ratio in favour of acute CAG or acute PCI
Retrospective    Anyfantakis Garot
were classified as  supporting PICO. Studies with non-significant Aurorea Hosmane
Buluta Knafelj
adjusted results were classified as  neutral to PICO. Studies with
Hovdenesa Mager
significant adjusted results favouring conservative treatment were
Keelana Reynolds
classified as  opposing . To be conservative, case-series without
Markusohn Richling
comparison groups were classified as  neutral despite high sur-
McCullough
vival rates due to possible selection bias. Siderisa
Wolfrum
Studies opposing acute CAG following OHCA
2.4. Statistics
Prospective     
Retrospective     
The statistical analysis was performed with a significance level
LOE = level of evidence (1  randomised controlled trials or meta-analyses of
of p < 0.05 (Stata 11, StataCorp LP, USA). Data was collected from the
randomised controlled trials; 2  studies using concurrent controls without ran-
result sections of the included articles. Comparison studies were
domisation for comparison; 3  studies using retrospective controls for comparison;
4
included in a meta-analysis estimating an unadjusted pooled OR  studies without a control group for comparison; and 5  studies not directly
related to the specific population; CAG = coronary angiography; OHCA = out-of-
for survival using a random-effect model. The heterogeneity of the
hospital cardiac arrest.
studies was evaluated by the I-squared measure, which describes
a
Studies not evaluated in the 2010 International Consensus on Cardiopulmonary
the percentage of variation across the studies due to heterogeneity
Resuscitation and Emergency Cardiovascular CareScience with Treatment Recom-
rather than chance.
mendations document.
J.M. Larsen, J. Ravkilde / Resuscitation 83 (2012) 1427 1433 1429
Records identified through Additional records identified
database searching through other sources
(n = 1484) (n = 5)
Records after duplicates removed Records excluded
(n = 1313) (n = 1249)
Reviews; case reports;
editorials; letters;
comments; conference
abstracts; studies not
relevant to PICO; non-
Records screened
English writing.
(n = 1313)
Full-text articles
Full-text articles
assessed for eligibility
excluded
(n = 64)
(n = 32)
Necessary prognostic
information not
available; double-
Studies included in
publication of prognostic
the systematic review
data; studies not
(n = 32)
relevant to PICO.
Studies included in
the meta-analysis
(n = 10)
Fig. 1. Flow chart of the selection of articles for the systematic review and meta-analysis. The database search included PubMed (n = 613 records), Embase (n = 866 records) and
SveMed+ (n = 5). The records from other sources were obtained by screening reference lists of the included studies and the 2010 International Consensus on Cardiopulmonary
Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations document and the peer review process. PICO = patient, intervention, comparison,
outcome.
presumed new LBBB after ROSC, mainly after OHCA.13 Acute in two other studies.10,20 A study, comparing direct admittance
coronary occlusion was found in 74%. The remaining patients had to a PCI centre and transfer from a referral hospital, demon-
severe chronic stenosis. The survival to hospital discharge was 55%. strated no significant difference in survival, but the proportion of
The survival at six months was 54%, primarily with a good neuro- non-transferred patients with preserved left ventricular ejection
logical status. A comparably good long-term prognosis was seen fraction was higher (61% vs 25%, p = 0.02).17 A comparison of
Table 2
Characteristics of studies on acute coronary angiography in patients with ST-segment elevation myocardial infarction.
Study Inclusion of patients N OHCA (%) Witnessed (%) VF/VT (%) Unconscious (%) TH (%) Survivala (%)
Kahn et al.,8 USA 1989 1994 11 100 NA 100 64 0 51
McCullough et al.,9 USA 1989 1996 22 100 100 91 NA 0 41
Bendz et al.,10 Norway 1998 2001 40b 100 100 90 90 0 73
Quintero-Moran et al.,11 Spain 2000 2003 63 43 100 81 NA NA 68 (30 days)
Markusohn et al.,12 Israel 1998 2006 25 100 92 84 72 8 76
Garot et al.,13 France 1995 2005 186 84 67 NA NA 18 55
Knafelj et al.,14 Slovenea 2000 2005 72 NA 100 100 100 56 61
Richling et al.,15 Austria 1991 2003 46b 98 98 100 NA 37 55 (6 months)
Pleskot et al.,16 Czech Republic 2002 2004 20b 100 NA 100 90 NA 70
Peels et al.,17 The Netherlands 2004 2005 44 100 NA NA NA NA 50
Mager et al.,18 Israel 2001 2006 21 NA NA NA 43 5 86 (30 days)
Wolfrum et al.,19 Germany 2003 2006 33 100 NA 100 100 48 70 (6 months)
Lettieri et al.,20 Italy 2005 99b 100 70 90 NA 12 78
Szymanski et al.,21 Poland NA 12b NA NA 100 NA NA 92 (30 days)
Hosmane et al.,22 USA 2002 2006 98 68 90 NA 75 NA 64
Total NA 792 NA NA NA NA NA 64 (mean)
N = number of patients; OHCA = out-of-hospital cardiac arrest; VF/VT = ventricular fibrillation or tachycardia; TH = therapeutic hypothermia; NA = not available.
a
Survival to hospital discharge unless otherwise stated.
b
Only data on patients with acute coronary angiography following cardiac arrest is reported from the study.
Identification
Screening
Eligibility
Included
1430 J.M. Larsen, J. Ravkilde / Resuscitation 83 (2012) 1427 1433
Table 3
Characteristics of studies on patients with systematic acute coronary angiography following out-of-hospital cardiac arrest without an obvious non-cardiac aetiology.
Study Inclusion of n ST-segment Significant Angiographic PCI (%) VF/VT (%) Unconscious (%) TH (%) Survivala (%)
patients elevation or CAD (%) ACS (%)
LBBB (%)
Spaulding et al.,23 France 1994 1996 84 63 71 69 33 93 NA 0 38
Anyfantakis et al.,24 France 2001 2006 72 49 64 45 33 50 94c NA 49
Dumas et al.,25 France 2003 2008 435 31d 70 46 41 68 NA 86 39
Sideris et al.,26 France 2002 2008 165 50 59 36 30 51 99c 76 31
Möllmann et al.,27 Germany 2003 2005 65b 55d NA NA 58 NA NA NA 81 (6 months)
Total 1994 2008 821 42 NA NA 39 NA NA NA 41 (mean)
N = number of patients; LBBB = left bundle branch block; CAD = coronary artery disease; angiographic ACS = recent occlusion or irregular lesion at angiography in resuscitated
patients; PCI = percutaneous coronary intervention; VF/VT = ventricular fibrillation or tachycardia; TH = therapeutic hypothermia; NA = not available.
a
Survival to hospital discharge unless otherwise stated.
b
Only data on patients with acute coronary angiography following cardiac arrest is reported from the study.
c
Unconscious and intubated.
d
ST-segment elevation but not left LBBB was reported in study.
thrombolysis and acute CAG demonstrated no significant The pioneering prospective study by Spaulding et al. included
difference in survival at six months (68% vs 55%, p = 0.13).15 84 patients with systematic acute CAG following OHCA without
Two studies indicated TH to be feasible in STEMI patients undergo- obvious non-cardiac aetiology.23 The positive and negative pre-
ing acute CAG with a probable positive effect on good neurological dictive values for recent coronary occlusion on angiography of
survival.14,19 chest pain and/or ST-segment elevation were 63% and 74%, respec-
tively. Survival to hospital discharge was 38%. Successful PCI was
an independent predictor of survival (adjusted OR 5.2, p = 0.04).
3.3. Studies on systematic acute coronary angiography in selected
The largest study including 435 patients from a prospective reg-
patients with out-of-hospital cardiac arrest of mixed aetiology
istry also demonstrated suboptimal but slightly better diagnostic
predictive values of ST-segment elevation, and successful PCI was
Table 3 illustrates five studies on systematic acute CAG in
an independent predictor of survival (adjusted OR 2.1, p = 0.01).25
patients following ROSC after OHCA without an obvious non-
Similar suboptimal diagnostic values of ST-segment elevation for
cardiac aetiology.23 27 The reported survival ranged from 31% to
identifying angiographic lesions comparable to ACS were also
81%. The patient characteristics were more varied compared to the
seen in two retrospective studies.24,26 The newest of these stud-
pure STEMI studies. TH was used in the majority of the patients
ies suggested an extended ECG criterion of ST-segment elevation
in the two largest studies.25,26 The prevalence of significant coro-
and/or depression and/or LBBB and/or unspecific wide QRS and/or
nary artery disease (CAD) was high ranging from 59% to 71%.
right bundle branch block.26 The extended criterion demonstrated
Angiographic signs comparable to acute myocardial infarction with
a lower positive predictive value of 48% but a negative predic-
recent occlusion or irregular lesions varied from 36% to 69%. ST-
tive value of 100% with a potential to reduce the needed acute
segment elevation or LBBB was seen in 31 63%. Shockable rhythms
procedures.
ranged from 50% to 93%.
Table 4
Studies including patients with and without acute coronary angiography following cardiac arrest.
Study Inclusion of N OHCA (%) ST-segment Acute PCI (%) VF/VT (%) Unconscious (%) TH (%) Survival with
patients elevation or CAG (%) and without
LBBB (%) acute CAG (%)a
Bulut et al.,30 The Netherlands NA 72 100 NA 14 11 69 69 0 40 vs 37
p = 1.00
Merchant et al.,31 USA 2000 2005 110 0 12 27 15 100 NA NA 80 vs 54
p = 0.02
Nielsen et al.,32 Multinational 2004 2008 986 100 NA 49 30 70 100 100 63 vs 50
p < 0.001
Reynolds et al.,33 USA 2005 2007 241 56 19 26 NA 39 NA 33 52 vs 31b
p = 0.004
Aurore et al.,34 France 2000 2006 445 100 28 30 16 42 NA NA 23 vs 10
p < 0.001
Cronier et al.,35 France 2003 2008 111 100 54 82 42 100 NA 70 59 vs 30
p = 0.02
Gräsner et al.,36 Germany 2004 2010 584 100 NA 26 NA 42 NA 31 52 vs 13b
p < 0.001
Mooney et al.,37 USA 2006 2009 140 100 49 72 40 76 100 100 62 vs 38
p = 0.01
Tömte et al.,38 Norway 2003 2009 174 100 NA 83 45 49 78 NA 52 vs 31b
p = 0.04
Strote et al.,39 USA 1999 2002 240 100 34 25 16 98 NA 0 72 vs 49
p = 0.003
Total NA 3103 92 NA 41 NA 62 NA NA 56 vs 32 (means)
p < 0.001
N = number of patients; OHCA = out-of-hospital cardiac arrest; LBBB = left bundle branch block; CAG = coronary angiography; PCI = percutaneous coronary intervention;
VF/VT = ventricular fibrillation or tachycardia; TH = therapeutic hypothermia; NA = not available.
a
Survival to hospital discharge unless otherwise stated with calculated p-values by Fischer s exact test or Chi-square test.
b
Survival to hospital discharge with good neurology.
J.M. Larsen, J. Ravkilde / Resuscitation 83 (2012) 1427 1433 1431
Fig. 2. Forest plot from a meta-analysis of studies including patients with and without acute coronary angiography. The odds ratios are unadjusted for possible selection
bias and should be interpreted with caution. The grey boxes covering the point estimate of the odds ratio illustrate the weight of the individual study in the pooled odds
ratio. These weights were defined by a random effect model due to heterogeneity of the studies as illustrated by a high I-squared. N = number of patients; OR = odds ratio;
CI = confidence interval; CAG = coronary angiography.
Two small studies with systematic acute CAG following OHCA CAG (>6 h) or no CAG before discharge.39 PCI was performed in 61%
with ventricular fibrillation were not included in Table 3 due to with acute CAG, which more often had ST-segment elevation and
more selective inclusion criteria. One of the studies included 15 pre-arrest symptoms indicating ACS. PCI before discharge was only
patients with a survival to hospital discharge of 73%.28 The other performed in 7% of patients without acute CAG. The crude survival
study included 50 comatose haemodynamic unstable patients and to hospital discharge was better in patients with acute CAG (72% vs
demonstrated an impressive six-month survival of 82%.29 49%, p = 0.003). To address possible selection bias, matching with
propensity score analysis was done indicating a survival benefit of
acute CAG in the patients with propensity scores with middle to
3.4. Comparison studies including patients with and without
high likelihood of undergoing acute CAG. No multivariate analysis
acute coronary angiography
of the prognostic effect of acute CAG including the propensity score
was reported in the study.
Table 4 illustrates 10 studies on patients resuscitated from car-
The crude prognostic information from the ten studies was com-
diac arrest of mixed aetiology with acute CAG only performed in
piled in a meta-analysis as illustrated in the forest plot in Fig. 2. All
selected patients.30 39 The indication for performing acute CAG was
studies, except the smallest and oldest, had a significant unadjusted
not specified in most of the studies. The use of acute CAG was very
OR for survival favouring acute CAG. The pooled unadjusted OR
heterogeneous in the studies ranging from 14% to 83%. Overall, the
was 2.78, 95% confidence interval (1.89; 4.10). The high I-squared
patients undergoing acute CAG had a better survival. The character-
illustrates heterogeneity in the studies.
istics of patients in the studies were heterogeneous, e.g. shockable
rhythms ranged from 39% to 100%. The prevalence of comatose sur-
4. Discussion
vivors was only sparsely reported, but TH was generally used more
common than in the pure STEMI-studies.
The largest study prospectively included 986 patients resus- The high rate of mortality associated with OHCA calls for opti-
mised treatment both before and after ROSC. No randomised trials
citated from OHCA at 38 centres in seven countries admitted to
exist evaluating the use of acute CAG following successful resusci-
intensive care units treated with TH.32 Sixty-three percent of the
tation from OHCA (Table 1).
patients presented with acute myocardial infarction, but only 49%
underwent acute CAG, 30% PCI, 5% thrombolytic treatment and
1% coronary artery bypass grafting. Initial shockable rhythm was 4.1. Acute coronary angiography in ST-segment elevation
predictive of a favourable outcome if acute CAG was performed myocardial infarction following resuscitation from out-of-hospital
(p < 0.001), whereas asystole was only predictive of a bad outcome if cardiac arrest
acute CAG was not performed (p < 0.001). Bleeding requiring trans-
fusion was more common in patients with acute CAG (6.2% vs 2.8%, Acute CAG with subsequent PCI compared to fibrinolysis in
p = 0.02). In three other studies, acute CAG was found to be an inde- STEMI patients without preceding cardiac arrest is favourable for
pendent predictor of survival with adjusted OR of 3.8 (p < 0.05), 5.7 survival and morbidity, when the transfer time to a PCI cen-
(p < 0.001) and 11.2 (p < 0.001), respectively.31,36,38 One study only tre is short.3,40 Cardiac arrest survivors are frequently excluded
demonstrated a significant independent predictive value on sur- from randomised studies because of unconsciousness and unsta-
vival with good neurology of performing CAG before discharge, but ble circulation due to post-cardiac arrest syndrome and potential
not acute CAG.33 Another study found a significant independent irreversible brain injury.41 The optimal management of espe-
predictive value of acute PCI, but not acute CAG.35 cially the comatose survivors of OHCA with a proper balance
The newest and only LOE2 study by Strote et al. included 240 between action and withdrawal of treatment is very challenging
resuscitated OHCA patients and compared acute CAG (d"6 h) to later both for the interventional cardiologist and the intensivist. The
1432 J.M. Larsen, J. Ravkilde / Resuscitation 83 (2012) 1427 1433
recommendation in the 2010 CoSTR and 2010 European guide- benefit of acute CAG when adjusting for selection bias by propensity
lines for resuscitation is that acute CAG should be considered in score analysis. Six other studies of our review also demonstrated
resuscitated OHCA patients with ST-segment elevation or new significant adjusted odds ratios in favour of either acute CAG or
LBBB.4,42 Several case series on selected resuscitated patients with acute PCI.23,25,31,35,36,38
ST-segment elevation or new LBBB demonstrate acute CAG to be The poor diagnostic properties of the ECG in resuscitated OHCA
feasible and with a relatively good survival (Table 2). The studies patients with a high prevalence of CAD emphasises the routine use
have poor evidence levels most often including patients with wit- of systematic acute CAG as part of a standard post-cardiac-arrest
nessed arrests and shockable rhythms. This selection of patients protocol. The use of routine acute CAG in conscious survivors is not
probably results in overoptimistic survival rates, but the studies do very controversial as most interventional cardiologists will con-
demonstrate that acute CAG with coronary intervention indeed is sider this as high risk acute coronary syndrome. Routine acute CAG
feasible in the post cardiac arrest setting. in comatose survivors is more debatable due to the poor evidence,
A small retrospective study comparing acute fibrinolysis and possible irreversible brain injury and an inherent slightly higher
acute CAG following OHCA demonstrated no significant difference risk of bleeding complications with concurrent TH.19,32 However,
in survival, but actually a non-significant trend favouring fibri- the studies in our review exclusively including comatose survivors
nolysis probably due to time delay before start of the invasive mainly treated with TH and acute CAG with coronary intervention
treatment.15 Time delay can also explain poorer left ventricu- did show relatively good survival rates.14,19,29,32,37 We recommend
lar ejection fraction in patients transfer from referral hospital future randomised studies including comatose survivors of OHCA
compared to direct admittance to a PCI centre for acute CAG fol- without STEMI or new LBBB undergoing TH. This will be clinical
lowing OHCA.17 This emphasises the need for speed in treatment of feasible and of importance both for the intensive care and inter-
patients with an acute coronary occlusion. If transfer to a PCI centre ventional cardiology communities.
is not possible in a reasonable time, an alternative reperfusion strat-
egy with acute fibrinolysis should still be considered in resuscitated
5. Limitations
patients with STEMI despite preceding chest compressions.
The search strategy only included three databases. Non-English
4.2. Acute coronary angiography in patients following
articles were excluded. Relevant articles could be missing in the
resuscitation from out-of-hospital cardiac arrest
review, but this is less likely as the reference lists of the included
articles and the 2010 CoSTR were screened. The classification of
The 2010 CoSTR and European guidelines on resuscitation rec-
the studies as supporting, neutral and opposing PICO is debat-
ommend acute CAG to be considered in selected resuscitated OHCA
able. We have used a more conservative approach than in the 2010
patients irrespective of ECG findings, if coronary ischaemia is
CoSTR evaluation process by only allowing studies to be classified
suspected to be the aetiology for cardiac arrest, and it may be rea-
as supporting if adjusted statistical evidence was present in order to
sonable to include acute CAG as part of a standardised post-cardiac
reduce confounding. The definition of acute CAG differed between
arrest protocol.4,42 This recommendation is based on observational
the studies from less than 6 h up to less than 24 h. This contributes to
studies with poor evidence levels. In our review, we identified sev-
the heterogeneity of the reported prognosis in the studies. It would
eral mainly newer but still low evidence level studies on patients
have been clinically relevant to make a separate more thorough
with cardiac arrest of mixed aetiology not evaluated in the 2010
prognostic analysis of conscious and comatose survivors, as their
CoSTR, adding further evidence on the topic.21,26 30,34 39
prognosis differ. This was not feasible with the available data. The
Systematic acute CAG in patient without an obvious non-cardiac
meta-analysis was based on prognostic data from heterogeneous
aetiology has demonstrated a high prevalence of significant CAD
studies. This was evident by the high I-squared value. A pooled OR
and a favourable survival (Table 3). Studies with systematic acute
seemed fair as the individual odds ratios all were pointing in the
CAG in patients resuscitated from OHCA with shockable rhythms
same direction. A random effect model was used due to the hetero-
also demonstrate very high survival rates.28,29 Several studies have
geneity. The meta-analysis was not adjusted for possible selection
examined the diagnostic properties of ST-segment elevation fol-
bias as the necessary data was not available. Therefore, the meta-
lowing OHCA compared to angiographic findings with variable
analysis should be interpreted with caution, but several individual
results. In general, the diagnostic values were suboptimal, espe-
studies with adjusted analysis do support the use of acute CAG in
cially the negative predictive value.23 26 In one study, the negative
the post cardiac arrest setting.
predictive value was increased to 100% on behalf of a much poorer
positive predictive value by using an extended ECG criterion for
6. Conclusions
triage with ST-segment elevation and/or depression and/or LBBB
and/or unspecific wide QRS and/or right bundle branch block.26
No randomised studies exist on acute CAG following OHCA. An
However, the author is cautious to recommend implementation of
increasing number of observational studies support feasibility and
this strategy for triage before completion of prospective studies, as
a possible survival benefit of an early invasive approach. Acute CAG
it is well known that the ECG can be without ischaemic findings
is associated to a better survival in studies on resuscitated patients
despite an acute occlusion in patients without preceding cardiac
with heterogeneous aetiology to OHCA. Systematic acute CAG fol-
arrest.
lowing OHCA without an obvious non-cardiac aetiology should be
Our meta-analysis comparing patients with and without acute
strongly considered irrespective of electrocardiographic findings
CAG in populations with mixed aetiology to the cardiac arrest
due to a high prevalence of CAD and unreliable diagnostic proper-
demonstrated a significant crude positive association between
ties of the electrocardiographic findings. Randomised multicentre
acute CAG and survival. Unfortunately, no data is available for an
studies with acute CAG following OHCA are warranted especially in
adjusted analysis to control for selection bias (Table 4 and Fig. 2).
comatose survivors for optimising the diagnostic and therapeutic
Therefore, the pooled OR in the meta-analysis should be inter-
strategy.
preted with caution. The risk of selection bias is emphasised by the
only LOE2 study in which age, bystander cardiopulmonary resus-
citation, daytime presentation, history of PCI or stroke and acute Conflict of interest statement
ST-segment elevation in ECG were positively associated to receiv-
ing an acute CAG.39 The study did however indicate a survival None.
J.M. Larsen, J. Ravkilde / Resuscitation 83 (2012) 1427 1433 1433
Acknowledgements 19. Wolfrum S, Pierau C, Radke PW, Schunkert H, Kurowski V. Mild therapeutic
hypothermia in patients after out-of-hospital cardiac arrest due to acute ST-
segment elevation myocardial infarction undergoing immediate percutaneous
The authors thank chief librarian Conni Skrubbeltrang and
coronary intervention. Crit Care Med 2008;36:1780 6.
librarian assistant Jacob Borg Andersen from the Medical Library 20. Lettieri C, Savonitto S, De Servi S, et al. Emergency percutaneous coronary inter-
vention in patients with ST-elevation myocardial infarction complicated by
at Aalborg University Hospital for valuable help on performing the
out-of-hospital cardiac arrest: early and medium-term outcome. Am Heart J
database search. We thank research secretary Hanne Madsen from
2009;157:569 75.
the Department of Cardiology at Aalborg University Hospital for
21. Szymanski FM, Grabowski M, Karpinski G, Hrynkiewicz A, Filipiak KJ, Opolski
G. Does time delay between the primary cardiac arrest and PCI affect outcome?
assisting in the final preparation of the manuscript.
Acta Cardiol 2009;64:633 7.
Funding: No external funding was used in the preparation of the
22. Hosmane VR, Mustafa NG, Reddy VK, et al. Survival and neurologic recovery
manuscript.
in patients with ST-segment elevation myocardial infarction resuscitated from
cardiac arrest. J Am Coll Cardiol 2009;53:409 15.
23. Spaulding CM, Joly L, Rosenberg A, et al. Immediate coronary angiography in
References
survivors of out-of-hospital cardiac arrest. N Engl J Med 1997;336:1629 33.
24. Anyfantakis ZA, Baron G, Aubry P, et al. Acute coronary angiographic findings in
1. Atwood C, Eisenberg MS, Herlitz J, Rea TD. Incidence of EMS-treated out-of-
survivors of out-of-hospital cardiac arrest. Am Heart J 2009;157:312 8.
hospital cardiac arrest in Europe. Resuscitation 2005;67:75 80.
25. Dumas F, Cariou A, Manzo-Silberman S, et al. Immediate percutaneous coro-
2. Davies MJ. Anatomic features in victims of sudden coronary death. Coronary
nary intervention is associated with better survival after out-of-hospital cardiac
artery pathology. Circulation 1992;85:19 24.
arrest: insights from the PROCAT (Parisian Region Out of Hospital Cardiac Arrest)
3. Van de Werf F, Bax J, Betriu A, et al. Management of acute myocardial infarction
registry. Circ Cardiovasc Interv 2010;3:200 7.
in patients presenting with persistent ST-segment elevation: the task force on
26. Sideris G, Voicu S, Dillinger JG, et al. Value of post-resuscitation electrocardio-
the management of ST-segment elevation acute myocardial infarction of the
gram in the diagnosis of acute myocardial infarction in out-of-hospital cardiac
European Society of Cardiology. Eur Heart J 2008;29:2909 45.
arrest patients. Resuscitation 2011;82:1148 53.
4. Bossaert L, O Connor RE, Arntz HR, et al. Part 9: acute coronary syndromes: 2010
27. Mollmann H, Szardien S, Liebetrau C, et al. Clinical outcome of patients treated
International Consensus on Cardiopulmonary Resuscitation and Emergency
with an early invasive strategy after out-of-hospital cardiac arrest. J Int Med Res
Cardiovascular Care Science with Treatment Recommendations. Resuscitation
2011;39:2169 77.
2010;81(Suppl. 1):175 212.
28. Keelan PC, Bunch TJ, White RD, Packer DL, Holmes DR. Early direct coro-
5. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in
nary angioplasty in survivors of out-of-hospital cardiac arrest. Am J Cardiol
epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies
2003;91:1461 3.
in Epidemiology (MOOSE) group. JAMA 2000;283:2008 12.
29. Hovdenes J, Laake JH, Aaberge L, Haugaa H, Bugge JF. Therapeutic hypother-
6. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for sys-
mia after out-of-hospital cardiac arrest: experiences with patients treated with
tematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol
percutaneous coronary intervention and cardiogenic shock. Acta Anaesthesiol
2009;62:1006 12.
Scand 2007;51:137 42.
7. Morley PT, Atkins DL, Billi JE, et al. Part 3: evidence evaluation process: 2010
30. Bulut S, Aengevaeren WRM, Luijten HJE, Verheugt FWA. Successful out-
International Consensus on Cardiopulmonary Resuscitation and Emergency
of-hospital cardiopulmonary resuscitation: What is the optimal in-hospital
Cardiovascular Care Science with Treatment Recommendations. Resuscitation
treatment strategy? Resuscitation 2000;47:155 61.
2010;81(Suppl. 1):32 40.
31. Merchant RM, Abella BS, Khan M, et al. Cardiac catheterization is underutilized
8. Kahn JK, Glazier S, Swor R, Savas V, O Neill WW. Primary coronary angioplasty
after in-hospital cardiac arrest. Resuscitation 2008;79:398 403.
for acute myocardial infarction complicated by out-of-hospital cardiac arrest.
32. Nielsen N, Hovdenes J, Nilsson F, et al. Outcome, timing and adverse events in
Am J Cardiol 1995;75:1069 70.
therapeutic hypothermia after out-of-hospital cardiac arrest. Acta Anaesthesiol
9. McCullough PA, Prakash R, Tobin KJ, O Neill WW, Thompson RJ. Application of
Scand 2009;53:926 34.
a cardiac arrest score in patients with sudden death and ST segment elevation
33. Reynolds JC, Callaway CW, El Khoudary SR, Moore CG, Alvarez RJ, Rittenberger
for triage to angiography and intervention. J Intervent Cardiol 2002;15:257 61.
JC. Coronary angiography predicts improved outcome following cardiac arrest:
10. Bendz B, Eritsland J, Nakstad AR, et al. Long-term prognosis after out-of-hospital
propensity-adjusted analysis. J Intensive Care Med 2009;24:179 86.
cardiac arrest and primary percutaneous coronary intervention. Resuscitation
34. Aurore A, Jabre P, Liot P, Margenet A, Lecarpentier E, Combes X. Predictive factors
2004;63:49 53.
for positive coronary angiography in out-of-hospital cardiac arrest patients. Eur
11. Quintero-Moran B, Moreno R, Villarreal S, et al. Percutaneous coronary inter-
J Emerg Med 2011;18:73 6.
vention for cardiac arrest secondary to ST-elevation acute myocardial infarction.
35. Cronier P, Vignon P, Bouferrache K, et al. Impact of routine percutaneous
Influence of immediate paramedical/medical assistance on clinical outcome. J
coronary intervention after out-of-hospital cardiac arrest due to ventricular
Invasive Cardiol 2006;18:269 72.
fibrillation. Crit Care 2011;15:R122.
12. Markusohn E, Roguin A, Sebbag A, et al. Primary percutaneous coronary inter-
36. Grasner JT, Meybohm P, Caliebe A, et al. Postresuscitation care with mild
vention after out-of-hospital cardiac arrest: patients and outcomes. Isr Med
therapeutic hypothermia and coronary intervention after out-of-hospital
Assoc J 2007;9:257 9.
cardiopulmonary resuscitation: a prospective registry analysis. Crit Care
13. Garot P, Lefevre T, Eltchaninoff H, et al. Six-month outcome of emer-
2011;15:R61.
gency percutaneous coronary intervention in resuscitated patients after
37. Mooney MR, Unger BT, Boland LL, et al. Therapeutic hypothermia after out-of-
cardiac arrest complicating ST-elevation myocardial infarction. Circulation
hospital cardiac arrest: evaluation of a regional system to increase access to
2007;115:1354 62.
cooling. Circulation 2011;124:206 14.
14. Knafelj R, Radsel P, Ploj T, Noc M. Primary percutaneous coronary intervention
38. Tomte O, Andersen GO, Jacobsen D, Draegni T, Auestad B, Sunde K. Strong and
and mild induced hypothermia in comatose survivors of ventricular fibrillation
weak aspects of an established post-resuscitation treatment protocol a five-
with ST-elevation acute myocardial infarction. Resuscitation 2007;74:227 34.
year observational study. Resuscitation 2011;82:1186 93.
15. Richling N, Herkner H, Holzer M, Riedmueller E, Sterz F, Schreiber W.
39. Strote JA, Maynard C, Olsufka M, et al. Comparison of role of early (less
Thrombolytic therapy vs primary percutaneous intervention after ventricular
than six hours) to later (more than six hours) or no cardiac catheterization
fibrillation cardiac arrest due to acute ST-segment elevation myocardial infarc-
after resuscitation from out-of-hospital cardiac arrest. Am J Cardiol 2012;109:
tion and its effect on F outcome. Am J Emerg Med 2007;25:545 50.
451 4.
16. Pleskot M, Babu A, Hazukova R, et al. Out-of-hospital cardiac arrests in patients
40. Andersen HR, Nielsen TT, Rasmussen K, et al. A comparison of coronary angio-
with acute ST elevation myocardial infarctions in the East Bohemian region over
plasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med
the period 2002 2004. Cardiology 2008;109:41 51.
2003;349:733 42.
17. Peels HO, Jessurun GA, van der Horst IC, Arnold AE, Piers LH, Zijlstra F. Outcome
41. Nolan JP, Neumar RW, Adrie C, et al. Post-cardiac arrest syndrome: epi-
in transferred and nontransferred patients after primary percutaneous coronary
demiology, pathophysiology, treatment, and prognostication. Resuscitation
intervention for ischaemic out-of-hospital cardiac arrest. Catheter Cardiovasc
2008;79:350 79.
Interv 2008;71:147 51.
42. Arntz HR, Bossaert LL, Danchin N, Nikolaou NI. European Resuscitation Coun-
18. Mager A, Kornowski R, Murninkas D, et al. Outcome of emergency percutaneous
cil Guidelines for Resuscitation 2010 Section 5. Initial management of acute
coronary intervention for acute ST-elevation myocardial infarction complicated
coronary syndromes. Resuscitation 2010;81:1353 63.
by cardiac arrest. Coron Artery Dis 2008;19:615 8.