Forensic Science, 5 (1975) 11-19 0 Elsevier Sequoia S.A., Lausanne - Printed in The Netherlands
UNEXPECTED NATURAL DEATH OF CEREBRAL IN MEDICOLEGAL PRACTICE
ORIGIN
J.C. de VILLIERS Department of Neurosurgery, (South Africa)
Groote Schuur Hospital,
Observatory,
Cape Town
SUMMARY Due to the rapid evolution of vascular lesions it is not surprising that most causes of sudden death of cerebral origin are due to vascular pathology. Of the traumatic causes extradural haemorrhage is a fairly common clinical entity but as a cause of death declining in its frequency. Sources of diagnostic error can be attributed to the fact that not all patients with extradural haematomas have marked external evidence of trauma and a significant number, particularly children and adolescents, show no radiological, clinical or I for that matter, post-mortem evidence of a fracture. Subdural haematomas of a chronic variety are usually produced by minor trauma and occur predominantly in the older person. Acute subdural haematomas are most frequently the result of trauma and may be rapidly fatal due to the associated massive cerebral damage. That intracranial aneurysm or angioma may rupture into the subdural space and cause an acute or chronic subdural haematoma, is less widely appreciated. The acute spontaneous arterial subdural haematoma due to the rupture of a cortical vessel, usually one affected by atheroma, into the subdural space is an uncommon entity. It should be looked for specifically in patients with minimal trauma and the clinical picture of an acute subdural haematoma. Subarachnoid haemorrhage due to aneurysmal rupture is still the common cause of unexpected rapid demise in young adults. There is very little evidence that antecedent trauma or exertion play a part as precipitating factors. Centrally placed aneurysms situated at the anterior communicating artery origin or terminal carotid seem to be particularly malignant in their effects. Cause of death is usually massive extrusion of blood into the intracranial cavity with increasing intracranial pressure, compressive haematoma formation and widespread arterial spasm with ischaemic consequences. Whether aneurysmal rupture can be caused by trauma cannot really be satisfactorily resolved. Intracerebral haemorrhage is most commonly due to hypertension but, as in the case of other haematomas, bleeding disorders may also be a cause. Intracerebral haematoma may, however, also result from rupture of micro-angiomata and the brain should be carefully examined for them in the young patient without evidence of hypertension. Hypertensive crises occurring in people on monoamine oxidase inhibitors should also be remembered as a cause of intracerebral haemorrhage.
The brain as the essential central regulatory system of the entire body is well-guarded, both anatomically and physiologically, to maintain its functions in a fluctuating and, at times, dangerously variable external and internal environment. Any insult which rapidly transgresses these safeguarding
12
mechanism will cause failure of the central regulatory processes and result in rapid demise of the patient. Rapid deterioration or sudden abeyance of brain function is usually vascular in origin due to the acute evolution of occlusive or haemorrhagic vascular lesions of the nervous system. Of the two haemorrhage is the more dramatic in onset and also the more frequent cause of sudden death, and spontaneous subarachnoid haemorrhage, due to aneurysmal rupture, is a common example of rapid death in young persons. Haemorrhage at various levels in the cranium must, however, be considered as a cause of rapid demise, i.e. death within a period of 24 hours. EXTRADURALHAEMORRHAGE This is always traumatic in origin and there is almost always an associated skull fracture. These two statements are so widely accepted that it is almost forgotten that an extradural haemorrhage may occur without a skull fracture and may be unassociated with gross external evidence of trauma. Most large series of extradural haematomas usually include a few cases in which no fracture is demonstrated radiologically, at operation or at autopsy. That the radiological absence of a fracture is not absolute evidence that no fracture is present is common knowledge. Of the 167 cases of extradural haematoma reported by Galagher and Browder [l] six patients had no clinical or autopsy evidence of fracture. Of particular interest in this respect are patients under the age of 15 years, in whom the absence of a fracture in the presence of an extradural haematoma is relatively common [Z] . This is also the age group in which extradural haemorrhage occurs particularly frequently [3]. Mealey [Z] found in his series that acute extradural haemorrhage without a skull fracture only occurred in older children and adolescents. In each instance, the head injury was of a direct, low velocity closed type and never particularly severe. In not one of these cases was there an initial period of unconsciousness, therefore the classical lucid interval was absent. The latent period varied from one to four hours in 4 of the 5 cases. An illustrative case of this type is reported by boy was thrown onto a heap of grass cuttings and Carter [ 41. An ll-year-old had no headache in the interim period until he developed a typical migraine attack, went to sleep and could not be woken up. He was found to have an extradural haematoma at operation, but no skull fracture. The pathogenesis of this type of extradural haemorrhage was first demonstrated by Sir Charles Bell [5], who showed that when the skull of a cadaver was struck with a wooden mallet the dura separated from the bone at the site of local impact. This work was repeated and confirmed experimentally by Galagher and Browder [ 11. They found that bleeding starts in the area where the dura separates from the bone and the dura is progressively stripped from the bone over a widening area as the haematoma increases in size. In this way, a haematoma can form without a major vessel being torn. This
13
seems to be particularly liable to occur in children where there is a greater elesticity of the skull than in the adult. Extradural haematomas can occur in newborn infants with haemorrhagic disease and in this instance there may not only be the compressive effects of the haematoma but the child may exanguinate and develop hypovolaemic shock due to the volume of blood lost intracranially. Other bleeding disorders, such as haemophilia, have also been reported as causes of extradural haematoma [6] in the absence of gross trauma. Summarizing, an extradural haematoma may be a surprise finding in a person dying after relatively minor trauma or where no external evidence of trauma is present. SUBDURAL
HAEMATOMA
This type of haemorrhage is also so frequently associated with trauma that other causes tend to be disregarded. The common chronic subdural haematoma occurs very frequently in the elderly and it is often not possible to elicit a history of previous head injury, but if one can, it is usual to find that it was of a minor nature. In the elderly there is often some degree of cerebral atrophy with increase in the subdural course of the bridging veins which may rupture after even minor trauma. This is almost acceptable as a natural cause of death where no history of trauma is at hand. Death results when a final episode of tentorial herniation occurs and there is irreversable brain stem damage due to distortion and local compression at the tentorial hiatus. Acute subdural haematoma is almost always equated with trauma and certainly is by far the commonest cause of this condition. It is always associated with underlying brain laceration. Rupture of an intracranial aneurysm or, less commonly, an angioma into the subdural space may give rise to an acute, subacute or chronic subdural haematoma. Logue [7] found that 8% of subdural haematomas in his experience were due to ruptured aneurysm. Boop et al. [S] state that subdural haematoma was an uncommon complication of aneurysmal rupture and found that only l--Z% of all subdural haematomas result from such a cause. They did, however, find that 2-S% of all ruptured aneurysms have an associated subdural haematoma. No figure is given for the incidence of this complication but its occurrence is mentioned in the cooperative study on intracranial aneurysms and subarachnoid haemorrhage [9]. In our experience it is uncommon, certainly not more than l-2% as mentioned by Clarke and Walton [lo] . These authors classified subdural haematoma due to aneurysmal rupture as follows: 1. Subdural heamatoma present after massive and rapidly fatal subarachnoid heamorrhage; 2. An insignificant subdural collection commonly found after severe subarachnoid haemorrhage;
14
3. Subdural haematoma behaving like the traumatic variety with acute compression. Bleeding into the subdural space can occur directly at the time of the first aneurysmal rupture, but previous bleeding which causes local adherence of the arachnoid membrane to the aneurysm predisposes to later rupture into the subdural space. It is well known that an aneurysm may be completely destroyed at the moment of rupture and a very careful search may have to be made for the origin of such a subdural haematoma which may occur with very little associated subarachnoid haemorrhage. Clinically, there may be a long interval from the time of the aneurysmal rupture to the time that the patient presents with his symptoms of a subdural haematoma but it may, on the-other hand, occur very rapidly. Haematoma may occur on the same side, the opposite side, or even bilaterally. An angioma may cause a subdural haematoma in the same way but far less commonly than an aneurysm. ACUTE SPONTANEOUS ARTERIAL
SUBDURAL HAEMATOMA
This is a condition which is not sufficiently well known. It has been fully reviewed by Talalla and McJSissock [ll]. The patient is usually a hypertensive male and over 50 years of age. There is no previous story of injury but a sudden onset of headache, loss of consciousness and signs of hemispheral dysfunction is characteristic. Blood in the cerebrospinal fluid is usually found on lumbar puncture. Of the 8 cases published by Talalla, there was ipsilateral limb paralysis in 7 patients in contradistinction to acute subdural haematoma where 13% in his experience had ipsilateral signs. In explanation they postulate that these false localising signs only occur if there is no intra-axial lesion. The cause for an artery rupturing into the subdural space is held to be a previous insignificant subdural haematoma which heals and is absorbed but leaves an adhesion between the surface of the brain and the inner surface of the dura. Such a filamentous adhesion may then tear with any sudden movement of the head and cause rupture of a small artery. We have dealt with 5 such cases in our department in the last year. The site of haemorrhage at operation was a small atherosclerotic vessel rupturing through the arachnoid and bleeding into the subdural space. Of the other non-traumatic causes of subdural haematoma, bleeding disorders and in particular leukaemia, thrombocytopaenic purpura and anticoagulants should be considered. Even less common but well documented, are subdural haematomas associated with infiltrative lesions of the dura such as carcinomatosis, sarcomatosis or an infiltrating meningioma. In these instances, the essential pathology seems to be intravascular tumour in the dural veins [ 121. In summary, it may be stated that, as with the extradural haematomas,
15
trauma is the common cause of a subdural haematoma. Only when indisputable evidence of trauma is present can this be held to be the cause, otherwise one of the other lesions must be searched for. SUBARACHNOID
HAEMORRHAGE
As stated before, this is the commonest cerebral cause of sudden, unexpected death in young people. In the majority of cases such subarachnoid haemorrhage is due to rupture of an aneurysm on one of the arteries at the base of the brain. Most of these aneurysms are developmental in origin but mycotic aneurysms are not uncommon in South Africa. Not all mycotic aneurysms are situated on the peripheral vessels on the surface of the brain as is classically described [ 131. Moritz and Zancheck [14] found in a military population that 91 cases of unexpected sudden death resulted from non-traumatic intracranial haemorrhage, and it was estimated that these 91 comprised approximately one-third of all cases of spontaneous intracranial haemorrhage that were reported to their institute between 1941 and 1945. In 40 instances, haemorrhage was proved to be due to a ruptured aneurysm, and in 21 instances it was inferred. Not one of these patients had any evidence of antecedent trauma or systemic disease. The highest mortality was in the age group 20-48 years. They had no statistical evidence that physical exertion predisposed to the rupture of an aneurysm although they imply this. Helpern and Rabson [15] found that 95 cases, or 4.7% of a total of 2030 necropsies, were due to spontaneous subarachnoid haemorrhage. This represented more than 25.7% of the 370 deaths from diseases of the nervous system. Spontaneous subarachnoid haemorrhage almost equalled spontaneous cerebral haemorrhage into the basal ganglia as a cause of sudden death in their experience. Dinning and Faulkner [16] coud find no significant difference as regards age, sex, location of the aneurysm and such predisposing factors as stress, size of aneurysms and pre-existing cardiovascular disease between two groups of autopsied cases with aneurysms of which one group presented clinically and the other forensically. The only striking difference concerned the duration of symptoms. In the hospital series the clinical course was usually sufficiently prolonged not only to allow the patient to reach hospital but also to permit a correct diagnosis to be made, whereas in a forensic case death was generally sudden, or rapid and unexpected. From the work of other authorities, however, it would seem that more centrally placed aneurysms on the anterior cerebral or terminal internal carotid artery have a more regular fatal outcome. Clinically, this would also correlate with our experience. Moritz and Zancheck [14] failed to correlate the position of the aneurysm and the type of haemorrhage with the time of survival. Housepain and Poole [17] found that anterior communicating artery aneurysms were particularly lethal. In Bebin and Courier’s [18] series
16
40% of patients who died had aneurysms in the anterior communicating region. They found that these aneurysms are in a more dangerous situation and are also more likely to rupture into the hemisphere and into the ventricle. Bebin and Courier [18] analysed a series of 35 autopsied cases in which there was indisputable evidence of death due to the ruptured aneurysm. Of these, 26 demonstrated both intracerebral and intraventricular haemorrhage: 6 ruptured primarily intraventricularly, 2 ruptured primarily subdurally and 1 intracerebrally. All of them died in the first episode. In general they found a good correlation between the amount of intraventricular bleeding and rapidity with which death occurred. In some of these cases blood distended and filled all the ventricles and the patient died within 6 hours of onset of his last episode. In others, all the ventricles were filled with blood but not distended and death occurred in 6 to 48 hours. In others there was blood in some of the ventricles and the patients lived 2-6 days. In some cases, however, there was no correlation. It is noteworthy that none of these 35 patients died of pure subarachnoid haemorrhage. The cause of death in subarachnoid haemorrhage has in the past been attributed to compression of the medulla but it is probable that diffuse cerebral arterial spasm when sufficiently widespread and involving vital areas is a more likely cause of sudden death. Massive cerebral compression due to a large volume of arterial blood suddenly released into the ventricles could also be a significant cause of death. Infarction plays a very prominent part in the problem and probably occurs secondary to spasm or local obstruction of vessels by a clot [ 191. The differential diagnosis of spontaneous subarachnoid haemorrhage is most commonly from traumatic subarachnoid haemorrhage. Trauma is the commonest cause of subarachnoid haemorrhage but the haemorrhage usually occurs on the convexity rather than at the base and is usually associated at its maximum site of occurrence with some local injury to scalp, skull or underlying brain. Whether trauma can cause rupture of an aneurysm is a moot point, but this can lead to medicolegal difficulties of considerable magnitude [15,20]. We have recently had two patients who gave a history of trauma in whom this problem is well illustrated. In the first case, a man suffered severe focal injury to the left frontotemporal region, was deeply unconscious and during his phase of recovery developed a fixed dilated pupil on the left side. This was attributed to a complication of his cerebral injury, probably an intracranial haemorrhage, and an arteriogram was performed. This revealed an aneurysm on the internal carotid artery at the level of the posterior communicating artery. It is difficult to decide whether this had any relationship to the injury. The chances are that this aneurysm ruptured irrespective of the cerebral injury because of the latent period between the onset of compression from the aneurysm to the time of his trauma. On the other hand, one may suggest
17
that he had minimal focal damage to the aneurysm at the time of injury with gradual distention afterwards. The second patient was a female who gave the story that she had been assaulted by her husband. All other evidence, however, was to the contrary. (It is not infrequent for a patient who suffers the sudden severe headache of a subarachnoid haemorrhage to be under the impression that something had struck him on the head and that this is the cause of the headache.) This woman apparently fell unconscious and when she woke up was confused, complaining bitterly of headache. In this fall she bruised her head and sustained a linear fracture of the skull. The whole sequence of events was so suspicious that arteriography was performed and this revealed bilateral aneurysms at the posterior communicating artery level. She subsequently died due to severe cerebral spasm and was never a candidate for surgery. One can well imagine to what difficulties this could have led to, had she died suddenly and the story that her husband had assaulted her had been taken seriously. SUBARACHNOID
HAEMORRHAGE
IN ALCOHOLIC
INTOXICATION
A number of cases have been reported in which subarachnoid haemorrhage resulted from rupture of a normal artery after a relatively minor injury [21]. This has been described in intoxicated as well as in non-intoxicated individuals but particularly in the former. The type of injury is often in the nature of a blow to the head striking the skull near its base. At autopsy a longitudinal tear of a normal vessel, usually the vertebral or basilar artery, is found, but no other abnormality or evidence of inherent weakness of disease of the vessel wall is evident. In a number of these cases it is impossible to demonstrate the origin of the haemorrhage at the base of the brain as one finds with aneurysmal rupture. The other significant finding is that in many of these cases there is no evidence of external violence. Death occurs virtually instantaneously. It has been suggested that the vascular dilatation and elevation of blood pressure which occurs with excessive intake of alcohol may be responsible for a set of circumstances in which the vessel is particularly liable to rupture if subjected to trauma. INTRACEREBRAL
HAEMORRHAGE
Hypertensive intracerebral haemorrhage is a common cause of death. Haemorrhage usually occurs from small vessels in the substance of the brain due to rupture of Charcot-Bouchard aneurysms. If massive, such a haemorrhage is usually fatal within 24 hours and may extend intracerebrally with or without secondary rupture into the ventricle. Of other causes of primary intracerebral haemorrhage, bleeding disorders such as haemophilia and thrombocytopenic purpura rank fairly high on the list. When one encounters what appears to be a “primary intracerebral
18
haematoma” in a young person, or one who is normotensive, the diagnosis should be reconsidered. One should look in particular for a micro-angioma of which there may be others scattered throughout the brain, thus indicating the primary pathology [22]. As mentioned previously, an intracerebral haemorrhage or haematoma may be due to a ruptured aneurysm which has bled primarily into the brain substance. This occurs particularly with aneurysms in the anterior communicating region but sometimes a sylvian haematoma resulting from a ruptured mid-cerebral aneurysm may simulate an intracerebral haematoma. In patients with a previous history of psychiatric problems and depression in particular, one should obtain an adequate history of drug and food intake. The patient may have been treated with monoamine oxidase inhibitor antidepressants and taken one of the foods known to cause a hypertensive crisis in the presence of monoamine oxidase inhibitors [23]. Traumatic intracerebral haematoma is common but usually there are other features to indicate their traumatic origin. It almost always extends from the surface inwards and at the surface there is usually local contusion. A common site for this to occur is at the sylvian fissure where the brain abuts against the sphenoid ridge and also subfrontally, but deep haematomas may also occur unrelated to surface trauma. One form of traumatic haemorrhage which is not uncommon in our environment, and which should be looked for specifically because it can very easily be missed, is that which results from a small perforating stab wound, either of the orbit or of the cranium. Such a wound inflicted through the scalp area, particularly in an individual with short, tightly curled hair, can easily be overlooked. The risk of misinterpreting this type of wound, has been stressed by Helpern [23] and others. ANOMALIES
AT THE CRANIOCERVICAL
JUNCTION
These are uncommon but they are usually missed unless specifically looked for. Assimilation of the atlas, atlanto-axial instability due to agenesis or pseudo-arthrosis of the odontoid and basilar impression are the common anomalies. These patients may be asymptomatic and perfectly normal until a trivial injury such as a sneeze or a slap on the back causes sudden, severe paralysis of all four limbs and respiration as well as impairment of consciousness. Haemorrhage into this region of the spinal cord or acute compression is usually the cause of death; unless looked for specifically this pathological process may be easily overlooked. REFERENCES
1 Galagher, J.P. & Browder, E.J. (1968) J. Neurosurg. 29: 1 2 Mealy, J. (1960) J. Neurosurgery 17: 27
19 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Jamieson, K.G. & Yelland, J.D.K. (1968) J. Neurosurg. 29: 13-23 Carter, A.E. (1961) J. Neurosurg. 18: 155 Bell, C. (1816) Surgical Observations, Ed. Shirst & Reed. London. Longmans Jamieson, K.G. (1954) Aust. N.Z.J. Surg. 24: 56-62 Logue, V. (1951) Modern Trends in Neurology Ed. A. Ferling London Butterworth Boop, W.C., Chou, S.N., & French, L.A. (1961) J. Neurosurgery 18: 834 Sahs, A.L., Perret, G.E., Locksley, H.B. & Nishioka, H. (1966) Intracranial Aneurysms & Subarachnoid Haemorrhage. Philadelphia. Lippincott Clarke, E., & Walton, J. (1953) Brain 76: 367-404 Talalla, A., & McKissock, W. (1971) Neurology 21: 19-25 Russell, D.S., & Cairns, H. (1934) Brain 57 : 32 Roach, M.R., & Drake, C.G. (1965) New Eng. J. Med. 273: 240-244 Moritz, A.R. & Zancheck, M. (1946) Arch. Path. 42: 459-494 Helpern, W., & Rabson, S.M. (1950) Am. J. Med. SC. 220: 262-271 Dinning, T.A.T., & Faulkner, M.A. (1953) Lancet 11: 799 Housepain, E.M., & Poole, L. (1958) J. Neuropath. Exp. Neurol. 17: 409-423 Bebin., J., and Courier, R.D. (1957) A.M.A. Arch. Int. Med. 99: 771 Tomlinson, B.E. (1959) J. Clin. Path. 12: 391-399 Courville, C.B., & Newbar, F.D. (1958) J. Forensic Sciences 3: 174-200 Simonsen, J. (1963) J. Forensic Sciences 8: 97 Gerlach, J. (1969) Progress in Neurological Surgery Vol. 3 Base1 & New York Karger Helpern, M. (1946) Ann. Int. Med. 24: 666-700 De Villiers, J.C. (1965) Brit. J. Psychiat. 12: 109-118
UNEXPECTED NATURAL DEATH OF CEREBRAL IN MEDICOLEGAL PRACTICE
ORIGIN
J.C. de VILLIERS Department of Neurosurgery, (South Africa)
Groote Schuur Hospital,
Observatory,
Cape Town
SUMMARY Due to the rapid evolution of vascular lesions it is not surprising that most causes of sudden death of cerebral origin are due to vascular pathology. Of the traumatic causes extradural haemorrhage is a fairly common clinical entity but as a cause of death declining in its frequency. Sources of diagnostic error can be attributed to the fact that not all patients with extradural haematomas have marked external evidence of trauma and a significant number, particularly children and adolescents, show no radiological, clinical or I for that matter, post-mortem evidence of a fracture. Subdural haematomas of a chronic variety are usually produced by minor trauma and occur predominantly in the older person. Acute subdural haematomas are most frequently the result of trauma and may be rapidly fatal due to the associated massive cerebral damage. That intracranial aneurysm or angioma may rupture into the subdural space and cause an acute or chronic subdural haematoma, is less widely appreciated. The acute spontaneous arterial subdural haematoma due to the rupture of a cortical vessel, usually one affected by atheroma, into the subdural space is an uncommon entity. It should be looked for specifically in patients with minimal trauma and the clinical picture of an acute subdural haematoma. Subarachnoid haemorrhage due to aneurysmal rupture is still the common cause of unexpected rapid demise in young adults. There is very little evidence that antecedent trauma or exertion play a part as precipitating factors. Centrally placed aneurysms situated at the anterior communicating artery origin or terminal carotid seem to be particularly malignant in their effects. Cause of death is usually massive extrusion of blood into the intracranial cavity with increasing intracranial pressure, compressive haematoma formation and widespread arterial spasm with ischaemic consequences. Whether aneurysmal rupture can be caused by trauma cannot really be satisfactorily resolved. Intracerebral haemorrhage is most commonly due to hypertension but, as in the case of other haematomas, bleeding disorders may also be a cause. Intracerebral haematoma may, however, also result from rupture of micro-angiomata and the brain should be carefully examined for them in the young patient without evidence of hypertension. Hypertensive crises occurring in people on monoamine oxidase inhibitors should also be remembered as a cause of intracerebral haemorrhage.
The brain as the essential central regulatory system of the entire body is well-guarded, both anatomically and physiologically, to maintain its functions in a fluctuating and, at times, dangerously variable external and internal environment. Any insult which rapidly transgresses these safeguarding
12
mechanism will cause failure of the central regulatory processes and result in rapid demise of the patient. Rapid deterioration or sudden abeyance of brain function is usually vascular in origin due to the acute evolution of occlusive or haemorrhagic vascular lesions of the nervous system. Of the two haemorrhage is the more dramatic in onset and also the more frequent cause of sudden death, and spontaneous subarachnoid haemorrhage, due to aneurysmal rupture, is a common example of rapid death in young persons. Haemorrhage at various levels in the cranium must, however, be considered as a cause of rapid demise, i.e. death within a period of 24 hours. EXTRADURALHAEMORRHAGE This is always traumatic in origin and there is almost always an associated skull fracture. These two statements are so widely accepted that it is almost forgotten that an extradural haemorrhage may occur without a skull fracture and may be unassociated with gross external evidence of trauma. Most large series of extradural haematomas usually include a few cases in which no fracture is demonstrated radiologically, at operation or at autopsy. That the radiological absence of a fracture is not absolute evidence that no fracture is present is common knowledge. Of the 167 cases of extradural haematoma reported by Galagher and Browder [l] six patients had no clinical or autopsy evidence of fracture. Of particular interest in this respect are patients under the age of 15 years, in whom the absence of a fracture in the presence of an extradural haematoma is relatively common [Z] . This is also the age group in which extradural haemorrhage occurs particularly frequently [3]. Mealey [Z] found in his series that acute extradural haemorrhage without a skull fracture only occurred in older children and adolescents. In each instance, the head injury was of a direct, low velocity closed type and never particularly severe. In not one of these cases was there an initial period of unconsciousness, therefore the classical lucid interval was absent. The latent period varied from one to four hours in 4 of the 5 cases. An illustrative case of this type is reported by boy was thrown onto a heap of grass cuttings and Carter [ 41. An ll-year-old had no headache in the interim period until he developed a typical migraine attack, went to sleep and could not be woken up. He was found to have an extradural haematoma at operation, but no skull fracture. The pathogenesis of this type of extradural haemorrhage was first demonstrated by Sir Charles Bell [5], who showed that when the skull of a cadaver was struck with a wooden mallet the dura separated from the bone at the site of local impact. This work was repeated and confirmed experimentally by Galagher and Browder [ 11. They found that bleeding starts in the area where the dura separates from the bone and the dura is progressively stripped from the bone over a widening area as the haematoma increases in size. In this way, a haematoma can form without a major vessel being torn. This
13
seems to be particularly liable to occur in children where there is a greater elesticity of the skull than in the adult. Extradural haematomas can occur in newborn infants with haemorrhagic disease and in this instance there may not only be the compressive effects of the haematoma but the child may exanguinate and develop hypovolaemic shock due to the volume of blood lost intracranially. Other bleeding disorders, such as haemophilia, have also been reported as causes of extradural haematoma [6] in the absence of gross trauma. Summarizing, an extradural haematoma may be a surprise finding in a person dying after relatively minor trauma or where no external evidence of trauma is present. SUBDURAL
HAEMATOMA
This type of haemorrhage is also so frequently associated with trauma that other causes tend to be disregarded. The common chronic subdural haematoma occurs very frequently in the elderly and it is often not possible to elicit a history of previous head injury, but if one can, it is usual to find that it was of a minor nature. In the elderly there is often some degree of cerebral atrophy with increase in the subdural course of the bridging veins which may rupture after even minor trauma. This is almost acceptable as a natural cause of death where no history of trauma is at hand. Death results when a final episode of tentorial herniation occurs and there is irreversable brain stem damage due to distortion and local compression at the tentorial hiatus. Acute subdural haematoma is almost always equated with trauma and certainly is by far the commonest cause of this condition. It is always associated with underlying brain laceration. Rupture of an intracranial aneurysm or, less commonly, an angioma into the subdural space may give rise to an acute, subacute or chronic subdural haematoma. Logue [7] found that 8% of subdural haematomas in his experience were due to ruptured aneurysm. Boop et al. [S] state that subdural haematoma was an uncommon complication of aneurysmal rupture and found that only l--Z% of all subdural haematomas result from such a cause. They did, however, find that 2-S% of all ruptured aneurysms have an associated subdural haematoma. No figure is given for the incidence of this complication but its occurrence is mentioned in the cooperative study on intracranial aneurysms and subarachnoid haemorrhage [9]. In our experience it is uncommon, certainly not more than l-2% as mentioned by Clarke and Walton [lo] . These authors classified subdural haematoma due to aneurysmal rupture as follows: 1. Subdural heamatoma present after massive and rapidly fatal subarachnoid heamorrhage; 2. An insignificant subdural collection commonly found after severe subarachnoid haemorrhage;
14
3. Subdural haematoma behaving like the traumatic variety with acute compression. Bleeding into the subdural space can occur directly at the time of the first aneurysmal rupture, but previous bleeding which causes local adherence of the arachnoid membrane to the aneurysm predisposes to later rupture into the subdural space. It is well known that an aneurysm may be completely destroyed at the moment of rupture and a very careful search may have to be made for the origin of such a subdural haematoma which may occur with very little associated subarachnoid haemorrhage. Clinically, there may be a long interval from the time of the aneurysmal rupture to the time that the patient presents with his symptoms of a subdural haematoma but it may, on the-other hand, occur very rapidly. Haematoma may occur on the same side, the opposite side, or even bilaterally. An angioma may cause a subdural haematoma in the same way but far less commonly than an aneurysm. ACUTE SPONTANEOUS ARTERIAL
SUBDURAL HAEMATOMA
This is a condition which is not sufficiently well known. It has been fully reviewed by Talalla and McJSissock [ll]. The patient is usually a hypertensive male and over 50 years of age. There is no previous story of injury but a sudden onset of headache, loss of consciousness and signs of hemispheral dysfunction is characteristic. Blood in the cerebrospinal fluid is usually found on lumbar puncture. Of the 8 cases published by Talalla, there was ipsilateral limb paralysis in 7 patients in contradistinction to acute subdural haematoma where 13% in his experience had ipsilateral signs. In explanation they postulate that these false localising signs only occur if there is no intra-axial lesion. The cause for an artery rupturing into the subdural space is held to be a previous insignificant subdural haematoma which heals and is absorbed but leaves an adhesion between the surface of the brain and the inner surface of the dura. Such a filamentous adhesion may then tear with any sudden movement of the head and cause rupture of a small artery. We have dealt with 5 such cases in our department in the last year. The site of haemorrhage at operation was a small atherosclerotic vessel rupturing through the arachnoid and bleeding into the subdural space. Of the other non-traumatic causes of subdural haematoma, bleeding disorders and in particular leukaemia, thrombocytopaenic purpura and anticoagulants should be considered. Even less common but well documented, are subdural haematomas associated with infiltrative lesions of the dura such as carcinomatosis, sarcomatosis or an infiltrating meningioma. In these instances, the essential pathology seems to be intravascular tumour in the dural veins [ 121. In summary, it may be stated that, as with the extradural haematomas,
15
trauma is the common cause of a subdural haematoma. Only when indisputable evidence of trauma is present can this be held to be the cause, otherwise one of the other lesions must be searched for. SUBARACHNOID
HAEMORRHAGE
As stated before, this is the commonest cerebral cause of sudden, unexpected death in young people. In the majority of cases such subarachnoid haemorrhage is due to rupture of an aneurysm on one of the arteries at the base of the brain. Most of these aneurysms are developmental in origin but mycotic aneurysms are not uncommon in South Africa. Not all mycotic aneurysms are situated on the peripheral vessels on the surface of the brain as is classically described [ 131. Moritz and Zancheck [14] found in a military population that 91 cases of unexpected sudden death resulted from non-traumatic intracranial haemorrhage, and it was estimated that these 91 comprised approximately one-third of all cases of spontaneous intracranial haemorrhage that were reported to their institute between 1941 and 1945. In 40 instances, haemorrhage was proved to be due to a ruptured aneurysm, and in 21 instances it was inferred. Not one of these patients had any evidence of antecedent trauma or systemic disease. The highest mortality was in the age group 20-48 years. They had no statistical evidence that physical exertion predisposed to the rupture of an aneurysm although they imply this. Helpern and Rabson [15] found that 95 cases, or 4.7% of a total of 2030 necropsies, were due to spontaneous subarachnoid haemorrhage. This represented more than 25.7% of the 370 deaths from diseases of the nervous system. Spontaneous subarachnoid haemorrhage almost equalled spontaneous cerebral haemorrhage into the basal ganglia as a cause of sudden death in their experience. Dinning and Faulkner [16] coud find no significant difference as regards age, sex, location of the aneurysm and such predisposing factors as stress, size of aneurysms and pre-existing cardiovascular disease between two groups of autopsied cases with aneurysms of which one group presented clinically and the other forensically. The only striking difference concerned the duration of symptoms. In the hospital series the clinical course was usually sufficiently prolonged not only to allow the patient to reach hospital but also to permit a correct diagnosis to be made, whereas in a forensic case death was generally sudden, or rapid and unexpected. From the work of other authorities, however, it would seem that more centrally placed aneurysms on the anterior cerebral or terminal internal carotid artery have a more regular fatal outcome. Clinically, this would also correlate with our experience. Moritz and Zancheck [14] failed to correlate the position of the aneurysm and the type of haemorrhage with the time of survival. Housepain and Poole [17] found that anterior communicating artery aneurysms were particularly lethal. In Bebin and Courier’s [18] series
16
40% of patients who died had aneurysms in the anterior communicating region. They found that these aneurysms are in a more dangerous situation and are also more likely to rupture into the hemisphere and into the ventricle. Bebin and Courier [18] analysed a series of 35 autopsied cases in which there was indisputable evidence of death due to the ruptured aneurysm. Of these, 26 demonstrated both intracerebral and intraventricular haemorrhage: 6 ruptured primarily intraventricularly, 2 ruptured primarily subdurally and 1 intracerebrally. All of them died in the first episode. In general they found a good correlation between the amount of intraventricular bleeding and rapidity with which death occurred. In some of these cases blood distended and filled all the ventricles and the patient died within 6 hours of onset of his last episode. In others, all the ventricles were filled with blood but not distended and death occurred in 6 to 48 hours. In others there was blood in some of the ventricles and the patients lived 2-6 days. In some cases, however, there was no correlation. It is noteworthy that none of these 35 patients died of pure subarachnoid haemorrhage. The cause of death in subarachnoid haemorrhage has in the past been attributed to compression of the medulla but it is probable that diffuse cerebral arterial spasm when sufficiently widespread and involving vital areas is a more likely cause of sudden death. Massive cerebral compression due to a large volume of arterial blood suddenly released into the ventricles could also be a significant cause of death. Infarction plays a very prominent part in the problem and probably occurs secondary to spasm or local obstruction of vessels by a clot [ 191. The differential diagnosis of spontaneous subarachnoid haemorrhage is most commonly from traumatic subarachnoid haemorrhage. Trauma is the commonest cause of subarachnoid haemorrhage but the haemorrhage usually occurs on the convexity rather than at the base and is usually associated at its maximum site of occurrence with some local injury to scalp, skull or underlying brain. Whether trauma can cause rupture of an aneurysm is a moot point, but this can lead to medicolegal difficulties of considerable magnitude [15,20]. We have recently had two patients who gave a history of trauma in whom this problem is well illustrated. In the first case, a man suffered severe focal injury to the left frontotemporal region, was deeply unconscious and during his phase of recovery developed a fixed dilated pupil on the left side. This was attributed to a complication of his cerebral injury, probably an intracranial haemorrhage, and an arteriogram was performed. This revealed an aneurysm on the internal carotid artery at the level of the posterior communicating artery. It is difficult to decide whether this had any relationship to the injury. The chances are that this aneurysm ruptured irrespective of the cerebral injury because of the latent period between the onset of compression from the aneurysm to the time of his trauma. On the other hand, one may suggest
17
that he had minimal focal damage to the aneurysm at the time of injury with gradual distention afterwards. The second patient was a female who gave the story that she had been assaulted by her husband. All other evidence, however, was to the contrary. (It is not infrequent for a patient who suffers the sudden severe headache of a subarachnoid haemorrhage to be under the impression that something had struck him on the head and that this is the cause of the headache.) This woman apparently fell unconscious and when she woke up was confused, complaining bitterly of headache. In this fall she bruised her head and sustained a linear fracture of the skull. The whole sequence of events was so suspicious that arteriography was performed and this revealed bilateral aneurysms at the posterior communicating artery level. She subsequently died due to severe cerebral spasm and was never a candidate for surgery. One can well imagine to what difficulties this could have led to, had she died suddenly and the story that her husband had assaulted her had been taken seriously. SUBARACHNOID
HAEMORRHAGE
IN ALCOHOLIC
INTOXICATION
A number of cases have been reported in which subarachnoid haemorrhage resulted from rupture of a normal artery after a relatively minor injury [21]. This has been described in intoxicated as well as in non-intoxicated individuals but particularly in the former. The type of injury is often in the nature of a blow to the head striking the skull near its base. At autopsy a longitudinal tear of a normal vessel, usually the vertebral or basilar artery, is found, but no other abnormality or evidence of inherent weakness of disease of the vessel wall is evident. In a number of these cases it is impossible to demonstrate the origin of the haemorrhage at the base of the brain as one finds with aneurysmal rupture. The other significant finding is that in many of these cases there is no evidence of external violence. Death occurs virtually instantaneously. It has been suggested that the vascular dilatation and elevation of blood pressure which occurs with excessive intake of alcohol may be responsible for a set of circumstances in which the vessel is particularly liable to rupture if subjected to trauma. INTRACEREBRAL
HAEMORRHAGE
Hypertensive intracerebral haemorrhage is a common cause of death. Haemorrhage usually occurs from small vessels in the substance of the brain due to rupture of Charcot-Bouchard aneurysms. If massive, such a haemorrhage is usually fatal within 24 hours and may extend intracerebrally with or without secondary rupture into the ventricle. Of other causes of primary intracerebral haemorrhage, bleeding disorders such as haemophilia and thrombocytopenic purpura rank fairly high on the list. When one encounters what appears to be a “primary intracerebral
18
haematoma” in a young person, or one who is normotensive, the diagnosis should be reconsidered. One should look in particular for a micro-angioma of which there may be others scattered throughout the brain, thus indicating the primary pathology [22]. As mentioned previously, an intracerebral haemorrhage or haematoma may be due to a ruptured aneurysm which has bled primarily into the brain substance. This occurs particularly with aneurysms in the anterior communicating region but sometimes a sylvian haematoma resulting from a ruptured mid-cerebral aneurysm may simulate an intracerebral haematoma. In patients with a previous history of psychiatric problems and depression in particular, one should obtain an adequate history of drug and food intake. The patient may have been treated with monoamine oxidase inhibitor antidepressants and taken one of the foods known to cause a hypertensive crisis in the presence of monoamine oxidase inhibitors [23]. Traumatic intracerebral haematoma is common but usually there are other features to indicate their traumatic origin. It almost always extends from the surface inwards and at the surface there is usually local contusion. A common site for this to occur is at the sylvian fissure where the brain abuts against the sphenoid ridge and also subfrontally, but deep haematomas may also occur unrelated to surface trauma. One form of traumatic haemorrhage which is not uncommon in our environment, and which should be looked for specifically because it can very easily be missed, is that which results from a small perforating stab wound, either of the orbit or of the cranium. Such a wound inflicted through the scalp area, particularly in an individual with short, tightly curled hair, can easily be overlooked. The risk of misinterpreting this type of wound, has been stressed by Helpern [23] and others. ANOMALIES
AT THE CRANIOCERVICAL
JUNCTION
These are uncommon but they are usually missed unless specifically looked for. Assimilation of the atlas, atlanto-axial instability due to agenesis or pseudo-arthrosis of the odontoid and basilar impression are the common anomalies. These patients may be asymptomatic and perfectly normal until a trivial injury such as a sneeze or a slap on the back causes sudden, severe paralysis of all four limbs and respiration as well as impairment of consciousness. Haemorrhage into this region of the spinal cord or acute compression is usually the cause of death; unless looked for specifically this pathological process may be easily overlooked. REFERENCES
1 Galagher, J.P. & Browder, E.J. (1968) J. Neurosurg. 29: 1 2 Mealy, J. (1960) J. Neurosurgery 17: 27
19 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Jamieson, K.G. & Yelland, J.D.K. (1968) J. Neurosurg. 29: 13-23 Carter, A.E. (1961) J. Neurosurg. 18: 155 Bell, C. (1816) Surgical Observations, Ed. Shirst & Reed. London. Longmans Jamieson, K.G. (1954) Aust. N.Z.J. Surg. 24: 56-62 Logue, V. (1951) Modern Trends in Neurology Ed. A. Ferling London Butterworth Boop, W.C., Chou, S.N., & French, L.A. (1961) J. Neurosurgery 18: 834 Sahs, A.L., Perret, G.E., Locksley, H.B. & Nishioka, H. (1966) Intracranial Aneurysms & Subarachnoid Haemorrhage. Philadelphia. Lippincott Clarke, E., & Walton, J. (1953) Brain 76: 367-404 Talalla, A., & McKissock, W. (1971) Neurology 21: 19-25 Russell, D.S., & Cairns, H. (1934) Brain 57 : 32 Roach, M.R., & Drake, C.G. (1965) New Eng. J. Med. 273: 240-244 Moritz, A.R. & Zancheck, M. (1946) Arch. Path. 42: 459-494 Helpern, W., & Rabson, S.M. (1950) Am. J. Med. SC. 220: 262-271 Dinning, T.A.T., & Faulkner, M.A. (1953) Lancet 11: 799 Housepain, E.M., & Poole, L. (1958) J. Neuropath. Exp. Neurol. 17: 409-423 Bebin., J., and Courier, R.D. (1957) A.M.A. Arch. Int. Med. 99: 771 Tomlinson, B.E. (1959) J. Clin. Path. 12: 391-399 Courville, C.B., & Newbar, F.D. (1958) J. Forensic Sciences 3: 174-200 Simonsen, J. (1963) J. Forensic Sciences 8: 97 Gerlach, J. (1969) Progress in Neurological Surgery Vol. 3 Base1 & New York Karger Helpern, M. (1946) Ann. Int. Med. 24: 666-700 De Villiers, J.C. (1965) Brit. J. Psychiat. 12: 109-118
