710788

research-article2017
ACC0010.1177/2048872617710788European Heart Journal: Acute Cardiovascular CareVan Den Berg and Body

EUROPEAN
SOCIETY OF
Review CARDIOLOGY ®

European Heart Journal: Acute Cardiovascular Care

The HEART score for early rule 2018, Vol. 7(2) 111­–119
© The European Society of Cardiology 2017
Reprints and permissions:
out of acute coronary syndromes sagepub.co.uk/journalsPermissions.nav
https://doi.org/10.1177/2048872617710788
DOI: 10.1177/2048872617710788
in the emergency department: a journals.sagepub.com/home/acc

systematic review and meta-analysis

Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021

Patricia Van Den Berg1 and Richard Body2,3

Abstract
Aims: The objective of this systematic review was to summarise the current evidence on the diagnostic accuracy of the
HEART score for predicting major adverse cardiac events in patients presenting with undifferentiated chest pain to the
emergency department.
Methods and results: Two investigators independently searched Medline, Embase and Cochrane databases between
2008 and May 2016 identifying eligible studies providing diagnostic accuracy data on the HEART score for predicting
major adverse cardiac events as the primary outcome. For the 12 studies meeting inclusion criteria, study characteristics
and diagnostic accuracy measures were systematically extracted and study quality assessed using the QUADAS-2 tool.
After quality assessment, nine studies including data from 11,217 patients were combined in the meta-analysis applying a
generalised linear mixed model approach with random effects assumption (Stata 13.1). In total, 15.4% of patients (range
7.3–29.1%) developed major adverse cardiac events after a mean of 6 weeks’ follow-up. Among patients categorised as
‘low risk’ and suitable for early discharge (HEART score 0–3), the pooled incidence of ‘missed’ major adverse cardiac
events was 1.6%. The pooled sensitivity and specificity of the HEART score for predicting major adverse cardiac events
were 96.7% (95% confidence interval (CI) 94.0–98.2%) and 47.0% (95% CI 41.0–53.5%), respectively.
Conclusions: Patients with a HEART score of 0–3 are at low risk of incident major adverse cardiac events. As 3.3%
of patients with major adverse cardiac events are ‘missed’ by the HEART score, clinicians must ask whether this risk is
acceptably low for clinical implementation.

Keywords
HEART score, acute coronary syndromes, diagnostic accuracy, sensitivity and specificity, emergency medicine, clinical
decision rules

Date received: 7 June 2016; accepted: 30 April 2017

Background
Chest pain is one of the most common reasons for emer-
gency hospital admission.1 Hospital and emergency depart-
ment (ED) crowding is a growing problem and is associated
with higher patient mortality.2 This highlights the pressing
need for diagnostic strategies that can rapidly ‘rule out’
acute coronary syndromes (ACSs) and avoid unnecessary
hospitals.3 The HEART score is a simple tool that can be
used at the time of a patient’s first presentation to the ED. It
1Maastricht University, Maastricht, The Netherlands
2The University of Manchester, Manchester, UK
3Central Manchester University Hospitals Foundation NHS Trust,

hospital admission. Manchester, UK

While it may be possible to achieve that within as little as
one hour of arrival using high sensitivity cardiac troponin
(hs-cTn) assays, such strategies still require two blood tests.
Furthermore, high sensitivity assays are not available in all
Corresponding author:
Richard Body, Emergency Department, Manchester Royal Infirmary,
Oxford Road, Manchester M13 9WL, UK.
Email: richard.body@manchester.ac.uk
112 European Heart Journal: Acute Cardiovascular Care 7(2)
Table 1. The HEART score.4
HEART score
History Highly suspicious
Moderately suspicious
Slightly suspicious
ECG Significant ST-depression
Non-specific repolarisation disturbance
Normal
Age ⩾65 years
>45–<65 years
⩽45 years
Risk factors ⩾3 risk factors,a or history of
atherosclerotic diseaseb
1 or 2 risk factors
No risk factors known
Troponin ⩾3× normal limit
>1–<3× normal limit
⩽ normal limit
aRisk factors include: currently treated diabetes mellitus, current or
recent smoker, diagnosed and/or treated hypertension, diagnosed
hypercholesterolaemia, family history of coronary artery disease,
obesity (body mass index >30).
bHistory of atherosclerotic disease includes: coronary revascularisation,
myocardial infarction, stroke, or peripheral arterial disease, irrespective
of the risk factors for coronary artery disease.
was designed to identify a group of patients that can be
immediately discharged from the ED following a single
blood test (Table 1).4 Patients with a HEART score of 0–3
are considered ‘low risk’ and eligible for potential immedi-
ate discharge.4
We aimed systematically to appraise the available evi-
dence to determine the diagnostic accuracy of the HEART
score for predicting major adverse cardiac events (MACEs)
in patients with suspected ACS in the ED.
Methods
Search strategy and eligibility criteria
This systematic review was performed in accordance with
the PRISMA guidelines and Cochrane methodology for
diagnostic test accuracy reviews.5,6 We searched the
Medline, Embase and Cochrane databases for the term
‘HEART score’ (in all fields) from 1 January 2008 to 15
May 2016. The time period for the publication was
restricted to start in 2008, the year the HEART score was
first derived and published.4 The reference lists of relevant
publications were also hand searched.
Retrospective and prospective cohort studies, as well as
randomised controlled trials investigating patients with
possible ACSs in the ED, were eligible. In order to be con-
sidered for inclusion, studies needed to evaluate the HEART
score at the time of arrival, and to report the prevalence of
acute myocardial infarction (AMI) and the incidence of
MACEs as outcomes. We excluded conference abstracts
not allowing for sufficient assessment of the methodologi-
cal approach. Likewise, publications in languages other
than English, Dutch and German were excluded.
2
1
0 Index test and outcome measures
2 The HEART score is defined as a composite score ranging
1 from 0 to 10 points. Each of the five acronym domains (his-
0 tory, electrocardiogram (ECG), age, risk factors and tro-
2 ponin) is scored with 0 to 2 points according to the original
1 definition (Table 1).
0
The primary outcome of MACEs was defined as a com-
2
Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021
posite of prevalent or incident AMI, percutaneous coronary
1
intervention, coronary artery bypass graft surgery and all-
0 cause death. Studies only reporting the prevalence of AMI
2 were also considered for the initial analyses. The third univer-
1 sal definition of AMI, consistent with a rise and/or fall of a
0 cardiac biomarker with minimally one result above the 99th
percentile upper reference limit in the context of a patient pre-
senting with cardiac ischaemia (ECG changes or imaging evi-
dence),7 was considered the optimal reference standard.
Patients with a HEART score of 0–3 points are consid-
ered at low risk of developing MACEs and therefore con-
stitute a patient group eligible for potential immediate
discharge. The remaining patients with a score between 4
and 10 points were considered at high risk of developing
MACEs.4,8,9
Data extraction
All results of the complete search on all databases were
screened based on title and abstract for potential eligibility
by two authors (PvdB and RB). Both authors then indepen-
dently undertook a full-text review, after which eligible
studies were short-listed for data extraction. Discrepancies
between investigators were solved by discussion. Trial
authors were contacted for missing data and clarifications
whenever necessary. We extracted and collated all data
required for a diagnostic accuracy assessment including
2×2 tables for the HEART scores 0–3 and 4–10 with respec-
tive MACEs and/or AMI rates, prevalence of MACEs and/
or AMI, proportion of low-risk patients with HEART score
of 0–3 and the percentage of missed MACEs and/or AMI.
The quality of eligible primary diagnostic accuracy
cohort studies was assessed with a modified version of the
QUADAS-2 (Quality Assessment of Diagnostic Accuracy
Studies – version 2) tool to suit the purpose of this system-
atic review.10 The modified QUADAS-2 tool is provided in
the Supplementary Appendix.
Statistical analysis
Studies deemed to have an acceptable risk of bias were
included for further meta-analysis. The analysis was con-
ducted using STATA Statistical Software IC package, ver-
sion 13.1 (StataCorp LP, College Station, TX, USA) applying
Van Den Berg and Body 113

Table 2.  Characteristics of included studies.

Study Year Country Study design Number of sites Study period

Backus et al.8 2010 Netherlands Retrospective cohort study 4 1 Jan–31 Mar 2006
Fesmire et al.14 2012 USA Retrospective cohort study 1 13-month period
Backus et al.9 2013 Netherlands Prospective cohort study 10 Oct 2008–Nov 2009
Six et al.15 2013 9 Asia-Pacific Retrospective cohort study 14 Nov 2007–Dec 2010
countries
Melki and 2013 Sweden Retrospective cohort study 1 1 Jan–12 Feb 2009
Jernberg16
Marcoon et al.17 2013 USA Retrospective cohort study 1 1999–2009
Visser et al.18 2014 Netherlands Prospective cohort study 1 1 Dec 2012–31 Jul 2013
Leite et al.19 2015 Portugal Retrospective cohort study 1 23–29 Jan 2012 and

Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021

23–29 Jul 2012
Carlton et al.20 2015 UK Retrospective cohort study 1 Jul 2012–Aug 2013
Bodapati et al.21 2016 Australia Retrospective cohort study 1 1 Jan 2013–16 May 2013
Sun et al.22 2016 USA Retrospective cohort study 8 1 Jun 1999–Aug 2001
Santi et al.23 2016 Italy Retrospective cohort study 1 1 Jan–30 Jun 2014

MIDAS and METANDI commands utilising a generalised
linear mixed model approach with random effects assump-
tion to generate pooled estimates of diagnostic accuracy.
Between-study heterogeneity was assessed with the
Cochrane Q chi-square test and the I2 statistics. With diag-
nostic test accuracy studies heterogeneity is to be expected,
therefore requiring a random effects model approach by
default.11 We generated a summary receiver operating char-
acteristic (ROC) curve, which allowed calculation of the area
under the curve (AUC) as a global measure of diagnostic test
performance as well as a visual evaluation of heterogeneity
and any potential threshold effect.12 To determine the poten-
tial effects of publication bias a Deeks funnel plot was
obtained, with P<0.10 for the slope coefficient considered to
indicate significant asymmetry and therefore a high likeli-
hood of publication bias.13
Results
Our literature search identified 218 articles of which 18
were considered potentially eligible for inclusion based on
title and abstract screening. After independent review 12
studies met the inclusion criteria for this review providing
the necessary data allowing for computation of diagnostic
accuracy measures.8,9,14–23 General study characteristics are
summarised in Table 2. A more detailed summary of inclu-
sion and exclusion criteria, precise definitions of outcomes,
the exact definitions utilised for the calculation of the
HEART score, as well as the cardiac troponin assays used
and the timing of troponin testing, is provided in the
Supplementary Appendix. Test characteristics for each
individual study including 2×2 tables, sensitivity, specific-
ity, positive predictive value (PPV), negative predictive
value (NPV), as well as the proportion of patients identified
as low risk (HEART score 0–3) are presented in Table 3.
After quality assessment, two further studies were
excluded from the final meta-analysis.17,22 The exclusion of
those two studies was based on a significant deviation from
the original HEART score definition in combination with a
vaguely defined reference standard allocated using rela-
tively old troponin assays as a result of early study periods.
Both authors, based on the QUADAS-2 assessment, agreed
that both studies were at significant risk of allocating inter-
mediate risk patients wrongly to the low-risk group. A sum-
mary of the quality assessment of all 12 initially eligible
studies with the QUADAS-2 tool is shown in Figure 1. The
methodological quality assessment of included studies
identified that most had important limitations. A detailed
assessment of each individual study is available in the
Supplementary Appendix. The major methodological limi-
tations identified in various degrees included deviating
from the original HEART score definition, frequent
unblinded determination of the HEART score or patient
outcome, the use of outdated definitions and insufficient or
lacking information on the timing of reference standard tro-
ponin testing, contributing to potential verification bias.
The study by Carlton et al. was excluded from the meta-
analysis as this study did not provide data about the inci-
dence of MACEs.20 This yielded a total of nine studies for
meta-analysis (Figure 2).
These nine studies included data from 11,217 patients.
The pooled prevalence of MACEs was 15.4% (95% confi-
dence interval (CI) 14.8–16.1%, range 7.3–29.1%) at a
mean follow-up time of 6 weeks. Across the various studies
4101 (36.6%, range 28.2–60.1%) had a HEART score of
0–3 and would therefore have been potentially suitable for
immediate discharge. In this low-risk group a combined
total of 1.6% (95% CI 1.2–2.0%, range 0.9–5.9%) of the
patients would have had a missed MACE. We found no evi-
dence of publication bias (P=0.58; see Supplementary
Appendix).
The pooled sensitivity estimate of the HEART score for
predicting MACEs in the nine studies included was 96.7%
(95% CI 94.0–98.2%) as summarised in Figure 3. The
114 European Heart Journal: Acute Cardiovascular Care 7(2)

Table 3.  Diagnostic and predictive indices of the HEART score for predicting major adverse cardiac events or acute myocardial
infarction (95% confidence intervals in parentheses).

Study Outcome N Sensitivity Specificity NPV PPV Proportion Outcome Missed

low riska prevalence outcomea
Backus 2010 MACE within 6 880 98.1% 41.6% 99.0% 26.9% 32.5% 18.0% 1.0%
weeks (94.6–99.6) (37.9–45.2) (97.1–99.8) (23.3–30.7)
Fesmire MACE within 2148 92.4% 48.5% 97.4% 23.9% 43.3% 14.7% 2.6%
2012 30 days (88.9–95.1) (47.2–51.8) (96.2–98.3) (21.5–26.4)
Backus 2013 MACE within 6 2388 96.3% 43.2% 98.3% 25.8% 36.4% 17.0% 1.7%
weeks (94.0–97.9) (41.0–45.4) (97.2–99.0) (23.6–28.1)
Six 2013 MACE within 2906 96.3% 31.8% 98.3% 17.3% 28.2% 12.9% 1.7%
30 days (93.8–97.9) (30.0–33.7) (97.2–99.0) (15.7–19.0)

Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021

Melki 2013 MACE within 3 410 96.7% 64.7% 99.6% 17.8% 60.2% 7.3% 0.4%
months (82.8–99.9) (59.7–69.6) (97.8–100) (12.3–24.5)
Marcoon MACE within 8252 75.5% 67.5% 96.9% 16.8% 64.1% 8.0% 3.1%
2013 30 days (72.0–78.7) (66.5–68.6) (96.4–97.4) (15.5–18.2)
Visser 2014 MACE within 6 255 93.3% 44.4% 94.1% 41.2% 33.3% 29.4% 5.9%
weeks (84.1–97.8) (37.1–52.0) (86.8–98.1) (33.7–49.0)
Leite 2015 MACE within 6 174 90.9% 63.2% 98.0% 26.3% 56.3% 12.6% 2.0%
weeks (70.8–98.9) (55.0–70.8) (92.8–99.8) (16.9–37.7)
Bodapati MACE within 678 98.6% 43.5% 99.2% 31.6% 34.7% 20.9% 0.9%
2016 30 days (95.0–99.8) (39.2–47.8) (97.0–99.9) (27.3–36.2)
Sun 2016 MACE within 8255 85.8% 51.2% 98.2% 10.3% 48.9% 6.2% 1.8%
30 days (82.5–88.7) (50.1–52.3) (97.8–98.6) (9.4–11.3)
Santi 2016 MACE within 1378 100% 43.7% 100% 23.8% 37.2% 15.0% 0.0%
30 days (98.2–100) (40.8–46.6) (99.3–100) (21.0–26.8)
Carlton 2015 (non)-fatal AMI 959 93.7% 33.9% 98.3% 11.3% 31.6% 8.2% 1.7%
(hs-cTnT) within 30 days (85.5–99.9) (33.1–34.2) (96.2–99.4) (10.3–11.8)
Carlton 2015 (non)-fatal AMI 867 97.0% 34.7% 99.3% 10.9% 35.2% 7.6% 0.7%
(hs-cTnI) within 30 days (88.7–99.5) (34.0–34.9) (97.3–99.9) (10.0–11.2)

N: number of patients; NPV: negative predictive value; PPV: positive predictive value; MACEs: major adverse cardiac events; AMI: acute myocardial
infarction; hs-cTnT: high-sensitivity cardiac troponin T; hs-cTnI: high-sensitivity cardiac troponin I.
aIn patients with HEART score of 0–3.

Figure 1.  QUADAS-2 assessment of eligible studies.

pooled specificity for the HEART score was 47.0% (95% CI
41.0–53.5%). The positive likelihood ratio (LR+) and nega-
tive likelihood ratio (LR–) were 1.82 (95% CI 1.62–2.06)
and 0.07 (95% CI 0.04–0.13), respectively. Depending on
the prevalence of MACEs in the various study cohorts the
NPV was fairly constant between 97.4% and 100%, while
the PPV was more variable between 17.3% and 41.2%.
On summary ROC curve analysis, the AUC of the
HEART score was 0.81 (95% CI 0.77–0.84), as shown in
Figure 4. While overall heterogeneity was relatively high,
Van Den Berg and Body 115
Figure 2.  Flow chart of study selection.5
12% was estimated to be attributed to a threshold effect
with a low between-study variation of 14% for sensitivity
and 5% for specificity. This was visually confirmed by
inspection of the summary ROC curve.
Sensitivity analysis
In a sensitivity analysis pooling all studies eligible for the
review, irrespective of the quality assessment, data from
27,724 patients were combined. The sensitivity of the
HEART score for predicting MACEs varied across individ-
ual studies from 75.5% to 100%. In comparison to the pri-
mary analysis, the pooled sensitivity was lower at 95.1%
(95% CI 90.5–97.5%) with a very high degree of heteroge-
neity (I2=96.09%, 95% CI 94.74–97.45%). The pooled
specificity was slightly higher at 49.3% (95% CI 42.9–
55.8%). The likelihood ratios were respectively calculated
at LR+ of 1.88 (95% CI 1.68–2.10) and LR– of 0.10 (95%
CI 0.05–0.18). The area under the summary ROC curve was
lower, at 0.75 (95% CI 0.71–0.78) with a more scattered
appearance representing higher between-study variability. A
Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021
Forest plot, summary ROC curve and Deeks funnel plot for
the sensitivity analysis are provided in the Supplementary
Appendix.
Sensitivity analysis with the HEART score at
a lower ⩽2 points cut-off
Five studies provided data for a potential lower 2 points or
less cut-off for the HEART score including 6397 patie
nts.14–16,18,21 The pooled sensitivity of the HEART score for
predicting MACEs at the lower 2 points or less cut-off was
higher at 99.4% (95% CI 96.8–99.9%) at the cost of a lower
pooled specificity of 22.0% (95% CI 14.2–32.5%). The
likelihood ratios were respectively a LR+ of 1.28 (95% CI
1.13–1.44) and a LR– of 0.03 (95% CI 0.00–0.18).
HEART score with high-sensitivity troponin
assays
Three studies evaluated the HEART score with hs-cTn
assays, only two of which reported data for MACEs at 30
116 European Heart Journal: Acute Cardiovascular Care 7(2)
Figure 3.  Forest plot of the HEART score sensitivity and specificity for prediciting major adverse cardiac events.
Figure 4.  Summary receiver operating characteristic curve of
the HEART score for prediciting major adverse cardiac events.
days (which rendered a separate meta-analysis inappro-
priate). The prospective study of Visser et al. used the
Roche Elecsys hs-cTnT assay, reporting a sensitivity of
93.3% (95% CI 84.1–97.8%) for MACEs at 6 weeks.18
Conversely, Santi et al. using the same troponin assay in
a recent retrospective study reported 100% sensitivity
Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021
(95% CI 98.2–100%) for MACEs within 30 days.23
Carlton et al. evaluated HEART with two hs-cTn assays,
although the primary outcome was non-fatal AMI within
30 days rather than MACEs.20 In that study, the cut-off
value used for the Roche Elecsys hs-cTnT assay deviated
from the original HEART score definition (requiring a
rise to three times the 99th percentile in order to score 2
points). The sensitivity of the HEART score with hs-cTnT
(93.7%, 95% CI 85.5–99.9%) was found to be lower than
with the Abbott Architect hs-cTnI assay (97.0%, 95% CI
88.7–99.5%).20 All remaining studies reported the usage
of various (contemporary) troponin assays only, with five
studies not specifying the assay used.
Discussion
Our findings demonstrate that the HEART score could be
used to identify patients with a suspected diagnosis of ACS
who are at low probability (1.6%) of developing MACEs
and who could potentially be discharged immediately from
the ED. However, this risk may not be acceptable to clini-
cians. Indeed, only 40% of emergency physicians would be
willing to discharge a patient from the ED if the probability
of MACEs exceeds 1%.24 Furthermore, the post-test prob-
ability of MACEs is heavily dependent on prevalence.
Sensitivity may therefore be a superior measure of diagnos-
tic accuracy, particularly on a meta-analysis of cohorts with
Van Den Berg and Body 117
varying prevalence. We found that the HEART score has a
sensitivity of 96.7%. In our sensitivity analysis (which
included two studies with lower methodological quality),
the pooled sensitivity was lower still, at 95.1%. In an addi-
tional sensitivity analysis of five studies at a lower 2 points
or less cut-off for the HEART score, suggested by the study
of Six et al.,15 suggested the pooled sensitivity of 99.4%
was raised to what is considered an acceptable risk of miss-
ing a MACE in the low-risk group. However, this increase
in sensitivity comes with a drastically decreased pooled
specificity of only 22.0%.
Our main analysis, although excluding two studies with
significant concerns, has been limited by the overall quality
of the remaining studies included. The retrospective nature
and unclear reporting of certain key methodological aspects
means that there is an uncertain degree of verification and
selection bias. It is also important to note the wide range of
troponin assays that have been used in studies evaluating
the HEART score. This should be considered when inter-
preting the results and highlights the need for further
research with improved standardisation of the troponin
assays used with the HEART score. It is also important to
note that the primary outcome (MACEs), which is widely
used in diagnostic research in this field, is a composite.
While it was not feasible or practical to run separate analy-
ses for each individual component of this composite out-
come, it is important to recognise that prevalent and
incident AMI, death and revascularisation have different
clinical significance.25
The specificity of the HEART score may appear low
(45%). However, it is important to recognise that patients
with a HEART score greater than 3 should not be consid-
ered to have ACS ‘ruled in’. Rather, these patients would be
expected to undergo further investigation with serial tro-
ponin testing in accordance with routine clinical protocols.
In doing so, the lack of specificity ought not to increase
resource utilisation, although this can only be robustly
evaluated in the context of a randomised controlled trial.
An obvious advantage of the HEART score is its sim-
plicity. It can be a paper-based score and is easy to calculate
without the use of a computer. With the name of the score
representing an acronym for the variables included in the
score it is easy to remember, although recalling the differ-
ent definitions of each category might still be challenging.
Because troponin results are expressed as multiples of
the 99th percentile, the HEART score was designed to be
used with any commercially available troponin assay. Our
findings suggest varying diagnostic performance when a
high sensitivity assay is used. Indeed, diagnostic sensitivity
is actually lower in two studies that evaluated the HEART
score with hs-cTnT (Roche Elecsys).18,20 On the contrary,
one retrospective study reported a very high sensitivity
with the same assay.23 Although the lack of direct compari-
sons between assays precludes the drawing of definitive
conclusions, this does suggest the need for more work to
validate the HEART score prospectively with hs-cTn
assays.
Other diagnostic strategies to rule out ACS following a
single blood test have also been described. For example,
the computer-based Manchester acute coronary syndromes
(MACS) rule. The MACS rule was shown effectively to
rule out MACEs within 30 days with high sensitivity
(97.8% and 100%) in two external observational studies,
identifying around 20% of patients as ‘very low risk’ and
therefore eligible for immediate discharge.26,27 Similar to
the HEART score, the MACS rule could be used not only to
rule out ACS following a single blood test but also to risk
Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021
stratify remaining patients, facilitating triage to the appro-
priate level of care. None of the studies included contained
sufficient data to enable direct comparison of MACS and
HEART but this is an important objective for future work.
With hs-cTn assays, it may also be possible to ‘rule out’
AMI in patients with initial hs-cTn concentrations below
the limit of detection (LoD) of the assay. While a definitive
rule out of AMI based on a single hs-cTnT at the LoD alone
was found to be inappropriate in a meta-analysis,28 a single
hs-cTnI measurement combined with a non-ischaemic
ECG has been reported to reach comparable diagnostic
accuracy to our findings for the HEART score, providing a
potentially very simplistic rule out strategy.29 Given the
increasing amount of data to back this strategy, the findings
of our meta-analysis may mean that clinical implementa-
tion of the HEART score is less likely at institutions where
hs-cTn assays are in use.
Finally, by adding in serial sampling it may be possible
to improve diagnostic accuracy. For example, the HEART
pathway uses the HEART score alongside troponin testing
at 0 and 3 hours. The addition of the 3-hour sample has
previously been shown to reduce the incidence of MACEs
in ‘low-risk’ patients from 1.8% to 0.0%.30 While the
requirement for serial troponin testing is a disadvantage,
other diagnostic protocols incorporating clinical risk strati-
fication and serial troponin testing over 2–3 hours have
shown similar promise. For example, the ADAPT-ADP and
EDACS-ADP equally showing 0.0% missed MACEs
within 30 days.31 Another observational study recently con-
firmed the potential of the one-hour rule in and rule out
algorithm when combined with a non-ischaemic ECG and
patient history, showing 0.5% missed MACEs within 30
days when including unstable angina and 0.0% missed
MACEs without unstable angina.32 Whether the HEART
score can provide additional value if used alongside the
one-hour rule in and rule out algorithm with hs-cTn should
be a focus for future work.3
Conclusion
Patients presenting with undifferentiated chest pain to the
ED obtaining a HEART score of 0–3 at initial assessment
are at low risk of incident MACEs. As 3.3% of patients
118 European Heart Journal: Acute Cardiovascular Care 7(2)
with MACEs are ‘missed’ by the HEART score, clinicians
must ask whether this risk is acceptably low for clinical
implementation, and should be carefully guided by local
circumstances influencing diagnostic performance. Future
work should focus on robust comparison with alternative
strategies.
Conflict of interest
Richard Body has previously undertaken research involving dona-
tion of reagents without charge by Roche, Abbott, Alere, Siemens
and Randox. Richard Body has accepted the provision of econ-
omy class travel and accommodation to present findings unrelated
to this work at two Roche-sponsored conferences and at a scien-
tific session sponsored by Randox.
Funding
This research received no specific grant from any funding agency
in the public, commercial, or not-for-profit sectors.
References
1. Goodacre S, Cross E, Arnold J, et al. The health care burden
of acute chest pain. Heart 2005; 91: 229–230.
2. Sun BC, Hsia RY, Weiss RE, et al. Effect of emergency
department crowding on outcomes of admitted patients. Ann
Emerg Med 2013; 61: 605–611 e6.
3. Mueller C, Giannitsis E, Christ M, et al. Multicenter evalua-
tion of a 0-hour/1-hour algorithm in the diagnosis of myocar-
dial infarction with high-sensitivity cardiac troponin T. Ann
Emerg Med 2016; 68: 76–87 e4.
4. Six AJ, Backus BE and Kelder JC. Chest pain in the emer-
gency room: value of the HEART score. Neth Heart J 2008;
16: 191–196.
5. Moher D, Liberati A, Tetzlaff J, et al. Preferred report-
ing items for systematic reviews and meta-analyses: the
PRISMA statement. J Clin Epidemiol 2009; 62: 1006–1012.
6. Leeflang MMG, Deeks JJ, Takwoingi Y, et al. Cochrane
diagnostic test accuracy reviews. Syst Rev 2013; 2: 82.
7. Thygesen K, Alpert JS, Jaffe AS, et al. Third universal
definition of myocardial infarction. Circulation 2012; 126:
2020–2035.
8. Backus BE, Six AJ, Kelder JC, et al. Chest pain in the emer-
gency room: a multicenter validation of the HEART score.
Crit Pathw Cardiol 2010; 9: 164–169.
9. Backus BE, Six AJ, Kelder JC, et al. A prospective validation
of the HEART score for chest pain patients at the emergency
department. Int J Cardiol 2013; 168: 2153–2158.
10. Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2: a
revised tool for the quality assessment of diagnostic accuracy
studies. Ann Intern Med 2011; 155: 529–536.
11. Macaskill P, Gatsonis C, Deeks JJ, et al. Chapter 10:
analysing and presenting results. In: Deeks JJ, Bossuyt
PM and Gatsonis C (eds) Cochrane handbook for system-
atic reviews of diagnostic test accuracy, Version 1.0. The
Cochrane Collaboration, 2010. Available at: http://srdta.
cochrane.org/
12. Jones CM and Athanasiou T. Summary receiver operating
characteristic curve analysis techniques in the evaluation of
diagnostic tests. Ann Thorac Surg 2005; 79: 16–20.
13. Deeks JJ, Macaskill P and Irwig L. The performance of tests
of publication bias and other sample size effects in system-
atic reviews of diagnostic test accuracy was assessed. J Clin
Epidemiol 2005; 58: 882–893.
14. Fesmire FM, Martin EJ, Cao Y, et al. Improving risk stratifi-
cation in patients with chest pain: the Erlanger HEARTS(3)
score. Am J Emerg Med 2012; 30: 1829–1837.
15. Six AJ, Cullen L, Backus BE, et al. The HEART score for
the assessment of patients with chest pain in the emergency
department: a multinational validation study. Crit Pathw
Cardiol 2013; 12: 121–126.
16. Melki D and Jernberg T. HEART score: a simple and useful
tool that may lower the proportion of chest pain patients who
Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021
are admitted. Crit Pathw Cardiol 2013; 12: 127–131.
17. Marcoon S, Chang AM, Lee B, et al. HEART score to fur-
ther risk stratify patients with low TIMI scores. Crit Pathw
Cardiol 2013; 12: 1–5.
18. Visser A, Wolthuis A, Breedveld R, et al. HEART score and
clinical gestalt have similar diagnostic accuracy for diag-
nosing ACS in an unselected population of patients with
chest pain presenting in the ED. Emerg Med J 2015; 32:
595–600.
19. Leite L, Baptista R, Leitao J, et al. Chest pain in the emer-
gency department: risk stratification with Manchester triage
system and HEART score. BMC Cardiovasc Disord 2015;
15: 48.
20. Carlton EW, Khattab A and Greaves K. Identifying patients
suitable for discharge after a single-presentation high-sensi-
tivity troponin result: a comparison of five established risk
scores and two high-sensitivity assays. Ann Emerg Med
2015; 66: 635–645 e1.
21. Bodapati SN, Gunnarsson R, McBride WJ, et al. Chest
pain risk assessment in Indigenous and non-Indigenous
Australians using HEART Score. Emerg Med Australas
2016; 28: 138–144.
22. Sun BC, Laurie A, Fu R, et al. Comparison of the HEART
and TIMI risk scores for suspected acute coronary syndrome
in the emergency department. Crit Pathw Cardiol 2016; 15:
1–5.
23. Santi L, Farina G, Gramenzi A, et al. The HEART score
with high-sensitive troponin T at presentation: ruling out
patients with chest pain in the emergency room. Intern
Emerg Med 2017; 12: 357–364.
24. Than M, Herbert M, Flaws D, et al. What is an acceptable
risk of major adverse cardiac event in chest pain patients
soon after discharge from the Emergency Department?: a
clinical survey. Int J Cardiol 2013; 166: 752–754.
25. Kip KE, Hollabaugh K, Marroquin OC, et al. The problem
with composite end points in cardiovascular studies: the
story of major adverse cardiac events and percutaneous coro-
nary intervention. J Am Coll Cardiol 2008; 51: 701–707.
26. Body R, Burrows G, Carley S, et al. The Manchester Acute
Coronary Syndromes (MACS) decision rule: validation with
a new automated assay for heart-type fatty acid binding pro-
tein. Emerg Med J 2015; 32: 769–774.
27. Carlton E, Body R and Greaves K. External validation of the
Manchester Acute Coronary Syndromes decision rule. Acad
Emerg Med 2016; 23: 136–143.
28. Zhelev Z, Hyde C, Youngman E, et al. Diagnostic accu-
racy of single baseline measurement of Elecsys Troponin
Van Den Berg and Body 119
T high-sensitive assay for diagnosis of acute myocardial
infarction in emergency department: systematic review and
meta-analysis. BMJ 2015; 350: h15.
29. Body R, Burrows G, Carley S, et al. High-sensitivity car-
diac troponin t concentrations below the limit of detection to
exclude acute myocardial infarction: a prospective evalua-
tion. Clin Chem 2015; 61: 983–989.
30. Mahler SA, Riley RF, Hiestand BC, et al. The HEART
Pathway randomized trial: identifying emergency department
patients with acute chest pain for early discharge. Circ
Cardiovasc Qual Outcomes 2015; 8: 195–203.
31. Than MP, Pickering JW, Aldous SJ, et al. Effectiveness of
EDACS versus ADAPT accelerated diagnostic pathways for
chest pain: a pragmatic randomized controlled trial embed-
ded within practice. Ann Emerg Med 2016; 68: 93–102.e1.
3 2. Mokhtari A, Borna C, Gilje P, et al. A 1-h combination algo-
rithm allows fast rule-out and rule-in of major adverse car-
diac events. J Am Coll Cardiol 2016; 67: 1531–1540.
Downloaded from https://academic.oup.com/ehjacc/article/7/2/111/5932678 by guest on 05 October 2021