Article Text

Download PDFPDF

Interpreting bruises at necropsy
  1. P Vanezis
  1. Department of Forensic Medicine and Science, University of Glasgow, Glasgow G12 8QQ, UK
  1. Dr Vanezis p.vanezis{at}


The accurate interpretation of bruising at necropsy is essential to understanding how a victim has been injured and assists the pathologist in a reliable reconstruction of the events leading to death. It is essential not only to assess the mechanism of production of a bruise, taking into account the type of impacting surface and the magnitude of force used, but also to estimate when the injury was caused. An account is given of the various methods used in the examination of bruises, particularly with respect to aging, as well as the factors that may affect their appearance. Differentiation from artefacts resulting from postmortem changes is also discussed in some detail.

  • bruising
  • necropsy
  • time of death
  • cause of death

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The interpretation of bruising for forensic purposes requires the pathologist to bear in mind the following three fundamental questions:

  • Is the discolouration seen a bruise?

  • When was it caused?

  • How was it caused?

A succinct definition of a bruise is that it is a collection of blood, visible to the naked eye as an area of discolouration, which has extravasated into the surrounding tissues after vascular disruption, principally as a result of trauma or occasionally spontaneously, as a result of a disease process. Typically, bruises are caused by blunt trauma, although they may be associated with any type of impact and can accompany many different types of wounds. Thus, they can provide much information with regard to their causation and can assist with the reconstruction of events leading to death.

Furthermore, when an injury is inflicted and blood extravasates from injured blood vessels the resultant bruising may be:

  • Located at the area of impact and visible shortly after infliction of the injury.

  • Located at the area of impact area but delayed in appearance.

  • Located at a site away from the impact area and result from tracking of blood from the impact area.

The word bruise is often used synonymously with the term haematoma and ecchymosis (or ecchymoma). Contusion is also another frequently used term, particularly in relation to internal trauma.

Although bruising is most often thought of as an extravasation of blood intradermally or subcutaneously, bruises can occur almost anywhere in the body.

Factors that influence the development and appearance of bruises

There are many variables that influence the development and absorption of bruises, as well as their appearance and extent of spread, thus adding to the difficulty in their interpretation.

Bruising occurs more easily where there is loose tissue—for example, over the eyebrow—rather than where the skin is more strongly supported. It also occurs more readily where there is an excess of subcutaneous fat. Because there is a greater skin deposition of fat in women, they tend to bruise more easily than do men.

The type of surface and force that impacts on the body will have a great effect on the intensity, size, shape, and pattern of the resultant bruising.

In infants and the elderly bruising tends to occur more easily. In the very young the skin is looser, more delicate, and there is an increased amount of subcutaneous fat. In old people, although there is loss of subcutaneous fat, blood vessels are also more poorly supported and bruises take longer to resolve.

Skin colouration modifies the appearance of a bruise to the naked eye. It is much easier to observe the extent and colour of bruising in lighter skinned individuals. Thus, it is particularly important to take extra care when examining dark skinned individuals so that any bruising is not overlooked.

It is also essential to be aware of conditions that could either give rise to spontaneous tissue haemorrhage/bruising or render the individual more likely to bruise easily (out of proportion to the force of impact). Such conditions fall into several categories and, apart from various bleeding disorders (table 1),1 one must also consider hypertension, cardiovascular degenerative changes, and disorders of collagen and other supporting tissues, which might result in an increase in the amount of blood extravasating from blood vessels.

Table 1

Some conditions associated with increased bleeding

Documenting and interpreting bruise characteristics

All bruises should be documented thoroughly, taking into account their shape, size, and colour (table 2). Photography and occasionally other visual media such as video are an essential part of the documentation process and provide a permanent visual record. However, the assessment of bruising from secondary images is second best to assessment at necropsy.

Table 2

The documentation of bruising includes the following

The nature of the agent causing a particular bruise is central to assessing the circumstances of injury and thus assisting in the reconstruction of events. On occasions, depending on the site of injury and nature of the impacting agent, it may be possible to match a patterned bruise with an object—for example, a heel mark with a corresponding shoe. Other bruises, although not producing a “mirror image” of the offending agent will nevertheless produce characteristics that are typical of the type of impacting agent. For example, round or discoid small bruises can be caused by fingertips, and tramline bruising can be caused by objects that have a longitudinal cylindrical surface (such as a rod or baseball bat).

A bruise should not be examined in isolation because on many occasions, particularly with shoe marks or blunt trauma from sticks, iron objects, belts and so on, there may be an accompanying characteristic abrasion or laceration. The size, intensity, and accompanying lacerations or abrasions will be useful pointers in assessing the force of an impact. Nevertheless, it should be appreciated that the extent of bruising will depend on several other factors other than force. These include site of injury (whether over bone or soft tissue), type of agent used, and factors intrinsic to the victim. One must not forget the effect of skin pigmentation on the appearance of bruises and thus their interpretation. It is important in dark skinned individuals to pay particular attention to the subdermal appearance and spread of a bruise.

The distribution and collective assessment of bruises, particularly in reconstructing events, is particularly pertinent to the investigation of the “battered child”. In such circumstances, where there are multiple bruises, possibly of varying ages, it is essential to assess the injuries as a whole to understand the circumstances and manner of causation, rather than to consider each injury in isolation (table 3).

Table 3

Assessment of bruising includes consideration of the following

Dating bruises

Any investigation into a death of forensic importance will need to establish when a particular trauma event occurred. However, the assessment of the timing of an injury is not a precise science and it is only possible to give an approximate time. Nevertheless, the pathologist should be aware of the various factors and techniques available that can assist him or her to arrive at a reasonable assessment of age, despite the inherent limitations.

For the clinical forensic medical examiner, the assessment of bruises in living subjects is obviously limited to their external gross appearance, including photographs and history given by the victim where available. The pathologist, on the other hand, has more options for an objective assessment, including microscopical examination and closer gross assessment by dissection. In addition, both the pathologist and his/her clinical colleague will need to corroborate or refute accounts given by eye witnesses where appropriate.

An assessment of a bruise to ascertain when it occurred can be made in several different ways, namely:

  • Direct gross examination of the victim (external and by dissection).

  • Conventional and special photography.

  • Conventional histological examination.

  • Various histochemical techniques.

  • Biochemical methods.

  • Objective colour assessment.


A large quantity of information can be obtained from gross examination of the body in the mortuary. For example, differentiating a fresh bruise from an older one is usually not difficult. It is important, however, to bear in mind that the location of a bruise also needs to be taken into account when assessing its age. For example, a deep bruise in thigh muscle might not appear for a day or two after its infliction and might have a fresh appearance, whereas a bruise over a bony prominence and where tissue is loose—for example, the eyebrow—will appear very quickly and with accompanying swelling.

Larger bruises remain visible for longer than smaller ones. The development and rate of disappearance of a bruise will also depend on the state of health and age of the person. Bruises take much longer to disappear in the elderly.


Studies of bruising carried out in living subjects that are based on their appearance in photographs illustrate the difficulties involved in timing such injuries. Langlois and Gresham,3 in their study of 369 photographs, found that the most important change was the development of a yellow colour, which was also found to develop much faster in those under 65 years of age. They concluded that bruises with a yellow colour are more than 18 hours old and that the appearance of the other colours is less important. In a similar study,4 assessing photographs of children who suffered accidental bruising, a yellow colouration was not observed in those bruises that were under one day old. Over the years, several authors of forensic medicine textbooks have made various observations on colour changes in bruises with time (table 4).510 It is clear that the red/blue/purple colour in bruises can persist for days and even longer. The green colour, however, is difficult to interpret because it can reflect a combination of blue and yellow, but is nevertheless seen as an intermediate stage between the initial appearance and the later yellow appearance. One of the interesting findings is the time at which the yellow colour first appears and can be seen with the naked eye. Most authors writing in textbooks and referring mainly to anecdotal evidence of their own personal experience state that the yellow colour does not appear until at least several days and usually one week after injury. One comprehensive study of bruises11 pointed out that deep bruises can take 12–24 hours to appear, and that a brown colour would indicate that the bruise was over 24 hours old. This contrasts with others,9 who did not find a brown colour until the end of the first week

Table 4

Summary of colour change in bruises with time


Most of the basic observations used for the timing of bruises have been known for over 100 years. Virchow in 1847 first described pathological pigments in old haemorrhagic areas inside and outside cells as diffuse, granular, and crystalline structures, which he called haematoidin,12 and which was later identified as bilirubin.13 In 1869 erythrophagocytosis was detected in haemorrhages,14 and in 1888 the iron containing pigment in old haemorrhages was named haemosiderin and differentiated from haematoidin.15

The earliest change seen microscopically in the development of a bruise is oedema, resulting in fluid exudation within the vicinity of the haemorrhage, together with widening of fibrous septa. This early, non-cellular exudative phase merges into the leucocyte reaction. It should be appreciated that it is not only the white blood cells extravasating with the haemorrhage that are detected, but also the granulocytes that migrate from neighbouring blood vessels. The earliest recorded leucocyte reaction was detected at 20–30 minutes,16 although more recent observations give times in both haemorrhages and wounds varying between one and 24 hours. With regard specifically to subcutaneous haemorrhages, the first influx of polymorphonuclear leucocytes has been noted after about four hours.17 The first leucocytes to migrate become increasingly necrotic after about 15–30 hours and are superseded by phagocytic mononuclear cells (macrophages).

Erythrocytes are seen in macrophages between 15 and 17 hours after infliction of the injury in the skin or brain, whereas the corresponding figure in the human lung is as early as 30 minutes.

Haemosiderin laden macrophages have been seen in the skin and subcutaneous tissue as early as 24–48 hours after the infliction of trauma, and more commonly from four to eight days. In brain, haemosiderin has been seen in macrophages as early as three to four days, but more commonly between five and 15 days. In human subdural haematomas, haemosiderin has been described as occurring after five days, whereas in the human lung haemosiderin is found much earlier (the earliest being at 17 hours).

In human postmortem material, haematoidin (bilirubin) has been found in haemorrhages of the skin and subcutaneous tissue from nine days. In haemorrhages of the human brain, haematoidin has been observed from 10 to 12 days onwards.

The Perls' Prussian blue reaction has been, and still is, the most useful stain for the detection of ferric iron. However, the presence of haematoidin is not so regular and its demonstration in paraffin wax embedded sections using the Van Gieson method is seldom successful. This could be because the crystals that are seen in frozen sections are not present in the paraffin wax embedded sections, presumably because they have been dissolved during the processing technique (for example, by xylol). The diffuse yellow material seen occasionally in old haematomas in paraffin wax embedded sections could represent partly dissolved crystals of bilirubin in some other form. Polarised light helps in the detection of haematoidin crystals in the frozen sections.

Haemosiderin deposits in more than 20% of the microscopic field are not seen until at least eight days after wounding,18 and thus the detection of considerable amounts of haemosiderin (arbitrarily defined as 20% or more of the evaluated area) indicate a minimum wound age of approximately one week. Because the extent of haemosiderin formation depends upon the extent of the initial haemorrhage and a “physiological” reduction in the amount of this pigment with advanced wound age, slight or absent haemosiderin deposits cannot provide information on the postinfliction interval.

The detection of erythrocytes and haemosiderin in macrophages and haematoidin have been the classic methods for age estimation of haemorrhages; however, we still cannot give an accurate timetable of events suitable for every haemorrhage. Despite attempts to improve matters, animal experimentation has revealed considerable interspecies variability and is of little use in drawing direct comparisons with human material.

As mentioned previously, different localities can result in differences in the time needed for the development of the various markers—for example, the speed of change is different in subcutaneous tissue and brain compared with lung and lymphoid tissue. Furthermore, it is common knowledge that subungal haemorrhages can remain for several months, without changing greatly.

Enzyme histochemical investigations of subcutaneous bruises from routine postmortem material revealed an increase in ATPase activity in the vascular walls after 2.5 hours.17 Aminopeptidase and esterase activity increased at 4.5 hours and seven hours, respectively, and polymorphonuclear and mononuclear cells appeared within four and nine hours, respectively. Using several histochemical methods (Alcan blue periodic acid Schiff (PAS) and dialysed iron PAS, both controlled by enzyme digestion), after initially decreasing, acid glycosaminoglycans were demonstrated in bruises that were several days old, together with an increase in the cellular elements of the connective tissue.19

It is also appropriate in discussing age changes in bruises to consider the morphological changes seen in open skin wounds and abrasions. Histological ageing, which is based on these changes, will generally follow a definite order and can be divided into three stages:

  • Inflammatory phase (one to three days after injury): vascular, haemostatic, and cellular response.

  • Proliferative phase (up to 10–14 days after injury): epithelial and connective tissue regeneration.

  • Reorganisation or remodelling phase (several months after injury).

Table 5 gives a brief timetable of changes suggested by Raekallio,20 applicable to the histological examination of skin wounds in general.

Table 5

Suggested schema for the histological estimation of the time interval from infliction of injury to death in open skin wounds and abrasions


Biochemical methods are of little practical use in routine forensic practice.

Although measurement of the increase of bilirubin content in haematomas is possible, and seems to correlate with the age of the haematoma (K Laiho at the 8th meeting of the Scandinavian Forensic Society, Vedbaek Denmark 1982),21 there are serious methodological difficulties with this approach. Attempts to separate bilirubin from haemoglobin in supernatants of tissue homogenates have been largely unsuccessful and have led to considerable losses of bilirubin. Furthermore, the turbidity of the supernatant is also a problem in spectrophotometric analysis.

Extraction of haem and extractable iron from homogenates is straightforward. However, the concentration of extractable iron alone cannot be used for age estimation because its concentration obviously depends on the amount of haemoglobin present and degraded in the tissue. Total iron concentration minus the control values is at least partly related to the initial concentration of haemoglobin present in the tissue. The ratio of extractable iron in relation to haem content reflects the change between the amount of haem or haemoglobin present in the tissue and the degraded extractable iron, but is influenced more by the amount of blood originally present than the ratio based on the total iron. When the haemorrhage in the tissue is very small and the values of extractable iron, total iron, and haem are nearly the same as in the control tissue, the evaluation of the age of a haematoma by ratio comparison would give unreliable results.


Dating a bruise from its colouration is problematical for various reasons, as discussed earlier. One of these reasons is that in general forensic practice the colour of a bruise on a cadaver or a living person is assessed by naked eye examination which, by its very nature, is subjective and conditions for assessment are virtually impossible to standardise.

Despite the limitations of relating colour change to date of trauma infliction, it is useful to apply a standardised method for measuring colour and intensity (lightness or luminance) of a bruise.

There are two standard techniques that are used to measure colour and that have been applied to bruising, namely: colorimetry and spectrophotometry.

Measurement of skin colour by colorimetry was initially used22, 23 to assess hypostasis in relation to time of death estimation and was subsequently applied to colour changes in bruising (our unpublished results, 1997).24 The colorimeter is designed to work on the same principle as the human eye, which interprets the pigmentation of an image by calculating the amount of light, as well as the proportions of its constituent red, green, and blue colours (fig 1).

Figure 1

Tristimulus colorimeter.

It allows an accurate, non-subjective assessment to be made, giving a numerical value based on an internationally recognised colour system: system CIE (Commission Internationale de l'Eclairage) L*a*b*. Thus, any object can be described in terms of its spatial position on a three dimensional axis (fig 2). The vertical axis L* describes the luminance of the image, the horizontal a* value describes the position of the red–green axis, and the b* value describes the position on the yellow–blue axis.

Figure 2

Three dimensional colour axis.

The technique is easy to perform, fast, gives reproducible results, and is non-invasive. The test surface is lit by a standard illuminant produced by a xenon flash lamp, which approximates to the same conditions as daylight, and the light is directed to three measuring filters for the three standard tristimulus values X, Y, and Z. A complicated series of formulae is required to convert the raw data into the L*a*b* format, which is the conventional form for communicating the results.

Using this technique, data recently presented (our unpublished results, 1997) were in broad agreement with naked eye observations in relation to colour changes. It was noted in this study of 93 live subjects examined in a casualty department that, although there was great variability in colour change, the results followed the general trend seen in other studies. Furthermore, in common with the observations of two of these3, 4 a yellow colour was not seen until towards the end of the first day.

Colour change can also be measured using spectrophotometry. This technique is used in many branches of science and its main use is to identify substances by their colour properties.

Spectrophotometry works on the principle that different substances absorb, reflect, or emit light in different ways. These changes in light intensity, relative to its wavelength after it has interacted with the sample, can be measured. To summarise the process, a spectrophotometer emits light from a lamp on to a test sample. The sample absorbs some of the light, with the rest passing through. In the spectrophotometer, a diffraction grating disperses the light from the sample by wavelength and the spectrum formed is observed with a detector. Reflectance spectrophotometry (light measured on the same side of the sample as the source of light and used on opaque samples such as solids) is now commonly used in dermatology. It was first used to analyse skin colour25 and has since been developed to measure colour changes in skin under various circumstances (such as the application of corticosteroids). The technique has also been applied to measure colour changes in subcutaneous bruises over time.26 These authors performed measurements between wavelengths 450 and 700 nm (visible wavelength 360/380–750/780 nm) and found the best results to be between 540 and 580 nm. Although they were not able to produce unambiguous results on the ageing of bruising, the authors concluded that the technique could be used as an objective non-invasive technique for bruise analysis. However, a recent study27 using both excised postmortem samples and live subjects found spectrophotometry to be of limited use because of the large number of variables involved.

Differentiation from artefacts and other postmortem appearances

Clearly, it is essential to establish whether an area that is discoloured and has the appearance of a bruise is in fact a bruise. This is generally straightforward in a non-putrefied cadaver, although even in such cases there are occasions when confusion may arise.


The extravasation of blood into the tissues after death, for whatever reason, in certain circumstances can lead to misinterpretation. I have avoided using the word bruising in relation to postmortem events because the forensic understanding of the word implies an antemortem phenomenon. The phrase “pseudo bruise” is perhaps preferable and could be applied to postmortem discolourations that resemble true bruises.

This problem is of more than academic interest and, as most forensic pathologists will confirm, the distinction between bruising and pseudo bruising can sometimes be difficult if not impossible, both at necropsy and even after routine histological and immunohistochemical examination.

At this point, it is appropriate to consider whether blunt trauma delivered after death and causing tissue disruption can result in extravasation of blood with an appearance indistinguishable from bruising produced before death. It has been reported in one study7 that the production of postmortem “bruises” required the application of considerable violence, with the resulting bruise almost always being small or wholly disproportionate to the force delivered. Interestingly, however, they found that a blow with a wooden mallet delivered after death to the occiput (moderate force was used so as not to fracture the skull) can cause blood extravasation involving the full thickness of the scalp up to one inch (2.54 cm) in diameter. Bearing in mind that the scalp is rich in blood vessels and that cadavers are stored in the supine position, this might not be an unusual phenomenon, and therefore such bruising requires careful interpretation.


Postmortem lividity or hypostasis, which results from pooling of blood after death as a result of gravity, gives a red/purple appearance to the areas where the blood is pooled. Usually there is no difficulty differentiating hypostasis from bruising, except where the hypostasis has a patchy appearance, or bruising is on the back (cadavers stored in a mortuary in a supine position are usually found with a substantial proportion of lividity on the back). The situation can be compounded by congestive changes to the body, particularly where congestion is seen in association with deaths involving some kind of trauma.

Cases involving a mechanical asphyxial mode of death, such as manual strangulation or postural (positional) asphyxia, might show areas of congestion and genuine bruises, the sizes of which are larger than would be expected because of the increased volume of blood in surrounding vessels, which will escape and contribute to the bruised area. In addition, congested areas may show postmortem ecchymoses that resemble petechial haemorrhages. These are often seen within areas of hypostasis.

Although it is recommended that a careful dissection of the involved area should be carried out to assess whether, as in the case of a bruise, the blood has escaped from blood vessels into the surrounding tissue, occasionally (particularly with pronounced congestion) this can still be problematical. It is sometimes useful to move the cadaver into another position to allow drainage of pooled blood to a secondary position; true bruising will remain in the same position.


All pathologists are familiar with the typical appearance of postmortem injuries, which have a tendency to a yellowish brown “bloodless” appearance and lack vital reaction. To the naked eye the bruising accompanying injuries (such as lacerations) demonstrate the appearance of the extravasation of blood (red/purple/blue when fresh) and other changes reflecting tissue reaction to injury, and which are dependent on the interval between the infliction of trauma and death. However, in many instances, where there is congestion of the cadaver, sufficient blood escapes from vessels damaged after death to give an appearance of bruising that is indistinguishable from a fresh injury occurring shortly before death.

The postmortem dissection procedure will also produce artefactual bruising that is indistinguishable from true bruising. This is particularly problematical when pronounced congestion is present or where the area in question is very vascular. The pathologist carrying out a second necropsy must be aware of this and not over interpret findings. Furthermore, postmortem dissection will also facilitate the migration of genuine bruising. Against this background, true bruises will be modified after death as discussed in the next section.

A particular situation that can lead to misinterpretation concerns the examination of the neck structures in asphyxial deaths involving compression—for example, manual strangulation. In such cases it is important to bear in mind that the neck structures will probably be congested. The assessment of bruising and its distribution is of paramount importance to the pathologist in his/her examination. As stated above, bleeding from vessels cut or pulled away from other structures occurs as part of the postmortem dissection and evisceration, and results in collections of blood in soft tissues and organs that might be thought erroneously to be true bruises. With regard to the neck, even with careful dissection of the anterior structures away from the cervical vertebral column, considerable extravasation of blood can still occur. Such artefactual production of bruising has been described as the Prinsloo-Gordon effect.28 These authors recommend that a bloodless field be produced in the neck region before examination and removal of the anterior structures. This is achieved by opening the skull as well as the chest and abdomen to allow free flow of blood from these areas before dissection of the neck.

The differentiation between haemorrhage that has occurred before or after death might be impossible where injuries are inflicted just before, during the agonal phase, or within a short time after death.

Clotted haemorrhages were originally thought to occur only during life, but it has been shown that coagulation in blood can occur as late as six hours after death.29 Furthermore, it has also been found30 that the standard histological staining methods for fibrin are not conclusive and that the absence of fibrin does not mean that the haemorrhage occurred after death. Even with immunohistochemical studies, the same author could not distinguish postmortem from antemortem fibrin with certainty. In contrast, however, well preserved fibrin networks found at necropsy performed two to three days after death pointed to an antemortem or agonal haemorrhage.31 This is because a large proportion of a postmortem subcutaneous haemorrhage undergoes fibrinolysis within one day of its production.


Another area of practical difficulty for the pathologist is the differentiation of bruises from marks caused by resuscitation. Because in most instances resuscitation occurs around the time of death with some degree of maintenance of a circulation, or at least intermittent forced movement of blood within vessels, there is every possibility of producing extravasation of blood to the tissues that is indistinguishable from true bruising. Therefore, care should be taken in the interpretation of injuries in areas of the body where resuscitation has taken place. These areas include the face, neck, and chest. Frequently, the question arises in cases involving pressure to the face and neck, especially manual strangulation, where fingertip type bruising, as well as other marks such as abrasions caused by fingernails, are commonly found. The problem is compounded in areas where there is intense congestion.


With the increasing postmortem interval, bruises become more diffuse and are frequently accentuated in intensity as a result of the degradation products of haemoglobin. Indeed, bruises can appear a day or two after the postmortem examination that were not visible at the first necropsy, or those that were seen initially can appear more pronounced. Fingertip bruises indicative of grip marks are a particularly good example of this phenomenon. With the onset of putrefaction, the body becomes discoloured and bruises become modified in their appearance, making their accurate assessment difficult. Immunological methods have demonstrated the usefulness of glycophorin A, a constituent of red blood cell membranes, as a marker to differentiate between true bruising and putrefactive discolouration.32 Although haemoglobin pigments readily filter through blood vessels, erythrocyte membranes do so less easily because of their molecular size. Therefore, bruises will contain a greater amount of erythrocyte membrane material than areas of discolouration resulting from putrefactive change. However, glycophorin A cannot help to differentiate between antemortem and postmortem injury because extravasated blood from vessels includes erythrocytes, regardless of whether the damage occurred before or after death.