Forensic Science International 233 (2013) 154–157

Contents lists available at ScienceDirect

Forensic Science International

journal homepage: www.elsevier.com/locate/forsciint

Cardiac troponin T in forensic autopsy cases

S. Remmer a,b,*, A. Kuudeberg a, M. Tõnisson a, D. Lepik a, M. Väli a,b
a
Institute of Pathological Anatomy and Forensic Medicine, University of Tartu, Ravila 19, 50411 Tartu, Estonia
b
Estonian Forensic Science Institute, Tervise 30, 13419 Tallinn, Estonia

A R T I C L E I N F O A B S T R A C T

Article history: The aim of the present study was to describe the findings of postmortem serum and pericardial fluid (PF)
Received 2 May 2013 cardiac troponin T (cTnT) in various causes of death with regard to the postmortem interval (PMI) and
Received in revised form 4 September 2013 comorbid cardiovascular disease, using 101 autopsy cases with PMI of 8–141 h divided into 9 groups:
Accepted 6 September 2013
cardiovascular disease (CVD), other diseases (OD), poisoning (P), asphyxia (A), drowning (D),
Available online 14 September 2013
hypothermia (H), thoracic trauma (TT), other trauma (OT) and fire fatalities (F). The results suggest
that cTnT levels may help to differentiate cardiovascular death from poisoning and non-thoracic trauma,
Keywords:
as well as to differentiate cardiovascular and other diseases as cause of death from drowning and
Postmortem biochemistry
hypothermia. However, the effect of PMI, unlike comorbid cardiovascular disease, has to be taken into
Cardiac troponin T
Blood account.
Pericardial fluid ß 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction taken into account. Due to rapid postmortem putrefaction of blood,

The use of cardiac troponins as a part of laboratory investiga-
tions in forensic autopsy has been analysed by several authors [1–
11]. Cardiac troponin T (cTnT), a marker for myocardial injury, has
been recommended for the evaluation of the presence and extent
of myocardial damage in various causes of death [12–15]. cTnT
may be used to confirm the diagnosis of cardiac-related death and
in distinguishing certain causes of deaths from others [12,13,15–
17].
In the living, an increase in the level of cTnT in serum can be
seen approximately 2–4 h after the onset of cardiac symptoms and
persists for up to 2 weeks [18]. Postmortem cTnT levels depend on
several factors, such as the severity and duration of myocardial
damage before death, the cause of death, the postmortem interval
(the time between death and the collecting of samples, PMI) and
the sampling site [6,8,9,12–14,19–21]; therefore, clinical serum
cTnT reference values cannot be directly used in interpreting
postmortem samples and cut-off values for postmortem cTnT
levels have been suggested [12,13,16]. Besides acute myocardial
infarction, other conditions, especially cardiovascular diseases, are
known to result in elevated cTnT levels [3,12,13,16,18], and the
effect of comorbid disease on postmortem cTnT levels should be
* Corresponding author at: Institute of Pathological Anatomy and Forensic
Medicine, University of Tartu, Ravila 19, 50411 Tartu, Estonia. Tel.: +372 7374290.
E-mail addresses: synne.remmer@ekei.ee (S. Remmer), anne.kuudeberg@ut.ee
(A. Kuudeberg), mailis.tonisson@ut.ee (M. Tõnisson), delia.lepik@ut.ee (D. Lepik),
marika.vali@ut.ee (M. Väli).
the use of alternative biological fluids, such as pericardial fluid (PF)
is advised [8,19,20,22,23].
There is no previous experience using postmortem cardiac
markers as a diagnostic tool in forensic autopsies in Estonia. The
aim of this study is to describe the findings regarding the possible
use of postmortem serum and pericardial fluid cTnT in investigat-
ing the cause of death.
2. Materials and methods
2.1. Materials
A total of 230 forensic autopsy cases at the Estonian Forensic Science Institute
were examined. At autopsy, blood was drawn from femoral/iliac veins and
pericardial fluid was collected after opening the pericardial sac. PF samples
contaminated with blood and all cases with evident decomposition were
excluded. The final sample consisted of 101 cases (87 males and 14 females), 25–
54 years of age with a median of 42 years of age. The estimated postmortem
interval was 8–141 h with a median of 34.5 h. The cause of death was grouped on
the basis of gross, histological and toxicologial findings as follows: cardiovas-
cular disease (CVD, n = 10), other diseases (OD, n = 12), poisoning (P, n = 20),
asphyxia (A, n = 25), drowning (D, n = 11), hypothermia (H, n = 5), blunt and
sharp thoracic trauma (TT, n = 7), other trauma (OT, n = 8) and fire fatalities
(carbon monoxide poisoning, F, n = 3). The CVD group included cases with
International Classification of Diseases codes I20–25 and I 30-52. The poisoning
group consisted of cases with acute lethal poisoning by the following agents:
ethanol, surrogate alcohol, diethyl ether, diazepam, phenobarbital, pentobar-
bital, thiopental, clozapine, tramadol, fentanyl, 3-methylfentanyl, methadone
and amphetamine. All cases were classified according to the presence or absence
of comorbid chronic cardiovascular disease (n = 36 and n = 65, respectively).
PMI was divided into 5 groups: less than 12 h (n = 8), 12–24 h (n = 20), 24–48 h
(n = 41), 48–72 h (n = 21) and over 72 h (n = 11). Case profiles are shown in
Table 1. Due to the small number of cases, fire fatalities were left out of further
statistical analysis.

0379-0738/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.forsciint.2013.09.010
 S. Remmer et al. / Forensic Science International 233 (2013) 154–157 155

Table 1
Case profiles (n = 101, 87 male and 14 female).

Cause of death n Age (years) PMI (h)

Range Median Range Median

CVD 10 28–53 48.0 16.5–100.0 34.3

OD 12 30–54 42.5 11.0–129.5 34.0
P 20 27–52 40.0 9.0–84.0 35.0
A 25 41–54 40.0 8.0–96.0 36.0
D 11 25–51 39.0 13.5–57.5 35.5
H 5 27–54 48.0 22.0–106.0 35.0
TT 7 26–54 37.0 8.5–66.3 16.0
OT 8 29–54 42.5 8.0–74.0 25.8
F 3 43–51 48.0 10.5–141.0 106.0

CVD, cardiovascular disease; OD, other diseases; P, poisoning; A, asphyxia; D,

drowning; H, hypothermia; TT, thoracic trauma; OT, other trauma; F, fire fatality;
PMI, postmortem interval (h) hours.

2.2. Biochemical analyses

Samples were stored at +4 8C during transportation to the laboratory (24 h).

Blood was centrifuged to separate the serum. Samples were stored at 20 8C until
Fig. 1. Serum cTnT levels according to PMI interval.
use. cTnT was measured using electro-chemiluminescence immunoassay (Roche
Diagnostics Elecsys). The upper limit for clinical reference range was 0.03 ng/ml.

(p-value = 0.7549 for serum, p-value = 0.3402 for pericardial fluid,

2.3. Statistical methods
Kruskal–Wallis) or among individual COD groups (p-val-
Spearman’s rank order correlation (rs) was used to evaluate the relationship ue = 0.7575 and p-value = 0.342 respectively, Mann–Whitney U-
between pairs of parameters. Non-parametric Kruskal–Wallis test was used to test, Table 3).
compare groups and the non-parametric Mann–Whitney U test was used for
comparisons between individual groups. Statistical analysis was performed using
statistical software R (version 2.14.0) and Statistica 12.0. A p-value of less than 0.05
was considered statistically significant.
4. Discussion

Cardiac troponin T (cTnT) has previously been assessed for the

3. Results postmortem diagnosis of myocardial injury with some controver-
sy.
We found no significant relationship between cTnT levels and Postmortem cTnT levels are generally lower in serum from
age or gender. peripheral blood compared to the levels in pericardial fluid, and the
There was a positive correlation between serum cTnT and proximity of the myocardium and direct leakage of cTnT into the
pericardial fluid cTnT levels (rs = 0.43, p < 0.001, Spearman). pericardium has been suggested as the cause [12,13,24]. Corre-
The level of pericardial fluid cTnT was higher compared to spondingly, pericardial fluid cTnT levels were generally higher
serum cTnT among all groups of causes of death, except for compared to serum cTnT levels in our study. We found a positive
asphyxiation and hypothermia. When compared pairwise, peri- correlation between the cTnT levels in serum and pericardial fluid,
cardial fluid cTnT levels were higher compared to serum cTnT which has not been previously reported.
levels in most cases (90/101) and equal in 1 case. In cases with A postmortem time-dependent elevation in cTnT levels in the
higher serum cTnT compared to pericardial fluid cTnT, the cause of first days after death has been shown previously [6,9,12,13],
death was asphyxiation in 5 out of 10.
We found a moderate positive correlation between pericardial
fluid cTnT level and PMI (rs = 0.41, p < 0.001, Spearman) and a
weak positive correlation between serum cTnT and PMI (rs = 0.22,
p = 0.02, Spearman). A significant postmortem time-dependent
elevation in serum and pericardial fluid cTnT levels emerged
between different PMI groups (p-value = 0.04 for serum and p-
value = 0.004 for pericadial fluid, Kruskal–Wallis, Figs. 1 and 2).
Among different groups of COD, postmortem time-dependent
elevation in pericardial fluid cTnT levels was found for asphyxia-
tion (rs = 0.80, p-value < 0.001, Spearman). No statistically impor-
tant postmortem elevation in serum cTnT levels for COD groups
was noticed.cTnT levels in serum and pericardial fluid regarding
different causes of death are shown in Table 2. Serum cTnT levels in
the CVD group were significantly higher compared to poisoning (p-
value < 0.01, Mann–Whitney U test). Pericardial fluid cTnT levels
showed significantly higher results in the CVD group compared to
drowning (p-value < 0.01), other trauma (p-value = 0.03) and
hypothermia (p-value < 0.01). Also, pericardial fluid cTnT levels
were significantly higher in OD group compared to drowning (p-
value = 0.02) and hypothermia (p-value = 0.03).
We found no significant difference in the levels of serum and PF
cTnT between cases with and without CVD comorbidity for all COD Fig. 2. Pericardial fluid cTnT levels according to PMI interval.
156 S. Remmer et al. / Forensic Science International 233 (2013) 154–157

Table 2 drowning presented 3/8 cases with serum cTnT values above the
Serum and pericardial fluid cTnT regarding cause of death.
suggested cut-off and none (0/8) with pericardial fluid cTnT above
Cause of death n Serum cTnT (ng/ml) PF cTnT (ng/ml) the cut-off; there were no drowning cases with PMI of less than
Range Median Range Median 12 h. For 12–48 h PMI, we found acute myocardial infarction (1/1)
with both serum and pericardial fluid cTnT above the suggested
CVD 10 0.03–84.36 10.56 4.98–250.00 51.35
OD 12 0.01–214.70 1.10 0.51–310.40 31.76
cut-off levels of 0.6 ng/ml and 100 ng/ml, respectively, there were
P 20 0.01–80.96 0.21 0.04–192.70 27.36 no carbon monoxide poisoning cases with PMI 12–48 h. Trauma
A 25 0.01–500.00 0.60 0.19–271.90 39.27 groups (TT and OT combined) presented cTnT values below the
D 11 0.01–17.54 0.83 0.38–65.10 4.91 suggested cut-off for serum in 6/10 and for pericardial fluid in 9/10
H 5 0.01–24.02 0.28 0.97–14.08 2.14
cases, and the respective number of cases for asphyxiation was 5/
TT 7 0.01–39.28 0.20 0.01–159.10 22.05
OT 8 0.01–13.66 0.55 0.17–62.33 10.40 12 and 11/12.
F 3 0.01–0.81 0.06 4.01–215.50 12.45 The role of cardiac contusion and thoracic injury in increased
postmortem troponin levels has been evaluated by several authors
CVD, cardiovascular disease; OD, other diseases; P, poisoning; A, asphyxia; D,
drowning; H, hypothermia; TT, thoracic trauma; OT, other trauma; F, fire fatality. with some controversy [12,21,24,25]. Many cardiovascular con-
ditions, such as congestive heart failure, hypertension, cardiomy-
opathy, myocarditis, aortic valve disease and cerebrovascular
although Ellingsen and Hetland [16] report serum cTnT being disease, are known to result in elevated cTnT levels [3,12,13,16,18].
stable in the first 3 days after death and Peter et al. [21] found no We did not notice significantly elevated cTnT levels in the thoracic
correlation between serum cTnT levels and autolysis time. Only a trauma group compared to other causes of death in the present
few reports exist on cTnT levels in case of PMI above 75 h [3,21]. In study, and comorbid cardiovascular disease also seemed to have no
our routine forensic practice, PMI tends to be relatively long; effect on postmortem cTnT levels.
therefore, cases with PMI up to 141 h were included in the current Elevated cTnT levels in postmortem serum from peripheral
study. Our findings suggest that the time-dependent elevation in blood may depend on the survival period following the onset of
both serum and pericardial fluid cTnT levels continues after 75 h myocardial injury [12,13], which was not taken into consideration
postmortem. in the current paper due to difficulties obtaining correct data.
It has been suggested that postmortem cTnT could be used to In conclusion, the results of the current study generally confirm
support the diagnosis of heart disease as the cause of death if other previous findings on postmortem serum and pericardial fluid cTnT.
causes of myocardial damage can be excluded [12,13,16] and to Postmortem cTnT, in addition to other forensic evidence, may help
differentiate drowning and fatal hypothermia from acute myocar- to define cause of death; however, the postmortem time-
dial infarction and methamphetamine abuse [12]. According to our dependent elevation in cTnT levels has to be taken into
findings, serum cTnT may be used as a supportive tool in consideration.
differentiating deaths caused by CVD from poisoning and
pericardial fluid cTnT in deaths caused by CVD from OT, as well References
as CVD and OD from drowning and hypothermia. However, the
variation in cTnT levels among individual cases is high and there is [1] M.D. Perez-Carceles, J. Noguera, J.L. Jimenez, P. Martinez, A. Luna, E. Osuna,
Diagnostic efficacy of biochemical markers in diagnosis post-mortem of ischae-
overlap between groups, which makes it difficult to apply the mic heart disease, Forensic Sci. Int. 142 (2004) 1–7.
results on individual cases. [2] F. Martı́nez Dı́az, M. Rodrı́guez-Morlensı́n, M.D. Pérez-Cárceles, J. Noguera, A.
As mentioned above, cTnT clinical reference values are not Luna, E. Osuna, Biochemical analysis and immunohistochemical determination of
cardiac troponin for the postmortem diagnosis of myocardial damage, Histol.
applicable for postmortem investigations [12,13,16]. In the current Histopathol. 20 (2005) 475–481.
study, all cases in the CVD group presented serum cTnT levels [3] S.J. Davies, D.C. Gaze, P.O. Collinson, Investigation of cardiac troponins in post-
equal to or higher than the clinical reference of 0.03 ng/ml. mortem subjects: comparing antemortem and postmortem levels, Am. J. Forensic
Med. Pathol. 26 (2005) 213–215.
Ellingsen and Hetland [16] evaluated the serum cTnT level of [4] G.L. de la Grandmaison, Is there progress in the autopsy diagnosis of sudden
0.6 mg/l as the cut-off value between evident/possible cardiac unexpected death in adults? Forensic Sci. Int. 156 (2006) 138–144.
deaths and negative controls. In our study, 8/10 cases in the CVD [5] B.L. Zhu, T. Ishikawa, T. Michiue, D.R. Li, D. Zhao, Y. Bessho, Y. Kamikodai, K. Tsuda,
S. Okazaki, H. Maeda, Postmortem cardiac troponin I and creatine kinase MB levels
group showed serum cTnT value above 0.6 ng/ml and 49/91 cases
in the blood and pericardial fluid as markers of myocardial damage in medicolegal
in all other groups combined represented with cTnT levels below autopsy, Leg. Med. 9 (2007) 241–250.
that level. Cut-off values have been suggested by Zhu et al. [12] as [6] B.L. Zhu, T. Ishikawa, T. Michiue, D.R. Li, D. Zhao, S. Tanaka, Y. Kamikodai, K. Tsuda,
0.2 ng/ml for serum from external iliac venous blood, 10 ng/ml for S. Okazaki, H. Maeda, Postmortem pericardial natriuretic peptides as markers of
cardiac function in medico-legal autopsies, Int. J. Legal. Med. 121 (2007) 28–35.
pericardial fluid in the early postmortem period (within 12 h) to [7] H. Maeda, B.L. Zhu, T. Ishikawa, L. Quan, T. Michiue, Significance of postmortem
differentiate between drowning and other groups, and 0.6 ng/ml biochemistry in determining the cause of death, Leg. Med. 11 (2009) S46–S49.
for serum from external iliac venous blood and 100 ng/ml for [8] A. Luna, Is postmortem biochemistry really useful? Why is not widely used in
forensic pathology?, Leg. Med. 11 (2009) S27–S30.
pericardial fluid 12–48 h postmortem to differentiate between [9] Q. Wang, T. Michiue, T. Ishikawa, B.L. Zhu, H. Maeda, Combined analyses of
higher cTnT level groups (hyperthermia, fatal methamphetamine creatine kinase MB, cardiac troponin I and myoglobin in pericardial and cerebro-
abuse, carbon monoxide poisoning, acute myocardial infarction) spinal fluids to investigate myocardial and skeletal muscle injury in medicolegal
autopsy cases, Leg. Med. 13 (2011) 226–232.
and intermediate groups (injuries, asphyxiation and fire fatalities). [10] C. Palmiere, P. Mangin, Postmortem chemistry update part II, Int. J. Legal Med. 126
In our study, with PMI within 12 h, other causes of death besides (2012) 199–215.
[11] C. Palmiere, et al., Biochemical markers of fatal hypothermia, Forensic Sci. Int. 226
(2013) 54–61.
Table 3 [12] B.L. Zhu, T. Ishikawa, T. Michiue, D.R. Li, D. Zhao, S. Oritani, Y. Kamikodai, K. Tsuda,
CVD comorbidity. S. Okazaki, H. Maeda, Postmortem cardiac troponin T levels in the blood and
pericardial fluid. Part 1: analysis with special regard to traumatic causes of death,
CVD comorbidity n Serum cTnT (ng/ml) PF cTnT (ng/ml) Leg. Med. 8 (2006) 86–93.
[13] B.L. Zhu, T. Ishikawa, T. Michiue, D.R. Li, D. Zhao, Y. Kamikodai, K. Tsuda, S.
Range Median Range Median
Okazaki, H. Maeda, Postmortem cardiac troponin T levels in the blood and
CVD+ 36 0.07–214.70 0.59 0.04–250.00 22.61 pericardial fluid. Part 2: analysis for application in the diagnosis of sudden cardiac
CVD 65 0.00–500.00 0.78 0.01–310.40 18.06 death with regard to pathology, Leg. Med. 8 (2006) 94–101.
[14] H. Maeda, T. Michiue, B.L. Zhu, T. Ishikawa, L. Quan, Analysis of cardiac troponins
CVD+ cases with comorbid cardiovascular disease, CVD cases without comorbid and creatine kinase MB in cerebrospinal fluid in medicolegal autopsy cases, Leg.
cardiovascular disease. Med. 11 (2009) S266–S268.
 S. Remmer et al. / Forensic Science International 233 (2013) 154–157
[15] A.B. Khalifa, M. Najjar, F. Addad, E. Turki, T. Mghirbi, Cardiac troponin T (cTnT) and the
postmortem diagnosis of sudden death, Am. J. Forensic Med. Pathol. 27 (2006) 175–177.
[16] C.L. Ellingsen, Ø. Hetland, Serum concentrations of cardiac troponin T in sudden
death, Am. J. Forensic Med. Pathol. 25 (2004) 213–215.
[17] B.L. Zhu, K. Ishida, M. Taniguchi, L. Quan, S. Oritani, K. Tsuda, Y. Kamikodai, M.Q.
Fujita, H. Maeda, Possible postmortem serum markers for differentiation between
fresh-, saltwater drowning and acute cardiac death: a preliminary investigation,
Leg. Med. 5 (2003) S298–S301.
[18] A.S. Jaffe, L. Babuin, F.S. Apple, Biomarkers in acute cardiac disease: the present
and the future, J. Am. Coll. Cardiol. 48 (2006) 1–11.
[19] B. Madea, F. Musshoff, Postmortem biochemistry, Forensic Sci. Int. 165 (2007)
165–171.
[20] H. Maeda, T. Ishikawa, T. Michiue, Forensic biochemistry for functional investi-
gation of death: concept and practical application, Leg. Med. 13 (2011) 55–67.
157
[21] J. Peter, A. Kirchner, E. Kuhlisch, M. Menschikowski, B. Neef, J. Dressler, The
relevance of the detection of troponins to the forensic diagnosis of cardiac
contusion, Forensic Sci. Int. 160 (2006) 127–133.
[22] A. Thierauf, F. Musshoff, B. Madea, Post-mortem biochemical investigations of
vitreous humor, Forensic Sci. Int. 192 (2009) 78–82.
[23] J.I. Coe, Postmortem chemistry update, Am. J. Forensic Med. Pathol. 14 (1993)
91–117.
[24] S.J. Cina, W.C. Thompson, R.J. Fischer, D.K. Brown, J.M. Titus, J.E. Smialek, A study
of various morphologic variables and troponin I in pericardial fluid as possible
discriminators of sudden cardiac death, Am. J. Forensic Med. Pathol. 20 (1999)
333–337.
[25] S.J. Cina, D.J. Li, D.W. Chan, J.K. Boitnott, R.H. Hruban, J.E. Smialek, Serum con-
centrations of cardiac troponin I in sudden death: a pilot study, Am. J. Forensic
Med. Pathol. 19 (1998) 324–328.