The post mortem temperature plateau and its role in the estimation of time
of death. A review
Jimmy L. Smart
, Michał Kaliszan
University of Kentucky, 4810 Alben Barkley Drive, Paducah, KY 42001, USA
Department of Forensic Medicine, Medical University of Gdan
´sk, ul. De˛bowa 23, 80-204 Gdan
Received 13 May 2011
Received in revised form 21 October 2011
Accepted 29 November 2011
Available online 28 January 2012
Time of death (TOD)
Temperature plateau effect (TPE)
Heat transport in human body
Postmortem cooling curve
Postmortem eye temperature
The purpose of this paper was to examine evidence to seek an explanation of the possible cause(s) or con-
tributing factors to the temperature plateau phenomenon and its inﬂuence on time of death (TOD) estima-
tion. The concept of the temperature plateau effect (TPE) is reviewed, and investigation is conducted into
its possible prediction under post mortem conditions. The conclusion of this paper is that the appearance of
a TPE in postmortem body core temperature decay curves is currently random and cannot be predicted.
This unpredictability is based upon the interindividual differences in states (core body temperature, hyper-
thermia, use of drugs, trauma, etc.) and biomarker concentrations (electrolytes, thyroxine, etc.) at antemor-
tem times, which will ultimately affect the shape of the postmortem temperature decay curve. However,
studies indicated that the TPE is diminished or even absent in the head tissues, including eye and ear. The
possibility of precise estimation of the TOD in the early post mortem period based on eye temperature mea-
surements is also commented.
Ó2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction . . . ....................................................................................................... 55
2. Different views on the origin of the temperature plateau effect................................................................. 56
2.1. Temperature control . . . . .......................................................................................... 56
2.2. Factors possibly influencing normal core body temperature . . . . . . . ....................................................... 57
2.3. Metabolism . . . . . . . . . . . .......................................................................................... 57
2.4. Endocrine system . . . . . . .......................................................................................... 57
2.5. Biomarkers. . . . . . . . . . . . ...................................................................................... .... 58
3. Experimental evaluation of body temperature gradients . . . . . ................................................................. 58
3.1. Experiments with wooden sphere . . . . . . . . . .......................................................................... 58
3.2. Experiments with human head . . . . . . . . . . . .......................................... ................................ 59
3.3. Other computer simulation models of body cooling. . . . . . . . . . . . . . ....................................................... 59
4. Process of death ....................................................................................................... 60
4.1. Determination of the moment of death . . . . . .......................................................................... 60
4.2. Supravitality . . . . . . . . . . .......................................................................................... 60
5. Summary . . . . . ....................................................................................................... 61
References . . . . ....................................................................................................... 61
Serious inquiries into estimation, including modeling, of time of
death (TOD) in humans has been ongoing for nearly two hundred
years. These techniques range from simple rules of thumb to elab-
orate mathematical formulas. Results of these models can be useful
to crime scene investigators, but even today, TOD approximations
are not strictly admissible into a court of law with certainties near
that of DNA analyses. Probably the best existing method for esti-
mating time of death during ﬁrst 24 h post mortem  claims a
maximum precision within 95% conﬁdence limits, of ±2.8 h either
1344-6223/$ - see front matter Ó2011 Elsevier Ireland Ltd. All rights reserved.
Corresponding author. Tel.: +48 583491264; fax: +48 583410485.
E-mail address: email@example.com (M. Kaliszan).
Legal Medicine 14 (2012) 55–62
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/legalmed
side of the actual TOD. Many of the existing mathematical models
of the postmortem temperature decay curve perform well in linear
or even single exponential regions of the curve. However, there are
two regions where all models encounter a problem: (1) start of the
cooling period, i.e., not immediately after death, and (2) near the
end of the cooling cycle where the temperature of the body
approaches that of the environment. Issues dealing with the ‘‘start’’
of the cooling period are addressed in this present paper.
Some investigators [2–4] in the mid 1800s reported a reduced
decrease in temperature soon after death, that later decayed more
rapidly. In 1880, Burman  recognized an apparent rise in axillary
temperature upon ‘‘apparent death’’. He went onto say that death
does not take place until the body commences to cool down. It
was not until 1965, that the observed phenomenon of signiﬁcantly
slower cooling of the body in the initial postmortem period was
ﬁrst called the ‘‘temperature plateau’’ . Over the years since that
time, various investigators have either charted a rise in tempera-
ture at TOD, a continuation of body temperature (plateau), or the
anticipated traditional temperature decay upon death [7–10].
Upon death, as a body begins to cool, and heat ﬂows from the
body to the surrounding environment, rectal temperatures do not
drop immediately. A short period of time elaspses as steady state
temperature gradients become established. This is normal temper-
ature decay phenomenon and has been described by many investi-
gators, including demonstration within inanimate geometries .
Some would call this normal temperature decay phenomenon a
temperature plateau. Other investigators  have declared that
there is no temperature plateau in the human body, but only nor-
mal slowed temperature loss in the very early post mortem period.
It is suggested that a true TPE must include increased/decreased
temperature that impact otherwise normal temperature gradients
Very early models of postmortem temperature decay applied
Newton’s law of cooling, and did not sufﬁciently describe the
TPE. Shapiro  proposed a ‘‘rule of thumb’’ to include the TPE. It
states that upon death, the body cools down approximately 1 °C
per hour, but 3 h must be added to the TOD because of the plateau
period. Marshall and Hoare  proposed a two-exponential mod-
el to describe the sigmoid shape of the postmortem temperature
decay proﬁle. This model was able to describe an initial ‘‘tempera-
ture lag’’ as the temperature gradient became established, as well
as a slowdown in temperature decay as the body temperature
approached that of the surrounding environment.
Part of the problem of trying to incorporate the TPE into a pre-
dictive model is its apparent random and erratic nature. Out of
117 forensic cases, Al-Alousi  reported that 26.7% of the total
cases based upon rectal temperatures displayed a temperature
plateau (only 7% based on brain temperatures). Schwarke 
found rectal temperatures above 38 °C in 16.4% of cases. Some-
times, the TPE is short, sometimes long, sometimes nonexistent
. One study  reported recorded rectal temperatures
remaining constant up to a period of 6 h after death. In experi-
ments with animals, the TPE also makes a random appearance.
Some experiments with pigs and dogs have shown a complete ab-
sence or only a residual TPE within the head [17–21]. It is be-
lieved that the TPE may be reduced or nonexistent when
studying temperature decay also within the human head. Baccino
 did not observe the presence of the TPE in humans when
performing experiments based upon outer ear temperature mea-
surements. Henssge  also reported that of all factors that
inﬂuence the early postmortem cooling rate of brain temperature
(weight, height, cranial bone width, head size, clothing, and hair),
only hair was to be considered.
Lundquist  recognized TPE phenomena when dealing with
rectal temperatures. He suggested that when there is evidence of
TPE, to extrapolate back from the steep part of the early temperature
decay curve to time zero (TOD). This would correspond to an initial
temperature of about 40 °C.
Based upon continuous temperature monitoring data, before
and after death, the TPE does not appear to be a step change in
body temperature occurring at the moment of death. Hutchins
 provided data that shows the slope of the body core tempera-
ture approaching death and extending beyond death to be positive,
negative, and ﬂat (plateau).
Obviously at death, blood circulation, vasomotor response, hor-
monal secretions, and respiration cease. However, some residual
effects may linger such as inertial effects of cellular metabolism
or slight skin temperature reduction by evaporation. Some brain
stem activity and nerve functioning have been shown to persist
for some time after death.
According to Henssge , the occurrence of a temperature pla-
teau in the human body is dependent upon the location of temper-
ature measuring site, ambient conditions, surface insulation (hair,
clothing, etc.), and size of the body. Though possibly, the cause of
death should be added as a factor in TPE consideration. Ambient
conditions will establish the temperature gradient between the
skin and the core body temperature and will inﬂuence the degree
and longevity of the TPE. Obese bodies have been shown to have a
much more retarded postmortem temperature decay curve than
lean bodies due to the insulating qualities of surrounding layers
of subcutaneous adipose tissue. Therefore, a longer TPE would be
expected in obese cases.
However, according to some investigators, the determination of
the TOD will never be more than a rough estimate [7,24]. This
assertion is based upon the facts that: (1) there is wide variation
in the initial temperature at TOD (which is unknown and can easily
vary as much as ±2 °C), and (2) the possible presence of the TPE.
In an effort to look for clues for possible sources of contributing
factors that may cause or inﬂuence the TPE, the existing literature
was examined. Topics dealing with temperature control, metabo-
lism, endocrine systems, factors that cause heat production/heat
reduction, hyperthermia/hypothermia, postmortem biomarkers,
supravitality were considered.
2. Different views on the origin of the temperature plateau
2.1. Temperature control
There are two mechanisms of heat regulation in humans: (1)
distinct hot and cold thermal receptors in the skin, and (2) thermo-
regulator regions in the brain that sense changes in blood
Regulation of skin blood ﬂow (vasoconstriction/vasodilation)
serves to adjust heat loss from the body by convection/radiation
and also to provide a ﬂow of heat to be lost during the evaporation
of water from the skin (sweating). Skin blood ﬂow is controlled by
core body temperature and mean skin temperature. With excep-
tion of blood vessels in the forehead, all skin areas vasoconstrict
when exposed to the cold. Certain skin areas on the body contain
more cold receptors or blood ﬂow, and therefore exert a greater
inﬂuence on body temperature regulation. This includes the scro-
tum in men and the mammary area in females.
There does exist evidence of the TPE in living subjects. In anes-
thetized humans, it has been demonstrated that vasoconstriction is
effective in causing a sequestration of metabolic heat to the core to
prevent loss of body core temperature for as much as 3 h .Itis
suggested that heat is redistributed from peripheral thermal com-
partments (legs and arms) to the core thermal compartment. Legs
and arms are considered to be thermal buffers and are estimated to
be able to maintain a core body temperature plateau for up to 6 h.
56 J.L. Smart, M. Kaliszan / Legal Medicine 14 (2012) 55–62
It is emphasized that in colder or warmer environments, the pres-
ence of vasoconstriction can still lead to a loss or elevation of core
body temperature. Unfortunately, this information concerning
anesthetized living humans does not translate into explaining a
postmortem TPE. Once a person dies, there is no evidence for this
same mechanism to continue to redistribute heat and sustain core
What is termed a ‘‘normal’’ body temperature in humans is based
upon what is known as the core temperature (T
). This temperature
is based primarily upon the lower torso and is measured as the rec-
tal temperature. In healthy individuals under normal conditions,
rectal temperatures can vary between 34.2 and 37.6 °C, with a mean
of 36.9 °C. Vital organs in the abdominal region produce about 70%
of the total body heat while at rest. Also, most individuals experi-
ence a diurnal (circadian) rhythm in which the body temperature
ﬂuctuates by ±0.5 °C around the person’s normal body temperature
Though temperatures do vary across locations throughout the
body, temperature differences between various sites within the
trunk do not exceed 0.5 °C. The temperature of any organ depends
upon its metabolic rate and the rate of blood perfusion. Control of
body core temperature appears to be an integrative process, with
signal input from sources throughout the body. These sources in-
clude the rostral hypothalmus, lower brain stem, spinal cord, abdo-
men, and heat/cold sensors distributed throughout the body,
including the skin. The hypothalmus accounts for 10–50% and
the spinal cord 0–15% of the total core temperature signal .
Changes in body heat content determines core body tempera-
ture. Heat production and heat loss both respond exponentially
to change, but the two different exponents have very different time
constants. Within seconds of when a person begins to exercise,
heart rate and oxygen uptake increases, and new steady states
are achieved in minutes. Studies that measure oxygen uptake show
that its response is exponential with a time constant of 30 s. On the
other hand, heat loss after exercise activity is slower to increase
and a new steady state plateau is achieved after 60 min. The time
constant in this case is 10 min. Because of the different time con-
stants of heat production vs. heat loss, there is an alternating inter-
play of accumulation and depletion of body heat. The overall effect
is that core body temperatures rise and fall over time with varying
activity . These antemortem activities could inﬂuence core
body temperatures and temperature gradients that may ultimately
carry over and contribute to a postmortem TPE.
2.2. Factors possibly inﬂuencing normal core body temperature
There are numerous conditions that will cause signiﬁcant devi-
ation from normal core temperature, i.e., conditions that prevent
heat loss or accelerate heat production. These include sunburn,
fainting, heat stroke, heat exhaustion, exercise, emotional stress,
fever, hyperthyroidism, circulatory disorders, and exposure to se-
vere cold/warm environments . Also, there are reported numer-
ous medical conditions that can cause hyperthermia/hypothermia
including infections, rashes, diseases, neoplasms, drugs, endocrine
disorders, and central nervous system disorders .
It is well documented that trauma, drug use, hyperthermia,
hypothermia, anapyrexia, fever, the aged , type 2 diabetes mel-
litus, Raynaud’s phenomena, erythromelalgia, hormonal issues of
postmenopausal women  will all affect the body’s ability to
sense temperature changes and may inﬂuence the postmortem
temperature decay curve. If any of these factors are known, they
may contribute to the presence of a TPE.
It is commonly known that drugs can affect the core body tem-
perature. These include steroids , narcotics, and antidepres-
sants . Though no reliable data is available, an argument
could be made that a signiﬁcant percentage of the population are
taking some medication, either prescribed or over-the-counter, at
any given time. Many of these medications will affect core body
temperature. In the United States, a 2005 report  stated that
11% of women and 5% of men in the non-institutionalized popula-
tion take antidepressants. Almost half of all Americans take at least
one prescription drug . Frequently, many other people take
over-the-counter drugs on a daily basis. In a retrospective study
of 744 cases involving violent death , hyperthermia leading
to elevated body core temperature was identiﬁed in 11% of the
cases. Hyperthermia was attributed to drugs, brain trauma, malig-
nant tumors, suffocation, and infectious diseases.
Sources of heat production in humans are due to muscular
movement and cellular metabolism. Most of the usual generated
heat is from the metabolic pathway, oxidation of carbohydrates
(Krebs cycle). For 85% of all normal individual groups of ages,
weights, and sex, basal metabolic rates remain within 10% of the
mean . However, depending upon internal and/or external con-
ditions, humans have the ability to conduct metabolic switches.
Similar to hibernating animals where body temperatures approach
0°C, humans can also exploit the low temperature activity of pan-
creatic triacylglycerol lipase to combust stored fatty acids for heat
production . Also, at ischemic conditions (temporary or at
TOD), humans can switch to anaerobic glycolysis as a short term
source of heat production.
Normal metabolic heat generated in man has been reported to
be equal to 0.0314 W/kg . Metabolic heat production in rela-
tion to surrounding conditions is 150 W at 10 °C, roughly linearly
dropping off to become ﬂat at 80 W in the 30–40 °C range, and ris-
ing again above 44 °C.
At rest, the musculature of man contributes 18–36% to the over-
all heat production in the body. Remaining heat production origi-
nates in the brain, heart, liver, and other viscera. During exercise
or shivering, these conditions reverse, with musculature providing
about 73% of the heat . To regulate temperature, heat is distrib-
uted throughout the body by blood circulation. Obviously, at death,
normal cellular metabolism (oxidation of carbohydrates) and cir-
culatory heat distribution ceases. Unless body temperature was
elevated at TOD (hyperthermia), these sources would offer no
residual contribution to a postmortem TPE. At death, heat produc-
tion by normal metabolism and muscle action has ended. Also,
there is no further heat loss by cutaneous vasodilation, though per-
haps there may be some very short term losses from the skin by
evaporation of residual sweat. Normal distribution of heat by blood
circulation ceases. If there is a source of heat in the early postmor-
tem period, e.g., anaerobic glycolysis, this energy would necessarily
have to be diffusely distributed by tissue conduction alone to
maintain an elevated core temperature.
2.4. Endocrine system
Horowitz [41,42] has shown that, in rats, either hyperosmolarity
(e.g., due to dehydration or hyperglycemia) or hypovolemia (de-
creased blood volume, e.g., due to hemorrhaging or dehydration)
elevates the body core temperature and decreases the basal meta-
bolic rate by 29%. On the other hand, recent research that controlled
for these confounding variables reported that hyperhydration
(water or glycerol) did not alter core temperature, skin temperature,
whole body sweating rate, local sweating rate, sweating threshold
temperature, sweating sensitivity, or heart rate responses com-
pared to normal water content (euhydration) conditions. If euhy-
dration is maintained during periods of exercise-heat stress, then
hyperhydration appears to have no meaningful advantage .
J.L. Smart, M. Kaliszan / Legal Medicine 14 (2012) 55–62 57
The thyroid gland secretes thyroxine whose presence/absence
can inﬂuence metabolic rate from 40–100% of normal cellular activ-
ity. Stimulation of the sympathetic nervous system with secretion
of epinephrine and norepinephrine can increase the metabolic rate
by as much as 15% in adults . Though deﬁnitive evidence is miss-
ing, it appears the endocrine system may play an important role in
long term acclimation changes in body temperature, rather than in
short term, or immediate response .
Various investigators have attempted to locate chemical bio-
markers in the body to estimate time of death. A good summary
of these efforts  examined blood chemistry (carbohydrates,
nitrogenous compounds, enzymes, electrolytes, and hormones),
electrolyte concentrations in vitreous humor (potassium, sodium,
calcium, chloride, urea, creatinine, and hypoxanthine), cerebrospi-
nal ﬂuid chemistry, pericardial ﬂuid, and synovial ﬂuid. Some of
these markers track well across the postmortem interval. However,
none of these markers are satisfactory in predicting time of death
because of the interindividual differences in initial starting concen-
trations. None of the biomarkers offer any insight into the possible
existence of a temperature plateau at time of death, again because
of interindividual differences.
3. Experimental evaluation of body temperature gradients
One can wonder if the temperature plateau effect is really a
manifestation of a continuous temperature function beginning
near TOD and continuing shortly after death. Various gradients of
the temperature proﬁle should be examined. These gradients can
be qualitatively depicted as shown in Fig. 1. Positive gradients 1
and 2 are attributed to conditions in the body that lead to an ele-
vated core body temperature, such as hyperthermia and elevated
metabolic rate. Gradient 3 represents a true temperature plateau,
where some unidentiﬁed source of heat is maintaining continuity
of postmortem temperature. Negative gradients 4 and 6 can be
caused by conditions that depresses the core body temperature,
such as sweating or hypothermia. Depending upon when a person
expires, normal daily changes in body core temperatures due to
circadian rhythms would offer various slight gradient changes that
might carry over into the postmortem phase. In this present study,
an experiment was designed to examine temperature gradients
and their inertial effects upon temperature decay.
Some investigators have suggested that the TPE is due merely to
the inertial effects of temperature in the body; that once the body
is heated, it tends to remain at that temperature for a short period
after death. Though untested, this seems to be a reasonable expla-
nation and may at least offer a partial contribution to the TPE phe-
nomenon. In an effort to test this idea, a wooden sphere with
similar dimensions and thermal diffusivity properties of a human
head was constructed as previously detailed .
3.1. Experiments with wooden sphere
A 20.3 cm diameter red oak sphere was ﬁtted with a thermocou-
ple at its center. The sphere was placed in the middle of a tempera-
ture-controlled oven. The oven door was closed and the oven
thermostat was adjusted so that the center of the oak sphere was
equilibrated near 37 °C. Separate thermocouples were used to mea-
sure temperatures at the sphere center, the inside oven tempera-
ture, and the outside ambient room temperature. Fig. 2 shows the
performance characteristics of the oven. Even though an analog
thermostat was used to set the inside oven temperature, note how
the internal oven temperature cycles between approximately
33.5–44 °C to maintain a constant temperature at the center of
the wooden sphere. Once an approximate steady state temperature
near 37 °C was achieved, the oak sphere was removed from the oven
and suspended over a laboratory bench. Over time, heat was trans-
ferred from the warm oak sphere (initially at 37 °C) to the cooler
environment of the surrounding air at ambient room conditions
(about 21 °C) in accordance with natural convection and radiation
effects. A temperature data capturing software program recorded
temperature conditions at 10–60 s intervals. This arrangement
was designed to simulate time of death conditions in the human
head and to generate an early postmortem temperature decay
The object of the experiment was to: (1) verify the existence of a
TPE, and (2) if it does exist, investigate the gradient, dT/dt, from
Fourier’s law of heat conduction. Does a cooling or heating gradient
within the sphere affect the nature of such a TPE? In other words,
at TOD, does the fact that the body temperature control mecha-
nism is heating or cooling the human body generate or inﬂuence
the extent of a TPE?
Fig. 3 shows the results of the upgradient temperature experi-
ment. Initially, the wooden sphere was at room temperature. It
was placed in the oven and allowed to gradually heat up to near
37 °C. The temperature gradient was a +0.011 °C/min. After
132 min, the oven door was opened and the sphere was removed
and relocated to ambient room conditions (20 °C). Once out of
the oven, since there was a positive temperature heating gradient
within the sphere, temperature inertia allowed the center of the
sphere to apparently continue heating and gain another 0.1 °C over
a period of 28 min. After 28 min, the temperature at the center of
the sphere began to decay.
Fig. 4 shows the results of the downgradient experiment. In
this case, the sphere was resting in the oven at a steady state tem-
perature of about 37.5 °C. A small reduction in the center sphere
Fig. 1. Qualitative hypothetical temperature gradients for various thermal condi-
tions in the human body that extend beyond time of death.
58 J.L. Smart, M. Kaliszan / Legal Medicine 14 (2012) 55–62
temperature was initiated by slightly reducing the thermostat
setting—as evidenced by the lower oven cycling temperatures. An
attempt was made to match the slope of the cooling gradient
(though in an opposite direction) with the slope of the heating
gradient in the previous experiment. The slope of the cooling
gradient was 0.009 °C/min. After 372 min, the oven door was
opened and the sphere was removed and relocated to ambient room
conditions (19.6 °C). After 18 min, the temperature at the center of
the sphere began to decay.
Though these experiments were repeated several times, it was
too difﬁcult to achieve consistency across experiments by control-
ling ambient room conditions and matching slopes of upgradient
heating/downgradient cooling to warrant statistical evaluation.
The primary purpose of these experiments was to examine trends.
It was demonstrated that when the sphere was heating (upgradi-
ent heating), that once the sphere was removed to ambient condi-
tions, the temperature continued to climb slightly, adding 0.1 °C.
During the downgradient experiment, when the sphere was re-
moved to ambient room conditions, the sphere began cooling
10 min, or 36% sooner than when removed from the upgradient
conditions. The temperature plateau effect was still evidenced,
though not for as lengthy a period .
What do these experiments with a wooden sphere have to do
with any temperature plateau effect that may be occurring in the
human body? The wooden sphere has dimensions and thermal dif-
fusivity properties similar to the human head. One investigator has
recorded temperature plateaus lasting as long as 1.5 h in the human
head . Upon death, due to the brevity of inertial effects, it is sug-
gested a true TPE cannot be signiﬁcantly sustained in the human
head by temperature inertial effects or by upgradient heating
caused by the body’s temperature control mechanism. They may
contribute to the TPE, but they are not its sole source. In conclusion,
though intuitive, it is interesting to demonstrate through these
experiments that a slightly more persistent temperature plateau
can be generated with upgradient heating than with downgradient
3.2. Experiments with human head
Recently, a ﬁnite element method utilizing Newton’s law of
cooling with an average radiation/convection coefﬁcient was pro-
posed to estimate time of death in humans based upon post mor-
tem temperature decay in the eyeball . With initial application
of the method, most of the postmortem temperature decay curve
of a recently deceased person was reconstructed. In the ﬁnal stage
of conforming the model temperature decay curve to be congruent
with the actual postmortem decay curve, slight adjustments were
made. These adjustments were necessary when either an invalid
starting temperature of 37 °C was assumed or there was a TPE.
Out of the small sample of ten experimental cases, 30% of the cases
were adjusted for TPE. Though the ﬁnite element method per-
formed reasonably well (R
= 0.9448–0.9980 congruency between
models and experimental cases), the use of a ﬁtting parameter is
a nagging issue.
3.3. Other computer simulation models of body cooling
Mall et al. [45,46] proposed a three dimensional heat-ﬂow ﬁnite
element model for estimation of the time of death. The model was
successfully validated using the experimental data from Marshall
and Hoare . The application of the ﬁnite element method to
postmortem cooling allows one to model the shape of the human
body, including various tissue compartments. The method also
can be adjusted to include layers of clothing, variation of the initial
Fig. 2. Performance characteristics of a convection oven used in thermal gradient
Fig. 3. Upgradient temperature experiment. Temperature continues to coast
upward (upgradient is +0.011 °C/min) for an additional 28 min before it starts to
decay. Over this period, the temperature within the sphere approximately increased
by an additional 0.1 °C.
J.L. Smart, M. Kaliszan / Legal Medicine 14 (2012) 55–62 59
body temperature, and calculation of the heat transfer in natural
and forced convection (including any thermal radiation effects).
In conclusion, Mall, et al. stated that results obtained from using
the ﬁnite element model of body cooling considerably improved
Hiraiwa et al.  in his studies on computer modeling of rectal
cooling was able to include variable environmental temperatures
in TOD calculations by applying the ﬁnite difference method.
Kanawaku et al.  performed computer simulation of post-
mortem processes in the outer ear based upon one experimental
case. Cooling patterns were analyzed using a 3D head model built
form CT images of the brain. He used a ﬁnite element model in TOD
calculations. When the heat transmission coefﬁcient for the outer
ear was set as 6 W/m
°C, the temperature proﬁle nearly over-
lapped with that of the case subject. He did not observe signiﬁcant
TPE during the outer ear cooling process.
4. Process of death
4.1. Determination of the moment of death
Historically, attempts to deﬁne the exact moment of a human’s
death have been fraught with problems. Death was once deﬁned as
the cessation of heartbeat (cardiac arrest) and of breathing, but the
development of cardiopulmonary resuscitation (CPR) and deﬁbril-
lation procedures have rendered that deﬁnition inadequate.
Breathing and heartbeat can sometimes be restarted. There are
many anecdotal references to people being declared dead by doc-
tors and then ‘‘coming back to life’’, sometimes days later in their
own cofﬁn, or when embalming procedures were about to begin.
In cases of electric shock, CPR performed for an hour or longer
can allow stunned nerves to recover, allowing an apparently dead
person to survive. People found unconscious under icy water may
survive if their faces are kept continuously cold until they arrive at
an emergency room.
Today, medical personnel usually rely upon the concept of
‘‘brain death’’ or ‘‘biological death’’ to deﬁne when a person is clin-
ically dead. In intensive care units, patients are considered dead
when the activity of their brain ceases. Hospitals have protocols
for determining brain death at widely separated intervals under
deﬁned conditions. In the ﬁeld, most coroners and medical exam-
iners do not have access to sophisticated medical instruments, so
usually, in the early postmortem period, they rely upon fundamen-
tal signs of cessation of heartbeat/respiration and obvious signs of
death including hypostasis and muscle rigidity. When death occurs
outside a hospital environment, in some cases, a patient or victim
may be pronounced dead, but there is continued lower brain stem
activity. Perhaps this brain activity and residual tissue metabolism
can inﬂuence metabolic activities or biochemical reactions that
might produce additional residual heat to create a postmortem
Supravitality is an area of research that deals with survival rates
of tissue after complete irreversible ischemia. Tissues can show re-
sponse to various stimuli even long after loss of blood ﬂow. Notice,
even though the issue of ischemia does ﬁgure into the deﬁnition of
death (circulatory response), it does not deﬁne ﬁnal death, as does
brain function. Madea  discussed various bodily functions that
continue post mortem. For example, he points out that the resusci-
tation period for the heart under normal conditions is up to 4 min
post ischemic, but supravital electrical excitability of cardiac mus-
cle can be preserved up to 2 h postmortem.
In another investigation , human stools were evaluated to
determine if intestinal bacteria could be a source of a post mortem
heat production. A mean temperature increase of 0.3 °C was attrib-
uted to anaerobic fermentation when stools were incubated at
37 °C for 6 h; the temperature rise was not believed to be signiﬁ-
cant and not a contributing factor to any postmortem TPE.
Most supravital activity (delayed muscle response, electrical
excitability of the iris, tissue response to various chemical agents,
etc.) does not belong into a discussion of the post mortem TPE.
However, one area worthy of investigation deals with anaerobic
glycolysis. One investigator  has speculated that glycogenolysis
might be responsible for maintaining temperature equilibrium in
the human corpse shortly after death. Glycogenolysis is the release
of energy (ATP) in the absence of oxygen, or anaerobic glycolysis.
When blood circulation ceases and oxygen becomes unavailable
for cellular oxidation of glucose, a small amount of energy can still
be released by the cells by glycolysis. This conversion is inefﬁcient
and only releases about 3% of the total energy that would be nor-
mally available from the complete oxidation of glucose .
Anaerobic glycolysis is one of four evolutionary anaerobic met-
abolic pathways available to invertebrates and vertebrates. Of the
four paths, anaerobic glycolysis is the least efﬁcient, but yields
the highest rate of energy production. From an evolutionary stand-
point, in arthropods, echinoderms, and chordates, it appears anaer-
obic glycolysis was selected based upon the development of
circulatory systems and efﬁcient methods of removing lactate. In
more developed animals, including man, anaerobic glycolysis also
has the advantage of providing an alternate pathway for high rates
of energy production for explosive and sustained muscular activi-
ties . The alternate pathway also provides a convenient safety
buffer for easy shuttle between aerobic and anaerobic conditions
caused by brief periods (5–10 min) of anoxia and ischemia .
Fig. 4. Downgradient temperature experiment. Temperature continues to maintain
downward gradient (downgradient is 0.009 °C/min) for an additional 18 min
before it starts to decay. Over this period, the temperature within the sphere
approximately decreased by 0.1 °C.
60 J.L. Smart, M. Kaliszan / Legal Medicine 14 (2012) 55–62
It has been reported that post mortem anaerobic glycolytic
metabolism can continue for periods up to 10 h after death .
Lundquist  and Mall  estimated excess energy available
from biochemical reactions during the supravital period. Lundquist
estimated the amount of heat evolved from glycolysis after death
sufﬁcient to raise the body temperature by 2 °C. Mall estimated ex-
cess thermal energy in the ﬁrst hour after death equivalent to a
3.1 °C rise for a lean person and a 2.4 °C rise for an obese person.
Even though Mall offers justiﬁcation for the presence of the TPE,
it is never explained why the TPE is not pervasive in all persons
and why it is reported to occur in only about 1/4 of all postmortem
rectal temperature decay proﬁles.
If anaerobic glycolysis is the primary source of the TPE, why is the
phenomenon inconsistent and unpredictable? Many studies [53,55]
have shown elevated concentrations of lactate and lactic acid accu-
mulating in tissues after death. However, these levels depend upon
tissue conditions, especially upon pH and glycogen stores. Appar-
ently, there are no absolute values of lactate concentrations that
would point toward high or low degrees of glycolysis; all values
are relative and vary depending upon interindividual differences.
Post mortem levels of lactic acid are also dependent upon cause of
death, with very high levels noted in victims of crime and trauma
(freezing, hanging, burning, drowning, and poisoning) [56,57].
The TPE may also be inﬂuenced by what is called by Maeda et al.
‘‘pathophysiological vital reactions’’  – systemic pathophysio-
logical changes involved in a death process that cannot be usually
detected by morphological methods.
"TPE generally observed in core body temperature is probably
generated primarily by anaerobic glycolysis. It is not clear
why TPE is not consistent and present in all cases, but there
may be a relationship between residual brain activity that
continues after death and glycolysis activity. Also, intensity
and continuation of glycolysis is dependent upon amount
of glycogen stores in the body at time of death. Currently,
there does not appear to be any biomarkers or other tests
identiﬁed that would indicate the degree of anaerobic glycol-
ysis after death.
"Core body temperature is not signiﬁcantly inﬂuenced over
the short run by hormonal secretions.
"Core body temperature can be inﬂuenced over short periods
of time by being reduced by sweating/vasodilation and
raised by vasoconstriction.
"There are many concentrations of various substances (elec-
trolytes, lactic acid, etc.) that continue to change after death
and these changes can often be regressed to time of death.
Initially, one might believe that someday, someone will dis-
cover a suitable chemical biomarker within the human body
whose concentrations can be quantitatively tracked over
time to use to predict time of death, e.g., potassium concen-
trations in vitreous humor. However, there is always the
problem of interindividual differences. The question will
always be: what was the initial concentration of the sub-
stance in a given person at their TOD? Although normal
ranges of these concentrations can be established in humans,
precise individual antemortem concentrations cannot be
known at post mortem conditions. There is too much vari-
ability in any biomarker concentrations throughout the pop-
ulation. Therefore, TOD cannot be determined and
temperature plateau effects, if any, cannot be predicted from
this approach. Also, it is not understood what affect concen-
trations of various molecules (electrolytes, proteins, precur-
sors, etc.) have upon the core temperature, which might
contribute to a temperature plateau effect. Although, indi-
vidual concentrations of various species may not be helpful
in determining time of death, perhaps ratios of specie con-
centration or even gradients of changing concentrations
may offer some value at a future date.
"It is known that various abnormal conditions do affect the
core body temperature. These include hypothermia, hyper-
thermia, fever, exercise, shivering, drugs, and physical/emo-
tional trauma. These and other factors could raise/lower the
core body temperature from a normal starting point of 37 °C
and could create/inﬂuence the presence of the TPE.
"In many cases, the medical history of the recently deceased
individual is not known. Generally, it is not known if the per-
son had a fever, was on antidepressants, or had recently
ingested some other drug (though this could be determined
in blood analysis). In most existing predictive models for
generation of postmortem temperature curves, it is assumed
that a person at death has normal core temperature and is
not suffering from any abnormal condition or has ingested
any substance that will inﬂuence the core temperature. If
any of these antemortem conditions were known to have
existed, they certainly could be incorporated into a predic-
tive model. Given, that so many drugs/substances and other
various conditions affect the core body temperature, it is sta-
tistically likely that a signiﬁcant portion of the population
will be experiencing a temperature elevation, a temperature
plateau, or even a temperature reduction just prior to death.
However, outside of a person implanted with a microchip,
similar to the black box in airplanes, that records body tem-
perature, precise antemortem conditions will never be
known at TOD for most deceased individuals. Therefore,
the existence of any temperature plateau effects will proba-
bly have to be estimated from ﬁeld collected data, whether it
be a portion of the actual postmortem temperature decay
curve or the measurement of concentration of some internal
biomarker molecule—today, as yet to be determined.
"It is preferred that any model that seeks to predict TOD of a
recently deceased individual should be based upon a number
of ﬁeld collected temperature measurements under similar
surrounding conditions. With this data, the postmortem
temperature decay curve can be reconstructed and the curve
can be extrapolated back to time zero to estimate TOD. This
procedure will offer the advantage of addressing issues of
temperature plateau effect and/or elevated core body tem-
perature (hyperthermia) .
"The TPE has been found to be diminished or absent in head
tissues, including the eyeball. Therefore, TOD estimation in
the early post mortem period can be improved when tem-
perature decay models are based upon temperature proﬁles
in head tissues vs. core body temperatures.
 Knight B, editor. The estimation of the time since death in the early
postmortem period. London: Arnold; 2002.
 Taylor A, Wilkes D. On the cooling of the human body after death. Guy’s Hosp
 Seydeler R. Nekrothermornetrie. Vierteljahrschrift fur die Praktische
 Rainy H. On the cooling of dead bodies as indicating the length of time since
death. Glasgow Med J 1869;1:323–30.
 Burman J. On the rate of cooling of the human body after death. Edin Med J
 Shapiro HA. The post-mortem temperature plateau. J Forensic Med
 Simonsen J, Jorgen J, Jeppsesn N. Determination of the time of death by
continuous post-mortem temperature measurements. Med Sci Law
J.L. Smart, M. Kaliszan / Legal Medicine 14 (2012) 55–62 61
 Olaisen B. Postmortem decrease in brain temperature. Z Rechtsmed
 Hutchins GM. Body temperature is elevated in the early postmortem period.
Hum Pathol 1985;16:560–1.
 Gulyas B, Dobai J, Szilagyi G, Csecsei G, Szekely G. Continuous monitoring of
post mortem temperature changes in the human brain. Neurochem Res
 Marshall T, Hoare F. Estimating the time of death - the rectal cooling after
death and its mathematical representation. J Forensic Sci 1962;7:56–81.
 Smart JL. Estimation of time of death with a fourier series unsteady state heat
transfer model. J Forensic Sci 2010;55:1481–7.
 Joseph AEA, Schickele MS. A general method for assessing factors controlling
postmortem cooling. J Forensic Sci 1970;15(3):364–91.
 Al-Alousi LM. A study of the shape of the post-mortem cooling curve in 117
forensic cases. Forensic Sci Int 2002;125:237–44.
 Schwarke R. Deutsche Zeitschrift für Thlermedizine. Gericht Med
 Lundquist F. Physical and chemical methods for the estimation of the time of
death. Acta Med Leg Soc 1956;9:205–13.
 Kaliszan M, Hauser R, Kaliszan R, Wiczling P, Buczynski J, Penkowski M.
Veriﬁcation of the exponential model of body temperature decrease after
death in pigs. Exp Physiol 2005;90:727–38.
 Baccino E, Cattaneo C, Jouineau C, Poudoulec J, Martrille L. Cooling rates of the
ear and brain in pig heads submerged in water. Am J Forensic Med Pathol
 Kuehn LA, Tikuisis P, Livingstone S, Limmer R. Body cooling after death. Aviat
Space Environ Med 1980:965–9.
 Kaliszan M. Does a draft really inﬂuence post mortem body cooling? J Forensic
 Kaliszan M, Hauser R, Buczynski J, Jankowski Z, Raczynska K, Kernbach-
Wighton G. The potential use of the eye temperature decrease in determining
the time of death in the early postmortem period. Am J Forensic Med Pathol
 Baccino E, De Saint Martin L, Schuliar Y, Guilloteau P, Le Rhun M, Morin JF,
et al. Outer ear temperature and time of death. Forensic Sci Int
 Henssge C, Beckmann ER, Wischhusen F, Brinkmann B. Todeszeibestimmung
durch Messung der zentralen Hirntemperatur. Z Rechtsmed 1984;93:1–22.
 Camps FE. Establishment of the time of death – a critical assessment. Forensic
 Kurz A, Sessler DI, Christensen R, Dechhert M. Heat balance and distribution
during the core-temperature plateau in anesthetized humans. Anesthesiology
 Jessen C. Thermal afferents in the control of body temperature. In: Schonbaum
E, Lomax P, editors. Thermoregulation physiology and biochemistry,
International Encyclopedia of Pharmacology and Therapeutics. Elmsford,
New York: Pergamon Press; 1990. p. 153–83. sect. 131.
 Webb P. The physiology of heat regulation. Am J Physiol 1995;268:R838–50.
 Braunwald E, Harrison TR. Harrison’s principles of internal medicine. 15th
ed. New York: McGraw-Hill; 2001.
 Enus J, Van Someren W. Thermoregulation and aging. Am J Physiol Regul Integr
Comp Physiol 2007;292:R99–R102.
 Charkoudain N. Skin blood ﬂow in adult human thermoregulation: how it
works, when it does not, and why. Mayo Clin Proc 2003;75:603–12.
 Silva NL, Boulant JA. Effects of testosterone, estradiol, and temperature on
neurons in preoptic tissue slices. Am J Physiol 1986;250:R632–35.
 http://www.anesthesia-analgesia.org/content/97/1/275.full, accessed Apr
 Stagnitti M. Antidepressant use in the US civilian non-insitutionalised
population, Statistical brief no. 77, Rockville, MD: Medical expenditure
panel, agency for healthcare research and quality; 2005.
 http://usgovinfo.about.com/od/healthcare/a/usmedicated.htm, accessed Apr
 Demierre N, Wyler D, Zollinger U, Bolliger S, Plattner T. Elevated body core
temperature in medico-legal investigation of violent death. Am J Forensic Med
 Andrews MT. Genes controlling the metabolic switch in hibernating animals.
Biochem Soc Trans 2004;32:1021–4.
 Hayward JS, Eckerson JD, Collis ML. Thermoregulatory heat production in man:
prediction equation based on skin and core temperatures. J Appl Physiol
 Werner J. Regelung der menschlichen korpertemperatur. Berlin: W. De
 Kleinebeckel D, Klussmann FW. Thermoregulation physiology and
biochemistry. In: Schonbaum E, Lomax P, editors. International Encyclopedia
of Pharmacology and Therapeutics. Elmsford, New York: Pergamon press;
1990. p. 235–53. sect. 131.
 Horowitz M, Meiri U. Thermoregulatory activity in the rat: effects of
hypohydration, hypovolemia, and hypertonicity and their interaction with
short-term heat acclimation. Comp Biochem Physiol 1985;82A:577–82.
 Horowitz M. Thermoregulation under conditions of impaired body ﬂuid/
osmotic balance in mammals. J Basic Clin Physiol Pharmacol 1990;1:
 Latzka WA, Sawka NW. Hyperhydration and glycerol: thermoregulatory effects
during exercise in hot climates. Can J Appl Physiol 2000;25:536–45.
 Guyton AC, Hall JE. Human physiology and mechanisms of disease. sixth
ed. Philadelphia: W.B. Saunders; 1997.
 J.L. Smart, M. Kaliszan. Use of a ﬁnite element model of heat transport in the
human eye to predict time of death. J Forensic Sci, accepted for publication.
 Mall G, Eisenmenger W. Estimation of the time since death by heat-ﬂow
Finite-Element model. Part I: method, model, calibration and validation. Leg
 Mall G, Eisenmenger W. Estimation of the time since death by heat-ﬂow
Finite-Element model. Part II:application to non-standard cooling conditions
and preliminary results in practical casework. Leg Med 2005;7:69–80.
 Hiraiwa K, Ohno Y, Kuroda F, Sebetan IM, Oshida S. Estimation of postmortem
interwal from rectal temperature by use of computer. Med Sci Law
 Kanawaku Y, Kanetake J, Komiya A, Maruyama S, Funayama M. Computer
simulation for postmortem cooling process in the outer ear. Leg Med
 Madea B. Importance of supravitality in forensic medicine. Forensic Sci Int
 Hooft P, van de Voorde J. The role of intestinal bacterial heat production in
confounding postmortem temperature measurements. Z Rechtsmed
 Nokes LDM, Hicks B, Knight B. The post-mortem temperature plateau - fact or
ﬁction? Med Sci Law 1985;25:263–4.
 Livingstone DR. Invertebrate and vertebrate pathways of anaerobic
metabolism: evolutionary considerations. J Geo Soc 1983;140:27–37.
 Schurr A. Brain anaerobic lactate production: a suicide note or a survival kit?
Dev Neurosci 1998;20:348–57.
 Mall G, Hubig M, Beier G, Buttner A, Eisenmenger W. Supravital energy
production in early post-mortem phase - estimate based on heat loss due to
radiation and natural convection. Leg Med 2002;4:71–8.
 K. Schourup. Determination of the time since death. Diss Med, Kophenhagen,
 Okada H. Effects of anoxia before cardiac arrest on postmortem changes in
heart muscle. III. Relation between causes of death and postmortem changes in
lactic acid concentration and pH level in heart muscle. Shikoku Igaku Zasshi
 Vorel F, Zemanova Z. Determination of the lactic acid concentration in
postmortem brain. Chekhoslovatskaya Fiziologiya 1964;13:425–31.
 Maeda H, Ishikawa T, Michiue T. Forensic biochemistry for functional
investigation of death: Concept and practical application. Leg Med
62 J.L. Smart, M. Kaliszan / Legal Medicine 14 (2012) 55–62