Pollanen, Michael S. MD, PhD; Perera, S D. Channa MD; Clutterbuck, David J.
One of the basic assumptions of forensic medicine is that the autopsy can be used to detect injuries, and thus unnatural deaths such as homicide. This premise is based on the proposition that detecting evidence of injury in the form of hemorrhage into the interstitial space of tissues allows the inference that force was applied to tissues during life, resulting in vascular damage and the interstitial extravasation of blood (eg, bruising). However, this assumption is not universally valid and has implications for the autopsy diagnosis of injury. The lack of validity relates to the fundamental postmortem process of hypostasis (lividity or livor mortis). Postmortem hypostasis is defined as the process where fluid blood settles in blood vessels under the influence of gravity, after death. Classic teaching indicates that lividity is intravascular only, thus differentiating postmortem hypostatic congestion or engorgement of blood vessels from the extravascular (interstitial) accumulation of blood that has been used to define tissue damage before death.1,2 However, while postmortem hypostasis is widely believed to be an intravascular process, there is anecdotal evidence that postmortem leakage of gravitationally distended blood vessels may give rise to extravascular blood, thus forming pseudo-bruises3–5 or hemorrhagic lividity. The significance of the postmortem hypostatic hemorrhage is that the process can lead to doubt about the significance of the autopsy findings of bruising, if the extravasated blood is present in areas of lividity. In particular, hemorrhagic lividity in the neck can lead to the over-diagnosis of injuries to the neck such as strangulation, which is defined by the presence of internal hemorrhages. The legal implications of such pseudo-bruising include misidentification of violent neck injury and wrongful conviction of an accused person for a nonexistent crime.
To determine whether the postmortem hypostasis and autolysis in the early postmortem period could result in a reproducible form of hemorrhagic lividity, we developed a human cadaveric model for the controlled induction of postmortem hypostasis in the neck. The development of postmortem hypostatic hemorrhage in the necks of cadavers was studied using macroscopic (dissection) and microscopic (histology after formalin-fixation and paraffin embedding of tissues) methods.
MATERIALS AND METHODS
The experimental design and use of postmortem human subjects were approved by the Health Sciences II Research Ethics Board, Office of Research Services, University of Toronto (protocol number 12082), and the Office of the Chief Coroner for the Province of Ontario.
The 5 subjects used for the research were human cadavers donated for medical research and anatomic dissection in the Faculty of Medicine, University of Toronto. All subjects died of natural disease either in a healthcare facility or at home. Each potential cadaver was screened upon entry into the Faculty of Medicine and assessed for inclusion in the study. The criteria for inclusion in the research project were: (i) nonfixed lividity (blanching to moderate digital pressure) distributed on the posterior torso, or lack of lividity at the time of initial assessment; (ii) the absence of injuries, needle punctures, intravenous cannulae, or other recent therapeutic interventions in the neck, supraclavicular fossae, or face; and (iii) the absence of facial or conjunctival petechial hemorrhages. The salient characteristics of the subjects used for the study are summarized in Table 1.
All subjects were placed prone on a wooden board and fixed into position using a canvass harness. The neck was gently extended such that the weight of the head and neck region was supported by the chin and the ventral neck was completely elevated from the surface of the board. Once the body was fixed to the board, the board was placed into the experimental position. The experimental position was inclination of the board at an angle of approximately 25°, such that the ventral neck was the lowest point (ie, prone and head down position) with the head resting on the chin. The body was kept in the experimental position for 24 hours at room temperature (∼22°C) and subsequently for an additional 24 hours while refrigerated (4°C), but still in the experimental position. Thus, each subject was placed in the prone-head down position for a total of 48 hours, with the first interval at room temperature and the second interval under refrigerated conditions. At the conclusion of the 48 hours of experimental positioning, the body was removed from the experimental apparatus and the neck was dissected. In all cases, dark purple fixed lividity had formed in the ventral neck.
Dissection of the Neck
In each subject, the neck was dissected after brain removal, evisceration of the torso, and elevation of the neck such that the venous system was adequately drained through the jugular foramina and subclavian veins. The neck dissection was performed in layers using incisions along the sides of the neck joined to the intermastoidal scalp incision (used to remove the brain), and the upper portion of the Y-shaped incision used to open the torso. The skin of the ventral and lateral neck was reflected by incising the subcutaneous fat/platysma layer. This is most readily done by first reflecting the skin at the sides of the neck to the level of the jaw and then reflecting the skin of the ventral neck to the level of the chin. Once the entire anterolateral neck was exposed, then the skin of the cheeks was also reflected. The facial skin is reflected by cutting through the external auditory canal, and continuing the subcutaneous incisions to the level of the nose and lateral canthus. Once the entire anterolateral neck and face were exposed, the fibrous membrane over the superficial strap muscles in the midline was removed. Each strap muscle was elevated and reflected superiorly (ie, cutting the origin of the muscle, but leaving the muscle attached to the anatomic insertion point). Each layer of the neck dissection was photographed in the same plane, including orientation and close-up images. The neck organs were eviscerated with the tongue by incising the floor of the mouth along the mandible. The neck organs were dissected ex situ using standard methods after immersion fixation in formalin.
Representative sections of each of the soft tissues of the neck, including the strap muscles were obtained. The larynx was prepared in parasagittal serial sections as described previously.6,7 All histologic sections were stained with hematoxylin and eosin.
In all subjects, intense fixed postmortem hypostasis developed in the anterior neck, supraclavicular fossae, and upper chest (Figs. 1–4). In 2 subjects (subject 1, Fig. 1, and subject 4, Fig. 4), patchy geographic cutaneous hemorrhages developed in the supraclavicular fossae region and/or thoracic inlet. Extensive punctate cutaneous hemorrhages or Tardieu spots developed in subject 3 (Fig. 3), and to a lesser extent in subject 2 (Fig. 2). In all subjects, there was blanching of the livor mortis due to pressure on the chin during the experimental positioning.
Hypostatic hemorrhages were found in the neck and/or scalp in all subjects. In subjects 1 to 3, hypostatic neck hemorrhages were present. Hypostatic scalp hemorrhages were found to some extent in all subjects, but were most prominent in subjects 2 and 5 (Fig. 5). In subject 1 (Fig. 1), the hypostatic neck hemorrhages were confluent and geographic, and involved the superficial soft tissue planes on the right anterior neck at the midline, the submental zone, and along the anterior border of the left sternomastoid muscle. Layered dissection also revealed extensive hemorrhage in the intermediate and deep strap muscle layers, mostly on the right side. In subject 2 (Fig. 2), there was diffuse intramuscular hemorrhage involving the right and left sternomastoid muscles with relative sparing of the other strap muscles of the neck. In subject 3 (Fig. 3), there was relatively minimal intramuscular hemorrhage with only focal hemorrhage in the left sternohyoid muscle and geographic hemorrhage into the subcutaneous fat of the left anterior shoulder. Subjects 4 (Fig. 4) and 5 failed to reveal any hypostatic neck hemorrhages.
The scalp hemorrhages were mostly petechial or punctate and uniformly distributed throughout the scalp (Fig. 5). Most of the hemorrhages were associated with markedly congested and distended blood vessels and some of the hemorrhages had a targetoid appearance (Fig. 5D).
The hypostatic neck hemorrhages had similar characteristics in the different subjects. The hypostatic neck hemorrhages involved both the endomysial and epimysial interstitial compartments of the strap muscles and were characterized by extensive extravasation of red blood cells (Fig. 6). In some areas, the hemorrhages had an angiocentric distribution with a dilated and congested vein in the center; however, most of the hemorrhages were confluent and extensively involved the interstitial space. In larger endomysial and perimysial hypostatic hemorrhages, there was a conspicuous increased concentration of interstitial neutrophils (Figs. 6C, D) simulating an acute inflammatory reaction. In areas of hypostatic hemorrhage with increased neutrophils, there were often extravascular aggregates of platelets (Fig. 6D).
The hypostatic scalp hemorrhages were found in all layers of the scalp but tended to concentrate in the reticular dermis (often with a perifollicular distribution), or at the galea (Fig. 7). The hypostatic hemorrhages of the reticular dermis were often associated with congested dermal blood vessels.
Postmortem hypostasis or lividity is defined as an intravascular phenomenon, but there is anecdotal evidence that postmortem hemorrhages can result from intense lividity. A detailed understanding of how and why this hemorrhagic form of lividity occurs is lacking, but it has significant medicolegal importance. For example, in cases with intense anterior lividity of the head and neck, pseudo-bruising could develop that confounds the diagnosis of subscalp bruising from blunt impact trauma and neck bruising in strangulation. This problem is not entirely theoretical—in some cases of disputed postmortem/antemortem neck hemorrhage, the issue has precipitated reviews of criminal convictions for murder and postconviction relief. Thus, the practical nature of the problems that arise from hypostatic hemorrhage makes the phenomenon worthy of careful consideration, including mechanistic analysis using experimental models. On this basis, the current cadaveric model was developed to create a system to study hypostatic hemorrhages, which can be induced under controlled conditions. It is hoped that this model will permit the macroscopic and microscopic features of hypostatic hemorrhages to be more fully defined and assist in developing methods to differentiate such hemorrhages from real bruising that occurs before death.
The initial general conclusion of this study is that postmortem hypostatic hemorrhages form after the progressive development of increasing gravitational hydrostatic pressure in an autolysing venous plexus. Specifically, as blood accumulates in the complex venous system of the neck, pressure increases within the vessel and blood leaks out of the autolysing blood vessel wall. However, it is highly probable that the development of extensive hypostatic hemorrhage requires an unusually rich venous plexus with complex anastomoses of thin-walled channels that are not well supported by extravascular connective tissues. If this hypothesis is correct, then the intense lividity of any part of the body with a venous plexus with these characteristics can form hypostatic hemorrhages. In fact, in addition to the jugular and scalp venous systems, there are other important venous plexuses that are of medicolegal interest, in terms of anatomic location, that can lead to the presence of forensically problematic hemorrhage: (i) pharyngeal (pharyngeoesophageal) venous plexus8; (ii) epidural (meningorachidian) venous plexus9,10; and (iii) communicating veins between the venous plexuses of the pelvic organs,11 including the rectal venous (hemorrhoidal) plexus.
The classic Prinsloo-Gordon hemorrhages (artifactual hemorrhages in the posterior pharyngeal wall) are well known to most pathologists and are probably the most common routinely encountered internal hypostatic hemorrhage.8 These hemorrhages are related to the perimortem distention and rupture of a series of veins in the prevertebral areas and in the adventitia of the posterior pharynx and superior esophagus known as the pharyngeal (pharyngoesophageal) venous plexus. Similarly, hemorrhages from postmortem distention and rupture of the meningorachidian venous plexus is the most likely explanation for the commonly observed spinal epidural hemorrhage12 that is found in some infants that die naturally and, to a lesser extent, adults. The meningorachidian venous plexus is a dense ramification of venous channels that almost entirely fills the anterior spinal epidural space. Intense lividity in the pelvic floor and perianal/perineal zone can also engorge the hemorrhoidal or rectal venous plexus causing a worrisome appearance of the anus after death. It is quite likely that factors beyond the geometry of the venous plexus (intercommunication of veins of the same caliber) are relevant to the development of hypostatic hemorrhages. At least 2 other anatomic concepts might be relevant. First, it seems likely that the angiosome concept might be useful to explain the distribution of hypostatic hemorrhages. The angiosome13–16 is a continuous 3-dimensional network of arteries and veins that roughly runs perpendicular to the skin surface, from superficial to deep (in contrast to a dermatotomy that represents innervation along the skin). Since blood pools under the influence of gravity, lividity fills blood vessels in adjacent angiosomes. Modeling blood pooling in angiosomes is probably more relevant than considering hypostatic congestion of named blood vessels when analyzing the patterns of hypostatic hemorrhage. In addition, venosomes (the venous portion of the angiosome)14 that are adjacent to each other may have watershed territories linked by communicating veins thereby enhancing hypostatic hemorrhage. Second, it may be that hypostatic hemorrhages principally occur at perivascular and interstitial spaces that have connective tissue that is not well supported by ground substance, matrix macromolecules, and collagen. This includes the simple intramuscular connective tissues such as the endomysium and perimysium of the strap muscles of the neck.17,18
Histologic examination of the postmortem hypostatic hemorrhages revealed true interstitial extravasation of red blood cells, in a pattern indistinguishable from antemortem hemorrhage. An entirely unanticipated finding was the presence of extravascular aggregates and interstitial “infiltrates” of neutrophils that closely resembled an acute inflammatory infiltrate associated with an early healing reaction. These extravascular collections of neutrophils likely represent the gravitational settling of nucleated cells in blood that is passively accumulating in the interstitial space. This is analogous to the formation of a “buffy coat” in blood samples in vitro, and postmortem blood clot in the heart chambers in situ where the cellular components of blood settles out from the fluid component under sedimentation. Another potential but less likely explanation would be postmortem leukocyte transmigration.
The reported model may allow for the development of histologic methods to differentiate real bruises from pseudo-bruising from hypostasis. Some areas for investigation include determining whether endothelial cell adhesion molecules are preferentially expressed only in injured blood vessels.19–21 In addition, there is evidence that platelets extravasated in antemortem bleeding are activated in the process and express adhesion molecules on the platelet membrane, whereas this may not occur in postmortem extravasation of blood. Therefore, a comparative immunohistochemical study of platelet and endothelial cell activation in real bruises and postmortem hypostatic hemorrhages could be elucidating and practically relevant.22
Although this experimental model of hypostatic hemorrhage has reasonable scope for further research, there are some potential limits. The model is limited by the types of human cadavers that are donated for medical research. Thus, it is necessarily the case that the subjects used in this research were elderly people with chronic disease. On this basis, the skin and soft tissue are not entirely representative of the entire age spectrum and not representative of the younger age range where difficulties with the postmortem diagnosis of neck compression often arise. Skin fragility and age-related changes in connective tissues may predispose to the development of hypostatic hemorrhages in the neck, similar to the predisposition to senile ecchymosis or purpura on the upper extremities during life. Despite these limitations, the model does allow the basic process of hypostatic hemorrhage to be probed. Another unexplained aspect of the present study was the fact that hypostatic neck hemorrhages could not be induced in all cadavers, despite the ability to concentrate the livor mortis in the neck and even form Tardieu spots. This indicates that additional poorly understood factors determine whether the hypostatic neck hemorrhages occur after postmortem prone positioning, ie, postmortem prone positioning and advancing postmortem interval is necessary but perhaps not sufficient to create hypostatic hemorrhages.
In conclusion, we have developed a model for the controlled induction of postmortem hypostatic hemorrhages in the neck. The macroscopic and microscopic appearances of the hypostatic neck hemorrhages can confound the postmortem diagnosis of strangulation. The model is a useful basis for studying parameters that will allow the differentiation of real bruising from postmortem pseudo-bruising such as epitopes related to platelet and endothelial cell activation. The results of the present study urge a caution when diagnosing strangulation in bodies with anterior neck lividity.
The authors thank Terry Irvine, Bill Wood, Jerry Topham, Tara Dunn, Patrick Kim, John Phu, and Dr. Noel McAuliffe for assistance.
1. Noriko T. Immunohistochemical studies on postmortem lividity. Forensic Sci Int. 1995;72:179–189.
2. Sannohe S. Change in the postmortem formation of hypostasis in skin preparations 100 micrometers thick. Am J Forensic Med Pathol. 2002;23:349–354.
3. Burke MP, Olumbe AK, Opeskin K. Postmortem extravasation of blood potentially simulating antemortem bruising. Am J Forensic Med Pathol. 1998;19:46–49.
4. Bockholdt B, Maxeiner H, Hegenbarth W. Factors and circumstances influencing the development of hemorrhages in livor mortis. Forensic Sci Int. 2005;149:133–137.
5. Vanezis P. Interpreting bruises at necropsy. J Clin Pathol. 2001;54:348–355.
6. Pollanen MS, McAuliffe DN. Intra-cartilaginous laryngeal haemorrhages and strangulation. Forensic Sci Int. 1998;93:13–20.
7. Pollanen MS. A triad of laryngeal hemorrhages in strangulation: a report of eight cases. J Forensic Sci. 2000;45:614–618.
8. Prinsloo I, Gordon I. Post-mortem dissection artifacts of the neck: their differentiation from ante-mortem bruises. S Afr Med J. 1951;25:358–361.
9. Okumura R, Asato R, Fukuyama H, et al. Epidural venous system (meningorachidian venous plexus) in juvenile amyotrophy of distal upper extremity: assessment with GD-DTPA enchanced volumetric MR study. Comput Med Imaging Graph. 1994;18:193–202.
10. Pearce JM. The craniospinal venous system. Eur Neurol. 2006;56:136–138.
11. Shafik A, Mohi-el-Din M. Pelvic organ venous communications. Anatomy and role in urogenital diseases; a new technique of cystovaginohysterography. Am J Obstet Gynecol. 1988;159:347–351.
12. Rutty GN, Squier WM, Padfield CJ. Epidural haemorrhage of the cervical spinal cord: a post-mortem artefact? Neuropathol Appl Neurobiol. 2005;31:247–257.
13. Taylor GI, Palmer JH. The vascular territories (angiosomes) of the body: experimental study and clinical applications. Br J Plast Surg. 1987;40:113–141.
14. Taylor GI, Caddy CM, Watterson PA, et al. The venous territory (venosomes) of the human body: experimental study and clinical implications. Plast Reconstr Surg. 1990;86:185–213.
15. Houseman ND, Taylor GI, Pan WR. The angiosomes of the head and neck: anatomic study and clinical applications. Plast Reconstr Surg. 2000;105:2287–2313.
16. Rabson JA, Hurwitz DJ, Futrell JW. The cutaneous blood supply of the neck: relevance to incision planning and surgical reconstruction. Br J Plast Surg. 1985;38:208–219.
17. Light N, Champion AE. Characterization of muscle epimysium, perimysium and endomysium collagens. Biochem J. 1984;219:1017–1026.
18. Sakamoto Y. Histological features of endomysium, perimysium and epimysium in rat lateral pterygoid muscle. J Morphol. 1996;227:113–119.
19. Dressler J, Bachmann L, Koch R, et al. Estimation of wound age and VCAM-1 in human skin. Int J Legal Med. 1999;112:159–162.
20. Dressler J, Bachmann L, Kasper M, et al. Time dependence of the expression of ICAM-1 (CD 54) in human skin wounds. Int J Legal Med. 1997;110:299–304.
21. Dressler J, Bachmann L, Strejc P, et al. Expression of adhesion molecules in skin wounds: diagnostic value in legal medicine. Forensic Sci Int. 2000;113:173–176.
22. Thomsen H, Krisch B. The postmortem activation status of platelets. Int J Legal Med. 1994;107:111–117.
© 2009 Lippincott Williams & Wilkins, Inc.