Providing an estimation of the time elapsed since death is a request almost invariably aimed at forensic specialists summoned to the scene of a death. Despite the fact that there has been an extensive literature on this topic for more than a century, this determination still remains difficult, even for experienced pathologists, and must be undertaken with extreme caution. In particular, the dogmatic application of rules and formulas based upon single and isolated observations (e.g. rectal temperature of the body, extent of lividity, stage of putrefaction)is a guarantee of inaccuracy because of the numerous factors of cadaveric or environmental origin that can influence the ‘normal’ rate of postmortem changes. The longer the time since death, the more important the weight of these factors and the more difficult it will be to estimate the death time with some accuracy.
The purpose of this article is to present a brief overview - focused on advantages and limitations – of the different techniques that can be applied to human corpses in order to determine the death time. As shown in Fig. 1, most of them concern the so-called ‘early postmortem period’ (usually defined as the time between death and the appearance of generalized putrefaction).

Body Temperature Models of body cooling

After death, body temperature progressively declines until it reaches the temperature of its surroundings. Although this phenomenon has been appreciated since prehistory, the first papers dealing with its potential interest to the field of forensic medicine date back to the mid-nineteenth century. In temperate climates this process generally takes about 8-12 h at the skin surface, but the core of the body is known to require two to three times as long to cool down.
The cooling of a body (also called ‘algor mortis’) is determined by physical rules and involves four different mechanisms of heat transfer: conduction, radiation, convection and evaporation. Their respective influence on the fall of body temperature after death strongly depends on environmental circumstances (temperature, hygroscopy, air agitation, etc.). According to a rule of thumb usually employed in the past, time since death could be roughly estimated by assuming a temperature loss of 1°C per hour during the first 24 h. However, the rate of cooling of a physical body is never a linear function of time: under experimental conditions this phenomenon is mainly determined by the difference in temperature between the body and its surroundings, and should be represented by an exponenential function of time (Newton’s law). As shown by many measurements on corpses, the reality is still much more complex, owing to several factors like the irregular shape  and inhomogenous composition of the human body, or a moderate postmortem heat production resulting from bacterial metabolism. Finally, when temperature is plotted against time, the postmortem cooling of the human body is best represented by a sigmoid or biexponential curve that comprises three successive periods:
• An initial phase, in which the body temperature remains relatively even – the so-called ‘temperature plateau’ or ‘lag period’. This period usually lasts 0.5-3 h, with important (and unpredictable)inter-individual variations.
• An intermediate phase, in which temperature drops rapidly and almost linearly.
• A terminal phase, in which this drop progressively slows as the core temperature approaches that of the environment.
Common estimators of time since death.
Figure 1 Common estimators of time since death.

Temperature measurements

It is essential to measure the central temperature of the body as well as the environmental (air)tempera-ture at the scene of death. Both must be taken at the same time and with the same instrument, and the time of measurement must be carefully noted. On the several anatomical sites that have been proposed for temperature determination, the rectum probably ranks first for convenience and reliability; however, this may be debatable if a sexual assault is obvious or suspected. The tympanic determination represents an interesting alternative for some authors. Endobuccal or external determinations (at the axilla or groin, or on the abdominal skin)must be avoided. Medical thermometers should not be used for rectal measurements because they are too short and their range of temperature is too narrow. The best instruments are thermocouple, electronic devices with digital display, which may usually be linked to a computer and/or printer. The thermometer must be equipped with a long, rigid probe that has to be inserted at least 10 cm into the rectum.

Advantages and limitations

If measured under appropriate conditions, body temperature may be held as one of the best estimators of the time since death during the first 24 h; however, it presents some limitations that should always be borne in mind:
• The technique is considered accurate only during the intermediate, pseudolinear phase of the sigmoid cooling curve. Otherwise, temperature measurements are not useful and may lead to inconsistent estimations during the very early postmortem period – the temperature plateau – or when the body temperature closely approaches that of the surroundings.
• All determinations of the time of death by the body cooling method assume that the body temperature at the time of death was within normal limits,37.2 + 0.4°C. Hypothermia or hyperthermia preceding death may result in mistakes and thus should be systematically searched for if antemor-tem data are available.
• Cooling equations also assume that the environmental temperature has remained constant over the postmortem interval. This may be the case when the death occurs in air-conditioned or heated buildings; however, problems may arise with corpses found outdoors, owing to the circadian changes of ambient temperature.
Furthermore there are a number of complicating factors, of endogenous (cadaveric)or exogenous (en-vironmental)origin, that may strongly affect the rate of postmortem cooling. These include:
• Body mass Heat loss is slower in obese subjects (owing to greater body mass and because fat may act as an insulator), whereas it is more rapid in slender ones.
• Movement of air Any air movement enhances heat losses by convection, thus accelerating the fall in temperature. At outdoor scenes of death it is therefore important to note if the weather is windy, and in buildings sources of air currents (open doors/windows)should be looked for.
• Humidity High atmospheric humidity will increase heat losses due to evaporation.
• Clothing Clothes will act as insulators; the thicker they are, the more the fall in temperature will be delayed. This is also true for any kind of covering found on or around the body (e.g. subjects dying in bed and found lying under sheets and blankets).
• Immersion in water A corpse immersed in still water will lose heat mainly by conduction, at a rate several times higher than in air (because water is a much better heat conductor); the temperature decay will be even more accelerated if the water is flowing.
Henssge’s nomogram probably constitutes the most elaborate and easy-to-use system developed to establish the time of death from body cooling, taking into account the main influencing factors. This three-parameter abacus (body temperature, ambient temperature, body mass)furnishes a crude estimation of the time since death, corresponding to the ‘average’ situation of a naked corpse lying in still air. Depending on the case, this preliminary value may be adjusted using corrective factors from 0.35 (for a naked body immersed in flowing water)to 2.40 (for a clothed body coved by a thick bedspread). Another major feature of this nomogram is that it provides the estimation together with a confidence interval (corresponding to a permissible variation of 95%); for an adult weighing 70 kg, this interval is +2.8h during the first 15 hours post mortem.
Many authors have proposed alternative solutions to enhance the accuracy of this technique, e.g. repeated or continuous postmortem temperature measurements over several hours, or the measurement of central temperature by invasive techniques (intra-hepatic or subhepatic temperature taken by abdominal stab, or intracerebral temperature measured by inserting the probe through the orbit). These methods are difficult to apply to the routine investigation of scenes of death; in addition, none of them has proved to be clearly better than the single measurement of rectal versus environmental temperature.


Cadaveric rigidity (rigor mortis)is the consequence of irreversible and complex physicochemical changes occurring in muscle proteins after death, including conversion of muscle glycogen to lactic acid. Based upon many observations, this phenomenon has been proposed as an estimator of the time since death for about 200 years (its first scientific description was that of Nysten in 1811).
Rigidity usually develops sequentially and follows a descending pattern, the so-called Nysten’s law: it affects successively the muscles of the face (especially those of the eyelids and lower jaw)and of the neck, then the trunk and upper limbs, and finally the lower limbs. It generally begins 3-4 h after death, is fully established after 8-12 h, remains unchanged for up to 36 h post mortem, and then disappears in 2-3 days -in most cases when putrefaction becomes patent. Breaking the rigidity manually (bending a joint by force, against the resistance of rigidity)will make it reappear if death occurred less than 8-12 h earlier; if the rigidity is complete, it will not reappear once broken down by force.
There are, however, many exceptions to this rule as far as both the sequential development and its time course are concerned. Numerous factors have been shown to affect, to a greater or lesser extent, the rate of onset and duration of rigor, or its intensity:
• Temperature is probably the more decisive parameter, as heat accelerates rigor whereas cold slows it down.
• Violent exercise prior to death may hasten the onset as well as disappearance of rigidity.
• Muscular development: the more muscular the subject, the greater the postmortem rigidity. On the other hand rigor mortis may be very moderate in children, emaciated people or in the elderly.
• Cause of death: rigidity may develop early, or even instantaneously (‘cadaveric spasm’), in fatalities involving convulsant drugs (strychnine), electrocution, traumatic lesions affecting the central nervous system, or when death is preceded by extreme psychological stress (homicides, violent suicides). On the contrary, it may be delayed in some as-phyxial deaths (carbon monoxide, hanging)or in deaths caused by massive hemorrhage.
In addition, the stage of rigor is generally evaluated subjectively, using criteria that vary from one author to another. Special devices based upon dynamometers have been proposed to measure rigidity with greater accuracy but they proved inconvenient and are never employed in casework.


Postmortem lividity (hypostasis, livor mortis)is a plurifocal staining of the skin, usually in the form of a more or less intense purple discoloration, due to the gravitational settling of blood in vessels after the circulation has ceased. It always develops at the lowest parts of the cadaver, which depends upon its posture after death: in a body lying in the usual supine position, lividity predominantly affects the nape of the neck and the posterior aspects of trunk and limbs. By contrast, it does not appear on skin areas exposed to pressure, e.g. areas in contact with the underlying supporting surface (shoulder blades, buttocks, calves and heels for a body in the supine position), or areas squeezed by tightly fitting clothing (belts, underwear elastic).
Lividity often becomes perceptible within 3-4 h of death and progressively develops in surface area and colour intensity to attain its maximum degree 8-12 h postmortem. During this early period it is still mobile, i.e. it may be locally displaced by thumb pressure on the skin, or it can move (partly or completely)to other regions of the corpse if the body position is altered. After 12-15 h, postmortem hypostasis becomes ‘fixed’ and thus can no longer be displaced by external action. It will then remain unchanged until masked by the generally darker discoloration resulting from putrefaction.
Although this sign is present in almost all bodies (with the exception of those dying of massive hemorrhage), the time course of lividity exhibits considerable intersubject variability – probably to a greater extent than most other estimators of time of death in the early postmortem period. In addition, its quantitative measurement is imprecise (some authors use colorimetric tables)and in most routine cases its estimation remains largely subjective.

Potassium Concentrations in Vitreous Humor

Vitreous humor - the transparent, gelatinous substance that fills the posterior chamber of the eye – is an interesting medium for postmortem biochemical studies owing to its anatomical location and its relative resistance to bacterial contaminations over the first week after death.
The relationship between vitreous potassium levels and the time since death has been well studied for about 35 years. During life, potassium concentration is low in the vitreous humor but much higher in the peripheral tissues of the eye (vitreous layer, retina). This electrolytic imbalance results from energy-consuming, vital cell activities (active membrane transport, selective membrane permeability), the postmortem cessation of which leads to progressive reversal of the potassium gradient, with the consequence of a rise in vitreous concentration. Based on experiments on large series of cadavers, many authors have established linear relationships between vitreous potassium and time since death. Of the different equations proposed, the most popular are:
• Sturner’s formula: TSD = 7.14 [K+] – 39.1
• Madea’s formula: TSD = 5.26 [K+] – 30.9
where TSD is the time since death (hours)and [K+]is the potassium concentration (mmoll-1).


Vitreous humor is easy to collect but some rules have to be followed to avoid perturbances of the biochemical results. The fluid should be taken by puncture of both eyes at the external angle, using a syringe with intravenous or intramuscular needle. Suction must be gentle (do not use Vacutainer-type vials)to prevent aspiration of fragments from the retina or choroid; the samples should be centrifuged to eliminate cell debris. The time of collection must be carefully noted. Only colorless, crystal-clear vitreous is suitable for potassium analysis. Presence of blood, cell remains, or a brown-greenish color due to putrefaction dictates the sample to be withdrawn. If the potassium determination cannot be performed immediately, the samples should be stored frozen at — 18°C.

Advantages and limitations

The main advantage of the vitreous potassium method is that it may be carried out up to 5-7 days after death, whereas most other estimators (especially body cooling)are useful only within the first 24 h. Unfortunately it also has some limitations:
• A major difficulty is presented by the 95% confidence interval of the technique which is always large and, in addition, somewhat different from author to author: for the first 24 h postmortem it ranges from + 4to +12 h, and for the period up to 100h postmortem from +9.5 to +40h.
• The rise of vitreous potassium concentrations may be strongly affected by the ambient temperature as well as by endogenous factors (age at death, duration of agony, etc.).
• At a given time, potassium concentrations may differ significantly between each eye (deviations can exceed 10% of the mean value of both eyes).
• All death time estimations by the vitreous potassium method assume that the antemortem period was not associated with electrolytic perturbations -which may be precisely the case in many deaths preceded by an agony of significant duration. Some authors have recommended the use of vitreous urea and/or creatinine in order to identify and reject subjects with possible antemortem electrolyte imbalance. By eliminating subjects with vitreous urea > 70 mg dl ~1 or vitreous creatinine > 1mgdl ~ 1,it has been stated that the 95% confidence interval for the period up to 120 h post mortem might be reduced to + 15 h. If confirmed, this would represent the most accurate estimation of the time since death that could be obtained by any existing method.

Other Biochemical Markers

Since the 1950s, a number of other biochemical markers have been investigated to assess their potential as estimators of time of death:
• Blood markers: electrolytes (sodium, potassium, calcium, magnesium, phosphorus, chloride), glucose, lactic acid, urea, creatinine, cholesterol, triglycerides, apolipoproteins, hormones (cortisol, adrenalin, thyroxin, thyroid-stimulating hormone, insulin), enzymes (phosphatases, amylases, phos-phoglucomutase), blood gases and pH, cleavage of the C3 component of complement.
• Vitreous markers: electrolytes (sodium, calcium, magnesium, chloride), urea, pH.
• Cerebrospinal fluid markers: electrolytes (potassium, magnesium, sodium, calcium, phosphorus, chloride).
• Pericardial fluid markers: electrolytes, cholesterol, glucose, lactic acid, enzymes (transaminases).
• Muscle markers: creatinine, enzymes (creatine phosphokinase).
• Lung markers: surfactant phospholipids.
Although experimental studies have shown that, in many cases, positive or negative correlations exist between the postmortem evolution of these markers and the time elapsed since death, their relative coarseness make them unsuitable for practical casework.

Supravital Reactions

Following cessation of the circulation, ischemia in organs and tissues leads to reversible, then irreversible changes affecting their structure and function. The time course of these phenomena is, however, very different, depending on the tissues; for example, brain cortex structures undergo definitive alterations after a few minutes, whereas other tissues (kidney, skeletal muscle)may tolerate prolonged ischemia for up to several hours. This intermediate period, beginning at brain death and lasting until cell activity has definitely ceased in the whole organism, is sometimes called the ‘supravital period’ because external stimuli applied to the corpse may induce life-mimicking, observable reactions. The supravital reactions that may be of interest in estimating the time since death are:
1. Mechanical excitability of the skeletal muscle, including:
a Tendon reaction (Zsako’s phenomenon), a contraction of the whole muscle due to propagated excitation following a mechanical stimulation. This can be obtained, for instance, by striking the lower third of the thigh 4-5 fingerbreadths above the patella with a reflex hammer, resulting in an upward movement of the patella due to contraction of the whole quadriceps muscle. Zsako’s phenomenon is particularly transient, as it usually cannot be observed beyond 2-3 h postmortem.
b Idiomuscular contraction, a localized muscular contraction (bulge)at the point of stimulation, demonstrated, for instance, by striking the biceps muscle of the arm with a reflex hammer. This can be observed several hours after cessation of Zsako’s phenomenon.
2. Electrical excitability of the skeletal muscle. This is generally investigated on the muscles of the face. Needle electrodes are inserted through the skin (at the nasal part of the upper eyelid, or on both sides of the mouth)and electrical impulses, provided by a portable generator, are applied. The reaction – a contraction of one or several muscles – may be semiquantitatively quoted according to the strength of the contraction and its extension to areas distant from the electrodes, both of them diminishing as time since death increases.
3. Pharmacological excitability of the iris muscle. Modifications of the pupil diameter can be observed following the administration of miotic (acetylcholine)or mydriatic (norepinephrine, at-ropine)solutions, which may be instilled on to the cornea or injected subconjunctivally at the limbus of the cornea.
Supravital reactivity may be of value for establishing the time of death, but only during the first hours of the postmortem period (none of the above-mentioned phenomena proved observable beyond 12-15 h postmortem). Since the muscular responses obtained can be estimated only semiquantitatively, their investigation requires very experienced operators. The equipment needed for electrical stimulation also limits the applicability of this technique for routine scene-of-death investigation.


Putrefaction is the destruction of the soft tissues of the cadaver by the action of bacteria and enzymes. Its evolution can be divided into five successive periods:
• Initial decay (up to 36-72 h post mortem): the corpse still appears fresh externally but internally it begins to decompose, owing to the combination of enzymatic autolysis and bacterial proliferation from the intestine.
• Early putrefaction or green putrefaction (up to 1 week post mortem): this period often begins with a greenish discoloration of the right iliac fossa, which subsequently spreads to the whole anterior abdominal wall and the rest of the trunk, then to the neck, face and finally to the limbs (upper and lower limb extremities typically putrefy last). Other signs include skin blisters; abdominal/scrotal swelling with gas; oozing of putrefactive fluids by mouth and nostrils (not to be mistaken for blood); and the typical odor of decaying flesh.
• Black putrefaction (up to 1 month post mortem): the cadaver exhibits a flesh of creamy consistency with exposed parts turning black, especially at the head and face. Skin decomposition results in generalized epidermal detachment, and the nails fall off. The abdomen collapses as gases escape. The odor of decay is at its maximum.
• Butyric fermentation (up to 2 months post mortem): this stage is marked by a progressive drying of the cadaver and the occurrence and proliferation of mold. The odor is less offensive and becomes typically cheesy.
• Dry decay, then skeletonization (months to years): final drying of the cadaver and progressive disappearance of the remaining soft tissues. Time-related changes affecting the remains at this stage are slow compared with those of the preceding periods.
Unfortunately, this chronology of putrefactive events is only indicative of what may occur, as enormous and generally unpredictable intersubject variability exists. Ambient temperature obviously influences the rate of putrefaction to a major extent, but many other factors may intervene, such as body corpulence (obese subjects putrefy more quickly than slender ones), ante-mortem diseases or circumstances of death (sepsis or edemas during the period preceding death may hasten decomposition). Furthermore, there are some variants of the ‘classical’ succession of decomposition stages, depending on environmental circumstances: adipo-cere formation (in warm, damp, preferably anaerobic conditions)and mummification (in hot and dry environments, especially if air is moving). For these reasons, and however great the experience of the forensic investigator, the putrefactive phenomena affecting a cadaver cannot be considered as reliable markers of the time elapsed since death.


After death, the tissues of animals become attractive to a large variety of insects and other invertebrates. These may be classified into four groups:
• Necrophagous species: invertebrates that feed on the corpse itself.
• Predators and parasites of the necrophagous species: these species do not feed directly on the corpse.
• Omnivorous species: invertebrates that feed both on the corpse and on the other arthropods present.
• Adventive or opportunistic species: invertebrates present on the corpse by chance, or using it as an extension of their usual environment (e.g. as a shelter, nest)without feeding on it.
The use of entomologic markers for determining the time of death is based upon the long-established observation (first scientifically reported by Megnin in 1894)that insects and other arthropods feeding on a corpse follow a specific faunal succession associated with the various stages of decay. Estimations of time of death require an accurate recognition of the sometimes numerous species present on a corpse, or its surroundings, in their different immature (eggs, larvae, puparia)or adult stages of growth, together with an extensive knowledge of their specific rates of development according to environmental parameters (season, temperature, humidity, etc.). Such investigations are difficult and can be undertaken only by experienced, full-time specialists in forensic entomology. Provided this is the case, this technique currently constitutes the only approach for estimating the time of death of a putrefied cadaver with some accuracy – in some cases, to within a few days, even in deaths obviously dating back for months.

Conclusions: Combined Methods

As shown by this brief survey, all methods proposed for the determination of the time since death remain relatively inaccurate, even when applied to the very early postmortem period. This limitation is a consequence of the huge interindividual variability affecting all postmortem changes, which is in large part attributable to the existence of many complicating factors, endogenous as well as exogenous: location of the body (exposed to atmosphere/buried/immersed); environmental temperature and other seasonal/climatic conditions (wind, humidity, exposure to sun); nature of soil; body corpulence; posture of the corpse; presence/nature of clothes or coverings; antemortem diseases; duration of the process of dying; causes and circumstances of death. Another reason for this poor level of accuracy is that many of the criteria used for the estimation are subjectively evaluated instead of being objectively measured. Finally, reporting to the police the time at which death is deemed to have occurred is of little interest, scientifically questionable and potentially dangerous if the crude estimation (usually a mean value obtained by computation of one or several tables or formulas)is not furnished together with an interval of confidence.
As emphasized by several authors, partial remedies may consist in:
• understanding and taking into account the major complicating factors acting on each criterion;
• using objective measurements instead of subjective evaluations;
• rejecting estimators that do not allow the calculation of confidence limits for the time of death (e.g. lividity, putrefaction), or using them only as adjustment parameters for more reliable methods.
The combination/integration of different estimators has been proposed to narrow down the margins of error associated with single methods. For practical casework, Henssge and coauthors use special charts in which they arrange the following parameters:
• body cooling (interpreted using Henssge’s nomo-gram with corrective factors);
• external examination of lividity and rigidity;
• electrical and mechanical excitability of facial muscles;
• chemical excitability of the iris.
Despite these recent improvements and unless some revolutionary technique – at present unforseeable -is introduced in the future, the great accuracy sometimes depicted in movies and detective novels (‘Your Honour, death occurred yesterday at 5.47 p.m.’)will remain an unrealistic mirage. The determination of the time since death is a difficult and complex task, frequently disappointing in the modest results it produces. Obviously, this work requires both scientific competence and field experience – but it also demands a large dose of humility.

Next post:

Previous post: