- Email: [email protected]
Spontaneous intracranial haemorrhage
NEUROSURGERY
Spontaneous intracranial haemorrhage
may be caused by blockage of the arachnoid villi by blood thus preventing absorption of CSF into the venous sinuses. Subarachnoid haemorrhage occurs when there is bleeding into the subarachnoid space from the medium-sized cerebral vessels which run between the pia mater and arachnoid mater. Intracerebral haemorrhage occurs when there is haemorrhage into the brain substance itself with destruction of brain tissue. We will concentrate on the most commonly encountered forms of spontaneous intracranial haemorrhage in this chapter article, that is, SAH and ICH.
Bassel Zebian Giles Critchley
Abstract Spontaneous intracranial haemorrhage may present as a surgical emergency requiring rapid assessment of the patient and rapid access to diagnostic imaging. The site of the haemorrhage will determine whether there is global cerebral dysfunction or the development of a focal deficit. Hydrocephalus may complicate intracranial haemorrhage. Immediate decisionmaking as to the timing and means of treatment of the patient and the appropriateness of any intervention may be necessary. In this article we discuss the common types of intracranial haemorrhage, the significant features of the assessment and how these are used to guide treatment. Recent multicentre clinical studies have influenced decision-making for this group of patients.
Epidemiology Spontaneous intracranial haemorrhage may be considered as a form of stroke and overall stroke mortality ranks third after cancer and heart disease in the UK. ICH may account for 20 e30% of stroke and SAH about 10%. The most common cause of SAH is an underlying cerebral aneurysm (85%). The global incidence of aneurysmal subarachnoid haemorrhage is 5e10 per 100,000 a year.1 The peak age is between 40 and 60 years with a median age of 59 years old. It is more common in females than males by a ratio of 2:1. SAH is more common in Scandinavian countries and Japan. Smoking is a risk factor. Hypertension may be present in 60% of patients. The second most common cause of spontaneous subarachnoid haemorrhage is an arteriovenous malformation (AVM) (5%). Patients with an AVM are often younger with an average age of around 30 years. Other much rarer causes include vasculitides, tumours, arterial dissection and pituitary apoplexy. In about 15% no cause is found. SAH may be associated with adult polycystic kidney disease, coarctation of the aorta, connective tissue disorders such as EhlerseDanlos syndrome, Marfan’s syndrome, and pseudoxanthoma elasticum. There may be a significant increase in family members where two or more first-degree relatives have proven aneurysmal SAH.
Keywords Cerebellar haematoma; cerebral aneurysm; intracerebral haemorrhage; intraventricular haemorrhage; subarachnoid haemorrhage; subdural haemorrhage
Definition Spontaneous intracranial haemorrhage occurs when there is spontaneous bleeding in the cranial cavity within the skull that is not caused by trauma. Intracranial haemorrhage may occur in any compartment within the cranial cavity and commonly consists of intraventricular haemorrhage (IVH), intracerebral haemorrhage (ICH), subarachnoid haemorrhage (SAH) and subdural haemorrhage (SDH). These forms of haemorrhage may co-exist and one may lead to one or more of the others (Figure 1). Spontaneous subdural haemorrhage may be due to an iatrogenic hypocoagulable state and may be associated with warfarin or aspirin use. Haemorrhage from an underlying vascular abnormality such as an aneurysm (such as a middle cerebral artery aneurysm) may also be a cause. Spontaneous intraventricular haemorrhage can either be due to extension of a subarachnoid or intracerebral haemorrhage into the ventricles or to bleeding from a lesion within the ventricle itself such as a tumour or arteriovenous malformation (AVM). In premature neonates IVH occurs as a consequence of fragile periventricular vessels. Haemorrhage into the ventricular system may cause obstructive hydrocephalus due to prevention of flow of cerebrospinal fluid (CSF) through the third ventricle, aqueduct or fourth ventricle. Communicating hydrocephalus
ICH may account for up to 30% of all strokes. The incidence is over twice that of SAH. Hypertension may be present in 70% of patients. Other risk factors include age (median age 73 years old), gender (more common in men), race (Afro-Caribbean twice as common), previous strokes, coagulopathies, alcohol consumption and use of illicit drugs such as cocaine. About 15% of ICH may be associated with warfarin use.2
Pathology and pathogenesis Subarachnoid haemorrhage: the natural history of SAH and aneurysm development and rupture is not well known. In post mortem studies 0.5e2% of the population may harbour aneurysms yet the incidence of SAH is 10 per 100,000. SAH is associated with cardiovascular risk factors and theories of aneurysm development assume the development of atherosclerotic changes developing due to turbulence in a pre-existing weak area at the bifurcation of a medium-sized cerebral blood vessel (Figure 2). When an SAH first occurs there is a transient rise in intracranial pressure and reduction in cerebral blood flow. Many patients may be unconscious as a result at the time of the bleed, known as the ‘ictus’ and subsequently improve. Following an SAH delayed
Bassel Zebian FRCS is a Consultant Neurosurgeon at Kings College Hospital, London, UK. Conflicts of interest: none declared. Giles Critchley MA MD FRCS(SN) is a Consultant Neurosurgeon at the South East Neurosurgery Centre, Brighton and Sussex University Hospitals NHS Trust, UK. Conflicts of interest: none declared.
SURGERY --:-
1
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
stiffness. The latter two are a direct result of meningeal irritation. A proportion of patients report a preceding headache in the previous few weeks which is usually of a lesser severity and is often referred to as a sentinel headache but still represents an SAH. Neurological examination of the patient must include an assessment of the conscious level using the Glasgow Coma Scale (GCS) and an assessment of any cranial nerve deficit or limb weakness. On fundoscopy subhyaloid, retinal and vitreal haemorrhages are also associated with a spontaneous SAH. Assessment of the patient should also include an assessment of respiratory function which may be compromised by neurogenic pulmonary oedema and blood pressure which may be elevated. Patients may be categorized according to a grading scale such as the World Federation of Neurological Surgeons (WFNS) (Table 1). This scale is based on conscious level and presence or absence of a neurological deficit (excluding cranial nerve abnormalities). A patient in category 1e3 is often referred to as being in a good grade and a poor grade those in category 4e5. Poor-grade patients may require ventilation and treatment of hydrocephalus. The grade of patient may influence the timing and success rates of treatment as it reflects the physiological impact of the haemorrhage on the brain. An unenhanced CT scan of the head is indicated in anyone presenting with a classical sudden onset headache irrespective of other symptoms and signs. This should be carried out as soon as possible as the sensitivity for SAH decreases with time (over 95% in the first 2 days). The CT scan (Figure 3) typically may show blood around the circle of Willis, in the basal cisterns, the interhemispheric fissure and the Sylvian fissure. There may be a coexisting ICH or SDH. Obstructive or communicating hydrocephalus may be present. If the CT scan is negative patients should undergo a lumbar puncture (LP) to determine whether there are breakdown products of blood. The gold standard is CSF spectrophotometry which would show a bilirubin peak if the patient had a SAH. The LP must ideally be carried out 12 hours after the onset of the headache to allow blood time to break down. It should be performed no later than 2 weeks as the sensitivity decreases significantly after that. Once the diagnosis of SAH has been confirmed then imaging of the cerebral vasculature is required to determine the cause of the haemorrhage and any underlying vascular abnormality. CT angiography has replaced formal catheter angiography as the first vascular investigation in many centres (Figure 4a). The angiogram is examined for the presence of cerebral aneurysms, presence of vasospasm and variations in the anatomy of the Circle of Willis (which may be relevant when planning treatment).
Figure 1
cerebral ischaemia ‘cerebral vasospasm’ may occur. This happens typically between 4 days and 10 days following SAH, but may last up to 3 weeks. There is vasoconstriction of mediumsized blood vessels which causes reduction of blood flow to the brain. This may cause neurological deficits which are potentially reversible. Cerebral vasospasm is caused by an imbalance between endothelin (a potent vasoconstrictor) and nitric oxide (NO) (a potent vasodilator). The chance of vasopasm resulting in an established stroke is reduced by administering 60 mg of nimodipine 4-hourly. Intracerebral haemorrhage: 70% of ICH is associated with hypertension. These haemorrhages commonly occur in the basal ganglia and thalamus. Arterial rupture occurs in the lenticulostriate branches of the middle cerebral artery which have been weakened by fat and fibrin deposition in the media of the arterial wall. This may cause micro-pseudoaneurysms. ICH in normotensive patients may occur in the subcortical white matter. These haemorrhages may be due to a tumour, AVM, coagulopathy, cerebral amyloid or recreational sympathomimetic drug use (cocaine, amphetamine). ICH may also occur due to reperfusion of an ischaemic stroke causing a haemorrhagic infarct. The ICH itself causes destruction of cerebral tissue. There may then be subsequent haematoma growth with further neurological deterioration (a feature in approximately 30%). Surrounding the haematoma may be a penumbra of tissue that is compressed and ischaemic and as such is at risk of further cell death. Treatment strategies for ICH aim to prevent deterioration of this penumbra as well as preventing the consequences of raised intracranial pressure such as uncal herniation and brainstem compression (‘coning’).
Intracerebral haemorrhage: a spontaneous ICH presents most commonly with a sudden onset headache and is associated with a focal deficit that depends on the location of the haematoma. It is important in the history to establish the presence of risk factors such as hypertension and anticoagulant use. In the examination it is important to determine the degree of disability and neurological deficit as well as the blood pressure. There is usually no immediate improvement in the patient’s condition following an ICH compared to an SAH.
Diagnosis (history, examination and investigations) Subarachnoid haemorrhage is characterized by a sudden onset headache. This is typically described as the worst headache ever and builds in intensity over seconds. Patients often liken the headache to suddenly being hit on the back of the head. The headache may be associated with a transient loss of consciousness and with nausea and vomiting, photophobia and neck
SURGERY --:-
2
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
Figure 2 Circle of Willis showing common sites of cerebral aneurysms.
intubation and ventilation may be necessary. Seizures should be controlled with an anticonvulsant where possible (intravenous phenytoin usually gives rapid control) or alternatively with intubation and ventilation.
A CT scan is the first investigation (Figure 5). The size of the haematoma and its location are assessed. It is important to determine if the haematoma is deep within the brain or close to the surface as this determines appropriateness for surgical treatment. Lobar haematomas may be amenable to surgical evacuation. If the haematoma is in the cerebellum then the presence of brainstem haematoma (associated with a negative prognosis) as well as hydrocephalus (usually treatable) needs to be considered also. If an underlying lesion vascular lesion is suspected then angiography should be performed to detect a cerebral aneurysm or AVM. A delayed MRI scan may be useful if there is a question of an underlying tumour.
Subarachnoid haemorrhage: the initial management of patients with SAH consists of bed rest, regular neurological observation, careful monitoring of blood pressure and electrolytes. Hyponatraemia is common and is usually due to cerebral salt wasting and as such should be treated with replacement rather than fluid restriction. Intravenous fluids, usually normal saline, should be given to ensure euvolaemia. Nimodipine, a lipophilic calcium channel blocker is given. To date this is the only drug proven to be neuroprotective that is in regular clinical use.3 Patients with hydrocephalus should be managed with CSF diversion via an external ventricular drain. A lumbar puncture or a lumbar drain may be used if the hydrocephalus is communicating (but preferably not if the patient is unconscious). Patients with SAH and an intracerebral haematoma should have this evacuated if there is significant mass effect causing a deterioration in neurological state. This is usually accompanied by surgically securing the aneurysm if appropriate. Once an aneurysm has been identified the next step in management is to secure it as the risk of rebleeding from an unprotected ruptured aneurysm is high (4% in the first day, then 1.5% every day for the first 2 weeks; 50% will rebleed within 6 months but after this period the risk decreases to around 3% a year) and the risk of death from a rebleed is extremely high (80% in some series).4,5 Aneurysms can be secured radiologically (endovascularly) or surgically (through a craniotomy and clipping) (Figure 4b). The International Cooperative Study on the Timing of Aneurysm Surgery4,5 demonstrated that in patients who were alert preoperatively outcome was best if intervention was within the
Management Initial management in patients with all forms of intracranial haemorrhage is to ensure an adequate airway. A decreased conscious level and focal deficits may lead to loss of protective reflexes which may lead to an obstructed airway with aspiration, hypoxia and hypercapnia. This may cause a secondary deterioration with raised intracranial pressure. Airway protection with
World Federation of Neurological Surgeons (WFNS) grading scale for subarachnoid haemorrhage WFNS grade
Glasgow Coma Scale score
Focal deficit
0 1 2 3 4 5
Intact aneurysm 15 13e14 13e14 7e12 3e6
No No Yes Yes or No Yes or No
Table 1
SURGERY --:-
3
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
SAH in the basal cisterns
Figure 3 A CT scan showing the typical distribution of blood following an aneurysmal subarachnoid haemorrhage (SAH). Dilatation of the temporal horns of the lateral ventricles indicates early hydrocephalus.
first 3 days from the SAH or after day 10. For those who were drowsy the outcome was best if the intervention was after day 10. This has led to the current practice of offering early intervention (within 3 days) to those who suffer a SAH with a WFNS
Figure 5 A CT scan showing an intracerebral haemorrhage involving the left basal ganglia with surrounding low-density ischaemia. There is marked midline shift.
grade 1e3 and delayed intervention (after day 10) to those with a poor grade (4 or 5). Since the early 1990s two main techniques have been utilized to secure aneurysms: surgical clipping through a craniotomy and coiling through an endovascular approach. The International Subarachnoid Aneurysm Trial (ISAT)6,7 compared endovascular coiling and surgical clipping and concluded that, at 1 year follow-up, patients treated endovascularly had a better outcome. Analysis of longer term follow-up showed that at 7 years the advantage of coiling over clipping was maintained, but that the risk of recurrence and rebleeding was higher in coiling group. Many centres will now offer endovascular coiling as first-line treatment reserving craniotomy and clipping for patients whose aneurysm cannot be coiled or who are less than 40 years old where the risk of recurrence is higher. Cerebral vasospasm occurs due to reduction in calibre of the medium-sized blood vessels. This is the leading cause of mortality and morbidity in patients surviving the initial bleed. Although up to 70% of patients show radiological evidence of vasospasm only 30% will experience clinical vasospasm (also known as delayed ischaemic neurological deficit). Clinical vasospasm may be defined as a drop in GCS of two points or more or the development of a new neurological deficit not attributable to other causes (potential other causes include new hydrocephalus, seizures, infection, hyponatraemia). The mainstay of treatment is to increase cerebral blood flow (CBF). This is achieved by increasing cerebral perfusion pressure (CPP) and reducing intracranial pressure (ICP). CPP is increased by driving mean arterial pressure (MAP) up through the use of fluids in the first instance and the addition of inotropes such as norepinephrine when this fails. This is termed ‘triple H therapy’
a
Anterior communicating artery aneurysm
b
Aneurysm clip
Figure 4 (a) CT angiogram showing anterior communicating artery aneurysm (Figure 2) and (b) postoperative CT angiogram showing titanium clip on aneurysm preserving vessels.
SURGERY --:-
4
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
hypertension, hypervolaemia and haemodilution.8 In refractory clinical vasospasm there is some evidence to suggest that endovascular salvage methods (such as intra-arterial nimodipine or nicardipine or balloon angioplasty in proximal short segment spasm) can be used effectively.
the haemorrhage as well as its size and location. In the STICH study overall mortality was 36% with 25% of patients in both early surgery and best medical management group having a favourable outcome. Prevention of intracranial haemorrhage Treatment of cardiovascular risk factors and cessation of smoking should have a preventative effect on intracranial haemorrhage. As yet there is no routine screening programme for the presence of cerebral aneurysms but screening with CT angiography, MR angiography or catheter angiography may be used for patients where there is a strong family history after adequate counselling and consent. Our knowledge of the natural history of unruptured aneurysms has been enhanced by the International Study of Unruptured Intracranial aneurysm (ISUIA)12 which reported in 2003 and included 4060 patients. This was a retrospective and prospective study to determine the natural history of unruptured aneurysms and the complication rate of treating them. Unoperated patients were divided into two groups: those where the unruptured aneurysm was not associated with previous SAH (group 1) and those where there was previous SAH from a separate aneurysm (group 2). In the operated group mortality and morbidity was recorded. The study concluded that aneurysm rupture rate was related to the size and location of the aneurysm. For aneurysms less than 7 mm the risk was increased if there had been a previous SAH from a separate aneurysm (group 2). The cumulative 5-year risk from rupture of a 7e12 mm aneurysm was 2.6% for anterior communicating artery aneurysm compared to 14.5% for posterior circulation or posterior communicating artery aneurysms. Risk of surgery was a 12% morbidity and up to 3.8% mortality. This was influenced by the patient’s age (over 50 years is worse), size of aneurysm, symptoms other than rupture, previous ischaemic stroke or posterior circulation location. This study has helped to inform decision-making as to treatment of aneurysm to prevent SAH as the benefits of treatment need to be balanced against the risks for each individual patient. A
Intracerebral haemorrhage Aggressive reduction of an elevated blood pressure may prevent haematoma growth and improve outcome.9 If the patient is anticoagulated with warfarin then rapid reversal with fresh frozen plasma, vitamin K and prothrombin complex concentrates may be used. Surgical treatment should be considered for young patients with large lobar haemorrhages who are deteriorating because of mass effect. Patients with a cerebellar haematoma and obstructive hydrocephalus may benefit from treatment of hydrocephalus and evacuation of the haematoma even if they are in coma. A patient with a large cerebellar haematoma greater than 3 cm in diameter is at risk of sudden deterioration and early surgery should be considered. Elderly patients in a poor neurological condition (GCS <8) with deep basal ganglia haematomas with intraventricular extension do not usually benefit from surgery. The role of surgery in the treatment of spontaneous ICH remains controversial and is still under investigation. The International Surgical Trial in Intracerebral Haemorrhage (STICH)10 concluded that there is no overall benefit from early surgery versus best medical management within 72 hours of onset. The study recruited patients in whom a state of equipoise prevailed with regard to best management (that is where the surgeon was uncertain as to the benefit of initial surgery versus medical management). Subgroup analysis demonstrated that when the haematoma was superficial (within 1 cm from cortex) there was a relative benefit, albeit not statistically significant, from early surgery. This resulted in the STICH II Trial11 of early surgery versus initial conservative treatment in patients with spontaneous supratentorial lobar intracerebral haematomas which again did not demonstrate a significant difference in outcome between the two groups.
REFERENCES 1 van Gijn J, Kerr RS, Rinkel GJE. Subarachnoid haemorrhage. Lancet 2007; 369: 306e18. 2 Mayer SA, Rincon F. Treatment of intracerebral haemorrhage. Lancet Neurol 2005; 4: 662e72. 3 Pickard JD, Murray GD, Illingworth R, et al. Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. Br Med J 1989; 298: 636e42. 4 Kassell NF, Torner JC, Haley Jr EC, et al. The International Cooperative study on the timing of aneurysm surgery. Part 1: overall management results. J Neurosurg 1990; 73: 18e36. 5 Kassell NF, Torner JC, Jane JA, et al. The International Cooperative study on the timing of aneurysm surgery. Part 2: surgical results. J Neurosurg 1990; 73: 37e47. 6 Molyneux A, Kerr R, Stratton I, et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 2002; 360: 1267e74. 7 Molyneux A, Kerr RS, Yu LM, et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular
Prognosis Subarachnoid haemorrhage: aneurysmal SAH is a devastating condition with an overall mortality approaching 50%. Almost one-third of those who survive are left with a moderate to severe disability and about two-thirds will never return to their premorbid quality of life. Prognostic factors include initial clinical condition (WFNS grade), age, admission blood pressure, preexisting medical conditions and the amount of blood on the CT scan (Fisher grade). Of those who survive the initial bleed the leading cause of morbidity and mortality is vasospasm (14%) followed by rebleeding (7%). Those patients in the good outcome groups may still have neuropsychological sequelae. Intracerebral haemorrhage: various mortality figures have been quoted in the literature with a generally accepted 1-year mortality of 40e50%. This depends on factors such as neurological state on admission, age, medical co-morbidities, underlying cause of
SURGERY --:-
5
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
10 Mendelow AD, Gregson BA, Fernandes HN, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised trial. Lancet 2005; 365: 387e97. 11 Mendelow AD, Gregson BA, Rowan EN, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial lobar intracerebral haematomas (STICH II): a randomised trial. Lancet 2013; 382: 397e408. 12 International Study of Unruptured Intracranial Aneurysms Investigators. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003; 362: 103e10.
coiling in 2143 patients with ruptured intracranial aneurysms: a randomised comparison of effects on survival, dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet 2005; 366: 809e17. 8 Kassell NF, Peerless SJ, Durwood QJ, Beck DW, Drake CG, Adams H. Treatment of ischaemic deficits from vasospasm with intravascular volume expansion and induced arterial hypertension. Neurosurgery 1982; 11: 337e43. 9 Anderson CS, Huang Y, Arima H, et al. Effects of early blood pressure lowering treatment on the growth of hematoma and perihematomal edema in Acute intracerebral hemorrhage. The Intensive Blood pressure reduction in Acute Cerebral Haemorrhage Trial (INTERACT). Stroke 2010; 41: 307e12.
SURGERY --:-
6
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
Spontaneous intracranial haemorrhage
may be caused by blockage of the arachnoid villi by blood thus preventing absorption of CSF into the venous sinuses. Subarachnoid haemorrhage occurs when there is bleeding into the subarachnoid space from the medium-sized cerebral vessels which run between the pia mater and arachnoid mater. Intracerebral haemorrhage occurs when there is haemorrhage into the brain substance itself with destruction of brain tissue. We will concentrate on the most commonly encountered forms of spontaneous intracranial haemorrhage in this chapter article, that is, SAH and ICH.
Bassel Zebian Giles Critchley
Abstract Spontaneous intracranial haemorrhage may present as a surgical emergency requiring rapid assessment of the patient and rapid access to diagnostic imaging. The site of the haemorrhage will determine whether there is global cerebral dysfunction or the development of a focal deficit. Hydrocephalus may complicate intracranial haemorrhage. Immediate decisionmaking as to the timing and means of treatment of the patient and the appropriateness of any intervention may be necessary. In this article we discuss the common types of intracranial haemorrhage, the significant features of the assessment and how these are used to guide treatment. Recent multicentre clinical studies have influenced decision-making for this group of patients.
Epidemiology Spontaneous intracranial haemorrhage may be considered as a form of stroke and overall stroke mortality ranks third after cancer and heart disease in the UK. ICH may account for 20 e30% of stroke and SAH about 10%. The most common cause of SAH is an underlying cerebral aneurysm (85%). The global incidence of aneurysmal subarachnoid haemorrhage is 5e10 per 100,000 a year.1 The peak age is between 40 and 60 years with a median age of 59 years old. It is more common in females than males by a ratio of 2:1. SAH is more common in Scandinavian countries and Japan. Smoking is a risk factor. Hypertension may be present in 60% of patients. The second most common cause of spontaneous subarachnoid haemorrhage is an arteriovenous malformation (AVM) (5%). Patients with an AVM are often younger with an average age of around 30 years. Other much rarer causes include vasculitides, tumours, arterial dissection and pituitary apoplexy. In about 15% no cause is found. SAH may be associated with adult polycystic kidney disease, coarctation of the aorta, connective tissue disorders such as EhlerseDanlos syndrome, Marfan’s syndrome, and pseudoxanthoma elasticum. There may be a significant increase in family members where two or more first-degree relatives have proven aneurysmal SAH.
Keywords Cerebellar haematoma; cerebral aneurysm; intracerebral haemorrhage; intraventricular haemorrhage; subarachnoid haemorrhage; subdural haemorrhage
Definition Spontaneous intracranial haemorrhage occurs when there is spontaneous bleeding in the cranial cavity within the skull that is not caused by trauma. Intracranial haemorrhage may occur in any compartment within the cranial cavity and commonly consists of intraventricular haemorrhage (IVH), intracerebral haemorrhage (ICH), subarachnoid haemorrhage (SAH) and subdural haemorrhage (SDH). These forms of haemorrhage may co-exist and one may lead to one or more of the others (Figure 1). Spontaneous subdural haemorrhage may be due to an iatrogenic hypocoagulable state and may be associated with warfarin or aspirin use. Haemorrhage from an underlying vascular abnormality such as an aneurysm (such as a middle cerebral artery aneurysm) may also be a cause. Spontaneous intraventricular haemorrhage can either be due to extension of a subarachnoid or intracerebral haemorrhage into the ventricles or to bleeding from a lesion within the ventricle itself such as a tumour or arteriovenous malformation (AVM). In premature neonates IVH occurs as a consequence of fragile periventricular vessels. Haemorrhage into the ventricular system may cause obstructive hydrocephalus due to prevention of flow of cerebrospinal fluid (CSF) through the third ventricle, aqueduct or fourth ventricle. Communicating hydrocephalus
ICH may account for up to 30% of all strokes. The incidence is over twice that of SAH. Hypertension may be present in 70% of patients. Other risk factors include age (median age 73 years old), gender (more common in men), race (Afro-Caribbean twice as common), previous strokes, coagulopathies, alcohol consumption and use of illicit drugs such as cocaine. About 15% of ICH may be associated with warfarin use.2
Pathology and pathogenesis Subarachnoid haemorrhage: the natural history of SAH and aneurysm development and rupture is not well known. In post mortem studies 0.5e2% of the population may harbour aneurysms yet the incidence of SAH is 10 per 100,000. SAH is associated with cardiovascular risk factors and theories of aneurysm development assume the development of atherosclerotic changes developing due to turbulence in a pre-existing weak area at the bifurcation of a medium-sized cerebral blood vessel (Figure 2). When an SAH first occurs there is a transient rise in intracranial pressure and reduction in cerebral blood flow. Many patients may be unconscious as a result at the time of the bleed, known as the ‘ictus’ and subsequently improve. Following an SAH delayed
Bassel Zebian FRCS is a Consultant Neurosurgeon at Kings College Hospital, London, UK. Conflicts of interest: none declared. Giles Critchley MA MD FRCS(SN) is a Consultant Neurosurgeon at the South East Neurosurgery Centre, Brighton and Sussex University Hospitals NHS Trust, UK. Conflicts of interest: none declared.
SURGERY --:-
1
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
stiffness. The latter two are a direct result of meningeal irritation. A proportion of patients report a preceding headache in the previous few weeks which is usually of a lesser severity and is often referred to as a sentinel headache but still represents an SAH. Neurological examination of the patient must include an assessment of the conscious level using the Glasgow Coma Scale (GCS) and an assessment of any cranial nerve deficit or limb weakness. On fundoscopy subhyaloid, retinal and vitreal haemorrhages are also associated with a spontaneous SAH. Assessment of the patient should also include an assessment of respiratory function which may be compromised by neurogenic pulmonary oedema and blood pressure which may be elevated. Patients may be categorized according to a grading scale such as the World Federation of Neurological Surgeons (WFNS) (Table 1). This scale is based on conscious level and presence or absence of a neurological deficit (excluding cranial nerve abnormalities). A patient in category 1e3 is often referred to as being in a good grade and a poor grade those in category 4e5. Poor-grade patients may require ventilation and treatment of hydrocephalus. The grade of patient may influence the timing and success rates of treatment as it reflects the physiological impact of the haemorrhage on the brain. An unenhanced CT scan of the head is indicated in anyone presenting with a classical sudden onset headache irrespective of other symptoms and signs. This should be carried out as soon as possible as the sensitivity for SAH decreases with time (over 95% in the first 2 days). The CT scan (Figure 3) typically may show blood around the circle of Willis, in the basal cisterns, the interhemispheric fissure and the Sylvian fissure. There may be a coexisting ICH or SDH. Obstructive or communicating hydrocephalus may be present. If the CT scan is negative patients should undergo a lumbar puncture (LP) to determine whether there are breakdown products of blood. The gold standard is CSF spectrophotometry which would show a bilirubin peak if the patient had a SAH. The LP must ideally be carried out 12 hours after the onset of the headache to allow blood time to break down. It should be performed no later than 2 weeks as the sensitivity decreases significantly after that. Once the diagnosis of SAH has been confirmed then imaging of the cerebral vasculature is required to determine the cause of the haemorrhage and any underlying vascular abnormality. CT angiography has replaced formal catheter angiography as the first vascular investigation in many centres (Figure 4a). The angiogram is examined for the presence of cerebral aneurysms, presence of vasospasm and variations in the anatomy of the Circle of Willis (which may be relevant when planning treatment).
Figure 1
cerebral ischaemia ‘cerebral vasospasm’ may occur. This happens typically between 4 days and 10 days following SAH, but may last up to 3 weeks. There is vasoconstriction of mediumsized blood vessels which causes reduction of blood flow to the brain. This may cause neurological deficits which are potentially reversible. Cerebral vasospasm is caused by an imbalance between endothelin (a potent vasoconstrictor) and nitric oxide (NO) (a potent vasodilator). The chance of vasopasm resulting in an established stroke is reduced by administering 60 mg of nimodipine 4-hourly. Intracerebral haemorrhage: 70% of ICH is associated with hypertension. These haemorrhages commonly occur in the basal ganglia and thalamus. Arterial rupture occurs in the lenticulostriate branches of the middle cerebral artery which have been weakened by fat and fibrin deposition in the media of the arterial wall. This may cause micro-pseudoaneurysms. ICH in normotensive patients may occur in the subcortical white matter. These haemorrhages may be due to a tumour, AVM, coagulopathy, cerebral amyloid or recreational sympathomimetic drug use (cocaine, amphetamine). ICH may also occur due to reperfusion of an ischaemic stroke causing a haemorrhagic infarct. The ICH itself causes destruction of cerebral tissue. There may then be subsequent haematoma growth with further neurological deterioration (a feature in approximately 30%). Surrounding the haematoma may be a penumbra of tissue that is compressed and ischaemic and as such is at risk of further cell death. Treatment strategies for ICH aim to prevent deterioration of this penumbra as well as preventing the consequences of raised intracranial pressure such as uncal herniation and brainstem compression (‘coning’).
Intracerebral haemorrhage: a spontaneous ICH presents most commonly with a sudden onset headache and is associated with a focal deficit that depends on the location of the haematoma. It is important in the history to establish the presence of risk factors such as hypertension and anticoagulant use. In the examination it is important to determine the degree of disability and neurological deficit as well as the blood pressure. There is usually no immediate improvement in the patient’s condition following an ICH compared to an SAH.
Diagnosis (history, examination and investigations) Subarachnoid haemorrhage is characterized by a sudden onset headache. This is typically described as the worst headache ever and builds in intensity over seconds. Patients often liken the headache to suddenly being hit on the back of the head. The headache may be associated with a transient loss of consciousness and with nausea and vomiting, photophobia and neck
SURGERY --:-
2
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
Figure 2 Circle of Willis showing common sites of cerebral aneurysms.
intubation and ventilation may be necessary. Seizures should be controlled with an anticonvulsant where possible (intravenous phenytoin usually gives rapid control) or alternatively with intubation and ventilation.
A CT scan is the first investigation (Figure 5). The size of the haematoma and its location are assessed. It is important to determine if the haematoma is deep within the brain or close to the surface as this determines appropriateness for surgical treatment. Lobar haematomas may be amenable to surgical evacuation. If the haematoma is in the cerebellum then the presence of brainstem haematoma (associated with a negative prognosis) as well as hydrocephalus (usually treatable) needs to be considered also. If an underlying lesion vascular lesion is suspected then angiography should be performed to detect a cerebral aneurysm or AVM. A delayed MRI scan may be useful if there is a question of an underlying tumour.
Subarachnoid haemorrhage: the initial management of patients with SAH consists of bed rest, regular neurological observation, careful monitoring of blood pressure and electrolytes. Hyponatraemia is common and is usually due to cerebral salt wasting and as such should be treated with replacement rather than fluid restriction. Intravenous fluids, usually normal saline, should be given to ensure euvolaemia. Nimodipine, a lipophilic calcium channel blocker is given. To date this is the only drug proven to be neuroprotective that is in regular clinical use.3 Patients with hydrocephalus should be managed with CSF diversion via an external ventricular drain. A lumbar puncture or a lumbar drain may be used if the hydrocephalus is communicating (but preferably not if the patient is unconscious). Patients with SAH and an intracerebral haematoma should have this evacuated if there is significant mass effect causing a deterioration in neurological state. This is usually accompanied by surgically securing the aneurysm if appropriate. Once an aneurysm has been identified the next step in management is to secure it as the risk of rebleeding from an unprotected ruptured aneurysm is high (4% in the first day, then 1.5% every day for the first 2 weeks; 50% will rebleed within 6 months but after this period the risk decreases to around 3% a year) and the risk of death from a rebleed is extremely high (80% in some series).4,5 Aneurysms can be secured radiologically (endovascularly) or surgically (through a craniotomy and clipping) (Figure 4b). The International Cooperative Study on the Timing of Aneurysm Surgery4,5 demonstrated that in patients who were alert preoperatively outcome was best if intervention was within the
Management Initial management in patients with all forms of intracranial haemorrhage is to ensure an adequate airway. A decreased conscious level and focal deficits may lead to loss of protective reflexes which may lead to an obstructed airway with aspiration, hypoxia and hypercapnia. This may cause a secondary deterioration with raised intracranial pressure. Airway protection with
World Federation of Neurological Surgeons (WFNS) grading scale for subarachnoid haemorrhage WFNS grade
Glasgow Coma Scale score
Focal deficit
0 1 2 3 4 5
Intact aneurysm 15 13e14 13e14 7e12 3e6
No No Yes Yes or No Yes or No
Table 1
SURGERY --:-
3
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
SAH in the basal cisterns
Figure 3 A CT scan showing the typical distribution of blood following an aneurysmal subarachnoid haemorrhage (SAH). Dilatation of the temporal horns of the lateral ventricles indicates early hydrocephalus.
first 3 days from the SAH or after day 10. For those who were drowsy the outcome was best if the intervention was after day 10. This has led to the current practice of offering early intervention (within 3 days) to those who suffer a SAH with a WFNS
Figure 5 A CT scan showing an intracerebral haemorrhage involving the left basal ganglia with surrounding low-density ischaemia. There is marked midline shift.
grade 1e3 and delayed intervention (after day 10) to those with a poor grade (4 or 5). Since the early 1990s two main techniques have been utilized to secure aneurysms: surgical clipping through a craniotomy and coiling through an endovascular approach. The International Subarachnoid Aneurysm Trial (ISAT)6,7 compared endovascular coiling and surgical clipping and concluded that, at 1 year follow-up, patients treated endovascularly had a better outcome. Analysis of longer term follow-up showed that at 7 years the advantage of coiling over clipping was maintained, but that the risk of recurrence and rebleeding was higher in coiling group. Many centres will now offer endovascular coiling as first-line treatment reserving craniotomy and clipping for patients whose aneurysm cannot be coiled or who are less than 40 years old where the risk of recurrence is higher. Cerebral vasospasm occurs due to reduction in calibre of the medium-sized blood vessels. This is the leading cause of mortality and morbidity in patients surviving the initial bleed. Although up to 70% of patients show radiological evidence of vasospasm only 30% will experience clinical vasospasm (also known as delayed ischaemic neurological deficit). Clinical vasospasm may be defined as a drop in GCS of two points or more or the development of a new neurological deficit not attributable to other causes (potential other causes include new hydrocephalus, seizures, infection, hyponatraemia). The mainstay of treatment is to increase cerebral blood flow (CBF). This is achieved by increasing cerebral perfusion pressure (CPP) and reducing intracranial pressure (ICP). CPP is increased by driving mean arterial pressure (MAP) up through the use of fluids in the first instance and the addition of inotropes such as norepinephrine when this fails. This is termed ‘triple H therapy’
a
Anterior communicating artery aneurysm
b
Aneurysm clip
Figure 4 (a) CT angiogram showing anterior communicating artery aneurysm (Figure 2) and (b) postoperative CT angiogram showing titanium clip on aneurysm preserving vessels.
SURGERY --:-
4
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
hypertension, hypervolaemia and haemodilution.8 In refractory clinical vasospasm there is some evidence to suggest that endovascular salvage methods (such as intra-arterial nimodipine or nicardipine or balloon angioplasty in proximal short segment spasm) can be used effectively.
the haemorrhage as well as its size and location. In the STICH study overall mortality was 36% with 25% of patients in both early surgery and best medical management group having a favourable outcome. Prevention of intracranial haemorrhage Treatment of cardiovascular risk factors and cessation of smoking should have a preventative effect on intracranial haemorrhage. As yet there is no routine screening programme for the presence of cerebral aneurysms but screening with CT angiography, MR angiography or catheter angiography may be used for patients where there is a strong family history after adequate counselling and consent. Our knowledge of the natural history of unruptured aneurysms has been enhanced by the International Study of Unruptured Intracranial aneurysm (ISUIA)12 which reported in 2003 and included 4060 patients. This was a retrospective and prospective study to determine the natural history of unruptured aneurysms and the complication rate of treating them. Unoperated patients were divided into two groups: those where the unruptured aneurysm was not associated with previous SAH (group 1) and those where there was previous SAH from a separate aneurysm (group 2). In the operated group mortality and morbidity was recorded. The study concluded that aneurysm rupture rate was related to the size and location of the aneurysm. For aneurysms less than 7 mm the risk was increased if there had been a previous SAH from a separate aneurysm (group 2). The cumulative 5-year risk from rupture of a 7e12 mm aneurysm was 2.6% for anterior communicating artery aneurysm compared to 14.5% for posterior circulation or posterior communicating artery aneurysms. Risk of surgery was a 12% morbidity and up to 3.8% mortality. This was influenced by the patient’s age (over 50 years is worse), size of aneurysm, symptoms other than rupture, previous ischaemic stroke or posterior circulation location. This study has helped to inform decision-making as to treatment of aneurysm to prevent SAH as the benefits of treatment need to be balanced against the risks for each individual patient. A
Intracerebral haemorrhage Aggressive reduction of an elevated blood pressure may prevent haematoma growth and improve outcome.9 If the patient is anticoagulated with warfarin then rapid reversal with fresh frozen plasma, vitamin K and prothrombin complex concentrates may be used. Surgical treatment should be considered for young patients with large lobar haemorrhages who are deteriorating because of mass effect. Patients with a cerebellar haematoma and obstructive hydrocephalus may benefit from treatment of hydrocephalus and evacuation of the haematoma even if they are in coma. A patient with a large cerebellar haematoma greater than 3 cm in diameter is at risk of sudden deterioration and early surgery should be considered. Elderly patients in a poor neurological condition (GCS <8) with deep basal ganglia haematomas with intraventricular extension do not usually benefit from surgery. The role of surgery in the treatment of spontaneous ICH remains controversial and is still under investigation. The International Surgical Trial in Intracerebral Haemorrhage (STICH)10 concluded that there is no overall benefit from early surgery versus best medical management within 72 hours of onset. The study recruited patients in whom a state of equipoise prevailed with regard to best management (that is where the surgeon was uncertain as to the benefit of initial surgery versus medical management). Subgroup analysis demonstrated that when the haematoma was superficial (within 1 cm from cortex) there was a relative benefit, albeit not statistically significant, from early surgery. This resulted in the STICH II Trial11 of early surgery versus initial conservative treatment in patients with spontaneous supratentorial lobar intracerebral haematomas which again did not demonstrate a significant difference in outcome between the two groups.
REFERENCES 1 van Gijn J, Kerr RS, Rinkel GJE. Subarachnoid haemorrhage. Lancet 2007; 369: 306e18. 2 Mayer SA, Rincon F. Treatment of intracerebral haemorrhage. Lancet Neurol 2005; 4: 662e72. 3 Pickard JD, Murray GD, Illingworth R, et al. Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. Br Med J 1989; 298: 636e42. 4 Kassell NF, Torner JC, Haley Jr EC, et al. The International Cooperative study on the timing of aneurysm surgery. Part 1: overall management results. J Neurosurg 1990; 73: 18e36. 5 Kassell NF, Torner JC, Jane JA, et al. The International Cooperative study on the timing of aneurysm surgery. Part 2: surgical results. J Neurosurg 1990; 73: 37e47. 6 Molyneux A, Kerr R, Stratton I, et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 2002; 360: 1267e74. 7 Molyneux A, Kerr RS, Yu LM, et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular
Prognosis Subarachnoid haemorrhage: aneurysmal SAH is a devastating condition with an overall mortality approaching 50%. Almost one-third of those who survive are left with a moderate to severe disability and about two-thirds will never return to their premorbid quality of life. Prognostic factors include initial clinical condition (WFNS grade), age, admission blood pressure, preexisting medical conditions and the amount of blood on the CT scan (Fisher grade). Of those who survive the initial bleed the leading cause of morbidity and mortality is vasospasm (14%) followed by rebleeding (7%). Those patients in the good outcome groups may still have neuropsychological sequelae. Intracerebral haemorrhage: various mortality figures have been quoted in the literature with a generally accepted 1-year mortality of 40e50%. This depends on factors such as neurological state on admission, age, medical co-morbidities, underlying cause of
SURGERY --:-
5
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005
NEUROSURGERY
10 Mendelow AD, Gregson BA, Fernandes HN, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised trial. Lancet 2005; 365: 387e97. 11 Mendelow AD, Gregson BA, Rowan EN, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial lobar intracerebral haematomas (STICH II): a randomised trial. Lancet 2013; 382: 397e408. 12 International Study of Unruptured Intracranial Aneurysms Investigators. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003; 362: 103e10.
coiling in 2143 patients with ruptured intracranial aneurysms: a randomised comparison of effects on survival, dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet 2005; 366: 809e17. 8 Kassell NF, Peerless SJ, Durwood QJ, Beck DW, Drake CG, Adams H. Treatment of ischaemic deficits from vasospasm with intravascular volume expansion and induced arterial hypertension. Neurosurgery 1982; 11: 337e43. 9 Anderson CS, Huang Y, Arima H, et al. Effects of early blood pressure lowering treatment on the growth of hematoma and perihematomal edema in Acute intracerebral hemorrhage. The Intensive Blood pressure reduction in Acute Cerebral Haemorrhage Trial (INTERACT). Stroke 2010; 41: 307e12.
SURGERY --:-
6
Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Zebian B, Critchley G, Spontaneous intracranial haemorrhage, Surgery (2015), http://dx.doi.org/10.1016/ j.mpsur.2015.05.005