Int J Legal Med (2014) 128:117–125
DOI 10.1007/s00414-013-0876-x
ORIGINAL ARTICLE
Markers of mechanical asphyxia: immunohistochemical
study on autoptic lung tissues
R. Cecchi & C. Sestili & G. Prosperini & G. Cecchetto &
E. Vicini & G. Viel & B. Muciaccia
Received: 8 March 2013 / Accepted: 15 May 2013 / Published online: 29 May 2013
# Springer-Verlag Berlin Heidelberg 2013
Abstract Forensic pathologists are often asked to provide
evidence of asphyxia death in the trial and a histological marker
of asphyxiation would be of great help. Data from the literature
indicate that the reaction of lung tissue cells to asphyxia may be
of more interest for forensic purposes than migrating cells. The
lungs of 62 medico-legal autopsy cases, 34 acute mechanical
asphyxia (AMA), and 28 control cases (CC), were immuno-
stained with anti-P-selectin, anti-E-selectin, anti-SP-A, and
anti-HIF1-α antibodies, in order to verify if some of them
may be used as markers of asphyxia death. Results show that
P- and E-selectins expression in lung vessels, being activated
by several types of trigger stimuli not specific to hypoxia,
cannot be used as indicator of asphyxia. Intra-alveolar granular
deposits of SP-A seem to be related to an intense hypoxic
stimulus, and when massively present, they can suggest, to-
gether with other elements, a severe hypoxia as the mechanism
of death. HIF1-α was expressed in small-, medium-, and large-
caliber lung vessels of the vast majority of mechanical asphyxia
deaths and CO intoxications, with the number and intensity of
positive-stained vessels increasing with the duration of the
hypoxia. Although further confirmation studies are required,
these preliminary data indicate an interesting potential utility of
HIF1-α as a screening test for asphyxia deaths.
R. Cecchi (*) : C. Sestili : E. Vicini : B. Muciaccia
Department of Anatomical, Histological, Legal Medical and
Orthopaedic Sciences, Faculty of Medicine and Pharmacology,
Sapienza University of Rome, Viale Regina Elena 336,
00161 Rome, Italy
e-mail: rossana.cecchi@uniroma1.it
G. Prosperini
CASPUR, Consortium for Supercomputing Applications, Via dei
Tizii 6/B,
00185 Rome, Italy
G. Cecchetto : G. Viel
Institute of Legal Medicine, Department of Molecular Medicine,
University of Padova, Via Falloppio 50,
35121 Padova, Italy
Keywords Forensic histopathology . Mechanical asphyxia .
Hypoxia . Selectins . Surfactant protein-A . Hypoxia-
induced factor 1-α . Evidence
Introduction
Asphyxiation can present challenging issues in terms of
expert opinions and may generate debate regarding evidence
in court when the death is homicidal. As is well-known from
the forensic literature [1], homicidal asphyxiation with soft
objects is currently a subject of discussion, and histological
examinations are required to exclude any relevant acute
disease because of the uncharacteristic morphological find-
ings. Toxicological effects have to be excluded, and differ-
entiation of asphyxia-related or hypoxic tissue and organ
damage from autolytic changes is not always easy.
Histomorphological changes in pulmonary tissue of in-
dividuals who die from asphyxia are considered of great
importance [2, 3] and constitute one of the central fields of
interest for forensic histopathology. Nevertheless, their cor-
relation to asphyxia has to consider ventilation and perfu-
sion alterations due to age, underlying diseases, and external
factors, such as smog and smoke. The hemorrhagic edema,
which occurs in cases of drug addiction, may also cause
problems in morphological differential diagnosis with as-
phyxia. For this reason, classical histological alterations of
lung tissue do not always solve problems of differential
diagnosis, especially when intensive care measures or slow
death has occurred, and have to be excluded.
Therefore, over the decades, thanks to newly available
methods, such as immunohistochemistry, attention has grad-
ually shifted to focus on the possibility of dating the time of
asphyxia or to characterizing differential diagnosis among
various types of asphyxia death [4–8].
Forensic pathologists are often asked to provide evidence
of asphyxia death in trial and a good marker of asphyxiation
118
should be of great help. In this case, an ideal parameter
should be one which is absent in physiological conditions
and appears regularly after asphyxia, growing with time
before death [9, 10]. This goal is far from being realized,
but in specific cases the immunohistochemical findings can
be of indicative significance, together with all other avail-
able information [1].
Chronic alveolar hypoxia induces an inflammatory reac-
tion with macrophage recruitment, an increase in albumin
leakage, and enhanced expression of inflammatory media-
tors, which seems to be mainly macrophage dependent [11].
Macrophage recruitment was investigated as a marker of
prolonged asphyxia, and authors supported the proliferation
and mobilization of interstitial and alveolar macrophages
and the increased number of giant cells. This argument is
still being debated, however, since it can also be found in
other pathological conditions (inhalation of dust, cigarette
smoke, etc.) [4–8].
At the present time, it is well-known that the lung ex-
hibits several adaptation mechanisms to conditions of low
O2 tension. Acute exposure to hypoxia results in vasocon-
striction of resistance-sized pulmonary arteries, increased
pulmonary arterial pressure, and a redistribution of the
blood flow from the basal to the apical portion of the lung
[12, 13]. It is therefore expected that the acute inflammatory
response to asphyxia of the lung involves mediators of the
vessel response to inflammatory insult, and that some of
them may represent a good marker for forensic purposes.
Ortmann and Brinkmann [14] compared the expression
of P-selectin in inflammatory and non-inflammatory lung
tissue. P-selectin is a glycoprotein stored for rapid release in
Weibel–Palade bodies of the endothelium and in the alpha
granules of platelets [15] and represents an adhesion mole-
cule involved in the process of leukocyte rolling on the
endothelium. The authors found that in acute deaths (hang-
ing, carbon monoxide and cyanide intoxication) lung ves-
sels showed an overall occurrence of P-selectin with an
intense homogeneous staining pattern. The cases of drown-
ing showed a slightly weaker intensity, whereas protracted
deaths, due to pneumonia and septic shock, showed an
irregular distribution and weak intensity of the marker.
Zhu et al. [16–18] and Maeda et al. [19] focused on
alveolar surface integrity in asphyxia cases and found that
granular staining of surfactant protein-A (SP-A) increases in
intra-alveolar spaces in cases of hyaline membrane syn-
drome, drowning, aspiration of amniotic fluid, and mechan-
ical asphyxia. SP-A is the major surfactant protein and its
deficiency causes respiratory distress syndrome [20, 21]. It
is produced by type II alveolar cells and normally covers the
alveolar surface. A hypoxic condition increases the secre-
tion of SP-A, probably because of the strong forced breath-
ing, and then SP-A precipitates in the alveolar space in
aggregates. Zhu et al. [16] showed no significant differences
Int J Legal Med (2014) 128:117–125
between mechanical asphyxia and control groups in the
intensity of staining of the alveolar surface and in the num-
ber of the type II pneumocytes, but many prominent massive
intra-alveolar aggregates stained positive with SP-A were
found prevalently in the mechanical asphyxia group (ap-
proximately 60 % of all cases in this group). It was con-
cluded that this may be a result of enhanced secretion of the
pulmonary surfactant.
The above-mentioned articles indicate that the reaction of
lung tissue cells to asphyxia may be of more interest for
forensic purposes rather than migrating cells. In this context,
interesting information about time of asphyxia could be
given by investigating the rule of E-selectin and hypoxia-
induced factor 1-α (HIF1-α) in lung tissue cells.
E-selectin is an adhesion molecule that recognizes and
binds to sialylated carbohydrates present on the surface
proteins of leukocytes and mediates their adhesion to the
endothelium. It is produced by the endothelium only after
cytokine activation and needs a protein synthesis [22]. It is
the first molecule activated by TNF-α. In the skin, its
expression is detected after 1–2 h from insult up until
48 h. It is not constitutively expressed in tissues and is
activated later with respect to P-selectin. For these reasons,
its expression in lung tissue is expected to give information
on more prolonged asphyxia. To our knowledge, no study
with E-selectin was performed in asphyxia deaths.
HIF-1a is a transcription factor expressed in response to
hypoxia that activates expression of the genes involved in
erythropoiesis, angiogenesis, glycolysis, and modulation of
vascular tone. It is a potential mediator of pulmonary re-
sponses to hypoxia. HIF-1 is a heterodimer consisting of
HIF1-α and HIF1-β subunits. HIF-1α accumulates in the
nuclei of mammalian cells exposed to reduced O2 tension in
a time and O2 concentration-dependent manner [23]. The
HIF1-β subunit is constitutively expressed, whereas the
expression and activity of the HIF-1α subunit is precisely
regulated by the cellular O2. The HIF1-α subunit is contin-
uously synthesized and degraded under normoxic condi-
tions, while it accumulates rapidly following exposure to
low oxygen tension.
Few experimental studies suggest that levels of HIF-1
mRNA increase in rats subjected to hypoxia and that hypoxia
induces HIF1-α but not HIF1-β mRNA expression in pulmo-
nary artery endothelial cells of rats exposed to chronic hyp-
oxia [24–26]. A study by Yu et al. [12] shows that HIF-1α
protein expression is induced maximally when the lungs are
ventilated with 0 or 1 % O2 for 4 h, and the highest expression
occurs in the bronchial epithelium, bronchial smooth muscle,
alveolar epithelium, and vascular endothelium. On
reoxygenation, HIF1-α is rapidly degraded. These findings
show that HIF1-α expression is tightly coupled to O2 concen-
tration in vivo and are consistent with the involvement of
HIF1-α in the physiological and pathophysiological responses
Int J Legal Med (2014) 128:117–125
to hypoxia in the lung. In contrast, HIF1-β levels do not
change with an increase in time of hypoxia or a decrease in
O2 concentration. Not only hypoxia, but also oxidative stress
might play an important role in HIF1 activity [27, 28]. HIF-1α
also has an important role in inflammatory response. To our
knowledge, HIF1-α was studied in human lung tissue only for
tumor or inflammation [29–32] and never in asphyxia death.
The present work focuses on the reaction of lung tissue
components to asphyxia, particularly the vessels and alveoli,
through the study of P- and E-selectins, SP-A, and HIF1-α,
in order to verify if some of them may be useful in the
diagnosis of asphyxia death.
Materials and method
Cases
Sixty-two (62) medico-legal autopsy cases performed at the
Department of Anatomical, Histological, Legal Medical and
Orthopaedic Sciences of the Sapienza University of Rome,
Italy, and at the Institute of Legal Medicine of the University
of Padua (Italy) between 24 and 72 h post-mortem or, in
cases of drowning, between 24 and 72 h after the discovery
of the corpse were examined.
Primary pulmonary diseases were excluded by histological
investigations. Causes of death were based on routine
macromorphological, micropathological, histological, and tox-
icological findings, and two groups were identified as: acute
mechanical asphyxia (AMA) [total n=34: drowning (n=15),
hanging (n=7), strangulation (n=3), smothering (n=3), and
aspiration (n=6)]; control cases (CC) total n=28: carbon mon-
oxide (CO) intoxication (n=6), heroin-related deaths (n=9),
pneumonia (n=2), head trauma (n=5), natural death (n=6).
Histology and immunohistochemical analysis
Lung samples from each case were fixed in 4 % PBS–
formaldehyde solution, dehydrated through alcohol and
paraffin-embedded; 5-μm-thick paraffin-embedded sections
were cut from each lung specimen and processed for mor-
phological evaluation using hematoxylin–eosin (HE)
staining and for immunohistochemistry (IHC).
IHC analysis was performed using the following specific
antibodies:
– Human anti-CD62P antigen (P-selectin) mouse mono-
clonal antibody (1:50 clone C34 by Novocastra);
– Human anti-CD62E antigen (E-selectin) mouse mono-
clonal antibody (1:100 clone 16G4 by Novocastra);
– Human anti-pulmonary surfactant protein A (SP-A)
mouse monoclonal antibody (1:100 clone 32E12 by
Novocastra);
119
– Human anti-hypoxia-induced factor 1-α (HIF-α) mouse
monoclonal antibody (1:1000 clone ESEE122 by Abnova).
For the IHC, samples were de-waxed, hydrated, and
subjected to heat-induced epitope retrieval (HIER) using
Target Retrieval Solution (Dako) (20′) in a thermostatic bath
at 98 °C. The staining procedure was performed using a
Dako Autostainer Instrument, as follows [10, 33]: incuba-
tion in peroxide solution (10′); incubation at room temper-
ature (RT) with primary antibody, for P-selectin and E-
selectin (30′) and for SP-A (60′), in Dako REAL™
EnVision™ Detection System, Peroxidase/DAB+,
Rabbit/Mouse; incubation with Dako DAB-AWAY detec-
tion system (5′); washed in distilled water; and nuclear
staining with hemalaum.
The primary antibody was replaced with Dako wash
buffer 10× in the negative control test, while tonsil samples
were used for the positive control experiment for selectins
and SP-A-positive staining of intra-alveolar surface was
used as internal control for SP-A.
A manual immunohistochemical technique was used, as
previously described [34], only for immunohistochemical
detection of HIF1-α antibody. Briefly, paraffin sections
were dewaxed, re-hydrated, subjected to the antigen retriev-
al procedure [citrate buffer pH 6.0 (15′) in microwave] and,
after quenching endogenous peroxidase and blocking non-
specific binding sites, incubation was performed (60′) with
the primary antibody at RT; (15′) with a biotinylated sec-
ondary antibody; (15′) with avidin-biotin-peroxidase-com-
plex (ABC of UltreTek HRP Anti-Polyvalent kit, ScyTek
Laboratories, USA); (1′) with DAB substrate (Roche, Italy);
then slides were washed and nuclear counterstaining (30″)
with hematoxylin was performed. Negative controls were
carried out using affinity-purified mouse-IgG or by omitting
the primary antibody. Optimal dilution of anti-HIF1-α anti-
body was set up and established using serial sections from
human placenta, as positive control tissue. Since the state of
morphological preservation of different samples varied
according to the type of asphyxia death, our experimental
protocol for HIF1-α immunodetection was designed to
avoid antibody overnight incubations and the color reaction,
with DAB substrate, was developed very quickly (max 1–
2 min). These precautions were able to greatly reduce back-
ground interference from all the lung samples.
Analysis of immunostaining
Five random microscopic fields of 1.76 mm2 for each im-
munostained tissue section were assessed by two indepen-
dent observers, using ×10 and ×40 original magnification.
The first observer using ×10 magnification randomly chose
the fields and marked them with a felt-tip pen, while the
second observer re-examined the marked areas in blind.
120 Int J Legal Med (2014) 128:117–125

Fig. 1 Presence of P-Selectin

(a) and E-selectin (b) A B
immunostaining in pulmonary
vessels (arrows) in a case
of hanging. The intensity
of the staining is
* *
scored +++. Nuclei were
counterstained using
*
hematoxylin solution *
(×200; bar=20 μm)

Grading of intensity and vessel counting was performed at different degrees of intra-alveolar hemorrhages with intra-
high-power view (×40 original magnification). No inter- alveolar edema intra- and interstitial alveolar edema were
observer variability was evidenced for grading of intensity, present (data not shown).
whereas only little inter-observer variability emerged in In 13 cases (six drowning, one strangulation, one smoth-
vessel counting. In these cases the medium value (between ering, two aspirations, one CO intoxication, and two heroin
the first and the second observer) was reported and utilized for intoxication cases), signs of putrefaction were visible. These
statistical analyses. Atelectatic areas were not considered in order cases were excluded from HIF1-α analysis.
to avoid interference with the results. Areas presenting an em-
physema aquosum were considered not to interfere with the total Immunohistochemistry
amount of vessels/HPF and were therefore evaluated. Since P-
selectin, E-selectin, and HIF1-α are absent in the capillaries, this P-selectin and E-selectin
vascular region was not considered in the assessment. Expression
of the three markers was scored by a semi-quantitative method, Intense overlapping and diffuse expression of P-selectin and
evaluating the intensity and number of positively stained vessels E-selectin were found in all types of vessels, predominantly
[4]. The intensity of the marker expression was graded as absent
(0), mild (+), moderate (++), and intense (+++). The number of Table 1 Grading of P- and E-selectins activation in AMA and CC
group. For the grading score see text. The number of vessels express-
reactive vessels was graded as absent (0), less than 10 (1), 10– ing P- and E-selectin and the intensity of the expression showed a
50 % (2), and more than 50 % (3). Because of possible false similar trend and there was an overlap of results between grading score
positives on the margins of lesions due to post-mortal artifacts, and cause of death
these were excluded from count.
Cause of death Number of Grading of Grading of
SP-A was scored according to Zhu et al. [16] evaluating cases=62 intensity the number
density and distribution of SP-A in intra-alveolar spaces as of activated
follows: negative (−); grade I (−/+), weakly positive; grade vessels
II (+), positive with a few massive aggregates of stained
Hanging (AMA) 7 III 3 (5 cases)
granules; grade III (++), intensely and diffusely positive
2 (2 cases)
with many massive aggregates of stained granules.
Strangulation 3 III 3
(AMA)
Statistical analysis Smothering 3 III 3
(AMA)
Aspiration 6 III 3
A student t test with Welch correction was applied to evaluate
(AMA)
the differences in grading and number of vessels stained with Drowning 15 III (4 cases) 3 (10 cases)
HIF1-α between CC versus AMA and CO intoxication versus (AMA) II (9 cases) 2 (5 cases)
AMA. Statistical significance was fixed at p<0.05. I (2 cases)
CO intoxication 6 III 3
(CC)
Results Heroin 9 II 2
intoxication
(CC)
Hematoxilin–eosin Pneumonia (CC) 2 II 2
Head trauma 5 II (4 cases) 3 (4 cases)
Local atelectasis and/or local emphysema and alveolar cap- (CC) I (1 case) 2 (1 case)
illary congestion were found in the lung samples from Natural death 6 II (4 cases) 3 (4 cases)
controls and asphyxia deaths. In the asphyxia group, CO (CC) I (2 cases) 2 (2 cases)
intoxication, heroin intoxication, and pneumonia cases
Int J Legal Med (2014) 128:117–125 121

Table 2 Immunohistochemical distribution of intra-alveolar SP-A. For SP-A

grading score, see text

Cause of death Total Grading In all cases, linear staining spreading on the intra-alveolar
inner surface and staining of the type II alveolar cells were
Hanging (AMA) 7 2 (5 cases) found. Significant differences in relation to the cause of
3 (2 cases) death were found in the distribution of SP-A-positive gran-
Strangulation (AMA) 3 2 (2 cases) ular precipitates in the alveolar spaces, which were classi-
3 (1 case) fied in three grades as previously explained (Table 2).
Smothering (AMA) 3 1 (2 cases) In the AMA group, grade III (intense and diffuse positive
2 (1 case) staining with many massive aggregates of granular stains)
Aspiration (AMA) 6 2 (4 cases) was found in 11/34 cases (32.4 %), whereas grade II (few
3 (2 cases) massive aggregates of stained granules) was found in 21/34
Drowning (AMA) 15 2 (9 cases) cases (61.8 %). In the CC group, grade II was found in 5/27
3 (6 cases) cases (18.5 %), while the weakly positive grade I in 23/27
CO intoxication (CC) 6 1 (5 cases) (cases 85.2 %; Fig. 2a–b).
2 (1 case)
Heroin intoxication (CC) 9 1 (6 cases)
HIF 1-α
2 (3 cases)
Pneumonia (CC) 2 1 (1 case)
HIF1-α proved to be very sensitive to putrefaction, thus the
2 (1 case)
13 cases (six drowning, one strangulation, one smothering,
Head trauma (CC) 5 1
two aspirations, one CO intoxication, and two heroin in-
Natural death (CC) 6 1
toxications) with histological signs of putrefaction were not
evaluated for HIF1-α. The analysis of HIF1-α refers, there-
fore, to a total of 49 cases in spite of 62: AMA group n=24
[drowning (n=9), hanging (n=7), strangulation (n=2),
in veins, apart from the capillaries, which were always smothering (n=2), and aspiration (n=4)] and CC group n=
negative. Where present, platelets and megakaryocytes 25 [CO intoxication (n=5), heroin intoxication (n=7), pneu-
stained positive with P-selectin and showed no difference monia (n=2), head trauma (n=5), and natural death (n=6)].
in staining intensity, acting as a further internal control. HIF1-α was found to be expressed on endothelial cells prev-
Endothelial cells of vessels in AMA group showed a alently in activated arteries and in few activated veins, never in
highly diffuse (grade 3) in 23/34 (67.6 %), intense (+++) capillaries (Fig. 3). In some cases, arteries expressed HIF1-α also
and homogeneous activation with P- and E-selectins in smooth muscle cells (Fig. 3b). Pneumocytes and leukocytes
(Fig. 1a–b). Only in drowning was the intensity of staining proved to be able to express HIF1-α (data not shown).
moderate (++; Table 1). The number of vessels activated with HIF1-α never
With regard to the CC group, in heroin intoxication and in exceeded 30 % of all vessels and was strongly correlated with
pneumonia the expression of P- and E-selectins was irregu- the cause of death. In CC group, 16/25 cases (64 %, corre-
larly distributed in 100 % of cases, less diffused (10–50 %, sponding to all natural deaths, head trauma, pneumonia, and
grade 2) and mild/moderate (+/++), while CO intoxication 4/7 heroin-related deaths) were scored as grade 0 (absent),
showed a 100 % of diffuse (grade 3) and intense (III) reaction, while the remaining 3/7 heroin-related deaths and all CO
and head trauma and natural death a grade 2/3 diffusion and a intoxication deaths were scored as grade 1 (<10 %). AMA
mild/moderate (I/II) intensity (Table 1). group was scored mainly as grade 1 and 2 (<10–50 %; Fig. 4).

Fig. 2 A representative pattern

of SP-A immunostaining A B
distribution in lung tissue of a
control case (a) with a
grading score of I/II (scattered),
and of a drowning case
(b) with a grading score
of II/III (dense but localized).
Nuclei were counterstained
using hematoxylin solution
(×100; bar=50 μm)
122 Int J Legal Med (2014) 128:117–125

Fig. 3 HIF1-α was

immunodetected in endothelial A B
and smooth muscle cells in
pulmonary arteries of
different sizes (arrows)
in a case of hanging
(b) and drowning (c)
while no staining was
found in control
cases (a). Sections
were hematoxylin-
counterstained [×200
(a) and (b); ×100 (c);
bar=50 μm (a), 100 μm (b),
20 μm (c)]
C

Intensity of staining was intense (+++) in 1/7 hanging;
moderate (++) in 6/7 hanging, 4/4 aspiration, 1/2 smothering,
3/5 CO intoxication, and 1/7 heroin intoxication; mild (+) in
4/9 drowning, 2/5 CO intoxication, and 2/7 heroin intoxication.
The mean values of HIF1-α positively stained vessels
and their differences in number and intensity were evaluated
using Student’s t test with Welch correction [35] to evaluate
the difference between CC versus AMA and CO intoxica-
tion versus AMA (reported in Tables 3 and 4, respectively),
and p values<0.05 were considered significant.
A comparison in relation to the caliber of vessels activated
shows in both groups a statistically significant higher number
Fig. 4 Vessel distribution: Number of vessels stained positive to HIF1-
α in relation to the causes of death (n=49). 1, hanging (n=7); 2,
strangulation (n=2); 3, smothering (n=2); 4, aspiration (n=4); 5,
drowning (n=9); 6, CO intoxication (n=5); 7, heroin intoxication
(n=7); 8, pneumonia (n=2); 9, head trauma (n=5); 10, natural death
(n=6). The 13 cases in which putrefaction was present are not included
in the table
of activated pre-capillary arterioles/post-capillary venules (de-
fined small-caliber vessels), as respect to arterioles/venules
(defined medium-caliber vessels) and arteries/veins (defined
large-caliber vessels; Table 3). A comparison of the results
concerning CO intoxication and the AMA group shows the
activation of HIF1-α also in CO intoxication; no statistical
significant difference was found between these two groups
(Table 4).
Discussion
In the present study, IHC was used in order to investigate
potential markers of asphyxia in a case history containing, in
addition to cases of mechanical asphyxia, also CO intoxica-
tion (in which there is a gradual depletion of oxygen in the
blood) and heroin intoxication (which are mostly deaths
from respiratory depression due to inhibition of the central
respiratory bulbar center). Natural deaths and head trauma
cases were used as a point of reference for defining the basic
levels of the markers.
P-selectin was used in a previous study for the differen-
tiation between inflammatory and non-inflammatory lung
diseases [14] and proved to be intensely expressed in a
higher percentage of vessels in non-inflamed tissue exposed
to lack of oxygen, such as cases of hanging, drowning, and
CO intoxication, which was confirmed by our results.
Nevertheless, the above-mentioned study did not investigate
healthy lungs and therefore it does not permit the verifica-
tion of whether P-selectin has a basic expression in lung
Int J Legal Med (2014) 128:117–125 123

Table 3 Intensity and number of vessels stained positive to HIF1-α: comparison between AMA and CC groups. The 13 cases in which
putrefaction was present are not included in the table

Vessel p value CC AMA

Mean SD Mean SD

Number Total 0.00031 3.40 6.19 15.38 13.22

Large caliber 0.00228 0.44 1.04 1.75 1.67
Medium caliber 0.00203 0.64 1.29 3.13 3.37
Small caliber 0.00066 2.32 4.88 10.58 9.68
Intensity Total 0.00005 0.64 0.76 1.54 0.66
Large caliber 0.00027 0.24 0.60 1.21 1.02
Medium caliber 0.00029 0.48 0.77 1.38 0.82
Small caliber 0.00052 0.60 0.76 1.54 0.98

vessels. Our casuistry contains 11 cases in which lungs were
healthy (head trauma and natural deaths) and shows in these
samples the diffused P-selectin immunodetection, with a
mild and moderate intensity, which corresponds to a basal
and constitutive expression of the marker in the lungs. This
prevents P-selectin from being used in the diagnosis of
mechanical asphyxia.
In our study, E-selectin was investigated for the first time
in healthy human lungs and in asphyxia cases, and showed a
comparable reaction as with P-selectin. Unfortunately, also
E-selectin cannot provide evidence of asphyxia deaths.
The constant and overlapping positivity of both markers
(P- and E-selectins) in lung vessels leads us to believe that,
in the lungs, internal and external factors may act as a
continuous trigger for the activation of selectins in the
endothelium of the lung vessels, resulting in a similar ex-
pression. Nevertheless, this would result in a continuous
recruitment of leukocytes that, however, does not occur.
Therefore, it can be argued that selectins may have a differ-
ent role in pulmonary endothelium, or that, at the end of life,
whatever the cause of death, there is an activation of the
pulmonary selectin system.
SP-A was confirmed to be a potential indicator of AMA
because massive intra-alveolar aggregates of granular stains
were found exclusively in the AMA group, while in the CC
group few precipitates were found in one case of pneumo-
nia, one of CO intoxication, and two cases of heroin intox-
ication. The absence of huge intra-alveolar aggregates in
healthy lungs (head trauma and natural deaths), makes SP-
A a promising marker of hypoxic conditions.
However, the most interesting results were provided by
the presence of HIF1-α in lung vessels. It is interesting that
all CC lung tissues (except CO intoxication cases and 3/7
heroin intoxications) did not show HIF1-α immunostaining
in the endothelium of arterioles, arteries, and veins, but only
in pre-capillary arterioles. On the contrary, different grades
of HIF1-α expression were found in AMA cases in small-,
medium-, and large-caliber vessels. This indicates that in
normal conditions, lung vessels, constitutively, do not ex-
press a detectable level of HIF1-α antigen, while after acute
lack of O2 they are activated and express, especially in
endothelial cells, a detectable HIF1-α level very quickly.
Activated vessels within AMA group cases expressed
HIF1-α in all vessel types. Pre-capillary arterioles/post-capillary

Table 4 Intensity and number of vessels stained positive to HIF1-α: comparison between AMA and CO intoxication cases. The 13 cases in which
putrefaction was present are not included in the table

Vessel p value CO intoxication AMA

Mean SD Mean SD

Number Total 0.31 10.2 8.76 15.38 13.22

Large caliber 0.52 1.2 1.64 1.75 1.67
Medium caliber 0.11 1.6 1.34 3.13 3.37
Small caliber 0.46 7.4 7.9 10.58 9.68
Intensity Total 0.84 1.6 0.55 1.54 0.66
Large caliber 0.22 0.6 0.89 1.21 1.02
Medium caliber 0.47 1.6 0.55 1.38 0.82
Small caliber 0.45 1.2 0.84 1.54 0.98
124
venules showed the highest number of vessels expressing HIF1-α,
followed by arterioles/venules and, finally, arteries/veins. This
particular and specific distribution pattern of immunostaining
could indicate a role of HIF1-α in the vasoconstriction of
resistance-sized pulmonary arteries induced by hypoxia in the
lungs [12, 13]. Arteries were activated particularly in hanging
deaths, in which the number of vessels and the intensity of the
staining was also higher in comparison with the other AMA
deaths. A reason for this may be that in hanging hypoxia is often
associated with hypoxemia, which emphasizes the reaction of the
endothelial cells. Apart from hanging, medium- and small-sized
vessels were activated particularly in CO intoxication, which may
be dependent on the progressive hypoxemia determined by the
increased accumulation of carboxyhemoglobin.
Conclusion
To evaluate different types of morphological changes in
combination is particularly useful when determining the
mechanism of death in suspected asphyxiations, especially
when external signs undergo atypical changes [36]. Our
results indicate that P- and E-selectin expression in lung
vessels, being activated by several types of trigger stimuli,
cannot be used as indicator of asphyxia. Intra-alveolar gran-
ular deposits of SP-A seem to be related to an intense
hypoxic stimulus, and when massively present, they can
suggest, together with other elements, a severe hypoxia as
the mechanism of death.
The most important results of our study concern, howev-
er, the presence, localization, and distribution pattern of
HIF1-α in lung vessels after hypoxic stimuli. HIF1-α was
expressed in small-, medium-, and large-caliber lung vessels
of the vast majority of mechanical asphyxia deaths and CO
intoxications, with the number and intensity of positive-
stained vessels increasing with the duration of the hypoxia.
Although further confirmation studies are required, these
preliminary data indicate an interesting potential utility of
HIF1-α as a screening test for asphyxia deaths.
References
1. Dettmeyer RB (2011) Forensic histopathology. Springer, New
York, 9783642206580
2. Strunk T, Hamacher D, Schulz R, Brinkmann B (2010) Reaction
patterns of pulmonary macrophages in protracted asphyxiation. Int
J Legal Med Nov 124(6):559–68
3. Schmeling A, Fracasso T, Pragst F, Tsokos M, Wirth I (2009)
Unassisted smothering in a pillow. Int J Legal Med Nov 123(6):517–9
4. Du Chesne A, Cecchi R, Püschel K, Brinkmann B (1996)
Macrophage subtype patterns in protracted asphyxiation. Int J
Leg Med 109:163–166
Int J Legal Med (2014) 128:117–125
5. Betz P, Beier G, Eisenmenger W (1994) Pulmonary giant cells and
traumatic asphyxia. Int J Leg Med 106:258–61
6. Vacchiano G, D’Armiento F, Torino R (2001) Is the appearance of
macrophages in pulmonary tissue related to time of asphyxia?
Forensic Sci Int 115:9–14
7. Vacchiano G (2002) Letter to the editor. Role of pulmonary mac-
rophages and giant cells in fatal asphyxia. Forensic Sci Int 127:245
8. Grellner W, Madea B (1996) Immunohistochemical characteriza-
tion of alveolar macrophages and pulmonary giant cells in fatal
asphyxia. Forensic Sci Int 79(3):205–13
9. Cecchi R (2010) Estimating wound age: looking into the future. Int
J Legal Med 124:523–536
10. Cecchi R, Aromatario M, Frati P, Lucidi D, Ciallella C (2012)
Death due to crush injuries in a compactor truck: vitality assess-
ment by immunohistochemistry. Int J Legal Med 126(6):957–60
11. Madjdpour C, Jewell UR, Kneller S, Ziegler U, Schwendener R,
Booy C, Kläusli L, Pasch T, Schimmer RC, Beck-Schimmer B
(2003) Decreased alveolar oxygen induces lung inflammation. Am
J Physiol Lung Cell Mol Physiol 284(2):L360–7
12. Yu A, Frid MG, Shimoda LA, Wiener CM, Stenmark K, Semenza
GL (1998) Temporal, spatial, and oxygen-regulated expression of
hypoxia-inducible factor-1 in the lung. Am J Physiol 275:818–826
13. Hansen TN, Le Blanc AL, Gest AL (1985) Hypoxia and angioten-
sin II infusion redistribute lung blood flow in lambs. J ApplPhysiol
58:812–818
14. Ortmann C, Brinkmann B (1997) The expression of P-Selectin in
inflammatory and non-inflammatory lung tissue. Int J Legal Med
110:155–158
15. Kuebler WM (2006) Selectins revisited: the emerging role of
platelets in inflammatory lung disease. J Clin Invest 116:3106–8
16. Zhu BL, Ishida K, Quan L, Fujita MQ, Maeda H (2000)
Immunohistochemistry of pulmonary surfactant apoprotein A in
forensic autopsy: reassessment in relation to the causes of death.
For Sci International 113:193–197
17. Zhu BL, Ishida K, Quan L, Fujita MQ, Maeda H (2000)
Immunohistochemistry of pulmonary surfactant-associated protein
A in acute respiratory distress syndrome. Leg Med 3:134–40
18. Zhu BL, Ishida K, Quan L, Fujita MQ, Maeda H (2000)
Immunohistochemical investigation of a pulmonary surfactant in
fatal mechanical asphyxia. Int J Legal Med 113:268–71
19. Maeda H, Fujita MQ, Zhu BL, Ishida K, Quan L, Oritani S,
Taniguchi M (2003) Pulmonary surfactant-associated protein A
as a marker of respiratory distress in forensic pathology: assess-
ment of the immunohistochemical and biochemical finding. Leg
Med 5:318–321
20. Schofield D, Cotran RS (1994) Respiratory distress syndrome in
newborn. In: Cotran RS, Kumar V, Robbins SL (eds) Robbins
pathologic basis of disease, 5th edn. Saunders, Philadelphia, pp
444–446
21. Lewis JF, Jobe AH (1993) Surfactant and the adult respiratory
distress syndrome. Am Rev RespirDis 147:218–233
22. Gearing AJH, Newman W (1993) Circulating adhesion molecules
in disease. ImmunolToday 14:506–512
23. Semenza GL, Agani F, Iyer N, Jiang BH, Leung S, Wiener C, Yu A
(1998) Hypoxia-inducible factor 1-α from molecular biology to
cardiopulmonary physiology. Chest 114(1 Suppl):40S–45S
24. Palmer LA, Semenza GL, Stoler MH, Johns RA (1998) Hypoxia
induces type II NOS gene expression in pulmonary artery endo-
thelial cells via HIF-1. Am J Physiol 274:L212–L219, Lung Cell
Mol Physiol 18
25. Wiener CM, Booth G, Semenza GL (1996) In vivo expression of
mRNAs encoding hypoxia-inducible factor 1. Biochem Biophys
Res Commun 225:485–488
26. BelAiba RS, Bonello S, Zahringer C, Schmidt S, Hess J, Kietzmann
T, Görlach A (2007) Hypoxia up-regulates hypoxia-inducible factor
1-α transcription by involving phosfphatidylinositol 3-kinase and
Int J Legal Med (2014) 128:117–125
nuclear factor kB in pulmonary artery smooth muscle cells. Mol Biol
Cell 18:4691–4697
27. Chandel NS, McClintock DS, Feliciano CE, Wood TM, Melendez
JA, Rodriguez AM, Schumacker PT (2000) Reactive oxygen spe-
cies generated at mitochondrial complex III stabilize hypoxia-
inducible factor-1alpha during hypoxia: a mechanism of O2 sens-
ing. J Biol Chem 275:25130–25138
28. Rabbani ZN, Mi J, Zhang Y, Delong M, Jackson IL, Fleckenstein
K, Salahuddin FK, Zhang X, Clary B, Anscher MS, Vujaskovic Z
(2010) Hypoxia inducible factor 1alpha signaling in fractionated
radiation-induced lung injury: role of oxidative stress and tissue
hypoxia. Radiat Res 173(2):165–74
29. Semenza GL (2003) Targeting HIF-1 for cancer therapy. Nat Rev
Cancer 3:721–732
30. Zhong H, DeMarzo AM, Laughner E, Lim M, Hilton DA, Zagzag
D, Buechler P, Isaacs WB, Semenza GL, Simons JW (1999)
Overexpression of hypoxia-inducible factor 1alpha in common
human cancers and their metastases. Cancer Res 59:5830–5831
31. Ioannou M (2009) Hypoxia inducible factor-1alpha and vascular
endothelial growth factor in biopsies of small cell lung carcinoma.
Lung 187:321–329
125
32. Dai C, Gao Q, Qiu S, Ju M, Cai M, Xu Y, Zhou J, Zhang B,
Fan J (2009) Hypoxia-inducible factor-1 alpha, in association
with inflammation, angiogenesis and MYC, is a critical prog-
nostic factor in patients with HCC after surgery. BMC Cancer
9(1):418
33. Cecchi R, Cipolloni L, Sestili C, Aromatario M, Ciallella C (2012)
Pulmonary embolisation of bone fragments from penetrating cra-
nial gunshot wounds. Int J Legal Med 126(3):473–6
34. Ferranti F, Muciaccia B, Ricci G, Dovere L, Canipari R, Magliocca
F, Stefanini M, Catizone A, Vicini E (2012) Glial cell line-derived
neurotrophic factor promotes invasive behaviour in testicular
seminoma cells. Int J Androl 35(5):758–768
35. Welch BL (1947) The generalization of “student’s” problem when
several different population variances are involved. Biometrika
34(1–2):28–35
36. Meyer FS, Trübner K, Schöpfer J, Zimmer G, Schmidt E, Püschel
K, Vennemann M, Bajanowski T, Asphyxia Study Group, Althaus
L, Bach P, Banaschak S, Cordes O, Dettmeyer SR, Dressler J, Gahr
B, Grellner W, Héroux V, Mützel E, Tatschner T, Zack F, Zedler B
(2012) Accidental mechanical asphyxia of children in Germany
between 2000 and 2008. Int J Legal Med 126:765–771