Heart type fatty acid binding protein: An overlooked cardiac biomarker

Abstract

Cardiac troponins (cTn) are currently the standard of carefor the diagnosis of acute coronary syndromes (ACS) in patients presenting to the emergency department (ED) with chest pain (CP). However, their plasma kinetics necessitatea prolonged ED stay or overnight hospital admission, especially in those presenting early after CP onset. Moreover, ruling out ACS in low-risk patients requires prolonged ED observationor overnight hospital admission to allow serialmeasurements of c-Tn, adding cost. Heart-type fatty acid-binding protein (H-FABP) is a novel marker of myocardial injury with putative advantages over cTn. Beingpresent in abundance in the myocellular cytoplasm, it is released rapidly (<1 hour) after onset of myocardial injury and could potentially play an important role in both earlier diagnosis of high-risk patients presenting early after CP onset, as well as in risk-stratifying low-risk patients rapidly. Like cTn, H-FABP also has a potential role as a prognostic marker in other conditions where myocardial injury occurs, such asacute congestive heart failure (CHF) and acute pulmonary embolism (PE). This review provides an overview of the evidence examining therole of H-FABP in early diagnosis and risk stratification of patients with CP and in non-ACS conditions associated with myocardial injury.

• Key messages:
• Heart type fatty acid binding protein is a biomarker that is elevated early in myocardial injury
• The routine use in the emergency department complements the use of troponins in ruling out acute coronary syndromes in patients presenting early with chest pain
• It also is useful in risk stratifying patients with other conditions such as heart failure and acute pulmonary embolism.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=iann20
Annals of Medicine
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/iann20
Heart-type fatty acid-binding protein: an
overlooked cardiac biomarker
Harsh Goel , Joshua Melot , Matthew D. Krinock , Ashish Kumar , Sunil K.
Nadar & Gregory Y. H. Lip
To cite this article: Harsh Goel , Joshua Melot , Matthew D. Krinock , Ashish Kumar , Sunil K.
Nadar & Gregory Y. H. Lip (2020) Heart-type fatty acid-binding protein: an overlooked cardiac
biomarker, Annals of Medicine, 52:8, 444-461, DOI: 10.1080/07853890.2020.1800075
To link to this article: https://doi.org/10.1080/07853890.2020.1800075
Published online: 04 Aug 2020.
Submit your article to this journal
Article views: 868
View related articles
View Crossmark data

REVIEW ARTICLE
Heart-type fatty acid-binding protein: an overlooked cardiac biomarker
Harsh Goel
a,b
, Joshua Melot
a
, Matthew D. Krinock
a
, Ashish Kumar
c
, Sunil K. Nadar
d
and Gregory Y. H.
Lip
e,f
a
Department of Medicine, St. Luke’s University Hospital, Bethlehem, PA, USA;
b
Luis Katz School of Medicine, Temple University,
Philadelphia, USA;
c
Department of Medicine, Wellspan York Hospital, York, PA, USA;
d
Department of Medicine, Sultan Qaboos
University, Muscat, Oman;
e
Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart & Chest Hospital,
Liverpool, UK;
f
Department of Clinical Medicine, Aalborg University, Aalborg, Denmark
ABSTRACT
Cardiac troponins (cTn) are currently the standard of care for the diagnosis of acute coronary
syndromes (ACS) in patients presenting to the emergency department (ED) with chest pain (CP).
However, their plasma kinetics necessitate a prolonged ED stay or overnight hospital admission,
especially in those presenting early after CP onset. Moreover, ruling out ACS in low-risk patients
requires prolonged ED observation or overnight hospital admission to allow serial measure-
ments of c-Tn, adding cost. Heart-type fatty acid-binding protein (H-FABP) is a novel marker of
myocardial injury with putative advantages over cTn. Being present in abundance in the myocel-
lular cytoplasm, it is released rapidly (<1 h) after the onset of myocardial injury and could
potentially play an important role in both earlier diagnosis of high-risk patients presenting early
after CP onset, as well as in risk-stratifying low-risk patients rapidly. Like cTn, H-FABP also has a
potential role as a prognostic marker in other conditions where the myocardial injury occurs,
such as acute congestive heart failure (CHF) and acute pulmonary embolism (PE). This review
provides an overview of the evidence examining the role of H-FABP in early diagnosis and risk
stratification of patients with CP and in non-ACS conditions associated with myocardial injury.
KEY MESSAGES
Heart-type fatty acid-binding protein is a biomarker that is elevated early in myocardial injury
The routine use in the emergency department complements the use of troponins in ruling
out acute coronary syndromes in patients presenting early with chest pain
It also is useful in risk stratifying patients with other conditions such as heart failure and
acute pulmonary embolism.
ARTICLE HISTORY
Received 25 May 2020
Revised 9 July 2020
Accepted 19 July 2020
KEYWORDS
HFABP; heart-type fatty
acid-binding protein;
review; acute coronary
syndrome; troponin;
congestive heart failure;
prognosis
Introduction
Chest pain (CP) is a common presenting complaint in
emergency departments (ED), accounting for >5% of
all visits, >7.5 million ED encounters/year in the US
alone [1]. The most pressing concern in patients with
CP is identifying acute coronary syndrome (ACS), i.e.
patients with acute myocardial infarction (AMI) or
unstable angina (UA), for a rapid institution of guide-
line-based therapy. Significant improvements in this
regard over the last two decades have led to rates of
missed AMI of less than 1–2% [2–4]. Conversely,
among all-comers with CP, over half have “non-
specific”CP, with 30% being admitted to the hospital
and barely5% eventually diagnosed with ACS, costing
billions of dollars in unnecessary diagnostic testing
and hospital stays [5].
Risk assessment of CP centers on history, electrocar-
diogram (EKG), and biomarkers. Aspartate transamin-
ase (AST) was the first biomarker used in defining AMI
in 1959 [6]. Since then several legacy biomarkers,
including lactate dehydrogenase (LDH), myoglobin,
creatine-kinase (CK), its cardiac-specific iso-enzyme CK-
MB, were used, but they have been superseded by
cardiac troponins (cTn) [7]. Though proven to be the
most sensitive and specific biomarker, cTn still leaves
important gaps. First, there is a 4–6 hours delay from
symptom-onset to first appearance of measurable cTn
in plasma. This often necessitates overnight stay for
many patients to allow serial measurements before
AMI can be reliably ruled out, thus increasing hospital-
isations and health care costs [8]. The use of high-
sensitivity cardiac troponin (hs-Tn), does offset this
CONTACT Harsh Goel harsh_goel@hotmail.com Department of Medicine-EW4, St. Luke’s University Hospital, 801 Ostrum Street, Bethlehem,
18015, PA, USA
ß2020 Informa UK Limited, trading as Taylor & Francis Group
ANNALS OF MEDICINE
2020, VOL. 52, NO. 8, 444–461
https://doi.org/10.1080/07853890.2020.1800075

delay to a certain degree, but at the cost of high
false-positives. Second, prolonged elevation of plasma
cTn (7–10 days) after an AMI complicates utility as a
marker of early re-infarction. To address these gaps, a
host of novel biomarkers-including structural proteins,
enzymes of energy metabolism, inflammatory markers,
cell-adhesion molecules, and extracellular matrix pro-
teins have been investigated. More prominent among
these include heart-type fatty acid-binding protein (H-
FABP), glycogen phosphorylase isoenzyme-BB (GPBB),
copeptin, and ischaemia-modified albumin, among
others [9,10]. Among these, (H-FABP) is oldest known,
and hence perhaps the most well-studied.
The current review aims to: (i) put in perspective
the current literature comparing H-FABP to cTn and
hs-Tn, (ii) offer insights as to whether H-FABP still has
a role as a marker of myocardial injury in the current
era of cTn, and if so, the population most suited for it,
and (iii) briefly review some emerging, non-ACS indica-
tions for H-FABP use.
Tissue distribution and plasma kinetics of
H-FABP
Fatty acid-binding proteins (FABP) are members of the
lipid-binding proteins superfamily. They are both
membrane-bound –aiding cellular long-chain fatty
acid (FA) uptake –and cytoplasmic, being crucial to
intracellular transport of FAs to sites of metabolic con-
version. Hence, FABPs are ubiquitous, though espe-
cially abundant in tissues with an active FA
metabolism, including heart, kidneys, brain, and mam-
mary glands, among others [11]. Among nine tissue-
specific cytoplasmic FABPs identified so far, FABP-3 is
predominantly distributed in cardiac myocytes and is
also named heart-type fatty acid-binding protein (H-
FABP) [12]. However, the myocardial tissue-specificity
of H-FABP is not absolute, significant amounts being
present in skeletal muscle, kidneys, mammary glands,
testes, lungs and stomach [13,14].
Plasma kinetics of H-FABP reflects small size
(15 kDa), and abundant existence in freely soluble
form in the cardiomyocytecytoplasm, in contrast to
cTn, which is largely bound to the contractile ele-
ments of the cardiomyocyte. Hence, significant myo-
cardial injury or even necrosis has to occur before cTn
is released into the plasma in quantities detectable by
standard assays. The abundance and freely soluble
cytoplasmic location of H-FABP are evidenced by the
fact that plasma H-FABP concentrations in response to
myocardial injury rise to >100 times the plasma con-
centration of cTn, hence the normal cut-off of 5–7 ng/
ml versus 0.05 for the latter (Tables 1 and 2). Whilst
CK-MB and cTn are undetectable for around 4–6h
after symptom-onset, peak at around 12 h, and return
to baseline at 24–72 h and 7–10 days, respectively [39],
plasma H-FABP levels start rising within one hour,
peak at 4–6 h, and return to baseline around 24 h after
myocardial injury, owing to rapid renal clearance
[40,41]. The distinct plasma kinetic profile offers two
theoretical advantages, i.e. (i) enhanced utility as an
earlier biomarker of AMI, and (ii) utility as a marker of
re-infarction. Moreover, given the predominant pres-
ence in soluble form, even minor myocardial ischae-
mia and injury should cause detectable plasma
elevations of H-FABP. Hence, beyond aiding early diag-
nosis of AMI, H-FABP may help identify troponin-nega-
tive high-risk patients with CP, and hence refine risk-
stratification of such patients.
H-FABP versus cTn as biomarker of AMI
H-FABP versus cTn: sensitivity, specificity and
accuracy
First recognised as a potential marker of myocardial
damage in the late 1980s–early 1990s [40,42,43], the
following decade saw H-FABP easily surpassing the
legacy markers (CK-MB and myoglobin), especially
early after symptom onset [44,45]. However, the rapid
development of cTn assays in the late 1990s, after
demonstration of excellent sensitivity and specificity in
the eventual diagnosis of AMI, led to the adoption of
cTn in the universal definition of AMI in 2000, and
relegated H-FABP to the background.
Early comparisons between H-FABP and cTn, before
the latter became part of universal definition of AMI,
revealed H-FABP far exceeding the sensitivity of cTn,
especially in those presenting 3-hours of symptom-
onset in both high-risk and low-risk cohorts [15–17].
Notably, using cTn to define AMI led to a decrease in
H-FABP’s sensitivity and an increase in that of cTn,
though the former remained significantly better [16].
As noted in Table 1, there is significant heterogeneity
in findings across studies, likely due to small sample
sizes, different cut-off values, specific cTn and H-FABP
assays used, population characteristics, definition of
end-points (AMI versus ACS), and time to symptom-
onset, among others. Nevertheless, there is certainly
consistency regarding the superior sensitivity of H-
FABP over cTn, especially early after symptom-onset,
with cTn catching up or exceeding H-FABP after about
hour 4–6. On the other hand, cTn remains more spe-
cific at all times. Consolidating the evidence, several
meta-analyses have confirmed a higher sensitivity for
ANNALS OF MEDICINE 445

Table 1. Comparative sensitivity and specificity of HFABP versus troponin by time of symptom onset in evaluation of acute chest pain.
First author, year.
(End-point) Population (N)
Plasma cut-off (ng/ml)
Time to
s/s onset
HFABP Tn HFABP þTn
Sens.
(NPV) %
Spec.
(PPV)% AUC
Sens.
(NPV)%
Spec.
(PPV)% AUC
Sens.
(NPV) %
Spec
(PPV) %Tn HFABP
Haastrup, 2000 [15] Typical CP (130) 0.5 8 2.3 (0–6) h 76 (95) 83 (46) NR 33 (88) 96 (64) NR NR NR
1.0 12 76 (95) 96 (80) 19 (86) 98 (67)
NSTEMI ¼12.3% 2.0 15 62 (93) 97 (81) 19 (86) 99 (80)
Seino, 2003 [16] (AMI) CP þNon-diagnostic EKG (371) NR (TnT) 6.2 <2 h 89 (80) 52 (69) 0.72 22 (50) 94 (80) NR NR NR
AMI¼49% 2–4 h 96 (91) 45 (68) 0.84 57 (65) 70 (91) NR
4–6 h 100 (100) 40 (57) 0.96 67 (71) 66 (61)
6–12 h 97 (97) 55 (55) NR 94 (95) 68 (62)
12–24 h 95 (90) 53 (70) NR 95 (92) 65 (76)
Seino, 2004 [17] (AMI) CPþ"ST or "Tn (129) NR (TnT) 6.2 <3 h 100 (100) 63 (44.4) NR 50 (86.7) 96.3 (80) NR NR NR
AMI ¼24% 3–6 h 75 (93.8) 93.8 (75) 0 (78.9) 93.8 (0) NR
6–12 h 100 (100) 72.7 (62.5) 60 (84.6) 100 (100)
>12 h 100 (100) 75 (62.5) 100 (100) 87.5 (76)
Ruzgar, 2006 [18] (ACS) Patients with ACS (40) 0.01 (TnT) 6.2 <6 h 95.2 100 NR 38.1 100 NR NR NR
STEMI ¼52% 6–24 h 91 100 100 100 NR
NSTEMI¼30%
Cavus, 2006 [19] (ACS) Typical CP <1 hour (67) 0.1 (TnT) 7 <1 h 97.6 38.5 NR 100 23.1 NR NR NR
STEMI ¼27%
NSTEMI ¼10%
4 h 97.6 88.5 100 88.5 NR
McCann, 2008 [20] (AMI) Ischaemic CP (415) 0.03 (TnT) 5 <4 h 73 (73) 71 (71) 0.77 55 (68) 95 (92) 0.78 85 (83) 69 (73)
STEMI ¼18%
NSTEMI ¼30%
4 h 78 (75) 56 (61) 0.74 (all) 88 (90) 94 (92) 0.88 (all) 98 (97) 55 (66)
Valle, 2008 [21]
(AMI þACS)
Suspected ACS (419) NR (TnT) 7 74 ± 51min 60 (80) 88 (72) NR 19 (69) 99 (97) NR NR NR
AMI ¼35%
Orak, 2010 [22] (ACS) Sudden CP þdyspnea/syncope/nausea /
vomiting <6 h duration (83)
0.01 (TnI) 2 <3 h 100 75 0.967 (<6h) 100 20 0.556 NR NR
3–6 h 97 68 75 21 (<6h) NR
STEMI ¼58%
NSTEMI ¼6%
<6 h (all) 98 71 77 20
Haltern, 2010 [23] (AMI) Ischaemic-type CP (94) 0.03 (TnT) 7.3 <4 h 86 (92) 66 (50) 0.76 (<4h) 42 (81) 100 (100) 0.71(<4h) 93 (96) 66 (52)
STEMI ¼12% >4 h 59 (72) 64 (50) 0.71 (all) 100 (100) 100 (100) 0.87(all) 100 (100) 64 (63)
NSTEMI ¼16% NR
Kim, 2010 [24] (AMI) CP suggesting AMI (117) 0.1 (TnT) 19 <2 h 60 77.4 (all) 0.78 (all) 20.2 98.1 (all) 0.82 (all) 33.3 NR
2–4 h 64.7 17.7 82.4 NR
AMI-55% 4–6 h 91.7 58.3 91.7
6–12 h 88.9 66.7 80
McMahon, 2012 [25] (AMI) CP considered cardiac (1128) 0.37 (TnI) 5.24 <3 h 64.3 (93) 84.2 (43) 0.84 50 (92) 93.3 (60) 0.76 71.4 (94)
AMI ¼10% 3–6 h 85.3 (97) 88.7 (55) 0.89 67.6 (95) 94.3 (66) 0.85 88.2 (98)
6–12 h 89.9 (98) 93.5 (70) 0.94 81 (97) 94.2 (70) 0.90 92.4 (99)
12–24 h 90.1 (99) 91.4 (56) 0.97 95.8 (99) 94.3 (71) 0.98 NR (100)
Garcia-Valdecasas,
2011 [26] (AMI)
Ischaemic CP within 6 h (165) 0.6 (TnI) 6.2 <3 h 81 (82) 53 (52) 0.73 6 (61) 98 (67) 0.66 NR NR
AMI ¼40% <6 h 81 (80) 50 (52) 0.72 25 (65) 91 (64) 0.66 NR
Aldous, 2012 [27] (AMI) CP s/o AMI w/o STEMI 0.028 (TnI) 60 4 h 50 (90.7) 89.8 (47.6) NR 85 (95.1) 94.7 (81) NR 86.7 (97.2) 87 (55.3)
NSTEMI ¼15.6%
Gerede, 2015 [28] (AMI) Ischaemic-type CP >30min
duration (48)
0.04 (TnI) 7 <3 h 89 (86) 100 (100) NR 33 (50) 100 (100) NR NR NR
NSTEMI ¼50% 3–6 h 70 (73) 89(88) 70 (67) 67 (70) NR
>6 h 100 (100) 89 (83) 100 (100) 89 (83)
Vupputuri, 2015 [29] (AMI) CP suggestive of ischaemia (77) 0.14 (TnI) 6.4 6 h 100 (100) 85.7 (88.2) NR 46.1 (63.6) 100 (100) NR NR NR
AMI >50% >6 h 78.6 (62.5) 45.5 (64.7) 78.6 (78.6) 100 (100)
Time to symptom onset reported as Mean (SD) or Median (IQR), as applicable. NPV and PPV were calculated in case of Seino 2003 using published raw data.
AM: Acute myocardial infarction; ACS: acute coronary syndrome; CP: Chest pain; NSTEMI: non ST-elevation myocardial infarction; s/o-symptoms of; STEMI: ST-elevation myocardial infarction; NR: not reported.
446 H. GOEL ET AL.

H-FABP in early diagnosis of AMI, though at the cost
of a lower specificity [46–48]. Combining the two
markers improved sensitivity, albeit at the cost of even
lower specificity [47,48]. The better sensitivity and
lower specificity of H-FABP translates to a similar or
only marginally better overall test accuracy over cTn
in most reports. Of note, almost all investigations after
2004 use cTn elevation to define AMI, making it diffi-
cult and perhaps impossible for a novel marker to
exceed cTn in test accuracy, and more importantly,
likely underestimating the true sensitivity of H-FABP.
H-FABP versus cTn: predictive values and role of
population characteristics
Though sensitivity, specificity, and receiver operating
characteristics-area under the curve (ROC-AUC) are
useful measures of a test’s fundamental credentials
when compared to a “gold standard”, they are little
help in guiding clinical decisions. To that end, nega-
tive and positive predictive values (NPV and PPV
respectively) are much more relevant, since they dir-
ectly depict the probability of a negative or positive
test indicating the absence or presence of a disease,
respectively [49]. In the report by McCann et al., for
example, H-FABP had a NPV of 73% in 415 patients
presenting within 4-hours of symptom-onset, given an
AMI prevalence of 50% [20]. Changing prevalence to a
more realistic 5%, while keeping sensitivity and specifi-
city unchanged, the NPV of H-FABP increases to
98.5%. Even a conservative 10% prevalence yields an
NPV of 95.5%.
To demonstrate this more clearly, we extracted raw
data of AMI prevalence, time to symptom-onset, and
H-FABP test characteristics from reports in Tables 1
and 2. To study whether, and to what extent the first
two could explain the observed heterogeneity in NPV
among these reports, we regressed AMI prevalence
(independent variable) against overall NPV of H-
FABP(dependent variable) using the Analyze-it Tool
Pak in Microsoft Excel 2016. As expected, there was a
moderate correlation (R¼–0.55, p¼.0003) between
AMI prevalence and NPV (Figure 1, upper panel), with
prevalence accounting for 1/3 (R
2
¼0.3042) of the
variability in NPV. Obviously, given the rapid rise and
decline of H-FABP, time from symptom-onset should
also influence NPV. To isolate the effects of time and
AMI prevalence, Figure 1 (lower panel) displays the
relation between prevalence of AMI and NPV only
among early presenters (<3–4 h), this time revealing
an even stronger correlation between the two (R¼
0.60, p¼.01), with prevalence accounting for over
one-third the variability in reported NPVs (R
2
¼0.36).
It should be noted that since most reports did not
report AMI prevalence in each time from symptom
onset sub-group, we assumed that AMI prevalence in
each sub-group was not significantly different from
the overall prevalence. Though a possible source of
error, we believe that the prevalence of AMI among
early presenters should indeed be higher than late
presenters. Hence our assumption, even if erroneous,
should result in an error on the conservative side, i.e.
underestimate the correlation between the two varia-
bles rather than overestimate it. Put another way,
using the same raw data, Figure 2 depicts the effect
of changing AMI prevalence to 10% on NPV among
early presenters, keeping sensitivity and specificity
unchanged. We again assumed a similar AMI preva-
lence between early presenters and the entire. As
expected, NPV uniformly increases markedly in each
case. Much more importantly, heterogeneity among
these reports almost disappears, revealing an NPV of
>95% consistently across reports.
H-FABP is most suited to rule out AMI in low-risk
early presenters
The clear association between AMI prevalence and
NPV, as well as the impact of a “real-world”prevalence
of AMI on NPV of H-FABP, as suggested by Figures 1
and 2, respectively, offer clues regarding causes of
heterogeneity in the literature, while aiding clinical
application of H-FABP. Indeed, H-FABP may be best
suited for ruling out AMI in low-risk patients present-
ing early after CP onset (<4 h).
Directly supporting this, a few large cohorts with
AMI prevalence 10% have consistently found very
high NPVs for H-FABP [25,50]. McMahon et al. reported
an NPV of 93% for H-FABP versus 92% for cTn in a
large cohort with an AMI prevalence of 10% among
those presenting within 3-hours of symptom-onset
[25]. In the RATPAC trial, a low-risk (8% ACS preva-
lence) cohort of 850 patients presenting to the ED at
a median 220 min after symptom onset, admission H-
FABP had an NPV of 97% versus 98% for cTn [50].
However, NPV for the early presenter (3 h) cohort
was not reported by the RATPAC authors, neither was
raw data available to allow calculation of NPV for this
sub-group.
Hence, when applied to a more real-world popula-
tion and early presenters, H-FABP does indeed seem
to consistently show a very high NPV, providing a
potential tool to “rule out”AMI during the early
ANNALS OF MEDICINE 447

window of cTn-negativity. To put this in further per-
spective, a clinically acceptable marker should not
exceed the current “accepted”rate of missed AMI, i.e.
1–2%, dictating the need for an NPV 98% [51]. In
this regard, Body et al. reported an NPV of 98.8%
when H-FABP and cTn were combined in clinically
low-risk patients, enabling AMI to be ruled out at pres-
entation in 45% of all patients, at the cost 6 AMIs
missed per 1000 patients, a miss rate of 0.6% [52].
Finally, given the staggered time course of the two
markers’rise in plasma, combining them may aid in
better assessing the onset time of ischaemia. Hence,
patients with an uncertain time of symptom-onset, a
positive H-FABP with a normal cTn will likely mean
duration of symptoms of 0–4 h, whereas a normal H-
FABP with elevated cTn would indicate the ischaemic
event having occurred >24 h previously. Such
knowledge/assessment could have important implica-
tions on individualising treatment strategies.
On the other hand, the specificity and PPV of H-
FABP is consistently lower than cTn, regardless of time
from symptom-onset. Given the high prevalence of
AMI in populations studied, PPV can only be expected
to be much lower in a real-world population, making
it largely unsuited, or at the very least inferior to c-Tn
for confirming AMI, patients deemed high risk based
on history and/or EKG. This low specificity is likely
multi-factorial, including known elevations of H-FABP
in those with renal disease, skeletal muscle disorders/
trauma, and myocardial injury of diverse aetiologies
like heart failure, pulmonary embolism (see later), etc.
Finally, two other factors need to be considered
when interpreting extant evidence. Firstly, using the
comparator biomarker (cTn) itself as the diagnostic
Study AMIPrevalence NPV
17.055.00102,miK
59.021.00002,purtsaaH
79.080.04102,nosnilloC
58.05.05102,edereG
28.033.00102,nretlaH
Garcia-Valdecasas, 2011 0.4 0.8
79
.01.02102,nohaMcM
58.072.03102,ffuR
49.081.01102,ydoB
45.044.03002,atakaN
47.084.08002,nnaCcM
89.042.04002
,onieS
18.038.06002,ragzuR
19.094.03002,onieS
8.053.08002,ellaV
77.025.01102,euonI
47.063.02102,sreggE
19.0
61.02102,suodlA
96.055.03102,arumatiK
9.090.03102,ttrohS
38.023.04102,regrebeneohS
87.032.05102,knaB
9.093
.05102,imaG
25.036.06102,snelleK
76.05.07102,ollengA
89.032.06002,suvaC
98.087.0karO
Study Prev<4h NPV<4h
Haastrup, 2000 0.12 0.95
Seino, 2003 0.54 0.86
Seino, 2004 0.23 1
Cavus, 2006 0.23 0.98
McCann, 2008 0.1 0.93
Valle, 2008 0.35 0.8
Orak, 2010 0.78 0.89
Haltern, 2010 0.29 0.92
Mcmahon, 2012 0.1 0.93
Garcia-Valdecasas,
2017 0.4 0.82
Aldous, 2012 0.16 0.91
Gerede, 2015 0.5 0.86
Inoue, 2011 0.61 0.43
Kitamura, 2013 0.65 0.59
Shoeneberger, 2014 0.32 0.83
Bank, 2015 0.33 0.76
y = -0.3409x + 0.9507
R² = 0.3042
R=0.55, p=0.003
0
0.2
0.4
0.6
0.8
1
1.2
0 0.2 0.4 0.6 0.8 1
AMI pevalence
NPV
y = -0.4257x + 0.9932
R² = 0.3603
R=0.60, p=0.01
0
0.2
0.4
0.6
0.8
1
1.2
0 0.2 0.4 0.6 0.8 1
AMI pevalence <4-h
NPV<4-h
Figure 1. Correlation between AMI prevalence and negative predictive value overall (upper panel), and among those presenting
<4-h after symptom-onset (lower panel).
448 H. GOEL ET AL.

gold standard for the outcome (AMI) being studied, as
has been the case in the majority of reports in the last
two decades is flawed. Intuitively, test characteristics
of cTn will have a direct influence on the test charac-
teristics of H-FABP, independent of all other factors. In
particular, low specificity of cTn will impact the sensi-
tivity of H-FABP, since some false-positives (due to low
specificity) on cTn-testing will be erroneously deemed
false-negatives (hence lower sensitivity)on H-FABP-test-
ing. Indeed, Seino et al. found a decline of H-FABP’s
sensitivity by about 5–10% for all time points after
symptom-onset [16], and a recent meta-analysis
reported H-FABP having a sensitivity of 76% when cTn
was sued to diagnose AMI, versus 91% otherwise [46].
The second issue pertains to the relative assay qual-
ity of the two markers. Given that cTn has become the
standard of care, and in fact now defines AMI, there is
obviously a greater commercial interest in advancing
cTn assays, rather than a novel test that bears signifi-
cantly greater burden of evidence to bring. As a result,
cTn assays have constantly improved in sensitivity and
precision, while H-FABP assays have seen little change
[53]. Indeed, most studies use point-of-care, semi-
quantitative H-FABP assays, which detect either nor-
mal or elevated H-FABP above a cut-off set at 6ng/
ml. This is problematic since such tests may suffer
inert-observer variability in result interpretation (usu-
ally colour development), as well as the inherent
inability to distinguish moderate from high levels of
the marker. Even with these early generation assays,
H-FABP has proven to be equally sensitive as even the
latest generation hs-Tn. Hopefully, novel, highly sensi-
tive and precise H-FFABP assays will aid this marker to
realise its full promise.
H-FABP vs cTn diagnosis and prognosis of
unstable angina (UA)
UA and NSTEMI represent a continuum, with the
boundary between them constantly changing as preci-
sion and sensitivity of biomarkers has advanced. In
essence, many patients who were classified as UA in
the era of CK/CK-MB, are now classified as NSTEMI,
due to the much higher sensitivity of cTn. Biomarker
release likely begins even with minor myocardial
injury, but may not cross the assay threshold or diag-
nostic cut-off. A soluble marker like H-FABP rises early
and to a greater degree than a structurally bound
marker like cTn, the latter requiring significant necrosis
before release. Given the continuum of severity and
amount of myocardial injury in patients with UA,
some patients will have enough biomarker leak to
cross the detection threshold of an assay, while some
may not. This would especially be true of semi-quanti-
tative assays as have mostly been used for H-FABP.
Seino et al. found admission H-FABP elevated in 24/
51 patients with UA (14/51 had elevated cTn), whereas
Cavus et al. found admission and 4-hour H-FABP
0.7
0.8
0.9
1
1.1
Negave Predicve Value (NPV)
Seino, 2003 (<2 h)
Seino, 2003 (2-4 h)
Seino, 2003 (4-6 h)
Seino, 2004 (<3 h)
Seino, 2004 (3-6 h)
Ruzgar, 2006 (<6 h)
McCann, 2008 (<4 h)
Haltern, 2010 (<4 h)
McMahon, 2010 (<3 h)
McMahon, 2010 (3-6 h)
Garcia-Valdecasas, 2011 (<3 h)
Garcia-Valdecasas, 2011 (<6 h)
Aldous, 2012 (<4 h)
Gerede, 2015 (<3 h)
Gerede, 2015 (3-6 h)
Vupputuri, 2015 (<6 h)
Reported
NPV
NPV at 10%
prevalence
Figure 2. Reported NPV and NPV if prevalence of AMI in each report were to be 10%. NPV at 10% prevalence calculated using
the formula: NPV¼ðspecificity 1prevalence
ðÞÞ
½specificity 1prevalence
ðÞðÞþ
1sensitivity
ðÞ
prevalence
ðÞ
:Raw data of AMI prevalence and NPV were obtained from stud-
ies in Table 1.
ANNALS OF MEDICINE 449

elevated in only 1/42 patients with UA [16,19].
Compared to hs-Tn, Eggers et al. found mean admis-
sion H-FABP levels not significantly different from
those without ACS, whereas mean hs-Tn levels were,
though even the latter were elevated in only 18/68
patients with UA [31]. Besides being generally small in
size, each of these reports had differing definitions of
UA, and variable proportions of patients with con-
founding illnesses, like renal failure, heart failure,
tachy-arrhythmias, etc. Valle et al. found the sensitiv-
ity/NPV of H-FABP to fall from 60%/80% to 47%/56%,
respectively, when ACS (AMI þUA) was used as out-
comes versus AMI [21]. Only 24.4% of patients with
UA had elevated H-FABP versus 13.2% for cTn.
However, this is controversial, as according to the uni-
versal definition of AMI, any significant rise in cardiac
enzymes would class the definition as NSTEMI rather
than UA.
Regardless of these barriers, H-FABP has repeatedly
been demonstrated to have prognostic value incre-
mentally to-and indeed independent of-cTn among
patients presenting to the ED with ACS (see below).
Though lacking direct evidence of its role in UA,
largely due to inherent issues with the clinical entity
itself, the enhanced prognostic ability of H-FABP likely
stems from its ability to identify patients with minor
myocardial injury. Definitive evidence of this would
require large prospective studies excluding patients
with any co-existent confounding co-morbidities, like
heart failure, renal disease, tachy-arrhythmias, myo-
pericarditis, etc. Indeed, maybe exquisitely sensitive
markers like H-FABP, or for that matter hs-Tn, could
be used to define UA when cTn is normal. Again, large
scale studies to determine cut-offs for normality, and
the development of high-precision assays are both
fundamental to achieving this.
H-FABP in the era of high-sensitivity troponin
(hs-Tn)
The last decade has seen significant improvements in
cTn assays, with current generation “highly-sensitive”
troponin (hs-Tn) assays able to detect very low con-
centrations of cTn in the plasma. In general, these
assays have <10% coefficient of variation at plasma
concentrations that are an order of magnitude lower
than conventional cTn assays. Although largely struc-
turally bound to the myocyte contractile elements,
about 5% of cTn is present in free form in the cyto-
plasm [54]. Akin to H-FABP, this cytoplasmic cTn is
released early after onset of ischaemic injury, but falls
below the detection limit of conventional assays.
Hence, hs-Tn assays, by detecting this miniscule rise
early after symptom-onset have proven more sensitive
than cTn, and displayed very high NPVs [55–57].
Obviously, this increased sensitivity comes at the cost
of poorer specificity. Intuitively, it follows that the
major improvement of hs-Tn over cTn lies largely in
rapidly “ruling out”, rather than “ruling in”AMI.
Therefore, hs-Tn has a role very analogous to H-FABP,
i.e. shortening the window of cTn-negativity.
In one of the earliest reports of hs-Tn, H-FABP was
noted to be the only biomarker among several to
have equivalent diagnostic accuracy as hs-Tn in those
with CP onset <3-hours, and superior to hs-Tn in
those with onset <2h [56]. Several investigations since
have compared the two markers, especially early after
symptom onset (Table 2). Importantly, the hs-Tn assay
itself has evolved rapidly since being introduced, mak-
ing it difficult to compare earlier reports to more
recent ones. As with the c-Tn studies, most reports
have a very high AMI prevalence. Nevertheless, and
especially in more recent reports, using the latest-gen-
eration assays, hs-Tn has generally demonstrated
superior sensitivity and NPV compared to H-FABP,
even early after symptom onset. This increased sensi-
tivity, however, comes at a cost of lower specificity.
Overall test accuracy, as measured by the ROC-AUC,
seems largely equivalent between the two markers.
In the largest cohort to date, the APACE study
enrolled 1074 consecutive patients with CP suggestive
of AMI [58]. H-FABP had a lower ROC-AUC than hs-Tn
in the overall cohort (0.84 vs 0.94), and in those pre-
senting <3-hours from symptom-onset (0.85 vs 0.92).
Combining the two markers yielded an even lower
accuracy than hs-Tn alone (ROC-AUC 0.88 vs 0.92).
However, both H-FABP and hs-Tn had very high NPV
(94% vs 98%, respectively), and poor PPV (41% vs
42%) with an AMI prevalence of 20%. Similar findings
were reported by Collinson et al. in another large
cohort of 850 low-risk patients presenting early to the
ED with CP [50]. Meta-analyses seem to confirm the
higher sensitivity of hs-Tn, and the relative lack of
improvement with H-FABP [59,60]. However, there was
significant heterogeneity among studies, and early
presenters (<3–4 h from symptom-onset) –the real
target population for both markers –were largely
missing in both analyses, still leaving questions about
the utility of H-FABP in the era of hs-Tn.
To help put these findings in a clinical context, a
novel risk scoring system, the Manchester Acute
Coronary Syndromes Rule (MACS), incorporating both
hs-Tn and H-FABP levels, along with clinical features
and EKG was developed and validated [61,62]. A
450 H. GOEL ET AL.

Table 2. Test characteristics of H-FABP versus hs-Tn in those presenting to ED with CP, stratified by time to symptom onset.
First author, Year
Population (N),
prevalence of AMI
Plasma cut-off (ng/ml) Time
to s/s
HFABP hs-Tn hs-Tn þH-FABP
Sens. (NPV) Spec. (PPV) AUC Sens. (NPV) Spec.(PPV) AUC Sens. (NPV) Spec. (PPV)hs-Tn HFABP
Inoue, 2011 [30] (ACS) CP >20 min (432) 0.014 6.2 <2 h 79 (42) 66 (87) 0.70 72 (40) 57 (84) 0.68 NR NR
STEMI ¼52% (hs-TnT) <3 h 73 (41) 61 (87) 0.69 77 (38) 48 (83) 0.67
NSTEMI ¼9% <4 h 76 (43) 73 (93) 0.74 83 (44) 52 (86) 0.72
<6 h 80 (49) 65 (88) 0.74 85 (52) 57 (86) 0.72
Eggers, 2012 [31] (AMI) CP þno STEMI (360) 0.014 5.8 <4 h 28.6 73.5
NSTEMI ¼35.6% (hs-TnT) <8 h 39.1 (73.8) 94.8 (94.8) 0.80 78.9 (86.5) 74.6 (63.1) 0.84 79.7 (86.9) 74.6 (63.4)
Aldous, []2012 [27] (AMI) CP s/o AMI, no STEMI (384) 0.014 6 <4 h 50 (90.7) 89.8 (47.6) NR 90 (97.7) 79.6 (45) NR 90 (97.5) 73.5 (38.6)
NSTEMI ¼15.6% (hs-TnT)
Kitamura, 2013 [32] (AMI) S/S suggestive of AMI (85) 0.014 6.2 <2 h 38 (57) 93 (86) 0.70 25 (40) 57 (40) 0.48 NR NR
NSTEMI ¼12% (hs-TnT) 2–4 h 88 (67) 75 (92) 0.95 100(100) 75 (93) 0.94
STEMI ¼43% >4 h 50 (84) 100 (100) 0.81 100 (100) 81 (67) 0.96
Shortt, 2013 [33] (AMI) Possible ACS s/s <6 h (163) 0.014 5.2 <6 h 43 (90) 80 (24) NR 86 (97) 63 (25) NR 86 (96) 54 (21)
AMI ¼8.6% (hs-TnI)
Shoenenberger, 2014
[34] (AMI)
CP s/o AMI (105) AMI ¼32.4% (hs-TnT) 5.76 <1 h 58.8 (83.3) 98.6 (95.2) 0.84 70.6 (85.9) 85.9 (70.6) 0.88 NR NR
Bank, 2015 [35] (ACS) CP s/o ACS, no STEMI (453) 0.014 (hs-TnT) 7 <3 h 47 (76) 85 (61) 0.73 63 (83) 92 (81) 0.86 69 (84) 83 (68)
3–6 h 68 (89) 79 (50) 0.78 64 (89) 89 (64) 0.86 71(86) 79 (59)
NSTEMI ¼23% >6 h 59 (77) 77 (59) 0.73 87(92) 92 (80) 0.91 87 (91) 72 (64)
Gami, 2015 [36] (AMI) CP <6 h (88) 0.014 (hs-TnT) 5 <6h 85 (90) 88 (82) 0.89 94 (94) 62 (61) 0.8 100 (100) 88.9 (85)
AMI ¼38.6%
Kellens, 2016 [37] (AMI) Typical CP (152) (hs-TnT) 5.3 158 min 54 (52) 84 (85) 0.79 72 (61) 73 (82) 0.83 82 (70) 71(83)
STEMI ¼33% (median)
NSTEMI ¼30%
Agnello, 2017 [38]CP<1-h duration þnormal cTn
(n ¼28 CP þ28 controls)
(hs-TnI) 6.1 <1 hour 55.5 (67) 89.2 (83) 0.65 34 (60.4) 100 (100) 0.80 NR NR
ACS: Acute Coronary Syndrome; AMI: Acute myocardial infarction; CP: Chest pain; NSTEMI: Non ST-Elevation Myocardial Infarction; S/o: Symptoms of; STEMI: ST-Elevation Myocardial Infarction; NR: Not Reported.
Data from Inoue et al. was extracted from graphs using the online graph reader tool (www.graphreader.com).
ANNALS OF MEDICINE 451

recent pilot RCT found that the MACS rule enabled
26% patients to be successfully discharged from the
ED within 4 h with no incident AMI in 30 days among
those discharged [63]. Similarly, an analysis of a single-
center arm of the multi-center ADAPT study found
that when combined with EKG, H-FABP orhs-Tn alone
had unacceptably low sensitivity [64]. However, in
combination H-FABP þhs-Tn þEKG changes maxi-
mised rule-outs (41% testing negative) while main-
taining >99% sensitivity. Other authors have also
demonstrated the benefits of this combination
approach [65].
To summarise, H-FABP may yet prove to be an
important adjunct to hs-Tn, enabling an optimal bal-
ance between sensitivity (and NPV) and specificity
(and PPV). Current evidence suggests that hs-Tn þH-
FABP combination strategy would maximize safe dis-
charges while minimising missed AMIs. Obviously,
adequately powered RCTs examining the optimised
cut-off values and timing in relation to symptom-onset
are needed.
Prognostic value of H-FABP in patients
with ACS
Since H-FABP may be released into the plasma follow-
ing myocardial injury even without myocardial necro-
sis, the prognostic value of H-FABP in those with
suspected ACS has been extensively studied, to iden-
tify cTn-negative patients who may be high risk, and
hence warrant observation or diagnostic workup.
Ishii et al. first reported that in 328 consecutive
patients with ACS (47% STEMI, 26.5% UA/NSTEMI),
serum H-FABP >9.8 ng/mL in first 6 h after CP onset,
but not elevated cTn, was a strong predictor of cardiac
death and non-fatal AMI within 6 months [66]. Several
subsequent reports have consistently demonstrated
superiority of H-FABP over c-Tn, and indeed other bio-
markers in this regard (Table 3). Viswanathan et al.
tested the prognostic value of H-FABP against hs-Tn
for the first time, and in a lower risk population than
prior studies (AMI prevalence 20.8%, STEMI excluded)
[72]. Of note, both were measured in the plasma >
12 h after symptom onset. Even so, H-FABP predicted
death or AMI within 18 months independent of hs-Tn
levels across the entire cohort. More importantly, H-
FABP >6.48 ng/ml strongly predicted 18-month death
and AMI in hs-Tn-negative patients, proving generaliz-
ability of previous findings to a more “real world”
cohort of unselected patients. Hence, H-FABP levels
during the first hours after symptom onset have con-
sistently been proven to identify a high-risk
population, regardless of cTn (or indeed hs-Tn) levels.
RCTs comparing outcomes using treatment/diagnostic
algorithms based on H-FABP, either alone or as part of
a multiple biomarker strategy, are needed to facilitate
adoption in clinical practice.
H-FABP in non-ACS disorders
Given the prognostic ability of H-FABP in ACS (Table
3), its utility to identify high-risk patients in other non-
ACS disorders known to cause myocardial strain, and
perhaps injury in severe cases, has attracted attention.
H-FABP in congestive heart failure (CHF)
Cardiac biomarkers have become integral to the man-
agement of CHF. Established and widely available bio-
markers including brain natriuretic peptide (BNP) and
N-terminal-pro-brain natriuretic peptide (NT-pro-BNP),
are useful in diagnosis (to rule out heart failure)
[73,74], ascertain prognosis [75,76], predict mortality
or re-hospitalization [77–79], and possibly guide early
lifestyle and pharmacologic interventions in asymp-
tomatic at-risk patients [80,81].
There seems to be little added value or improve-
ment with H-FABP over natriuretic peptides in con-
firming the diagnosis of CHF. A post hoc analysis of
the MANPRO cohort reported that H-FABP levels corre-
lated with CHF clinical severity, and with echocardio-
graphic indices of systolic and diastolic dysfunction
[82]. However, though H-FABP plus NT-proBNPhad a
significant improvement in PPV compared to NT-
proBNP alone (58% vs 45%, p<0.0001), it was well
short of being recommended for clinical use. There
was no improvement in NPV over NT-proBNP alone.
Importantly, almost 25% patients in the “no acute
CHF”group had a history of chronic CHF, making it
difficult to interpret findings since H-FABP and NT-
proBNP are known to be raised in those with chronic
stable CHF, each to a variable degree. More recently,
Lichtenauer et al., in a cohort of 124 patients with sys-
tolic CHF (ischaemic and non-ischaemic), showed H-
FABP as having the highest AUC (0.80, 95% CI ¼
0.74–0.86) among several novel inflammatory markers,
though no comparison to natriuretic peptides was per-
formed [83].
Majority of the studies investigating H-FABP in CHF
have focussed on prognostic utility, both in acute
decompensated and chronic stable CHF (Table 4).
Notably, these reports span widely varying popula-
tions in terms of the degree of systolic dysfunction,
aetiology of CHF, and clinical severity as assessed by
452 H. GOEL ET AL.

New York Heart Association Classification (NYHA). As
noted in Table 4, reports have consistently found H-
FABP to be the only, or at the very least, best pre-
dictor of outcomes in those with chronic stable CHF,
when compared to other commonly used biomarkers
[84,90–92]. In the only prognostic study restricted to
HFpEF, Kutsuzawa et al. found H-FABP to be the sole
predictor of CV events among a host of clinical (age,
NYHA class, hypertension, diabetes, renal function)
and biochemical markers (BNP, cTn, hs-CRP) [91].
Physiologically, it may be that since BNP is a surrogate
for myocardial “stretch”engendered by pressure/vol-
ume overload, while H-FABP directly depicts myocar-
dial injury, the latter is a better predictor of adverse
outcomes by virtue of indicating ongoing myocardial
damage. In a small study comparing hs-Tn, H-FABP
and NT-proBNP between 49 patients with HFpEF, 51
patients with asymptomatic diastolic dysfunction, and
30 controls with normal diastolic function, all three
markers were elevated in HFpEF, only hs-Tn and H-
FABP were elevated in asymptomatic diastolic dysfunc-
tion, indicating that subtle myocardial injury precedes
the development of CHF [96].
In acute decompensated CHF (Table 4), H-FABP has
again been found to exceed BNP and NT-proBNP as a
predictor of mortality and readmissions [89]. In fact,
among admission and discharge BNP and H-FABP,
only discharge H-FABP was found to predict cardiac
death and CHF readmission [95]. Hence, the admission
H-FABP, as well as the H-FABP response to therapy in
Table 3. Prognostic utility of H-FABP in patients with ACS.
First Author, Year N
Population (Time of
blood sampling) Follow up Primary outcome Biomarkers Findings
Ishii, 2005 [66] 328 ACS (<6-h after s/s
onset)
6-mo Cardiac death
Cardiac events
H-FABP
c-Tn
H-FABP predicted 6-month
cardiac events (RR 8.92,
1.15–69.4) but not cTn.
Suzuki, 2005 [67] 90 ACS (Admission) 30-d All-cause death
Cardiac event
H-FABP
Troponin T
CK-MB
HFABP predicted cardiac
events at 30-days (RR 44.98,
1.48–1364.88) but not cTn
or CK-MB
O’Donoghue,
2006 [68]
2,287 ACS
(41 ± 20 h after s/s
onset)
10-mo All-cause death
Nonfatal AMI
New/worsening CHF
H-FABP
cTn
BNP
Myoglobin
H-FABP predicteddeath (HR
4.1, 2.6–6.5), CHF (HR 4.5,
2.6–7.8), MI (HR 1.6,
1.0–2.5), or all (HR 2.6,
1.9–3.5)10 months
independent of cTn/BNP. H-
FABP incremental to cTn/
BNP for prognosis.
Kilcullen, 2007 [69] 1,448 ACS (12-24 h after s/s) 12-mo All-cause death H-FABP
cTn
H-FABP 5.8 ng/ml predicted
1-yr mortality (HR 11.35,
2–64.34) in cTn-negative
(UA) patients, as well in
those with "cTn (NSTEMI)
(HR 3.11, 1.45–6.7).
Ilva, 2009 [70] 293 Suspected ACS
(Median 4.7 h after
s/s onset)
6-mo All-cause death
Recurrent MI
cTnI
H-FABP
cTnI independently predicted
6-mo death þAMI (RR 3.02,
1.62–5.63) but not H-FABP.
McCann, 2009 [71] 550 Suspected ACS
(Median 6 h after s/s
onset)
12-mo All-cause death
Recurrent AMI
H-FABP
cTn
NT-pro-BNP
hs-CRP
MPO
MMP-9
others
AmissionH-FABP(OR 2.7,
1.1–6.4), admission NT-pro-
BNP (OR 2.7, 1.4-5.2), and
peak cTn(OR 3.6,
1.4–9.0)independently
predicted 1-yr mortality.
H-FABP, cTn, NT-pro-BNP had
incremental prognostic
value
Viswanathan,
2010 [72]
955 hs-Tn-negative
suspected ACS (NR)
18-mo All-cause death
Recurrent AMI
H-FABP
hs-Tn
H-FABP predicted 18-month
death/MI in hs-Tn (-)
patients incrementally when
stratified by degree of H-
FABP elevation.
Garcia-Valdecasas,
2011 [26]
165 Chest pain (<6 h after
s/s onset)
6-mo All-cause death
Recurrent ACS/AMI
Other cardiac events
cTnI
H-FABP
CK-MB
Increased H-FABP (HR 2.50,
1.31–4.80) and cTnI(2.53,
1.19–5.38) independently
predicted 6 month
outcomes.
Reiter, 2013 [58] 955 Chest pain suggestive
of MI (<12-h after
onset/peak of
symptoms)
12-mo All-cause death H-FABP
Copeptin
hs-Tn
H-FABP predicted 1-yr
mortalityirrespective of
hs-Tn
ANNALS OF MEDICINE 453

acute CHF assessed at the time of discharge, akin to
similar findings with BNP, may indeed identify candi-
dates for aggressive therapy and close follow-up.
In summary, H-FABP may be a robust prognostic
marker, both in chronic stable CHF, in acute decom-
pensated CHF and also HFpEF, and indeed seems
superior to BNP, NT-proBNP, and cTn.
Low and intermediate risk pulmonary embolism
In-hospital and early mortality in acute PE varies
widely depending on the severity at presentation [97].
Those with severe or “massive”PE, as indicated by
hemodynamic instability, are clearly recommended to
receive immediate mechanical or chemical thromboly-
sis [98]. However, optimal management of non-high-
risk, hemodynamically stable patients has been some-
what challenging, since this sub-group itself varies
widely in terms of risk of adverse outcomes. In par-
ticular, identifying those with subclinical right ven-
tricular strain or myocardial injury, a group with an
intermediate mortality risk, has attracted much focus
[99,100].
Recent guidelines recommend using a combin-
ation of clinical assessment, imaging evidence of
right ventricular dysfunction (RVD), and biomarkers
(cTn) to further stratify this group into low, inter-
mediate-low and intermediate-high risk, with differ-
ent management strategies for each group, i.e.
home discharge with anti-coagulation, inpatient
observation, or close monitoring and rescue reperfu-
sion if needed, respectively [98]. The guidelines also
opine that “the optimal, clinically most relevant
combination (and cut-off levels) of clinical and bio-
chemical predictors of early PE-related death remain
to be determined, particularly about identifying
Table 4. Studies investigating prognostic value of H-FABP in acute decompensated CHF and chronic stable CHF.
First Author, Year Population (n) Primary outcome Follow up Biomarker Risk estimate, 95% CI
Setsuta, 2002 [84] Stable chronic CHF (56) All-cause death
CHF readmission
16 ± 12-mo H-FABP HR 2.6, 1.1–6.5 per 3.86ng/ml
increase
cTn HR 7, 1.1–44
BNP NS
ANP NS
CK-MB NS
Arimoto, 2005 [85] Acute CHF (179) Cardiac death
CHF readmission
20-mo H-FABP HR 7.39, p ¼0.0065
LDH NS
CK NS
Niizaki, 2005 [86] Acute CHF (186) Cardiac death
CHF readmission
534 ± 350 days H-FABP HR 5.42, 2.20–13.32
BNP HR 2.41, 1.02–5.73
Komamura, 2006 [87] Chronic stable
non-ischaemic DCM (92)
Cardiac death
Heart transplant
LV assist device
48 months H-FABP RR 7.5, 0.7–36.1
BNP RR 10.9, 3.5–35.3
cTn NS
Niizeki, 2007 [88] Acute CHF (126) Cardiac death
CHF readmission
474 ± 328-days H-FABP HR 15.7, 3.8–64.5
BNP HR 2.6, 0.87–7.8
cTn NS
Niizek, 2008 [89] Acute CHF (113)
(Admission þdischarge)
Cardiac death
CHF readmission
624 ± 299 days H-FABP (at discharge) HR 5.7, 2–9.5
BNP (at discharge) OR 4.62, 1.49–14.33
a
Setsuta, 2008 [90] Chronic stable CHF (103) All-cause death
CHF readmission
28 ± 26 mo H-FABP HR 2.24, 1.21–4.14
cTn HR 1.95, 1.02-3.71
Kutsuzawa, 2012 [91] Chronic CHF with
preserved EF (151)
Cardiac death
CHF readmission
694 (29-2000) days H-FABP HR 1.165, 1.034–1.314
cTn NS
BNP NS
hs-CRP NS
Hoffmann, 2015 [82] Acute CHF (122) All-cause death
CHF readmission
5-yrs H-FABP ACM-NS
CHF readmit-HR 1.07, 1.02-1.13
cTn CHF readmit /ACM-NS
NT-proBNP CHF readmit/ACM-NS
Otaki, 2014 [92] Chronic stable
CHF ± AF (402)
All-cause death
Cardiac death
CHF readmission
643 days-AF
488 days-SR
H-FABP-AF HFABP-SR HR 1.57, 1.2–2
HR 1.28, 1.04–1.58
cTn-AF
cTn-SR
HR 1.4, 1.13, 1.74
NS
BNP-AF/SR NS
Shirakabe, 2015 [93] CHF ± AKI admitted to
ICU (NYHA III/IV)
All-cause death
CHF readmission
90-days H-FABP HR 5.1, 1.86–14.17
hs-Tn NS
BNP NS
Kadowaki, 2017 [94] Acute CHF (322) Cardiac death
CHF readmission
534 (203-1014) days H-FABP HR 1.745, 1.088–2.7903
BNP NS
Kazimierczyk, 2018 [95] Acute NYHA III/IV CHF (71)
Admission þdischarge
CV death
CHF readmission
9.2 ± 7.3-mo H-FABP (at discharge) (OR 1.3, 1.06–1.68)
BNP NS
Unless specified, risk estimates in last column are for composite end-point, and are those achieved after multi-variate analyses, including co-markers
checked in each study.
a
OR for Niizeki, 2008 calculated from reported raw data.
454 H. GOEL ET AL.

possible candidates for reperfusion treatment among
patients with intermediate-risk PE”, hence the
ongoing search for novel markers.
The role of H-FABP in PE was first demonstrated by
Kaczynska et al. in 2006, in a prospective cohort of 77
patients, including 9 with massive, 43 with sub-mas-
sive, and 25 with non-massive PE [101]. Compared to
cTn, NT-pro-BNP, and myoglobin, H-FABP emerged as
the only predictor of 30-day PE-related as well as all-
cause mortality. These findings were quickly replicated
by Puls et al. the following year, in a cohort of 107
patients [102]. Again, H-FABP was superior to cTn or
NT-proBNP even when 24-hour peak levels of the lat-
ter were considered, and had additional prognostic
ability over echocardiographic assessment of RV dys-
function (Table 5) summarizes subsequent reports
investigating the prognostic value of H-FABP alone
and against other markers in sub-massive/normoten-
sive PE, whereby H-FABP appears to be a strong
marker of adverse clinical outcomes in this population.
A meta-analysis of 9 studies including 1680 patients
found that elevated H-FABP levels were associated
with an increased risk of RVD (OR 2.57; 95% CI,
1.05–6.33), complicated clinical course (OR 17.67; 95%
CI, 6.02–51.89), and30-day PE-related mortality (OR,
32.94; 95% CI, 8.80–123.21) [110]. Compared to hs-Tn,
brain natriuretic peptide (BNP), and N-terminal-pro-
BNP (NT-pro-BNP), H-FABP was the strongest predictor
of short-term PE-related and all-cause mortality, and
had the lowest negative likelihood ratio for mortality.
H-FABP was tested as part of the European Society
of Cardiology (ESC) guidelines algorithm for risk-strati-
fying patients with acute PE [109]. In 271 patients
assessed to be low-risk by the simplified PE severity
index (sPESI), 30-day complication rate (death, cat-
echolamine use, mechanical ventilation, resuscitation)
was 1.1%; however, those with an elevated H-FABP
had a 4.3% complication rate, compared with 0.4% for
those with normal H-FABP, thereby achieving signifi-
cantly enhanced precision over clinical assessment
alone. Hence, H-FABP seems to be a promising bio-
marker for risk-stratifying low-intermediate risk
patients with acute PE. From the standpoint of triag-
ing patients for thrombolysis, one small observational
study did not find a difference in 30-day mortality
between H-FABP-positive patients who received
thrombolysis versus those who did not [107], although
interventional RCTs are awaited.
Other conditions
In patients undergoing coronary artery bypass grafting
(CABG), the slow rise and fall of traditional biomarkers
like CK-MB and cTn makes them unsuitable to discrim-
inate between early graft failure, on the one hand,
and the expected myocardial injury resulting from the
surgery itself or ischaemia-reperfusion injury on the
Table 5. Prognostic performance of H-FABP in low-intermediate risk acute PE.
First Author, Year Sample size Primary outcome Biomarkers Findings (Risk estimate, 95%CI)
Boscheri, 2010 [103] 101 All-cause mortality at 6-mo H-FABP
Troponin I
H-FABP alone predicted 30-day PE-related
mortality (OR 37, 5–266).
Dellas, 2010 [104] 126 All-cause mortality at 30-days CPR
Endotracheal intubation
Catecholamine use
H-FABP
cTnT
NT-proBNP
H-FABP alone predicted 30-day composite
outcome (OR 36.6, 4.3–308)
H-FABP alone predicted long term (median
499 days) mortality (HR 4.5, 2.0–9.8)
Gul, 2012 [105] 61 All-cause mortality at 30-days H-FABP
Troponin I
CK-MB
H-FABP alone predicted 30-day mortality (OR
7.27, 1.78–29.7)
Lankeit, 2013 [106] 257 30-day adverse outcome (death,
catecholamine use,
endotracheal intubation, CPR)
H-FABP
cTn
NY-pro-BNP
H-FABP (OR 6.79, 2.4–19.26) stronger predictor
of 30-day adverse outcomes than cTn (OR
3.47, 1.21–9.90), or NT-proBNP (OR 3.79,
1.20–11.95)
Gul, 2014 [107] 80 IHM and 30-day mortality H-FABP
cTn
H-FABP alone predicted in–hospital (HR 6.63,
1.33–33.34) & 30-day mortality (HR 7.81,
1.59–38.34)
Thrombolysis in patients with "H-FABP did
not improve outcomes.
Langer, 2016 [108] 161 All-cause mortality at 30-days H-FABP
CK-MB
Troponin I
HFABP (OR 27.1, 2.1–352.3) stronger predictor
of 30-day mortality than CK-MB (OR 5.3,
1.3–23.3). cTn did not predict outcomes
after adjusting for other variables.
Dellas, 2018 [109] 716 Death, Resuscitation, Intubation or
Catecholamine use in 30 days
H-FABP
sPESI
Multidetector CT
H-FABP had incremental prognostic value in
low risk (sPESI ¼0) and intermediate-risk
(sPESI 1 or RVD on MDCT) patients.
ACM: all cause mortality; CK-MB: creatinine kinase MB; IHM: in-hospital mortality; RVD: right ventricular dysfunction; sPESI: simplified pulmonary embol-
ism severity index; MDCT: multidetector computed tomography; NT-proBNP: N-terminal pro b-type natriuretic protein.
ANNALS OF MEDICINE 455

other. Consistent with its presence in freely soluble
form in the cardiac myocellular cytoplasm, H-FABP has
been repeatedly found to be the earliest marker
(as early as 60–90 min post-operatively) to increase
post-operatively after CABG, even excluding patients
with post-operative AMI [111–113]. Not surprisingly, in
a prospective cohort of 1298 patients undergoing
CABG, H-FABP peaked earlier, and was superior to c-
Tn and CK-MB as a predictor of long-term mortality
and ventricular dysfunction [114]. Hence, H-FABP may
be a marker of high-risk patients and predict the
requirement of closer post-operative monitoring, or
more aggressive application of strategies to reduce
ischaemia-reperfusion injury.
Given known tissue distribution patterns, H-FABP
was also thought to have potential value in diagnosis
and prognostication in neurologic disorders, most
prominently ischaemic stroke and traumatic brain
injury (TBI). In the context of ischaemic stroke, a small
pilot study in 2004 indicated H-FABP may be signifi-
cantly more sensitive and specific than the hitherto
most extensively studied markers, neuron-specific eno-
lase and S100B [115]. H-FABP seems to peak 3h
after symptom onset and remain elevated for up to
5 days, and more importantly, peak H-FABP values cor-
related with the severity of neurological deficit at
10–12 days (r
2
¼0.49), and functional outcomes at
90 days (r
2
¼0.56) [116]. However, the relation between
H-FABP levels and infarct volume was non-linear, with
markedly elevated levels of H-FABP restricted to those
with an infarct volume of >150 ml [116]. Subsequent
small cohorts, as well as a very recent meta-analysis
indicate that though H-FABP as a single marker has
modest diagnostic and prognostic value in acute
ischaemic stroke, it falls short of clinical applicability,
though it may add value as part of a biomarker panel
[117–119]. Whether H-FABP alone, or as part of a bio-
marker panel, has value identifying late-presenting
stroke patients who might benefit from thrombolysis
remains to be investigated.
The major role of biomarkers in traumatic brain
injury (TBI) pertains to their role in improving initial
triage in those with clinically mild TBI in order to
reduce the need for costly and potentially harmful
(radiation exposure) neuroimaging. This is especially
important since the incidence of clinically significant
imaging abnormalities in this sub-group is quite low,
and computed tomography (CT) imaging is overused
in this context [120,121]. Hence, the ideal biomarker in
this case should have a very high NPV, in order to reli-
ably mitigate the need of CT imaging in clinically mild,
low-risk TBI. A screening study examining 87
biomarkers in 110 patients with clinically mild TBI
found a predictive model with 6 of the markers
including H-FABP to have an NPV of 98.6%, though a
PPV of just 60% [122]. In a larger cohort of 261
patients with mild TBI, both H-FABP and S100B dis-
played high sensitivity and NPV, but the former had a
higher specificity (6% vs 29% with sensitivity set at
100%), though the improvement in specificity hardly
made H-FABP a clinically usable positive predictor of
CT findings [123]. Most recently a panel of 8 bio-
markers in TBI of all severities identified H-FABP com-
bined with two other markers to constitute the best
biomarker panel in terms of sensitivity to predict CT
abnormalities [124]. Sensitivity, specificity, and predict-
ive values of all markers individually were found sub-
par for clinical use. Hence, H-FABP’s role in predicting
CT-negativity in those with mild TBI seems best suited
as part of a panel of biomarkers, a field that remains
rapidly evolving, given the large number of potential
markers being investigated.
More recently, myocardial injury, as defined by an
elevated cTn, has been found to predict severe cor-
onavirus disease 2019 (COVID-19) in some reports,
raising speculation as to the high incidence of myocar-
ditis in these patients [125,126]. Intriguingly, though
hardly surprising, a recent small cohort reported sig-
nificantly higher H-FABP levels in those with severe
versus non-severe COVID-19 infection [127].
Conclusion
To summarise, H-FABP remains a biomarker of high
interest, even in the era of highly sensitive troponin
assays, particularly in the context of ruling out AMI in
low-risk early presenters, allowing earlier discharge of
such patients from the ED and reducing cost. Lack of
specificity, as has been seen with other biomarkers
obviously makes it unsuitable for confirming AMI,
especially as the sole marker. Lack of data, lack of pro-
gress in improving assay kits, iii) paucity of studies
incorporating H-FABP with or without cTn or hs-Tn as
part of clinical decision pathways.
By virtue of the fact, it is elevated in many cardio-
vascular conditions, helps identify patients at higher
risk of complications, similar to the hs-cTn. Enough
evidence now exists to directly investigate outcome-
oriented clinical decision-making algorithms incorpo-
rating H-FABP, for example, whether these patients
warrant further inpatient observation or testing. Its
role in other non-cardiac conditions are also being
appreciated. Improvement in assays and more studies
would help make clinicians more aware of its utility
456 H. GOEL ET AL.

and perhaps it would find its rightful place in manage-
ment algorithms.
Disclosure statement
No potential conflict of interest was reported by the
author(s).
References
[1] Centers for Disease Control and Prevention,
Ambulatory and Hospital Care Statistics Branch.
National Hospital Ambulatory Medical CareSurvey:
2016 emergency department summary tables.
Available from: https://www.cdc.gov/nchs/data/
nhamcs/web_tables/2016_ed_web_tables.pdf.
[2] Schull MJ, Vermeulen MJ, Stukel TA. The risk of
missed diagnosis of acute myocardial infarction asso-
ciated with emergency department volume. Ann
Emerg Med. 2006;48(6):647–655.
[3] Moy E, Barrett M, Coffey R, et al. Missed diagnoses
of acute myocardial infarction in the emergency
department: variation by patient and facility charac-
teristics. Diagnosis (Berl)). 2015;2(1):29–40.
[4] Pope HJ, Aufderheide TP, Ruthazer R, et al. Missed
diagnoses of acute cardiac ischemia in the emer-
gency department. N Engl J Med. 2000;342(16):
1163–1170.
[5] Hsia RY, Hale Z, Tabas JA. A national study of the
prevalence of life-threatening diagnoses in patients
with chest pain. JAMA Intern Med. 2016;176(7):
1029–1032.
[6] World Health Organization. Technical report series
no. 168. Hypertension and coronary heart disease:
Classification and criteria for epidemiological studies.
[cited 2020 May 25]. Availabel from: http://whqlib-
doc.who.int/ trs/WHO_TRS_168.pdf. Published 1959.
[7] Alpert JS, Thygesen K, Antman E, et al. Myocardial
infarction redefined –a consensus document of The
Joint European Society of Cardiology/American
College of Cardiology Committee f or the redefin-
ition of myocardial infarction. J Am Coll Cardiol.
2000;36(3):959–969.
[8] Tucker JF, Collins RA, Anderson AJ, et al. Early diag-
nostic efficiency of cardiac troponin I and troponin T
for acute myocardial infarction. Acad Emerg Med.
1997;4(1):13–21.
[9] Aydin S, Ugur K, Aydin S, et al. Biomarkers in acute
myocardial infarction: current perspectives. Vasc
Health Risk Manag. 2019;15:1–10.
[10] Wang J, Tan GJ, Han LN, et al. Novel biomarkers for
cardiovascular risk prediction. J Geriatr Cardiol. 2017;
14(2):135–150.
[11] Glatz JFC, Van Nieuwenhoven FA, Luiken JJFP, et al.
Role of membrane-associated and cytoplasmic fatty
acid-binding proteins in cellular fatty acid metabol-
ism. Prostaglandins Leukot Essent Fatty Acids. 1997;
57(4–5):373–378.
[12] Schaap FG, Van der Vusse GJ, Glatz JFC. Fatty acid-
binding proteins in the heart. Mol Cell Biochem.
1998;180(1-2):43–51.
[13] Watanabe K, Wakabayashi H, Veerkamp JH, et al.
Immunohistochemical distribution of heart-type fatty
acid-binding protein immunoreactivity in normal
human tissues and in acute myocardial infarct. J
Pathol. 1993;170(1):59–65.
[14] Zschiesche-Appie Kleine WH, Spitzer Jacques
Veerkamp-Jan FC, Glatz EH, et al. Histochemical
localization of heart-type fatty-acid binding protein
in human and murine tissues. Histochem Cell Biol.
1995;103(2):147–156.
[15] Haastrup B, Gill S, Risom Kristensen S, et al.
Biochemical markers of ischaemia for the early iden-
tification of acute myocardial infarction without ST
segment elevation. Vol 94.; 2000. Available from:
www.karger.com.
[16] Seino Y, Ogata K. i, Takano T, et al. Use of a whole
blood rapid panel test for heart-type fatty acid-bind-
ing protein in patients with acute chest pain:
Comparison with rapid troponin T and myoglobin
tests. Am J Med. 2003;115(3):185–190.
[17] Seino Y, Tomita Y, Takano T, et al. Office
Cardiologists Cooperative Study on whole blood
rapid panel tests in patients with suspicious acute
myocardial infarction: comparison between heart-
type fatty acid-binding protein and troponin T tests.
Circ J. 2004;68(2):144–148.
[18] Ruzgar O, Bilge AK, Bugra Z, et al. The use of human
heart-type fatty acid-binding protein as an early
diagnostic biochemical marker of myocardial necro-
sis in patients with acute coronary syndrome, and its
comparison with troponin-T and creatine kinase-
myocardial band. Heart Vessels. 2006;21(5):309–314.
[19] Cavus U, Coskun F, Yavuz B, et al. Heart-type, fatty-
acid binding protein can be a diagnostic marker in
acute coronary syndromes. J Natl Med Assoc. 2006;
98(7):1067–1070.
[20] McCann CJ, Glover BM, Menown IBA, et al. Novel
biomarkers in early diagnosis of acute myocardial
infarction compared with cardiac troponin T. Eur
Heart J. 2008;29(23):2843–2850.
[21] Valle HA, Riesgo LGC, Bel MS, et al. Clinical assess-
ment of heart-type fatty acid binding protein in
early diagnosis of acute coronary syndrome. Eur J
Emerg Med. 2008;15(3):140–144.
[22] Orak M, Ust€
unda
gM,G
€
ulo
glu C, et al. The role of
the heart-type fatty acid binding protein in the early
diagnosis of acute coronary syndrome and its com-
parison with troponin i and creatine kinase-MB iso-
form. Am J Emerg Med. 2010;28(8):891–896.
[23] Haltern G, Peiniger S, Bufe A, et al. Comparison of
usefulness of heart-type fatty acid binding protein
versus cardiac troponin T for diagnosis of acute
myocardial infarction. Am J Cardiol. 2010;105(1):1–9.
[24] Kim Y, Kim H, Kim SY, et al. Automated heart-type
fatty acid-binding protein assay for the early diagno-
sis of acute myocardial infarction. Am J Clin Pathol.
2010;134(1):157–162.
[25] McMahon CG, Lamont JV, Curtin E, et al. Diagnostic
accuracy of heart-type fatty acid-binding protein for
ANNALS OF MEDICINE 457

the early diagnosis of acute myocardial infarction.
Am J Emerg Med. 2012;30(2):267–274.
[26] Garcia-Valdecasas S, Ruiz-Alvarez MJ, De Tena JG,
et al. Diagnostic and prognostic value of heart-type
fatty acid-binding protein in the early hours of acute
myocardial infarction. Acta Cardiol. 2011;66(3):
315–321.
[27] Aldous S, Pemberton C, Troughton R, et al. Heart
fatty acid binding protein and myoglobin do not
improve early rule out of acute myocardial infarction
when highly sensitive troponin assays are used.
Resuscitation. 2012;83(2):e27–e28.
[28] Gerede DM, G€
ulec S, Kilic¸kap M, et al. Comparison of
a qualitative measurement of heart-type fatty acid-
binding protein with other cardiac markers as an
early diagnostic marker in the diagnosis of non-ST-
segment elevation myocardial infarction. CVJA. 2015;
26(6):204–209.
[29] Vupputuri A, Sekhar S, Krishnan S, et al. Heart-type
fatty acid-binding protein (H-FABP) as an early diag-
nostic biomarker in patients with acute chest pain.
Indian Heart J. 2015;67(6):538–542.
[30] Inoue K, Suwa S, Ohta H, et al. Heart fatty acid-bind-
ing protein offers similar diagnostic performance to
high-sensitivity troponin T in emergency room
patients presenting with chest pain. Circ J. 2011;
75(12):2813–2820.
[31] Eggers KM, Venge P, Lindahl B. High-sensitive car-
diac troponin T outperforms novel diagnostic bio-
markers in patients with acute chest pain. Clin Chim
Acta. 2012;413(13-14):1135–1140.
[32] Kitamura M, Hata N, Takayama T, et al. High-sensitiv-
ity cardiac troponin T for earlier diagnosis of acute
myocardial infarction in patients with initially nega-
tive troponin T test-comparison between cardiac
markers. J Cardiol. 2013;62(6):336–342.
[33] Shortt CR, Worster A, Hill SA, et al. Comparison of
hs-cTnI, hs-cTnT, hFABP and GPBB for identifying
early adverse cardiac events in patients presenting
within six hours of chest pain-onset. Clin Chim Acta.
2013;419:39–41.
[34] Schoenenberger AW, Stallone F, Walz B, et al.
Incremental value of heart-type fatty acid-binding
protein in suspected acute myocardial infarction
early after symptom onset. Eur Heart J Acute
Cardiovasc Care. 2016;5(2):185–192.
[35] Bank IE, Dekker MS, Hoes AW, et al. Suspected acute
coronary syndrome in the emergency room: Limited
added value of heart type fatty acid binding protein
point of care or ELISA tests: The FAME-ER (Fatty Acid
binding protein in Myocardial infarction Evaluation
in the Emergency Room) study. Eur Hear Journal
Acute Cardiovasc Care. 2016;5(4):364–374.
[36] Gami BN, Patel DS, Haridas N, et al. Utility of heart-
type fatty acid binding protein as a new biochemical
marker for the early diagnosis of acute coronary syn-
drome. J Clin Diagn Res. 2015;9(1):BC22–BC24.
[37] Kellens S, Verbrugge FH, Vanmechelen M, et al.
Point-of-care heart-type fatty acid binding protein
versus high-sensitivity troponin T testing in emer-
gency patients at high risk for acute coronary
syndrome. Eur Heart J Acute Cardiovasc Care. 2016;
5(2):177–184.
[38] Agnello L, Bivona G, Novo G, et al. Heart-type fatty
acid binding protein is a sensitive biomarker for
early AMI detection in troponin negative patients: a
pilot study. Scand J Clin Lab Invest. 2017;77(6):
428–432.
[39] Bertinchant JP, Larue C, Pernel I, et al. Release kinet-
ics of serum cardiac troponin I in ischemic myocar-
dial injury. Clin Biochem. 1996;29(6):587–594.
[40] Kleine AH, Glatz JFC, Van Nieuwenhoven FA, et al.
Release of heart fatty acid-binding protein into
plasma after acute myocardial infarction in man. Mol
Cell Biochem. 1992;116(1–2):155–162.
[41] Glatz JFC, Renneberg R. Added value of H-FABP as
plasma biomarker for the early evaluation of sus-
pected acute coronary syndrome. Clin Lipidol. 2014;
9(2):205–220.
[42] Glatz JFC, van Bilsen M, Paulussen RJA, et al. Release
of fatty acid-binding protein from isolated rat heart
subjected to ischemia and reperfusion or to the cal-
cium paradox. Biochim Biophys Acta (BBA)/Lipids
Lipid Metab. 1988;961(1):148–152.
[43] Tanaka T, Hirota Y, Sohmiya K-I, et al. Serum and
urinary human heart fatty acid-binding protein in
acute myocardial infarction. Clin Biochem. 1991;
24(2):195–201.
[44] Tsuji R, Tanaka T, Sohmiya K, et al. Human heart-
type cytoplasmic fatty acid-binding protein in serum
and urine during hyperacute myocardial infarction.
Int J Cardiol. 1993;41(3):209–217.
[45] Glatz JFC, Van Der Vusse GJ, Simoons ML, et al.
Fatty acid-binding protein and the early detection of
acute myocardial infarction. Clin Chim Acta. 1998;
272(1):87–92.
[46] Bruins Slot MHE, Reitsma JB, Rutten FH, et al. Heart-
type fatty acid-binding protein in the early diagnosis
of acute myocardial infarction: a systematic review
and meta-analysis. Heart. 2010;96(24):1957–1963.
[47] Carroll C, Khalaf MA, Stevens JW, et al. Heart-type
fatty acid binding protein as an early marker for
myocardial infarction: systematic review and meta-
analysis. Emerg Med J. 2013;30(4):280–286.
[48] Lippi G, Mattiuzzi C, Cervellin G. Critical review and
meta-analysis on the combination of heart-type fatty
acid binding protein (H-FABP) and troponin for early
diagnosis of acute myocardial infarction. Clin
Biochem. 2013;46(1–2):26–30.
[49] Trevethan R. Sensitivity, specificity, and predictive
values: foundations, pliabilities, and pitfalls in
research and practice. Front Public Heal. 2017;
5(November):1–7.
[50] Collinson P, Gaze D, Goodacre S. Comparison of con-
temporary troponin assays with the novel bio-
markers, heart fatty acid binding protein and
copeptin, for the early confirmation or exclusion of
myocardial infarction in patients presenting to the
emergency department with chest pain. Heart. 2014;
100(2):140–145.
[51] MacGougan CK, Christenson JM, Innes GD, et al.
Emergency physicians’attitudes toward a clinical
prediction rule for the identification and
458 H. GOEL ET AL.

earlydischarge of low risk patients with chest dis-
comfort. Can J Emerg Med. 2001;3(02):89–94.
[52] Body R, McDowell G, Carley S, et al. A FABP-ulous
’rule out’strategy? Heart fatty acid binding protein
and troponin for rapid exclusion of acute myocardial
infarction. Resuscitation. 2011;82(8):1041–1046.
[53] Pelsers MMAL, Hermens WT, Glatz JFC. Fatty acid-
binding proteins as plasma markers of tissue injury.
Clin Chim Acta. 2005;352(1–2):15–35.
[54] Wu AHB. Release of cardiac troponin from healthy
and damaged myocardium. Front Lab Med. 2017;
1(3):144–150
[55] Carlton E, Greenslade J, Cullen L, et al. Evaluation of
high-sensitivity cardiac troponin I levels in patients
with suspected acute coronary syndrome. JAMA
Cardiol. 2016;1(4):405–412.
[56] Keller T, Zeller T, Ojeda F, et al. Serial changes in
highly sensitive troponin I assay and early diagnosis
of myocardial infarction. JAMA. 2011;306(24):
2684–2693.
[57] Rubini Gim
enez M, Hoeller R, Reichlin T, et al. Rapid
rule out of acute myocardial infarction using
undetectable levels of high-sensitivity cardiac tropo-
nin. Int J Cardiol. 2013;168(4):3896–3901.
[58] Reiter M, Twerenbold R, Reichlin T, et al. Heart-type
fatty acid-binding protein in the early diagnosis of
acute myocardial infarction. Heart. 2013;99(10):
708–714.
[59] Liou K, Ho S, Ooi SY. Heart-type fatty acid binding
protein in early diagnosis of myocardial infarction in
the era of high-sensitivity troponin: a systematic
review and meta-analysis. Ann Clin Biochem. 2015;
52(3):370–381.
[60] Xu LQ, Yang YM, Tong H, et al. Early diagnostic per-
formance of heart-type fatty acid binding protein in
suspected acute myocardial infarction: evidence
from a meta-analysis of contemporary studies. Hear
Lung Circ. 2018;27(4):503–512.
[61] Body R, Carley S, McDowell G, et al. The Manchester
Acute Coronary Syndromes (MACS) decision rule for
suspected cardiac chest pain: derivation and external
validation. Heart. 2014;100(18):1462–1468.
[62] Carlton E, Body R, Greaves K. External validation of
the Manchester Acute Coronary Syndromes decision
rule. Acad Emerg Med. 2016;23(2):136–143.
[63] Body R, Boachie C, McConnachie A, et al. Feasibility
of the Manchester Acute Coronary Syndromes
(MACS) decision rule to safely reduce unnecessary
hospital admissions: A pilot randomised controlled
trial. Emerg Med J. 2017;34(9):586–592.
[64] Young JM, Pickering JW, George PM, et al. Heart
fatty acid binding protein and cardiac troponin:
development of an optimal rule-out strategy for
acute myocardial infarction. BMC Emerg Med. 2016;
16(1):34.
[65] Van Hise CB, Greenslade JH, Parsonage W, et al.
External validation of heart-type fatty acid binding
protein, high-sensitivity cardiac troponin, and elec-
trocardiography as rule-out for acute myocardial
infarction. Clin Biochem. 2018;52:161–163.
[66] Ishii J, Ozaki Y, Lu J, et al. Prognostic value of serum
concentration of heart-type fatty acid-binding
protein relative to cardiac troponin T on admission
in the early hours of acute coronary syndrome. Clin
Chem. 2005;51(8):1397–1404.
[67] Suzuki M, Hori S, Noma S, et al. Prognostic value of
a qualitative test for heart-type fatty acid-binding
protein in patients with acute coronary syndrome.
Int Heart J. 2005;46(4):601–606.
[68] O’Donoghue M, de Lemos JA, Morrow DA, et al.
Prognostic utility of heart-type fatty acid binding
protein in patients with acute coronary syndromes.
Circulation. 2006;114(6):550–557.
[69] Kilcullen N, Viswanathan K, Das R, et al. Heart-type
fatty acid-binding protein predicts long-term mortal-
ity after acute coronary syndrome and identifies
high-risk patients across the range of troponin val-
ues. J Am Coll Cardiol. 2007;50(21):2061–2067.
[70] Ilva T, Lund J, Porela P, et al. Early markers of myo-
cardial injury: cTnI is enough. Clin Chim Acta. 2009;
400(1-2):82–85.
[71] McCann CJ, Glover BM, Menown IBA, et al.
Prognostic value of a multimarker approach for
patients presenting to hospital with acute chest
pain. Am J Cardiol. 2009;103(1):22–28.
[72] Viswanathan K, Kilcullen N, Morrell C, et al. Heart-
type fatty acid-binding protein predicts long-term
mortality and re-infarction in consecutive patients
with suspected acute coronary syndrome who are
troponin-negative. J Am Coll Cardiol. 2010;55(23):
2590–2598.
[73] Ponikowski P, Voors AA, Anker SD, et al. 2016 ESC
Guidelines for the diagnosis and treatment of acute
and chronic heart failure: The Task Force for the
diagnosis and treatment of acute and chronic heart
failure of the European Society of Cardiology
(ESC)Developed with the special contribution of the
Heart Failure Association (HFA) of the ESC. Eur Heart
J. 2016 ;37(27):2129–2200.
[74] Yancy CW, Jessup M, Bozkurt B, et al. 2017 ACC/
AHA/HFSA Focused Update of the 2013 ACCF/AHA
guideline for the management of heart failure: a
report of the American College of Cardiology/
American Heart Association Task Force on Clinical
Practice Guidelines and the Heart Failure Society of
America. J Am Coll Cardiol. 2017;70(6):776–803.
[75] Anand IS, Fisher LD, Chiang YT, et al. Changes in
brain natriuretic peptide and norepinephrine over
time and mortality and morbidity in the Valsartan
Heart Failure Trial (Val-HeFT). Circulation. 2003;
107(9):1278–1283.
[76] Berger R, Huelsman M, Strecker K, et al. B-type natri-
uretic peptide predicts sudden death in patients
with chronic heart failure. Circulation. 2002;105(20):
2392–2397.
[77] Kociol RD, Horton JR, Fonarow GC, et al. Admission,
discharge, or change in B-type natriuretic peptide
and long-term outcomes: data from Organized
Program to Initiate Lifesaving Treatment in
Hospitalized Patients with Heart Failure (OPTIMIZE-
HF) linked to Medicare claims. Circ Heart Fail. 2011;
4(5):628–636.
[78] Logeart D, Thabut G, Jourdain P, et al. Predischarge
B-type natriuretic peptide assay for identifying
ANNALS OF MEDICINE 459

patients at high risk of re-admission after decom-
pensated heart failure. J Am Coll Cardiol. 2004;43(4):
635–641.
[79] Verdiani V, Ognibene A, Rutili MS, et al. NT-ProBNP
reduction percentage during hospital stay predicts
long-term mortality and readmission in heart failure
patients. J Cardiovasc Med. 2008;9:694–699.
[80] Huelsmann M, Neuhold S, Resl M, et al. PONTIAC
(NT-proBNP selected prevention of cardiac eveNts in
a population of diabetic patients without a history
of cardiac disease): a prospective randomized con-
trolled trial. J Am Coll Cardiol. 2013;62(15):
1365–1372.
[81] Ledwidge M, Gallagher J, Conlon C, et al. Natriuretic
peptide–based screening and collaborative care for
heart failure: the STOP-HF randomized trial. Vol 310;
2013. Available from: https://jamanetwork.com/.
[82] Hoffmann U, Espeter F, Weiß C, et al. Ischemic bio-
marker heart-type fatty acid binding protein (hFABP)
in acute heart failure –diagnostic and prognostic
insights compared to NT-proBNP and troponin I.
BMC Cardiovasc Disord. 2015;15(1):50.
[83] Lichtenauer M, Jirak P, Wernly B, et al. A compara-
tive analysis of novel cardiovascular biomarkers in
patients with chronic heart failure. Eur J Intern Med.
2017;44:31–38.
[84] Setsuta K, Seino Y, Ogawa T, et al. Use of cytosolic
and myofibril markers in the detection of ongoing
myocardial damage in patients with chronic heart
failure. Am Jourmal Med. 2002;113(9):717–722.
[85] Arimoto T, Takeishi Y, Shiga R, et al. Prognostic value
of elevated circulating heart-type fatty acid binding
protein in patients with congestive heart failure. J
Card Fail. 2005;11(1):56–60.
[86] Niizeki T, Takeishi Y, Arimoto T, et al. Combination
of heart-type fatty acid binding protein and brain
natriuretic peptide can reliably risk stratify patients
hospitalized for chronic heart failure. Circ J. 2005;69:
922–927.
[87] Komamura K, Sasaki T, Hanatani A, et al. Heart-type
fatty acid binding protein is a novel prognostic
marker in patients with non-ischaemic dilated car-
diomyopathy. Heart. 2006;92(5):615–618.
[88] Niizeki T, Takeishi Y, Arimoto T, et al. Heart-type
fatty acid-binding protein is more sensitive than
troponin t to detect the ongoing myocardial dam-
age in chronic heart failure patients. J Card Fail.
2007;13(2):120–127.
[89] Niizeki T, Takeishi Y, Arimoto T, et al. Persistently
increased serum concentration of heart-type fatty
acid-binding protein predicts adverse clinical out-
comes in patients with chronic heart failure. Circ J.
2008;72(1):109–114.
[90] Setsuta K, Seino Y, Kitahara Y, et al. Elevated levels
of both cardiomyocyte membrane and myofibril
damage markers predict adverse outcomes in
patients with chronic heart failure. Circ J. 2007;72(4):
569–574.
[91] Kutsuzawa D, Arimoto T, Watanabe T, et al. Ongoing
myocardial damage in patients with heart failure
and preserved ejection fraction. J Cardiol. 2012;60(6):
454–461.
[92] Otaki Y, Arimoto T, Takahashi H, et al. Prognostic
value of myocardial damage markers in patients
with chronic heart failure with atrial fibrillation.
Intern Med. 2014;53(7):661–668.
[93] Shirakabe A, Hata N, Kobayashi N, et al. Serum
heart-type fatty acid-binding protein level can be
used to detect acute kidney injury on admission and
predict an adverse outcome in patients with acute
heart failure. Circ J. 2015;79(1):119–128.
[94] Kadowaki S, Watanabe T, Otaki Y, et al. Combined
assessment of myocardial damage and electrical dis-
turbance in chronic heart failure. World J Cardiol.
2017;9(5):396–469.
[95] Kazimierczyk E, Kazimierczyk R, Harasim-Symbor E,
et al. Persistently elevated plasma heart-type fatty
acid binding protein concentration is related with
poor outcome in acute decompensated heart failure
patients. Clin Chim Acta. 2018;487:48–53.
[96] Dinh W, Nickl W, F€
uth R, et al. High sensitive tropo-
nin T and heart fatty acid binding protein: Novel
biomarker in heart failure with normal ejection frac-
tion?: a cross-sectional study. BMC Cardiovasc
Disord. 2011;11(1):41.
[97] Kasper W, Konstantinides S, Geibel A, et al.
Management strategies and determinants of out-
come in acute major pulmonary embolism: results of
a multicenter registry J Am Coll Cardiol
1997;30:165–171.
[98] Konstantinides SV, Meyer G, Becattini C, et al. ESC
Guidelines for the diagnosis and management of
acute pulmonary embolism developed in collabor-
ation with the European Respiratory Society (ERS).
Eur Heart J. 2020;41(4):543-603.
[99] Barco S, Mahmoudpor SH, Planquette B, et al.
Assessing the severity of acute pulmonary embolism:
back to the future?. Eur Heart J. 2019;40(11):
911–913.
[100] Kucher N, Rossi E, De Rosa M, et al. Prognostic role
of echocardiography among patients with acute pul-
monary embolism and a systolic arterial pressure of
90 Mm Hg or higher. Arch Intern Med.
2005;165:1777–1781.
[101] Kaczy~
nska A, Pelsers MMAL, Bochowicz A, et al.
Plasma heart-type fatty acid binding protein is
superior to troponin and myoglobin for rapid risk
stratification in acute pulmonary embolism. Clin
Chim Acta. 2006;371(1–2):117–123.
[102] Puls M, Dellas C, Lankeit M, et al. Heart-type fatty
acid-binding proteins (H-FABP): A reliable tool for
initial risk stratification of pulmonary embolism?. Eur
Heart J. 2007;28(2):146–147.
[103] Boscheri A, Wunderlich C, Langer M, et al.
Correlation of heart-type fatty acid-binding protein
with mortality and echocardiographic data in
patients with pulmonary embolism at intermediate
risk. Am Heart J. 2010;160(2):294–300.
[104] Dellas C, Puls M, Lankeit M, et al. Elevated heart-
type fatty acid-binding protein levels on admission
predict an adverse outcome in normotensive
patients with acute pulmonary embolism. J Am Coll
Cardiol. 2010;55(19):2150–2157.
460 H. GOEL ET AL.

[105] G€
ul EE, Can I, G€
uler I, et al. Association of pulmonary
artery obstruction index with elevated heart-type
fatty acid binding protein and short-term mortality
in patients with pulmonary embolism at intermedi-
ate risk. Diagn Interv Radiol. 2012;18(6):531–536.
[106] Lankeit M, Dellas C, Benz V, et al. The predictive
value of heart-type fatty acid-binding protein is
independent from symptom duration in normoten-
sive patients with pulmonary embolism. Thromb
Res. 2013;132(5):543–547.
[107] Gul EE, Can I, Kayrak M, et al. Thrombolysis in
patients with pulmonary embolism and elevated
heart-type fatty acid-binding protein levels. J
Thromb Thrombolysis. 2014;37(4):483–489.
[108] Langer M, Forkmann M, Richter U, et al. Heart-type
fatty acid-binding protein and myocardial creatine
kinase enable rapid risk stratification in normoten-
sive patients with pulmonary embolism. J Crit Care.
2016;35:174–179.
[109] Dellas C, Lobo JL, Rivas A, et al. Risk stratification of
acute pulmonary embolism based on clinical param-
eters, H-FABP and multidetector CT. Int J Cardiol.
2018;265:223–228.
[110] Bajaj A, Rathor P, Sehgal V, et al. Risk stratification in
acute pulmonary embolism with heart-type fatty
acid-binding protein: a meta-analysis. J Crit Care.
2015;30(5):1151.e1–e7.
[111] Fransen EJ, Maessen JG, Hermens WT, et al.
Demonstration of ischemia-reperfusion injury separ-
ate from postoperative infarction in coronary artery
bypass graft patients. Ann Thorac Surg. 1998;65:
48–53.
[112] Thielmann M, Pasa S, Holst T, et al. Heart-type fatty
acid binding protein and ischemia-modified albumin
for detection of myocardial infarction after coronary
artery bypass graft surgery. Ann Thorac Surg. 2017;
104(1):130–137.
[113] Hayashida N, Chihara S, Akasu K, et al. Plasma and
urinary levels of heart fatty acid-binding protein in
patients undergoing cardiac surgery. Jpn Circ J.
2000;64(1):18–20.
[114] Muehlschlegel JD, Perry TE, Liu KY, et al. Heart-type
fatty acid binding protein is an independent pre-
dictor of death and ventricular dysfunction after cor-
onary artery bypass graft surgery. Anesth Analg.
2010;111(5):1101–1109.
[115] Zimmermann-Ivol CG, Burkhard PR, Le Floch-Rohr J,
et al. Fatty acid binding protein as a serum marker
for the early diagnosis of stroke: a pilot study. Mol
Cell Proteomics. 2004;3(1):66–72.
[116] Wunderlich MT, Hanhoff T, Goertler M, et al. Release
of brain-type and heart-type fatty acid-binding pro-
teins in serum after acute ischaemic stroke. J Neurol.
2005;252(6):718–724.
[117] Park SY, Kim MH, Kim OJ, et al. Plasma heart-type
fatty acid binding protein level in acute ischemic
stroke: Comparative analysis with plasma S100B level
for diagnosis of stroke and prediction of long-term
clinical outcome. Clin Neurol Neurosurg. 2013;115(4):
405–410.
[118] Park SY, Kim J, Kim OJ, et al. Predictive value of cir-
culating interleukin-6 and heart-type fatty acid bind-
ing protein for three months clinical outcome in
acute cerebral infarction: multiple blood markers
profiling study. Crit Care. 2013;17(2):R45.
[119] Dolmans LS, Rutten FH, Koenen NCT, et al.
Candidate biomarkers for the diagnosis of transient
ischemic attack: a systematic review. Cerebrovasc
Dis. 2019;47(5–6):207–216.
[120] Borg J, Holm L, Cassidy JD, et al. Diagnostic proce-
dures in mild traumatic brain injury: results of the
WHO Collaborating Centre Task Force on Mild
Traumatic Brain Injury. J Rehabil Med Suppl. 2004;
36:61–75.
[121] Melnick ER, Szlezak CM, Bentley SK, et al. CT overuse
for mild traumatic brain injury. Jt Comm J Qual
Patient Saf. 2012;38(11):483–489.
[122] Sharma R, Rosenberg A, Bennett ER, et al. A blood-
based biomarker panel to risk-stratify mild traumatic
brain injury. PLOS One. 2017;12(3):e0173798.
[123] Lagerstedt L, Egea-Guerrero JJ, Bustamante A, et al.
H-FABP: a new biomarker to differentiate between
CT-positive and CT-negative patients with mild trau-
matic brain injury. PLOS One. 2017;12(4):e0175572.
[124] Posti JP, Takala RSK, Lagerstedt L, et al. Correlation
of blood biomarkers and biomarker panels with trau-
matic findings on computed tomography after trau-
matic brain injury. J Neurotrauma. 2019;36(14):
2178–2189.
[125] Du RH, Liang LR, Yang CQ, et al. Predictors of mor-
tality for patients with COVID-19 pneumonia caused
by SARSCoV-2: a prospective cohort study. Eur
Respir J. 2020;55(5):2000524.
[126] Shi S, Qin M, Shen B, et al. Association of cardiac
injury with mortality in hospitalized patients with
COVID-19 in Wuhan. JAMA Cardiol. 2020;5(7):802.
[127] Yin L, Mou H, Shao J, et al. Correlation between
heart fatty acid binding protein and severe COVID-
19: a case-control study. PLOS One. 2020;15(4):
e0231687.
ANNALS OF MEDICINE 461