ORIGINAL ARTICLE
Temperature Measurement From the Brain
and Rectum in Charred Corpses
A Pilot Study on an Animal Model
Daniele Gibelli, MD, PhD,* Angela Cantatore, PhD,Þ Davide Porta, BSc,* Pasquale Poppa, BSc, PhD,*
Remo Sala, PhD,Þ Grandi Marco, MD,* and Cristina Cattaneo, MD, PhD*
Abstract: Measurement of body temperature provides relevant data on
postmortem interval, and different studies have been so far attempted
to apply temperature assessment methods also under extreme environ-
mental conditions; however, none of them has been performed yet on
charred or heated bodies, where temperature measurement is pre-
sumed to be unreliable because of the possible influence of heating. This
study aimed at verifying any possible early-stage alterations of rectal
and endocranial temperature due to fire on an animal model during
the charring process.
Three pigs, 2 adults (pigs 1 and 2) whose weight was about 50 kg
each and 1 piglet weighing 3 kg, were heated and burnt on a natural fire
lit on top of a wooden stack, without the use of accelerants; 2 thermo-
couples were positioned in the rectum and in the cranium to record
second-by-second rectal and endocranial temperature values. Results
demonstrate that the rectal temperature does not seem to increase in
adult pigs for 40 to 50 minutes after the body has been exposed to fire,
probably because of the thermal insulating characteristics of the adi-
pose tissue. Therefore, temperature may still be of some help for es-
timating postmortem interval on heated or burnt cadavers.
Key Words: forensic pathology, carbonization, postmortem interval,
endocranial temperature, rectal temperature
(Am J Forensic Med Pathol 2014;35: 34Y37)
D etermination of postmortem interval (PMI) is a matter
of the utmost importance in forensic science. Among the
main postmortal signs, body cooling is supposed to be the most
reliable, because it can be measured by a scalar quantity, unlike
rigor and livor, which are evaluated by subjective parameters;
the measurement of environmental and body temperature allows
one to estimate time since death by different methods,1Y3 also
accounting for other parameters such as body mass, coverage
or not of the corpse being examined, exposition to streams of
air, and so on. The most complete method of time-since-death
estimation seems to be Henssge nomogram, which considers
different individual and environmental factors that can influence
Manuscript received October 15, 2012; accepted April 21, 2013.
From the *LABANOF, Laboratorio di Antropologia e Odontologia Forense,
Sezione di Medicina Legale, DMUYDipartimento di Scienze
Biomediche per la Salute, Università degli Studi di Milano; and
†Dipartimento di Meccanica, Sezione Misure Meccaniche e Termiche,
Politecnico di Milano, Milan, Italy.
The authors report no conflicts of interest.
Reprints: Cristina Cattaneo, MD, PhD, LABANOF, Laboratorio di
Antropologia e Odontologia ForenseSezione di Medicina Legale,
DMUYDipartimento di Scienze Biomediche per la Salute, Universita
degli Studi di Milano, V. Mangiagalli 37, Milan, Italy.
E-mail: cristina.cattaneo@unimi.it.
Copyright * 2014 by Lippincott Williams & Wilkins
ISSN: 0195-7910/14/3501Y0034
DOI: 10.1097/PAF.0000000000000066
34 www.amjforensicmedicine.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
temperature decrease.4 A sigmoidal shape or rectal temperature
degree curve was originally derived, expressed by a double ex-
ponential equation,5 and recently improved by the triple ex-
ponential formula method, which allows one to estimate time
since death by a single rectal temperature measurement throughout
a thermographic system.6 Rectal temperature is considered the
easiest and most reliable measurement at the crime scene, but
several experimental projects have shown that it is affected by
a wide plateau phase, and temperature may even increase be-
cause of residual cellular activity,7,8 as demonstrated by the
sigmoidal cooling trend.
Another temperature measurement site is the brain, where
the first temperature variations are supposed to be less in-
fluenced by environmental variables, with a higher reliability
of PMI estimation. This method seems to be reliable, especially
under 15 hours after death.9 In detail, brain temperature is
supposed to be less influenced by individual and environmental
variables; literature has shown that brain cooling is faster than
rectal cooling and is affected by a limited plateau phase during
the early cooling period,10 with a higher correlation with time
since death.
However, the ‘‘crime scene’’ almost never matches stan-
dard environmental conditions; sometimes, the forensic patholo-
gist is asked to fix a time-since-death estimation on submerged
or charred bodies; methods for determining time since death,
based on temperature measurement, have been applied so far
to submerged corpses, by experimental11 and animal models,12
with conflicting results, but nothing at all actually exists for
bodies affected by carbonization.
No experimental study has been carried out so far to verify
brain and rectal temperature alteration in burning bodies. In
fact, it is a common assumption that body temperature in
corpses that have been exposed to heat cannot be used. The
only data available from literature deal with cases of fire vic-
tims, and estimation of time since death is provided by means
of rectal temperature measurement13; this experience pointed
out a high feasibility of rectal temperature measurements and
showed a strong influence of the individual mass.
The current study aims at testing for the first time the
modifications of temperature affecting the rectum and brain due
to the external heating, to verify if it is feasible to consider ap-
plying methods related to cooling even on cadavers exposed to
heat sources. From this point of view, this study can be consid-
ered as a first pilot experiment to evaluate if the thermal analysis
of brain and rectum temperature in charred bodies may poten-
tially still provide useful information for the estimation of PMI.
MATERIALS AND METHODS
Carbonization Process
Three pigs, 2 adults (pigs 1 and 2) whose weight was about
50 kg each and 1 piglet weighing 3 kg, were chosen for the
Am J Forensic Med Pathol & Volume 35, Number 1, March 2014
Am J Forensic Med Pathol & Volume 35, Number 1, March 2014
experimental study. All animals had died of natural reasons,
independent from the needs of the experimental study.
Two thermocouples were positioned in the rectum and in
the cranium to record rectal and endocranial temperature values
second-by-second during the burning process; the endocranial
thermocouple was forced throughout a small hole in the frontal
region made by a drill. The contact between the brain and the
external envelope of the thermocouple was adequate to guar-
antee an optimal thermic conduction; in addition, the thermo-
couple was inserted to prevent any contact with the external
environment and to avoid any modification of brain tempera-
ture, which may derive from the cranial hole.
The pigs were put on a natural fire lit on top of a wooden
stack: the carbonization process was performed without the
use of accelerants. The same carbonization process was applied
to all the 3 samples. For adult pig 1, the carbonization pro-
cess lasted until the fire naturally extinguished, approximately
3 hours 46 minutes after the beginning of the experiment;
for adult pig 2 and the piglet, the carbonization process was
stopped after 1 hour, when the skin was completely charred and
broken, with the exposition of underlying bone structures. In the
case of adult pig 1, the measurements of rectal and endocranial
temperature went on for 4 hours 46 minutes after the beginning
of the experiment (approximately 1 hour after the end of the
carbonization process). For what concerns the pig 2 and the
piglet, the measurements were stopped 1 hour after the begin-
ning of the experiment, with the end of the charring process.
The starting temperature was 5-C for pig 1 (taken out from
the refrigerator) and 20-C for pig 2 (to have a check on the
initial temperature influence, before the burning process); for
the piglet, the starting temperature was again 20-C.
Thermocouples
The choice of the most suitable type of sensor was per-
formed following the needs of the specific case, to find a safe
and cheap solution and at the same time grant the needed
sensitivity, with small probes for a prompt response and easy
insertion in the animals. The main requirements were a mea-
surement range between 0-C and 1100-C (the estimated maxi-
mum during the charring process), with 5-m-long connection
cables to preserve the operator safety; the chosen probes can
FIGURE 1. Rectal and endocranial temperature measurement variations with time observed in the 2 adult pigs: on the x axis,
the time interval from the beginning of the experiment; on the y axis, the temperature recorded by the thermocouples.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Brain and Rectum Temperature in Charred Corpses
bear high and sudden thermal stress (eg, from the environmental
temperature to fire) and maintain specific characteristics even
in case of chemical aggressions, with a declared uncertainty
lower than 4-C.
Thermocouples are the most commonly used sensors in
the industrial field, thanks to their wide use, short response
times, high mechanical and chemical resistance, and limited
size. Thermocouples are made by joints between 2 metallic
wires of different material, which provide temperature mea-
surements by the Seebeck thermoelectric effect. If 1 of the
2 junctions is kept at a reference temperature, the unknown
temperature of the other junction temperature can be determined
by measuring the voltage produced by the 2 wires at the 2
junctions. The thermoelectric effect at the junctions depends
on the material constituting the 2 wires and the temperature
difference between the measurement and the reference junc-
tions. The chosen solution is a type K thermocouple system
(Chromel, Ni-Cr/Alumel, Ni-Al), which grants a 50-C to 1100-C
measurement range, high resistance in oxidizing environment,
and high reliability, with insulated measurement junction and
a 5-m extension wire.
Temperature data acquisition was achieved by means of
a 2-channel data logger, with 2.2-C accuracy, which was con-
sidered a reasonable uncertainty for the type of experimenta-
tion. The recording unit allows one to register temperature data
from 2 channels, with a sampling interval between 1 second and
1 hour. The chosen sampling frequency was 1 Hz, which per-
mitted 10-hour recording series. Between the different experi-
ments, data were downloaded.
RESULTS
Data obtained from thermocouples during the carbonization
process of pigs 2 and 1 after 1 hour are compared in Figure 1 (on
the x axis, the time interval from the beginning of the carboni-
zation process; on the y axis, the temperature recorded by the
thermocouples).
It is possible to observe that endocranial temperature ex-
hibits a similar trend for both adult pigs, although there is a
relevant delay in the onset of the higher derivative part for pig 1,
which started from 5-C: in addition, while steady-state conditions
www.amjforensicmedicine.com 35
Gibelli et al Am J Forensic Med Pathol & Volume 35, Number 1, March 2014

FIGURE 2. Pig 1: variations in rectal and endocranial temperature during the complete experimental process (until the flames were
completely extinguished): on the x axis, the time interval from the beginning of the experiment; on the y axis, the temperature recorded
by the thermocouples. The arrow shows the end of the experiment.

seem to be reached for pig 2 after 1 hour of the carbonization
process, this condition seems still far to be reached for pig 1; in
both cases, a discontinuity in temperature measurements seems
to mark the onset of a trend change.
The rectal temperatures also had similar trends, but they
showed lower increments with respect to endocranial ones
all through the test. In detail, for pig 1, the rectal temperature
showed a very small variation during the first hour, with a vis-
ible change in slope after 50 minutes from the beginning of the
experimental process (this change is in the order of a few
degrees) (Fig. 2). Conversely, for the same pig, the endocranial
temperature showed a constant increase from 6.4-C to 37.35-C,
which started after 20 minutes from the beginning of the char-
ring process, up to the end of the experiment (4 hour 40 minutes
after the beginning of the experiment).
Both temperature measurements for pig 1 started from 0-C
to 6-C. The main difference between the 2 pigs is that the sec-
ond one began the charring process from ambient temperature
(20-C), whereas the first one from a lower temperature of 5-C.
Also for pig 2, the rectal temperature underwent a slight
variation through the whole test, with an increase from 17.5-C
to 20.6-C during the entire experiment, which lasted 1 hour; the
endocranial temperature showed a continuous increase from 20-C
at the test start, up to 100-C after 26 minutes 31 seconds, then
staying almost constant up to the end of the experiment.
In the case of pig 1, the carbonization process went on for
3 hours 46 minutes and therefore lasted for a longer time than
the pig 2; at end of the first hour, the rectal temperature began
to increase and went up to 57.75-C after 4 hours 30 minutes:
then the temperature remained almost constant until the end
of the experiment. The endocranial temperature after the first
hour went on increasing at a slower rate, reaching 88.2-C after
2 hours 13 minutes, and then slightly decreased to reach the
temperature of 39.05-C at the end of the experiment, when the
fire was completely extinguished (Fig. 2). The anomaly shown
at 1:30:05 was a bias due to an accidental removal of the en-
docranial thermocouple, which was immediately placed in the
correct position. This did not considerably change the overall
thermic analysis.
Rectal and endocranial temperature values from the piglet
are shown in Figure 3; in this case, the rectal temperature
showed a faster increase than the endocranial, which is visible
2 minutes after the start of the charring process; after approxi-
mately 20 minutes, the observed trend is more irregular than

FIGURE 3. Rectal and endocranial temperature variations recorded in the piglet: on the x axis, the time interval from the beginning of
the experiment; on the y axis, the temperature recorded by the thermocouples.

36 www.amjforensicmedicine.com * 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Am J Forensic Med Pathol & Volume 35, Number 1, March 2014
in the case of adult pigs, with variations between 90-C and
170-C; endocranial temperature began to increase 2 minutes
after the onset and reached 100-C after approximately 40 minutes.
In this case, the endocranial temperature showed a smoo-
ther trend all through the entire experimental period, whereas
rectal temperature showed a higher fast rate during the first
10 minutes from the beginning of the experiment, with the
consequent irregular trend discussed above.
DISCUSSION
Burning tests and temperature measurements from the 2
adult pigs and the piglet provided relevant, although prelimi-
nary, information about thermal modifications of bodies during
heating.
In detail, the rectal temperature remained almost unchanged
during the first phase (lasting 1 hour) of the charring process for
both adult pigs, regardless of the starting temperature. This
means that within the first phases of heating/burning the rectal
temperature is very likely not or less affected by heat, probably
because of the natural thermal insulation characteristics pro-
vided by adipose tissue; the endocranial temperature instead
showed wider variations, probably because of the more limited
thermal protection provided by soft tissues. Therefore, rectal
temperature proved to be affected by a higher inertia. This in-
formation points out that rectal temperature is less influenced by
external temperature modifications during the charring process
and underlines that the rectal measurement usually gives de-
layed temperature variations with respect to the corresponding
endocranial measurement. The endocranial temperature there-
fore seems not to show a regular trend. The experiment carried
out on the piglet showed that the sample size is important in
determining the endocranial and rectal temperature variations
during the first hour: this is justified by the lower mass and
consequently by the lower thermal inertia; this can also explain
why in the piglet differences between the 2 measurements
were less visible than in the case of both adult pigs. In detail, the
smaller mass may have induced a faster charring process and
more limited heat insulation by the adipose tissue.
The rectal temperature during the charring process showed
limited variations during the first hour, which are independent
from the starting temperature (5-C or 20-C); this piece of in-
formation demonstrates the actual thermal insulating charac-
teristic of soft tissues surrounding the rectum not only during
the cooling process in well-preserved bodies, but also in the
opposite direction for heated and burnt corpses, and confirms
the physical behavior of rectal temperature in case of thermal
variations.7,8 No precise information concerning the estimation
of PMI in charred bodies can be drawn, but the present results
indicate that the rectal temperature of a body that has been
burnt or heated only for a few minutes may still be of some help
in evaluating time since death; endocranial temperature, on the
other hand, seems to be quickly affected by heat. Widening
the experimental analysis to more samples may provide useful
information concerning the correlation between the thermal mod-
ifications and environmental variables.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Brain and Rectum Temperature in Charred Corpses
This limited experiment clearly cannot alone be used for a
standardization of the phenomenon of bodily cooling in burnt
cadavers; however, the study is, in our knowledge, the first at-
tempt at verifying the temperature changes in charred bodies
and provides some insight concerning this field of research.
The data suggest that the thermal modifications of corpses
during the charring process and the following cooling phase
may be analyzed and quantified to reach useful information
concerning PMI.
These technical data, although not enough to fill the gap in
literature, justify further analysis of the effects of heating on
cadavers. The study represents only an initial but necessary step
toward understanding the role of cooling as a PMI indicator in
bodies exposed to fire or heat.
REFERENCES
1. Simonsen J, Voict J, Jeppesen N. Determination of the time of
death by continuous post mortem temperature measurements.
Med Sci Law. 1977;17:121Y122.
2. Morgan C, Nokes LDM, Williams JH, et al. Estimation of the post
mortem period by multiple site temperature measurements and the use
of a new algorithm. Forensic Sci Int. 1988;39:89Y95.
3. Nokes LDM, Flint T, Williams JH, et al. The application of eight
reported temperature-based algorithms to calculate the post mortem
interval. Forensic Sci Int. 1992;54:109Y125.
4. Knight B. The evolution of methods for estimating the time of death
using the double exponential cooling model. Med Sci Law.
1988;25:223Y227.
5. Al Alousi M, Anderson RA. Microwave thermography in forensic
medicine. Police Surg. 1986;30:30Y42.
6. Hutchins GM. Body temperature is elevated in the early post mortem
period. Hum Pathol. 1985;16:560Y561.
7. Henssge C. Death time estimation in case work, I: the rectal temperature
time of death nomogram. Forensic Sci Int. 1988;38:209Y236.
8. Henssge C, Madea B. Estimation of the time since death in the early
post-mortem period. Forensic Sci Int. 2004;144:167Y175.
9. Kanawaku Y, Kanetake J, Komiya A, et al. Computer simulation for
postmortem cooling processes in the outer ear. Legal Med.
2007;9:55Y62.
10. Joseph AEA, Schickele E. A general method for assessing factors
controlling post mortem cooling. J Forensic Sci. 1970;15:364Y391.
11. Baccino E, Cattaneo C, Jouineau C, et al. Cooling rates of the ear and
brain in pig heads submerged in water: implications for postmortem
interval estimation of cadavers found in still water. Am J Forensic Med
Pathol. 2007;28:80Y85.
12. Riepert T, Lasczkowski G, Becker J, et al. Extremely varied rectal
temperatures in 10 fatalities after a house fireVa contribution to the
significance of body weight in cadaver cooling. Arch. Kriminol.
1993;191:107Y113.
13. Al-Alousi LM, Anderson RA, Worster DM, et al. Multiple-probe
thermography for estimating the postmortem interval: I. Continuous
monitoring and data analysis of brain, liver, rectal and environmental
temperatures in 117 forensic cases. J Forensic Sci. 2001;46:317Y322.
www.amjforensicmedicine.com 37