Neurosurg Focus 15 (4):Article 1, 2003, Click here to return to Table of Contents

Spontaneous intracerebral hemorrhage: a review

MATTHEW E. FEWEL, M.D., B. GREGORY THOMPSON, JR., M.D., AND JULIAN T. HOFF, M.D.
Department of Neurosurgery, University of Michigan Health System, Ann Arbor, Michigan

Spontaneous intracerebral hemorrhage (SICH) is a blood clot that arises in the brain parenchyma in the absence of
trauma or surgery. This entity accounts for 10 to 15% of all strokes and is associated with a higher mortality rate than
either ischemic stroke or subarachnoid hemorrhage. Common causes include hypertension, amyloid angiopathy, coag-
ulopathy, vascular anomalies, tumors, and various drugs. Hypertension, however, remains the single greatest modifi-
able risk factor for SICH. Computerized tomography scanning is the initial diagnostic modality of choice in SICH, and
angiography should be considered in all cases except those involving older patients with preexisting hypertension in
thalamic, putaminal, or cerebellar hemorrhage. Medical management includes venous thrombosis prophylaxis, gastric
cytoprotection, and aggressive rehabilitation. Anticonvulsant agents should be prescribed in supratentorial SICH,
whereas the management of hypertension is controversial.
To date, nine prospective randomized controlled studies have been conducted to compare surgical and medical man-
agement of SICH. Although definitive evidence favoring surgical intervention is lacking, there is good theoretical
rationale for early surgical intervention. Surgery should be considered in patients with moderate to large lobar or basal
ganglia hemorrhages and those suffering progressive neurological deterioration. Elderly patients in whom the Glasgow
Coma Scale score is less than 5, those with brainstem hemorrhages, and those with small hemorrhages do not typical-
ly benefit from surgery. Patients with cerebellar hemorrhages larger than 3 cm, those with brainstem compression and
hydrocephalus, or those exhibiting neurological deterioration should undergo surgical evacuation of the clot. It is
hoped that the forthcoming results of the International Surgical Trial in IntraCerebral Hemorrhage will help formulate
evidence-based recommendations regarding the role of surgery in SICH.

KEY WORDS • intracerebral hemorrhage • epidemiology • pathophysiology •

surgery • medical management

An SICH is defined in this review as a blood clot that EPIDEMIOLOGY OF SICH

arises in the brain parenchyma in the absence of trauma or
surgery. An SICH can be classified as either primary or
secondary depending on the underlying cause of the hem-
orrhage. Primary ICH accounts for approximately 70 to
80% of cases and is due to spontaneous rupture of small
vessels damaged by hypertension or amyloid angiopathy.
Secondary ICH is associated with a number of congenital
and acquired conditions such as vascular anomalies, coag-
ulopathies, tumors, and various drug therapies. Table 1
provides a summary of the important causes of SICH. Its
most frequent sites are the basal ganglia, thalamus, sub-
cortical white matter of the cerebral lobes, cerebellum,
and brainstem.
Abbreviations used in this paper: AVM = arteriovenous malfor-
mation; BBB = blood–brain barrier; CAA = cerebral amyloid an-
giopathy; CBF = cerebral blood flow; CI = confidence interval;
CPP = cerebral perfusion pressure; CT = computerized tomogra-
phy; GCS = Glasgow Coma Scale; ICP = intracranial pressure;
ICH = intracerebral hemorrhage; MABP = mean arterial blood
pressure; MR = magnetic resonance; SAH = subarachnoid hemor-
rhage; SBP = systolic blood pressure; SICH = spontaneous ICH;
SPECT = single-photon emission CT; tPA = tissue plasminogen
activator.
Incidence of SICH
Stroke is the third leading cause of death in the US.
Each year in the US, an estimated 37,000 to 52,400 peo-
ple suffer an ICH.13,119 This rate is expected to double by
2050 as a result of an aging population and changing
racial demographics. Spontaneous ICH accounts for 10 to
15% of all strokes and is associated with a higher mortal-
ity rate than either ischemic stroke or SAH, with only 38%
of affected persons surviving 1 year.31 In a study of all
cases of SICH and SAH in the greater Cincinnati area dur-
ing the year 1988, the investigators found that ICH is
more than twice as common as SAH.15 They determined
the incidence of SICH to be 15 per 100,000 individuals in
the population compared with six per 100,000 for SAH.
Variation in the incidence of SICH is seen among dif-
ferent sexes, age groups, and races. In African Americans
(50 cases per 100,000) and Japanese (55 per 100,000) the
incidence of SICH is twofold that in caucasians.16,118 Men
are more likely to suffer an SICH than women and the
likelihood increases with age.16
A circadian and circannual pattern of SICH onset in the
morning and winter has been reported.40 A similar circa-
dian rhythm has been reported for many cardiovascular

Neurosurg. Focus / Volume 15 / October, 2003 1

 M. E. Fewel, B. G. Thompson, Jr., and J. T. Hoff

and cerebrovascular diseases. Peak incidence in the morn-

ing has been postulated to occur coincidentally with an
increase in sympathetic tone or arterial blood pressure. A
circannual winter peak may be due to responsiveness of
the vascular system to environmental changes in temper-
ature.

CAUSES OF SICH
Table 1 provides an outline of the major causes of
SICH. Cerebrovascular damage to small arteries and arte-
rioles due to chronic hypertension is recognized as the
most significant cause of primary SICH (Fig. 1).37 Racial
differences in hypertension and socioeconomic factors
may explain the high incidence of SICH in certain ethnic
groups.
Cerebral amyloid angiopathy is the other major cause of Fig. 1. Axial CT scan obtained in a 57-year-old man with a his-
primary SICH and an important cause of lobar SICH in tory of hypertension who presented with left hemiplegia and obtun-
elderly populations.126 In CAA,
-amyloid protein is de- dation. A typical location for hypertensive SICH is demonstrated
posited in the media and adventitia of cortical and lep- in the right basal ganglia.
tomeningeal blood vessels. Histological examination of
brain tissue obtained during surgery or at autopsy is nec-
essary for the definitive diagnosis of CAA. Although some investigators have emphasized the problem of intra-
operative hemostasis in these patients, neurosurgery can
be performed safely in those with CAA-related SICH.57
TABLE 1 Hereditary ICH with amyloidosis, Dutch type, is a rare,
Summary of primary and secondary causes of SICH* autosomal-dominant type of CAA that results in recurrent,
sometimes multiple, lobar hemorrhages.
primary Vascular anomalies are the second most common cause
hypertension of SICH overall (Fig. 2). Aneurysms, AVMs, cavernomas,
amyloid angiopathy dural arteriovenous fistulas, and venous malformations
secondary
aneurysm all can result in secondary SICH. The hemorrhage due to
saccular a ruptured aneurysm almost always has a subarachnoid
fusiform component and often extends into the ventricles. There
mycotic should be a high index of suspicion in young patients with
vascular malformation frontal or temporal lobe clots. Arteriovenous malforma-
AVM tions are associated with an estimated mean annual hem-
cavernous malformation
venous angioma orrhage risk of 4%.
dural arteriovenous fistula Cerebral venous malformations are assumed to be con-
neoplasm genital lesions resulting from aberrant venous maturation.
primary These entities have also been termed venous angiomas or
metastatic developmental venous anomalies. Although rarely associ-
coagulopathy ated with SICH, venous malformations are the most fre-
acquired
anticoagulation (Coumadin, heparin) quently occurring vascular lesion demonstrated in autopsy
thrombolytics (tPA, urokinase) and radiological series.42 Their clinical significance re-
blood dyscrasias (DIC, leukemia, thrombocytopenia) mains controversial, with some authors reporting a neg-
hepatic failure ligible hemorrhage risk42 and others suggesting a higher
platelet dysfunction (renal failure, medications) risk.76 In a series of 63 patients with known venous mal-
congenital formations followed longitudinally by Naff, et al.,91 the
hemophilia
platelet disorders risk of hemorrhage was 0.15% per lesion-year. We believe
drugs or alcohol a conservative approach to their management should be
sympathomimetics (ephedrine, phenylpropanolamine, adopted, because the risks of radiosurgery or resection
pseudoephedrine) probably outweigh the risk of hemorrhage.
cocaine, amphetamine, Ecstasy Spontaneous ICH is not an uncommon presentation in
hemorrhagic ischemic stroke cases of previously unsuspected brain tumor. Hemorrhage
dural venous sinus thrombosis
vasculitis/vasculopathy resulting from brain tumors can occur in up to 10% of all
moyamoya disease primary or metastatic tumors. Hemorrhage is more likely
arterial dissection with certain types of tumors, including glioblastoma, he-
pregnancy mangioblastoma, oligodendroglioma, and metastatic tu-
eclampsia, venous sinus thrombosis mors. Metastatic tumors with a high propensity to hemor-
other/unknown rhage are malignant melanoma, renal cell, prostate, and
* DIC = disseminated intravascular coagulation. lung cancer. An indentation of the hematoma’s surface

2 Neurosurg. Focus / Volume 15 / October, 2003

Spontaneous intracerebral hemorrhage
Fig. 2. Imaging studies obtained in a 26-year-old woman who
presented with severe headache, vomiting, rapid progression to
unconsciousness, and a blown left pupil. Left: Axial CT scan re-
vealing a large left temporoparietooccipital SICH with midline
shift. At craniotomy for emergency clot evacuation, an AVM was
observed, but no attempt at resection was made. Right: Left ver-
tebral artery angiogram obtained after clot evacuation confirmed a
left posterior temporal AVM with a 2-cm nidus. Feeding arteries
from the posterior cerebral artery and drainage into the transverse
sinus are shown. The patient was allowed to recover from the ini-
tial hemorrhage and then underwent resection of the AVM 1 month
later.
revealed on pre–contrast CT scanning should prompt sus-
picion of a tumoral hemorrhage.56 Contrast-enhanced CT
scanning or contrast-enhanced MR imaging usually iden-
tifies the tumor, although small areas of contrast enhance-
ment can be obscured by hemorrhage. In young patients
without hypertension follow-up imaging is warranted if
initial studies are unequivocal regarding cause (Fig. 3).
Patients with liver failure or coagulopathy from any
cause are at higher risk for SICH.94 The liver is responsi-
ble for synthesis of most clotting factors and fibrinogen, as
well as playing a major role in metabolism of platelets.
Excluding neonatal intraventricular hemorrhage, struc-
tural vascular lesions are the most common cause of SICH
in children. Hematological disorders and coagulopathies,
as well as brain tumors, are also important causes. In a
series of 68 children presenting with SICH to a single in-
stitution, vascular anomaly or underlying brain tumor was
identified in 38 (56%), whereas hematological or coagula-
tion disorders were present in 22 (32%).4 Because sys-
temic hypertension is rare in this population, hypertensive
SICH is also rare. An underlying structural lesion should
be sought in all pediatric patients presenting with SICH.
We recommend that MR imaging and angiography be
undertaken in all pediatric patients suffering unexplained
SICH as well as follow-up imaging if no cause is found.
Moyamoya disease is a rare disorder characterized by
progressive occlusive changes in the anterior circulation
of the circle of Willis and the abnormal development
of collateral moyamoya vessels in the basal ganglia. Al-
though cerebral ischemia is usually the initial symptom in
children with the disease, SICH is the most common ini-
tial presentation in adults.117 Rupture of microaneurysms
and failure of the abnormally dilated moyamoya vessels
from hemodynamic stress are considered the cause of
Neurosurg. Focus / Volume 15 / October, 2003
these hemorrhages. Long-term prospective evidence that
revascularization surgery significantly prevents rebleed-
ing in patients with hemorrhagic moyamoya disease is
lacking.134
Abuse of sympathomimetic drugs such as cocaine, am-
phetamine, or Ecstasy has been reported in association
with SICH. The mechanism is believed to be secondary to
induced arterial hypertension. Drug abuse may represent
the most common cause of stroke in young adults. These
drugs may also predispose unsuspected vascular anom-
alies to hemorrhage. In a study of 11 patients (mean age
31 years) with a history of abuse of one of these agents
leading to SICH, 10 were found to harbor intracranial
aneurysms or AVMs.83 Therefore, the traditional teaching
that these were hypertension-related hemorrhages that did
not require investigation for underlying vascular abnor-
mality is no longer valid. Cerebral angiography or MR
angiography should be a standard part of the workup of
SICH in young adults, even in the presence of positive
toxicology screen findings.
The most commonly implicated sympathomimetic
medication related to SICH is phenylpropanolamine.8
Ephedrine, pseudoephedrine, and phencyclidine have also
been described in association with SICH.109 Ginkgo bilo-
ba is an over-the-counter herbal extract used by many peo-
ple in the US and Europe as a “memory aid;” several case
reports have described an association between its use and
ICH.78 A possible antiplatelet effect and Coumadin poten-
tiation mechanism has been postulated.
Pregnancy and the puerperium can be associated with
SICH due to dural venous sinus thrombosis and eclamp-
sia. SICH occurs in one to five per 10,000 pregnancies.82
There are numerous other rare causes of SICH including
cerebral artery dissection,41 infections, and a host of un-
usual bleeding diatheses.
Risk Factors
With a mortality rate of 40 to 50% in SICH, identifica-
tion of modifiable risk factors might help lower its inci-
dence. The role of hypertension as a risk factor for SICH
is well established. Hypertension continues to be the sin-
gle greatest modifiable risk factor for SICH.15,122 The in-
cidence of SICH has, in fact, declined coincident with
improvements in hypertension control. Authors of the Sy-
stolic Hypertension in the Elderly Program reported that
treatment of isolated systolic hypertension in the elderly
decreased the risk of SICH by 50%.113 The relative risk for
SICH in hypertensive individuals, however, remains 3.9
to 13.3 times that for normotensive individuals.20
Juvela and colleagues62 have reported that recent mod-
erate and heavy alcohol intake, hypertension, and antico-
agulant treatment appear to be independent risk factors for
SICH. Cigarette smoking seems to increase the risk for
SAH, but is less strongly associated with SICH.122 Dia-
betes mellitus is more commonly associated with SICH
than with SAH.59
Anticoagulant treatment is a well-established risk factor
for SICH, especially in elderly patients (Fig. 4).38 Conven-
tional anticoagulation therapy increases the risk of ICH
seven- to 10-fold.47 The use of thrombolytic agents can be
complicated by ICH. In a metaanalysis of trials assessing
treatment for ischemic stroke with tPA, the authors report-
ed a 6.5% risk of SICH.44
3
 M. E. Fewel, B. G. Thompson, Jr., and J. T. Hoff

Fig. 3. Neuroimaging studies acquired in a 4-year-old girl who presented with acute headache, vomiting, and seizures.
Upper Left: Axial CT scan revealing a large right frontal SICH with subdural extension. Both MR imaging and angiog-
raphy were negative for underlying tumor or vascular anomaly. The patient underwent craniotomy for clot evacuation,
but no definitive diagnosis was established. Upper Center: Follow-up CT scan obtained 5 months after initial presen-
tation is unremarkable except for demonstration of a small residual right frontal hypodensity. Upper Right: The patient
returned 1 month later with progressive left hemiparesis and headache. An axial CT scan demonstrated a recurrent mixed-
attenuation right frontal hemorrhagic lesion. Lower Left: Further evaluation with coronal T1-weighted MR imaging
demonstrated a mixed–signal intensity right frontal hemorrhagic mass. Lower Center: Coronal enhanced T1-weighted
MR image revealing enhancement of the mass, suspicious for a hemorrhagic tumor. Lower Right: The patient under-
went craniotomy for resection of the lesion. Examination of an intraoperative tissue specimen suggested a supratentori-
al primitive neuroectodermal tumor. A postoperative coronal enhanced T1-weighted MR image demonstrated no obvious
residual tumor.

In a recent metaanalysis of aspirin trials involving
55,462 participants, aspirin was associated with a slight
absolute risk increase in hemorrhagic stroke. This was
outweighed, however, by the overall risk reduction in my-
ocardial infarction and ischemic stroke.48 Use of nonas-
pirin nonsteroidal antiinflammatory drugs does not seem
to be associated with an overall increased risk of SICH.58
Sympathomimetic drugs such as phenylpropanolamine
and ephedrine, which are structurally similar to ampheta-
mines, have been implicated in cases of SICH.21
The apolipoprotein E genotype appears to have an im-
portant role in the pathogenesis of CAA, and this geno-
type is associated with an increased risk of recurrent lobar
ICHs.95
Recent attention has been focused on the association of
hypocholesterolemia and an increased incidence of SICH.
The mechanism underlying this relationship is not clear,
but some investigators have proposed that the interaction
of high diastolic pressure and low cholesterol levels weak-
ens the endothelium of intracerebral arteries.54 Alterna-
tively, there is some evidence that high plasma cholesterol
levels may be associated with reduced risk of SICH.122
Hepatitis C virus infection may also be a risk factor for
SICH as a result of subclinical clotting disorder or vessel
wall friability resulting from hypocholesterolemia.66
Prevention of SICH should emphasize treatment of hy-
pertension, control of alcohol intake and sympathomimet-
ic drug abuse, careful monitoring of anticoagulation levels
in patients taking warfarin, and diligent selection of pa-
tients for thrombolytic treatment for myocardial infarction
and acute ischemic stroke.13
CLINICAL FEATURES
Initial Presentation
The classic presentation of SICH is sudden onset of
focal neurological deficit progressing over hours with
accompanying headache, nausea, vomiting, altered con-
sciousness, and elevated blood pressure. Supratentori-
al hemorrhage is commonly accompanied by vomiting
and altered consciousness but is rarely seen in ischemic
stroke.22 Elevation in blood pressure occurs in as many as
90% of patients with SICH. Seizures occur in approxi-
mately 10% of patients.10
The type of focal neurological deficit, of course, de-
pends on hematoma location. Patients with supratentorial
hemorrhage often present with contralateral sensory or
motor deficits, aphasia, neglect, gaze deviation, and hemi-
anopia. Infratentorial hemorrhages manifest with signs of
brainstem dysfunction, cranial nerve abnormalities, atax-
ia, nystagmus, and dysmetria.
Blood may rupture into the ventricles and cause hydro-
cephalus. Rarely, blood finds its way into the subarach-
noid space. A large hemorrhage can raise ICP to the level

4 Neurosurg. Focus / Volume 15 / October, 2003

Spontaneous intracerebral hemorrhage
of the blood pressure until bleeding is tamponaded. De-
pending on clot location, this can result in brain hernia-
tion, compression of the brainstem, and death. Cerebellar
clots greater than 3 cm in diameter have a poor prognosis
if left untreated.
Risk of Hematoma Enlargement
In nearly one quarter of initially alert patients present-
ing with SICH, secondary deterioration in level of con-
sciousness occurs within the first 24 hours after onset.
Hematoma expansion and edema formation are believed
to be the major factors involved (Fig. 4).81 In several large
recent prospective and retrospective studies, investigators
have evaluated the rate of hematoma enlargement after
initial presentation and report rates ranging from 14 to
38% within the first 24 hours of admission.19,39,68
In their review of 627 patients with SICH, Fujii, et al.,39
reported that CT scanning within 24 hours of admission
demonstrated enlargement of the hematoma in 14% of
patients. Five factors were found to be associated with
enlargement: admission shortly after onset of symptoms,
heavy alcohol consumption, irregularly shaped hemato-
ma, reduced level of consciousness, and low level of fib-
rinogen.
Kazui, et al.,67 used multiple logistic regression analysis
of data obtained in 186 patients with SICH to identify fac-
tors related to the risk of hematoma enlargement. They
identified four factors predisposing to hematoma enlarge-
ment: history of brain infarction, liver disease, fasting glu-
cose level greater than 141 mg/dl with SBP greater than
200 mm Hg, and HgbA1c greater than 5.1% with SBP
greater than 200 mm Hg. Enlargement of the hematoma
was demonstrated in 22%. A long time interval between
onset of symptoms to the first CT scanning study as well
as small hematoma volume ( 25 cm3) strongly reduced
the risk of hematoma enlargement.
In another report by Kazui, et al.,68 the authors showed
that, although CT scanning–documented expansion of ICH
was common in the hyperacute stage, 17% of hematoma
Fig. 4. Axial CT scans obtained in a 72-year-old woman who
developed right hemiparesis after being hospitalized and undergo-
ing heparin-based anticoagulation therapy for acute myocardial
infarction. Left: The scan demonstrates a small left posterior
frontal SICH. The heparin therapy was stopped and reversed with
protamine. Right: A scan was obtained several hours later when
the patient deteriorated neurologically. Significant hematoma en-
largement is demonstrated. The patient subsequently died.
Neurosurg. Focus / Volume 15 / October, 2003
expansion occurred even after 6 hours of onset. Hema-
toma enlargement after 24 hours was rare. Early repeated
CT scanning appears to increase the rate of detection of
hematoma enlargement.68 It is our practice to perform re-
peated CT scanning within 24 hours of admission in pa-
tients in whom SICH is revealed on the initial CT scan.
In the only prospective study of this phenomenon,
Brott, et al.,19 evaluated 103 patients with SICH and found
no significant predictor of hemorrhage growth. Overall
hemorrhage growth occurred in 38% of patients in whom
CT scanning was conducted within 3 hours of symptom
onset.
Mayer79 has proposed using ultra-early hemostatic ther-
apy with recombinant activated factor VII as a potential
method of minimizing this early hematoma growth. The
author postulated that ultra-early hemostatic therapy for
SICH could be used as the counterpart to thrombolytic
intervention in cases involving acute ischemic stroke to
prevent early neurological deterioration due to hematoma
growth and late deterioration related to secondary effects
of edema, ischemia, and inflammation. We have recently
reported our experience using recombinant activated fac-
tor VII for rapid correction of coagulopathy in a small
series of coagulopathic patients (including two with intra-
parenchymal hemorrhages) requiring urgent neurosurgical
intervention (Fig. 5).96
Outcome After SICH
Despite advances in neurocritical care and neurosur-
gery, the prognosis for patients with SICH remains sober-
ing. Of the estimated 37,000 Americans who suffered an
SICH in 1997, 35 to 52% were dead by 1 month postictus
and only 20% were living independently by 6 months.15
The volume of ICH, initial GCS score, and intraventricu-
lar extension of the hemorrhage are powerful predictors of
30-day mortality and morbidity rates in patients with
SICH.18,123
The 30-day mortality rate in one series of 188 patients
was 44%, with most deaths occurring in the first 2 days
after onset.15 Univariate logistic regression analysis in this
series demonstrated that ICH volume, intraventricular
hemorrhage volume, and initial GCS score were signifi-
cant predictors of 30-day mortality, whereas age, sex,
race, SBP, and location of hemorrhage were not. In pa-
tients in whom the hematoma volume was 60 cm3 or
greater and the GCS score was 8 or less, the predicted 30-
day mortality rate was 91% compared with only 19% in
those in whom the volume was less than 30 cm3 and the
GCS score was 9 or more.
In another study of 156 consecutive patients with SICH,
potential risk factors for death and impaired outcome were
studied prospectively.60 Risk of poor outcome was pre-
dicted by GCS score, volume of hematoma, age, amount
of alcohol consumed within 1 week before hemorrhage,
and presence of cerebellar hematoma. Of these factors,
only alcohol consumption before SICH is potentially
modifiable.
DIAGNOSTIC FACTORS
Computerizd Tomography Scanning
Computerized tomography scanning is the initial diag-
5
 M. E. Fewel, B. G. Thompson, Jr., and J. T. Hoff

Fig. 5. Axial CT scans acquired in a 78-year-old man receiving Coumadin for atrial fibrillation who presented with a
right temporoparietal SICH. Initial coagulation parameters were a prothrombin time of 28, international normalized ratio
of 2.8, and an activated partial thromboplastin time of 36; after a 90-g/kg dose of recombinant activated factor VII, the
values immediately corrected to, respectively, 9.6, 0.9, and 32. The patient then underwent an uncomplicated craniotomy
for clot evacuation. Left: Noncontrast scan obtained at presentation. Center: Scan obtained on postoperative Day 1.
Right: Follow-up scan obtained 6 weeks postoperatively.

nostic procedure of choice in acute stroke. Several clinical
stroke scoring systems have been developed in an effort to
help clinicians distinguish between hemorrhagic and
ischemic stroke on the basis of clinical presentation alone.
These stroke scoring systems, however, are inadequate for
accurately differentiating between the two stroke types.128
Hematomas, even just a few millimeters in diameter, are
rapidly and accurately identified on CT scans. Not only
does CT scanning demonstrate the size and location of
the SICH, but it can also suggest potential causes such
as tumor, vascular malformation, or aneurysm. Related
complications such as hydrocephalus, edema, herniation,
and intraventricular extension are easily identified. The
change in the CT scanning appearance of intracerebral
hematomas over time has been well studied.28
As mentioned previously, hemorrhage volume is one of
the strongest predictors of outcome from SICH.18 Using a
model of 30-day mortality based on GCS score and hem-
orrhage volume, outcome has been correctly predicted
with 97% sensitivity and specificity. Many modern CT
scanners are able to calculate hematoma volume directly
by using special software. If direct volume measurements
are not possible, a rapid, simplified method of determin-
ing hematoma volume has been described and validated.71
The formula used—(A  B  C)/2—is an approximation
for the volume of an ellipsoid where A is the greatest hem-
orrhage diameter on axial CT scans, B is the largest diam-
eter 90° to A, and C is the number of CT slices with hem-
orrhage multiplied by the slice thickness.
Magnetic Resonance Imaging
Magnetic resonance imaging and MR angiography are
becoming increasingly useful in the diagnosis of SICH
and may eventually replace CT scanning as the initial di-
agnostic test of choice in stroke. Magnetic resonance im-
aging is helpful for identifying brain tumors and is more
sensitive in detecting cavernous malformations than CT
scanning or angiography. The MR imaging appearance of
SICH is complex and depends on hemoglobin breakdown
6 Neurosurg. Focus / Volume 15 / October, 2003
products over time.25,33 Table 2 provides a summary of the
sequential MR imaging signal intensity changes of ICH.
Although SICH can be classified into distinct stages, there
is considerable overlap and variation in duration of these
stages.
Although MR imaging has been previously believed to
be poorly sensitive for detecting hyperacute hemorrhage,
the authors of more recent reports have shown that multi-
sequence MR imaging is as reliable as CT scanning in the
assessment of hyperacute SICH.98,112 In older studies a 24-
hour time period was suggested before detectable amounts
of paramagnetic deoxyhemoglobin would be present. The
results of several small studies support the hypothesis
that deoxyhemoglobin is present within the early hours
after SICH and may be detected best by using suscepti-
bility-weighted sequences.75,98,112 Multimodal MR imag-
ing techniques such as diffusion- and perfusion-weighted
sequences may add to its diagnostic reliability in hypera-
cute SICH.
Cerebral Angiography
Despite advances in imaging technology, conventional
cerebral angiography remains the gold standard for diag-
nosis of vascular abnormality. A negative CT angiogram,
MR image, and MR angiogram cannot completely ex-
clude a vascular lesion and angiography is often needed
for definitive diagnosis. Although the risk of angiography
is low, the diagnostic yield must be weighed against the
procedural risk at each institution. The site of hemorrhage,
patient age, preexisting hypertension, and clinical status
must all be considered collectively before choosing an-
giography. Elderly patients in whom the risk of surgery is
unacceptable or neurological disability is severe should
not undergo angiography if identification of an underlying
vascular abnormality would have no effect on patient
management. Surgery in a young patient with SICH and
evidence of herniation should not be delayed by angiogra-
phy (Figs. 2 and 6).
Potentially treatable structural abnormalities cannot be
Spontaneous intracerebral hemorrhage
TABLE 2
Evolution of ICH MR imaging–evidenced signal intensities*
MRI Signal Intensity†
ICH Stage Hemoglobin State T1-Weighted T2-Weighted
hyperacute oxyhemoglobin = to ↓
(0–6 hrs) (intracellular)
acute deoxyhemoglobin = to ↓
(6 hrs–3 days) (intracellular)
early subacute methemoglobin ↑↑
(3 days–1 wk) (intracellular)
late subacute methemoglobin ↑↑
(1 wk–1 mos) (extracellular)
chronic hemosiderin & ↓ ↓↓
(1 mo) ferritin
* “= to ↓" = "equal to decreased;” ↓ = decreased; ↓↓ = significantly
decreased; ↑ = increased; ↑↑ = significantly increased.
† Compared with gray matter.
excluded on the basis of hematoma location, patient age,
or a history of preexisting hypertension. In fact, a signifi-
cant number of elderly hypertensive patients are found to
harbor potentially surgically treatable lesions. In a pro-
spective study of 100 consecutive cases of nontraumatic
SICH, the authors identified aneurysms or AVMs in 49%
of cases.43 When a temporal lobe hematoma extended into
the sylvian fissure or when it was associated with SAH,
structural abnormalities were identified in more than 90%
of cases. Although the diagnostic yield was lower in hy-
pertensive patients, angiography still identified an AVM
or aneurysm in 25% of hypertensive patients.
In another prospective study of angiography in 206
patients with SICH, the angiographic yield was 45% (66
of 148) in normotensive patients compared with 9% (five
of 58) in hypertensive patients.137 A positive finding on
angiography was identified in 53 (50%) of 105 patients
younger than 45 years of age, compared with 18 (18%) of
101 patients older than 45 years. In grouping putaminal,
thalamic, and posterior fossa hemorrhages together, the
angiographic yield was 12 (48%) of 25 in younger nor-
motensive patients compared with zero (0%) of 29 in hy-
pertensive patients. Based on their results, Zhu, et al.,137
recommended that the use of angiography should be con-
sidered in all patients with SICH except those older than
45 years of age with preexisting hypertension in thalamic,
putaminal, or posterior fossa hemorrhage.
Imaging features that suggest an underlying structural
abnormality include SAH, extracerebral hemorrhage, tem-
poral lobe or perisylvian location, and intraventricular
extension (Fig. 7). Halpin, et al.,45 reported that CT scan-
ning features in patients with SICH that warrant angiogra-
phy include the presence of SAH, abnormal intracranial
calcification, prominent vascular structures, and perisyl-
vian hemorrhage location. In their study, angiography was
positive in 32 (84%) of 38 patients in whom there was one
or more of these CT findings. Follow-up angiography and
MR imaging/angiography should be considered in these
high-risk patients if initial studies are negative.
Laboratory Evaluation
Laboratory evaluations should include a complete
Neurosurg. Focus / Volume 15 / October, 2003
↑
↓
↓↓
↑↑
Fig. 6. Imaging studies obtained in a 19-year-old man with sud-
den-onset headache and rapidly progressive loss of consciousness.
Left: Axial CT scan demonstrating a large left cerebellar hematoma
with near-complete effacement of the fourth ventricle and an
enlarged right temporal horn. The patient underwent emergency
ventriculostomy and craniectomy for clot evacuation. A presumed
AVM was identified at surgery. Right: Anteroposterior left ver-
tebral artery angiogram confirming an AVM in the far-lateral
aspect of the cerebellum. Following clot evacuation, the patient
suffered multiple lower cranial nerve deficits and bilateral sixth
nerve palsies, which improved markedly over time. Cognitive sta-
tus normalized and he underwent resection of the AVM 6 weeks
after initial presentation.
blood count, liver panel, coagulation profile, electrolytes,
and toxicology screen. Underlying renal or liver failure,
acquired or iatrogenic coagulopathy, infection, or drug use
should be investigated.
Histopathological Examination
In a significant percentage of surgically treated cases,
the cause of the SICH is unknown prior to surgical inter-
vention. Often these hemorrhages are attributed to hyper-
tension. Histological examination of surgical specimens
obtained in a series of 84 cases of SICH in which the
cause was unknown revealed a 25% rate of diagnosis
involving specimens in which surrounding tissue was
included.50 Biopsy sampling of adjacent hematoma wall or
preservation of tissue fragments identified at the time of
surgery can, therefore, be of important diagnostic value.
PATHOPHYSIOLOGICAL MECHANISMS
Intracerebral hemorrhage commonly occurs in the basal
ganglia, thalamus, cerebral lobes, brainstem, and cerebel-
lum. Primary tissue damage and distortion occur at the
time of hematoma formation when the blood spreads be-
tween planes of white matter cleavage. Hemorrhage most
commonly results from rupture of the small penetrating
arteries damaged by the degenerative effects of chronic
hypertension. In 1868, Charcot and Bouchard26 described
the rupture of “microaneurysms” as the cause of SICH.
Recent investigators have questioned the existence of
such microaneurysms, suggesting that many of these
structures are injection-related artifacts, complex vascular
coils, subadventitial hemorrhages, or extravascular clots
resulting from endothelial damage by the hematoma.24,37
7

Fig. 7. Imaging studies obtained in a 39-year-old woman who
presented with sudden-onset severe headache, somnolence, and
left hemiparesis. Left: Axial CT scan revealing a large right tem-
poral SICH. Basal cistern SAH was not prominent, but the clot
extended inferiorly to the temporal floor and sylvian fissure. These
features, combined with the presence of intraventricular hem-
orrhage, should raise the suspicion of aneurysmal hemorrhage.
Right: Three-dimensional CT angiogram confirming the presence
of a right middle cerebral artery bifurcation aneurysm.
The vasculopathy of chronic hypertension affects the
perforating arteries 100 to 400 m in diameter, resulting
in lipohyalinosis or focal necrosis. This helps explain the
distribution of hypertensive hemorrhages in territories
supplied by the lenticulostriate arteries (basal ganglia), the
thalamoperforating arteries (thalamus), the perforating ar-
teries of the basilar artery (pons), and the superior and an-
terior inferior cerebellar arteries (cerebellum). Lobar hem-
orrhages in the elderly are often considered to be a result
of amyloid angiopathy, which preferentially involves the
cortical and leptomeningeal blood vessels.
Unlike primary tissue injury from the hematoma for-
mation, secondary brain injury and edema represent po-
tential therapeutic targets. The roles of edema, ischemia,
mass effect, direct cellular toxicity, inflammation, and
apoptosis are being evaluated in experimental studies of
SICH. There is controversy whether secondary ischemia
contributes to brain injury after SICH.51,80,111,114,136 Cush-
ing29 believed that brain injury due to SICH was the result
of local pressure compressing the microcirculation and
causing ischemia around a hematoma. Astrup, et al.,5 pro-
posed the term “ischemic penumbra” to describe brain
with CBF values above a lower limit of membrane dys-
function and cell death, but below an upper limit corre-
sponding to electrophysiological dysfunction.
There are a number of animal models that have been
used to study SICH and its effects on ICP, CBF, and is-
chemia.92 Findings included increased ICP associated with
hematoma formation, decreased CBF adjacent to the he-
matoma, and histological ischemic changes in perihema-
tomal brain tissue. Yang, et al.,133 have suggested that the
primary cause of edema formation in experimental mod-
els of SICH may be local tissue ischemia.
In more recent clinical studies, evidence does not clear-
ly support the concept of perilesional ischemia in SICH.124
8 Neurosurg. Focus / Volume 15 / October, 2003
M. E. Fewel, B. G. Thompson, Jr., and J. T. Hoff
The authors of several studies have suggested that perile-
sional hypoperfusion can occur without ischemia as a
result of reduced metabolic demand or diaschisis.111,136 In
a study of 23 patients with SICH studied using SPECT
scanning, perilesional blood flow was found to be lowest
during the first 24 hours postictus, but it normalized as
edema formed over the next 2 to 3 days.80 In a small study
in which positron emission tomography was used, no zone
of tissue hypoxia or “ischemic penumbra” was identi-
fied.51 Siddique and colleagues,114 however, analyzed ser-
ial SPECT scans obtained in patients with SICH and iden-
tified perilesional hypoperfusion. They were also able to
demonstrate that some of the ischemic perilesional brain
recovered its perfusion in the long term. There may be a
window of opportunity early after SICH during which re-
moval of the hematoma allows recovery of CBF and pre-
vention of secondary ischemic insult.
Blood and plasma products mediate many secondary
processes initiated after SICH.131 After SICH occurs, in-
flammatory mediators from the blood can induce an
inflammatory reaction in and around the hematoma. Neu-
trophils, macrophages, leukocytes, and activated micro-
glia can be found. The release of cytotoxic enzymes, free
oxygen radicals, nitric oxide, and products of the phos-
pholipid cascade is thought to contribute to secondary
neural injury and cell death.133 Both necrotic and apoptot-
ic neuronal death appear to play a role.49,104
Although the mechanisms of edema formation follow-
ing ICH are not fully clarified, several mechanisms may
be responsible. These include hydrostatic pressure during
hematoma formation and clot retraction; coagulation cas-
cade activation and thrombin production; erythrocyte lysis
and hemoglobin toxicity; complement activation; mass
effect; secondary perihematoma ischemia; reperfusion
brain injury; and BBB disruption.52,73,80,129,131 Evidence in-
dicates that thrombin formation plays an important role in
inducing brain injury after ICH.131 Cell toxicity and
thrombin-related BBB disruption are triggering mecha-
nisms for edema formation.74 In a rat SICH model, throm-
bin inhibition was shown to reduce brain edema and neu-
rological deficits.130
The expression of matrix metalloproteinase–9, a prote-
olytic enzyme involved in reorganization of the extracel-
lular matrix, has been shown to be increased following
SICH.1 In patients with deep-seated hemorrhages, this in-
creased expression of matrix metalloproteinase–9 appears
to be associated with perihematoma edema and neurolog-
ical worsening.
Both vasogenic and cytotoxic edema result from dis-
ruption of the BBB, sodium pump failure, and neuronal
death. In experimental animal models of SICH, brain ede-
ma increases during the first 24 hours, peaks at 4 to 5 days,
and then begins to resolve.133 A delayed phase of edema
formation and injury in the 2nd or 3rd weeks after SICH
in humans may be due to hemoglobin and its degradation
products.129 Hemoglobin-induced brain edema is partially
blocked by the iron chelator deferoxamine in animal mod-
els, supporting the hypothesis that iron has a role in hem-
orrhagic brain injury.53
Breakdown of the hematoma consists of invasion by
macrophages, progression of surrounding edema, devel-
opment of microvessels at the clot margin, and, eventual-
Spontaneous intracerebral hemorrhage
ly, gliosis.69 The end result is a hemosiderin-stained scar
or a cavity containing old blood surrounded by fibrous tis-
sue. An unusual form of chronic intracerebral hematoma
has been defined by Roda and colleagues.106 Their term
“encapsulated intracerebral hematoma” refers to a chroni-
cally encapsulated intraparenchymal hematoma found in
normotensive patients, usually caused by vascular malfor-
mations. These lesions exhibit progressive mass effect,
edema, and ring enhancement. The self-perpetuating na-
ture of these lesions has been compared with chronic sub-
dural hematomas.
TREATMENT OF SICH
Macewan reported the first successful surgery for SICH
in 1883. In general, there has been a pessimistic attitude
among many medical professionals, including neurosur-
geons, regarding treatment of SICH. As reported by
Broderick and colleagues,14 this is reflected in the fact that
more than 315 randomized clinical therapeutic trials for
acute ischemic stroke and 78 trials for SAH were com-
pleted or ongoing as of 1995, whereas only the results of
four small randomized surgical trials6,9,61,84 and four small
medical trials of SICH had been published.55,99,121,135 The
American Heart Association formed a task force that pub-
lished practice guidelines in 1999 for the management of
SICH and emphasized the desperate need for future ran-
domized controlled trials.13
Medical Treatment
Critical Care. Because the risk of clinical deterioration
after SICH is greatest during the 1st day, patients in whom
this entity has been diagnosed should be monitored in an
intensive care unit for a minimum of 24 hours. The goal
of fluid status is to maintain euvolemia with the use of iso-
tonic crystalloid intravenous solutions. Enteral nutrition,
via a feeding tube if necessary, should begin as soon as
possible. Pneumatic sequential compression devices and
elastic stockings decrease the risk of venous thrombosis
and pulmonary embolism. Gastric cytoprotection should
also be considered. Physical therapy, occupational thera-
py, speech therapy, and rehabilitation services should be
consulted as soon as medically feasible.
Seizure Prophylaxis. The authors of retrospective stud-
ies have suggested that seizures are more frequent in hem-
orrhagic stroke than ischemic stroke and that the risk of
epilepsy is greater in patients with late onset of the initial
seizure after stroke.23,116 Most seizures occur within the
first 24 hours after SICH.10
Cervoni, et al.,23 reviewed data obtained in 298 patients
with SICH and found an 18% rate of seizures. Hematoma
location was important because 35% of patients with lobar
hemorrhages suffered seizures compared with only 8% of
those with deep-seated hematomas. Epilepsy developed in
43% of patients affected by seizures occurring more than
2 weeks after the onset of SICH, whereas it developed in
only 17% of those with seizures appearing within 2 weeks
of onset. Passero, et al.,97 retrospectively reviewed data
acquired in 761 patients with supratentorial SICH and re-
ported a 30-day actuarial risk of post–SICH seizures of
8.1%. They found that the risk of early seizures in lobar
Neurosurg. Focus / Volume 15 / October, 2003
SICH was reduced by prophylactic antiepileptic drug
therapy.
Investigators recently reported the results of a large,
prospective multicenter study designed to determine the
incidence, outcome, and risk factors for seizures after both
ischemic and hemorrhagic stroke. In 1897 patients ana-
lyzed, seizures occurred in 28 (10.6%) of 265 patients
with hemorrhagic stroke compared with 140 (8.6%) of
1632 with ischemic stroke.10 The Kaplan–Meier survival
analysis indicated the 1-year actuarial risk of seizures to
be 20% in survivors of hemorrhagic stroke. Fifty-seven
percent of seizures after hemorrhagic stroke occurred in
the first 24 hours. Based on multivariate analysis, the only
risk factor for seizures after hemorrhagic stroke was cor-
tical location. Epilepsy occurred in 47 (2.5%) of 1897
patients. Epilepsy occurred in seven (100%) of seven pa-
tients with late-onset ( 2 weeks after SICH) seizures
after hemorrhagic stroke.
The effect of surgical treatment for SICH on the occur-
rence and development of seizures has not been well stud-
ied. Given the results of the aforementioned report,10 we
recommend undertaking anticonvulsant therapy in all pa-
tients who present with supratentorial SICH. No consen-
sus exists on when anticonvulsant therapy can be discon-
tinued following SICH, but most medications can be
withdrawn after the 1st month in patients in whom no fur-
ther seizure activity occurs. Those with late-onset seizures
may require long-term anticonvulsant prophylaxis.23
Blood Pressure Management. Hypertension is common
in patients who suffer SICH and in the early clinical
course. There is disagreement as to the mechanism of this
acute hypertensive state; however, major factors appear to
be catecholamine release and the Cushing response.
Proponents of lowering blood pressure argue that doing
so reduces secondary injury due to rebleeding and edema
formation.39 Opponents, who favor leaving hypertension
untreated, argue that preserving CPP is paramount to
areas in the ischemic penumbra. Although induced hyper-
tension can increase edema in experimental models, hy-
pertension following SICH is not clearly related to the
development of edema. Broderick, et al.,11 found no rela-
tionship between blood pressure and edema growth or
neurological worsening in a study of 115 patients. Be-
cause of these complex issues, the management of hyper-
tension in SICH remains controversial.3
The changes that occur in CBF in the setting of SICH
are complex. An excellent review of the existing laborato-
ry and clinical reports on this subject was recently provid-
ed by Qureshi and colleagues.103 Cerebral autoregulation
refers to the ability to maintain constant CBF over a wide
range of CPPs varying from approximately 60 to 150 mm
Hg.115 Autoregulation can become deranged after a num-
ber of cerebral insults including ischemia, SAH, traumat-
ic brain injury, and SICH. When CPP decreases below the
lower limit of autoregulation, oxygen extraction increases
in a compensatory fashion. After oxygen extraction is
maximized, ischemia occurs. In most patients with chron-
ic hypertension, the autoregulatory curve is at higher pres-
sures, which may increase the risk of cerebral ischemia if
blood pressure is normalized.103
There are no prospective randomized placebo-con-
trolled studies in which investigators have compared the
9

effects of lowering blood pressure on immediate clinical
status. Qureshi and colleagues102 reported, in a retrospec-
tive analysis of 105 patients admitted with SICH, that a
rapid decline in MABP within 24 hours after presentation
was independently associated with an increased mortality
rate. These authors postulated two mechanisms for the
deleterious effect of a rapid decline in MABP. First, a
rapid decline in blood pressure may lead to a reduction in
CPP and exacerbation of ischemic injury. Second, in re-
gions with intact autoregulation, a compensatory vasodi-
lation in response to the blood pressure reduction may
result in exacerbation of intracranial hypertension.
In a study in which they used SPECT scanning, Ku-
wata, et al.,72 found that a 20% reduction in SBP resulted
in decreased mean CBF in the acute period following hy-
pertensive ICH. Kaneko, et al.,65 suggested a goal of SBP
reduction of less than 20% in the acute period on the basis
of CBF studies. In two other studies investigators have
demonstrated that a controlled pharmacologically mediat-
ed reduction in blood pressure does not have deleterious
effects on CBF.100,105
In several studies improved outcomes have been report-
ed with the use of antihypertension agents. Meyer and
Bauer86 published a prospective nonrandomized unblind-
ed trial involving use of the antihypertension agent reser-
pine in acute SICH. They concluded that antihypertension
drugs improved the mortality rate; however, there were
important limitations to the study. Dandapini, et al.,30 also
demonstrated improved mortality and morbidity rates
with blood pressure reduction, although this study was ret-
rospective and did not consider confounding variables
such as hematoma volume and initial GCS score. In a ret-
rospective review of 188 patients with ICH, Broderick and
Brott14 found no difference in outcome in patients who
received antihypertension agents compared with those
who did not.
The intracranial hemodynamic effects of most antihy-
pertension agents are poorly understood. Data acquired in
clinical studies of patients with SICH are limited, but most
vasodilators have the potential to raise ICP by increasing
intracerebral blood volume. Short-acting, mixed - and
-
blockers such as labetalol should be used as first-line
agents. Nitroprusside, the most commonly used agent for
severe hypertension, is a vasodilator with the potential to
raise CBF and ICP, although this effect has not been dem-
onstrated in a clinical study. Arterial vasodilators, such as
hydralazine or angiotensin-converting enzyme inhibitors,
as well as calcium channel blockers may have less of a
tendency to raise ICP and maintain CBF.103 Venodilators,
diazoxide, and ganglionic blockers should not be used in
patients with SICH and elevated ICP.
In summary, there is no convincing evidence that low-
ering blood pressure in the acute period after SICH alters
the course or prognosis. Modest reductions in SBP may be
well tolerated, especially in patients with systemic hyper-
tension complications or those at greater risk for hemor-
rhage expansion (such as, in coagulopathic patients). To
balance the two theoretical risks of hemorrhage extension
compared with worsening ischemia, the American Heart
Association recommended maintaining MABP below 130
mm Hg in patients with preexisting hypertension and CPP
above 70 mm Hg.13 A large randomized controlled trial
10
M. E. Fewel, B. G. Thompson, Jr., and J. T. Hoff
should be conducted to determine the effect of lowering
blood pressure in acute SICH.
Management of Intracranial Hypertension. Elevated
ICP from SICH can result in herniation syn-dromes and
death. Intuitively, monitoring and treatment of elevated
ICP should reduce secondary injury. Improved outcomes,
however, have not been clearly linked with monitoring
and treatment of elevated ICP in SICH. Nevertheless,
most authors recommend treatment when ICP exceeds 20
mm Hg with a goal CPP greater than 60 to 70 mm Hg. In
general, ICP monitoring devices should be considered in
all patients who present with a GCS score less 9 or in
those patients in whom deteriorating clinical condition is
believed to be due to elevated ICP.
Elevated ICP can have deleterious effects on CPP and,
therefore, global CBF. Reduction in CPP and perihema-
toma ischemia may occur locally even in the absence of
globally increased ICP. In experimental studies investi-
gators have shown that a reduction in CPP leads to com-
pensatory vasodilation, which increases intracranial blood
volume, in turn, increasing ICP and further decreasing
CPP. The physiological correction of this cycle is the sys-
temic hypertension response that increases CPP, leading to
arteriolar vasoconstriction, decreased intracranial blood
volume, and lowered ICP. Rosner and Becker108 used the
term “ischemic response” to describe this phenomenon
that Cushing proposed in 1902.29 The impact of ICP treat-
ment and hematoma evacuation on CBF has not been well
studied, but some evidence exists that surgical evacuation
and elevated ICP treatment can improve CBF.103,127
Common strategies for managing elevated ICP include
elevation of the head of the bed, hyperventilation thera-
py, osmotic therapy, sedation, and cerebrospinal fluid
drainage. Mannitol is often given when surgery is not con-
sidered appropriate. Mechanical ventilation parameters
should be adjusted to a PCO2 goal of 30 to 35 mm Hg.
Intraventricular blood is associated with a high mortality
rate.123 Ventriculostomy, rather than fiberoptic ICP de-
vices, should be used in patients with hydrocephalus or
intraventricular hemorrhage (Fig. 8). Although ventricu-
lostomy reduces ICP, frequent clots in the catheter and
infections often diminish the beneficial effect on hydro-
cephalus and neurological status.2 We prefer to use pro-
phylactic antibiotic medications in patients with ICP mon-
itoring devices. Sedation with benzodiazepines, narcotic
analgesics, or propofol can help control elevated ICP.
Barbiturate coma or hemicraniectomy may be necessary if
all other measures fail.
Unhelpful Therapies. Based on the results of two ran-
domized trials that failed to demonstrate the efficacy of
corticosteroid therapy in ICH, its use is not indicated.99,121
The authors of two other small randomized medical trials
evaluating hemodilution and therapy with intravenous
glycerol also did not find any significant benefit.55,135
Surgical Treatment
Despite a nihilistic attitude among clinicians regarding
SICH, most neurosurgeons believe evacuation of certain
intracerebral hematomas can reduce the mortality rate and
improve outcomes. The fact that approximately 7000 op-
erations are performed annually in the US for SICH sup-
Neurosurg. Focus / Volume 15 / October, 2003
Spontaneous intracerebral hemorrhage
Fig. 8. Axial CT scan acquired in an 82-year-old man with a
history of hypertension who presented with headache, vomiting,
and progressive deterioration in consciousness. A small left cere-
bellar SICH extending into the fourth ventricle is demonstrated.
Obstructive hydrocephalus was successfully treated by ventricu-
lostomy. The patient did not require evacuation of the hematoma
and made a good recovery.
ports this belief.12 The efficacy of surgery for SICH, how-
ever, remains unproven in the year 2003.
Most of the published reports regarding surgical re-
moval of these hemorrhages, many of which are in the
Japanese literature, are nonrandomized case series for
which definitive recommendations or guidelines cannot
be made. There is considerable international variation in
the indications and rate of surgery for SICH. In part be-
cause of its high incidence in Japan, there has been an
increase in the number of clinical and experimental stud-
ies being reported over the past two decades.
The Cochrane Database of Systematic Reviews pub-
lished a metaanalysis of surgery for primary supratentori-
al ICH in 2001.101 The authors analyzed the randomized
and quasirandomized trials of medical management com-
bined with intracranial surgery compared with medical
management alone. Only four surgical trials comprising
260 patients met their criteria for inclusion.6,9,61,84 No trial
had blinded outcome assessment and each trial used dif-
ferent scales for assessing functional status. The authors
of the review concluded that the available evidence to date
was insufficient to make a recommendation concerning
the safety and efficacy of surgical treatment.101
In a review of supratentorial SICH by Hankey and
Hon46 published in 1997, the authors found a nonsignifi-
cant increase in the likelihood of death and dependency at
6 months in surgically treated patients. In a metaanalysis
conducted by Fernandes, et al.,34 the authors suggested a
benefit to surgery, with a reduction in the chances of death
and dependency after surgery by a factor of 0.63. This
analysis excluded a pre–CT scanning era trial and a Chi-
nese trial because of methodological issues.
Patients with relatively normal consciousness (GCS
Scores 13–15) rarely require surgery, whereas deeply co-
matose patients (GCS Scores 3–5) rarely benefit from
surgery. Surgery is therefore usually considered to have
the most potential benefit for the group of patients with
Neurosurg. Focus / Volume 15 / October, 2003
GCS scores between 6 and 12 or in patients with deterio-
rating status.35 The optimal surgical technique for he-
matoma evacuation is not agreed upon, although cranioto-
my remains the most common.35 Traditional stereotaxy or
frameless navigational systems, as well as intraoperative
ultrasonographic guidance, allow more precise clot local-
ization and minimization of injury to normal brain. Com-
pared with craniotomy, minimally invasive techniques
such as stereotactic or endoscopic clot evacuation may
offer the potential for a reduced incidence of surgery-relat-
ed complications and improved efficacy, but this has yet
to be proven.
There is a neurosurgical bias toward more aggressive
surgery for nondominant hemispheric hemorrhages, al-
though the authors of outcome studies have indicated that
despite language disability associated with dominant
hemispheric lesions, functional outcome is not necessari-
ly worse. Standard craniotomy for extirpation of primary
brainstem or thalamic hemorrhages has been all but aban-
doned because of poor outcomes.17,46 Apparently success-
ful cases of stereotactic aspiration of pontine hematomas
have been reported, but the effect on prognosis remains
unproven.
Although evidence favoring surgical intervention is
lacking, there is good theoretical rationale for early sur-
gery. Early hematoma evacuation may decrease the toxic
effects of blood and plasma products, diminish surround-
ing edema and ischemia, and prevent hematoma expan-
sion. Liquefaction of the clot by instillation of tPA and
subsequent aspiration at 3 hours has been shown to reduce
mass effect and perihematomal edema markedly at 24
hours in a pig model.127 Clot evacuation might be com-
bined with pharmacological therapy targeting the inflam-
matory response shown to develop around the hematoma
and leading to delayed cellular death in experimental ani-
mal models.49 Early evacuation might also be combined
with hemostatic therapy such as recombinant activated
factor VII to help prevent the risk of rebleeding (Fig. 5).79,89
Randomized Trials of Surgical and Medical Manage-
ment. To date, there are nine prospective randomized con-
trolled studies in which surgical and medical management
of SICH has been compared.6,9,27,61,84,89,90,120,138 Overall,
these studies fail to demonstrate a superiority of either
medical or surgical management. The results are also lim-
ited by considerable methodological differences among
the trials. A brief review of these trials follows.
McKissock, et al.,84 reported the first randomized study
of surgery for SICH in 1961. They randomized 180 pa-
tients in a series of 303 potential cases. Ninety-one pa-
tients were “randomly allocated” to best medical treat-
ment and 89 to surgery and best medical treatment, but the
method of randomization was not described. Because this
study was conducted well before the advent of CT scan-
ning, diagnosis was established using angiography or ven-
triculography. Although the proportion of surgically treat-
ed patients who were dead or totally disabled 6 months
after discharge was higher than the medically treated
group (80 and 66%, respectively), the results may not be
relevant to current practice. Almost no patients underwent
surgery before 24 hours. In addition, neurosurgical, anes-
thesia, and critical care techniques have improved sub-
stantially since that time.
11

Juvela and colleagues61 reported a prospective study of
52 patients with supratentorial SICH randomized to sur-
gery or best medical therapy for SICH. The median time
from onset to craniotomy for evacuation was 14.5 hours.
There were no statistically significant differences in mor-
tality or morbidity rates between the two treatment
groups, and these authors concluded that supratentorial
SICH should be treated conservatively. Although the mor-
tality rate of semicomatose or stuporous patients (GCS
Scores 7–10) was statistically significantly lower in the
surgical group, all surviving patients in this subgroup were
severely disabled.
Auer, et al.,6 randomly assigned 100 patients with su-
pratentorial SICH to undergo stereotactic endoscopic
evacuation of the blood clot within 48 hours after admis-
sion or to undergo medical therapy alone. At 6 months, the
outcome was better in the surgically treated group with no
or minimal deficits in 40% in this group compared with
25% in the medically treated group. The surgically treated
group had statistically nonsignificant lower odds of death
or dependency (odds ratio 0.45; 95% CI 0.19–1.04).
Batjer, et al.,9 randomized 17 hypertensive patients with
putaminal SICH of at least 3 cm to surgery or medical
therapy. The results showed a trend toward a lower rate of
death or level of dependence in the surgically treated pa-
tients (odds ratio 0.86; 95% CI 0.09–8.1), but the 95% CI
included one.
Chen, et al.,27 randomized 127 patients with hyperten-
sive supratentorial and cerebellar hemorrhages to surgery
or medical management. These authors included cerebel-
lar hemorrhages, and although there was no statistically
significant difference in mortality rates between the two
groups, there was a trend toward an increased risk of death
or dependency in those treated with surgery.
Zuccarello, et al.,138 reported a small randomized feasi-
bility study in which they evaluated the use of surgical
evacuation, craniotomy, or stereotactic aspiration com-
pared with medical treatment alone. Twenty patients were
randomized; the median time from symptom onset to
surgery was 8.5 hours. The likelihood of a good outcome
(defined as 3-month GOS score  3) was 56% in the sur-
gical group and 36% in the medical treatment group, al-
though this result did not reach statistical significance.
The authors concluded that early clot removal in supra-
tentorial SICH was feasible and warranted further investi-
gation in randomized clinical trials.
Morganstern, et al.,90 prospectively randomized 34
patients within 12 hours of onset. They reported a mod-
est benefit in 6-month mortality rates in patients treated
with craniotomy (19%) compared with medical therapy
(24%).90 Outcome was measured by an investigator un-
blinded to treatment allocation. In addition, only one pa-
tient underwent surgery within 4 hours and surgically
treated patients harbored a significantly larger-volume
hemorrhage, had lower GCS scores, and more frequently
presented with deep-seated ICH.
In a later publication Morganstern, et al.,89 also report-
ed preliminary results of a prospective single-center pilot
study of surgery performed within 4 hours of SICH; their
goal was to test the feasibility and safety of ultra-early clot
evacuation compared with results of their previous 12-
hour trial. This study was stopped for safety reasons after
12
M. E. Fewel, B. G. Thompson, Jr., and J. T. Hoff
interim analysis of 11 patients in the 4-hour surgery arm
showed a 6-month mortality rate of 36% compared with
only 18% in the 12-hour surgical group. Because of the
small number of patients in these pilot studies, the differ-
ences in mortality rates were not statistically significant.
Rebleeding was more common in patients undergoing sur-
gery within 4 hours (40%) than 12 hours (12%). Instead of
preventing early hematoma expansion by coagulating the
bleeding vessel, ultra-early surgery proved more difficult
to achieve hemostasis. Morganstern and colleagues rec-
ommended using craniotomy within the first few hours
after SICH with extreme caution because of the risk of
rebleeding. This study did demonstrate the feasibility of
ultra-early triage and surgery for patients with SICH much
like protocols of thrombolytic therapy for acute ischemic
stroke. In addition, patients with the smallest postopera-
tive hemorrhage volumes experienced the best outcomes,
arguing that removal of hemorrhage may ultimately be
beneficial.
Teernstra and colleagues120 recently reported results
from a multicenter randomized controlled trial of stereo-
tactic drainage involving plasminogen activator compared
with medical therapy. They randomized 36 patients to
surgery and 35 to medical therapy, and found no statisti-
cally significant difference in mortality and morbidity
rates at 6 months. A stereotactically placed catheter was
used to install urokinase to liquefy and drain the hema-
toma in 6-hour intervals over 48 hours and resulted in a
modest 18 ml reduction in hematoma volume over 7 days.
The International Surgical Trial in IntraCerebral Hem-
orrhage, the largest multicenter randomized controlled
trial of surgery for SICH, closed recruitment on February
25, 2003.85 Investigators randomized 1033 patients within
72 hours of supratentorial SICH to a surgery or medical
therapy group. Although this important trial should help
better define the role of surgery in SICH, it is likely to
generate additional questions about patient selection, tim-
ing, and surgical technique.
Nonrandomized Surgical Studies. Although nonran-
domized series of surgically treated patients with SICH
are limited by methodological flaws and lack definitive
conclusions, they do offer valuable information. In a non-
randomized report, Kaneko and colleagues64 reported
good results after performing craniotomy for the evacua-
tion of 100 putaminal hemorrhages within 7 hours of
onset and 60 hemorrhages within 3 hours of onset. They
reported only a 7% mortality rate and complete recovery
or minor disability in 50%.
In a large nonrandomized study, Kanaya and Kuroda63
evaluated 7010 patients with putaminal hemorrhage. The
mortality rate in alert and confused patients was signifi-
cantly lower in medically treated groups; however, in
patients who were stuporous or worse, the rate was signif-
icantly lower in the surgically treated group. They recom-
mended surgical treatment if the hematoma volume was
larger than 30 ml and the level of consciousness was som-
nolent to semicomatose.
In 1978 Backlund and von Holst7 proposed a new sur-
gical technique for subtotal hematoma removal using CT
scanning–guided aspiration with a special cannula. The
authors of several case series have reported encouraging
results when using stereotactic aspiration with fibrinoly-
Neurosurg. Focus / Volume 15 / October, 2003
Spontaneous intracerebral hemorrhage
sis.77,87,88,93,110 Doi, et al.,32 from Japan were the first to use
stereotactic aspiration with a hematoma lysis method.
Successful use of both urokinase and recombinant tPA has
been reported. The risk of recurrent hemorrhage in cases
involving the use of thrombolytic agents has not been
established but has been estimated to occur in up to 15%.93
Infectious complications of catheter placement and fibri-
nolysis vary between 0 and 5%.87,110
Rohde and colleagues107 reported the first use of frame-
less stereotaxy in a series of 27 patients with SICH who
underwent catheter placement and fibrinolysis. Frameless
stereotaxy has the potential advantage of allowing optimal
trajectory and entry point selection, which is sometimes
technically limited in frame-based systems. Thrombolytic
agents have also been successfully used for hemorrhage in
the ventricular system.36
Cerebellar SICH. Unlike the controversy regarding
surgery for supratentorial SICH, there seems to be a gen-
eral consensus regarding the role of surgery in patients
with infratentorial hematomas. This agreement exists de-
spite that fact that there are no randomized controlled tri-
als evaluating surgical methods in posterior fossa SICH.
Of the eight retrospective case studies involving surgery
for infratentorial SICH reviewed by Hankey and Hon,46
two were studies reporting a beneficial effect of surgery in
certain patients, one was a study suggesting that surgery
had an overall beneficial effect; and five were series re-
porting either inconclusive results or no beneficial effect
of surgery. Authors of several series have reported good
outcomes associated with surgical evacuation for patients
with cerebellar hemorrhages greater than 3 cm, or with
brainstem compression and hydrocephalus.13,125 Medical
management alone in these patients is often unsatisfacto-
ry. Once patients have become comatose, the prognosis is
poor, but not hopeless if urgent surgery can be performed
(Fig. 6).132
Kirollos, et al.,70 prospectively applied a management
protocol in 50 consecutive patients with spontaneous cere-
bellar hemorrhage. They found that evacuation may not be
required for hematomas greater than 3 cm if the fourth
ventricle was not completely obliterated, but that con-
scious patients with fourth ventricular obliteration should
undergo emergency clot evacuation. Most neurosurgeons,
however, recommend surgery for patients with cerebellar
hemorrhages greater than 3 cm, with compression of the
brainstem, or with obstructive hydrocephalus.
Summary of Surgical Treatment Recommendations.
Patients with small hemorrhages or minimal neurological
deficit generally do well by undergoing medical treatment
alone. Elderly patients in whom the GCS score is less than
5 and those with brainstem hemorrhages also do not typ-
ically benefit from surgery. Patients with cerebellar hem-
orrhages greater than 3 cm in whom are symptoms or
neurological deterioration have occurred, or in whom
brainstem compression and hydrocephalus are present,
should undergo evacuation of the clot. Evacuation should
be considered in patients with moderate- or large-sized
lobar hemorrhages, those with large-sized basal ganglia
hemorrhages, and those exhibiting progressive neurologi-
cal deterioration. Ultra-early removal of the hematoma by
localized minimally invasive surgical procedures is
promising but unproven.
Neurosurg. Focus / Volume 15 / October, 2003
CONCLUSIONS
Despite the grim prognosis for patients suffering SICH,
there are reasons for optimism. There have been improve-
ments in medical treatment of hypertension and neurocrit-
ical care. Secondary brain injury, edema formation, and
hematoma enlargement represent potential therapeutic tar-
gets. The roles of edema, ischemia, mass effect, direct cel-
lular toxicity, inflammation, and apoptosis are being eval-
uated in experimental studies of SICH. Neurosurgical
advances in minimally invasive techniques, endoscopy,
and frameless stereotactic systems should improve results
of surgery. It is hoped that the forthcoming results of the
International Surgical Trial in IntraCerebral Hemorrhage
will help formulate evidence-based recommendations re-
garding the role of surgery in SICH, although this trial is
unlikely to end the controversy. Future randomized trials
will still be needed to assess the effects of patient age,
level of consciousness, comorbidities, hematoma volume,
hematoma location, timing of surgery, and different surgi-
cal procedures.
References
1. Abilleira S, Montaner J, Molina CA, et al: Matrix metallopro-
teinase–9 concentration after spontaneous intracerebral hemor-
rhage. J Neurosurg 99:65–70, 2003
2. Adams RE, Diringer MN: Response to external ventricular
drainage in spontaneous intracerebral hemorrhage with hydro-
cephalus. Neurology 50:519–523, 1998
3. Adams RE, Powers WJ: Management of hypertension in acute
intracerebral hemorrhage. Crit Care Clin 13:131–161, 1997
4. Al-Jarallah A, Al-Rifai MT, Riela AR, et al: Nontraumatic brain
hemorrhage in children: etiology and presentation. J Child
Neurol 15:284–289, 2000
5. Astrup J, Siesjo BK, Symon L: Thresholds in cerebral ische-
mia—the ischemic penumbra. Stroke 12:723–725, 1981
6. Auer LM, Deinsberger W, Niederkorn K, et al: Endoscopic sur-
gery versus medical treatment for spontaneous intracerebral he-
matoma: a randomized study. J Neurosurg 70:530–535, 1989
7. Backlund EO, von Holst H: Controlled subtotal evacuation of
intracerebral haematomas by stereotactic technique. Surg Neu-
rol 9:99–101, 1978
8. Barinagarrementeria F, Mendez A, Vega F: [Cerebral hemor-
rhage associated with the use of phenylpropanolamine.] Neu-
rologia 5:292–295, 1990 (Spn)
9. Batjer HH, Reisch JS, Allen BC, et al: Failure of surgery to im-
prove outcome in hypertensive putaminal hemorrhage. A pro-
spective randomized trial. Arch Neurol 47:1103–1106, 1990
10. Bladin CF, Alexandrov AV, Bellavance A, et al: Seizures after
stroke: a prospective multicenter study. Arch Neurol 57:
1617–1622, 2000
11. Broderick J, Brott T, Kothari R, et al: Very early edema growth
not significantly associated with neurologic deterioration in
patients with intracerebral hemorrhage. Stroke 26:184, 1995
(Abstract)
12. Broderick JP: Intracerebral hemorrhage, in Gorelick PB, Alter
M (eds): Handbook of Neuroepidemiology. New York: Mar-
cel Dekker, 1994, pp 141–167
13. Broderick JP, Adams HP Jr, Barsan W, et al: Guidelines for the
management of spontaneous intracerebral hemorrhage: a state-
ment for healthcare professionals from a special writing group
of the Stroke Council, American Heart Association. Stroke 30:
905–915, 1999
14. Broderick JP, Brott T: Predictions of outcomes after intracere-
bral hemorrhage. Stroke 24:1761, 1993
15. Broderick JP, Brott T, Tomsick T, et al: Intracerebral hemor-
13

rhage more than twice as common as subarachnoid hemor-
rhage. J Neurosurg 78:188–191, 1993
16. Broderick JP, Brott T, Tomsick T, et al: The risk of subarach-
noid and intracerebral hemorrhages in blacks as compared with
whites. N Engl J Med 326:733–736, 1992
17. Broderick JP, Brott T, Zuccarello M: Management of intracere-
bral hemorrhage, in Batjer HH (ed): Cerebrovascular Disease.
Philadelphia: Lippincott-Raven, 1997, pp 611–627
18. Broderick JP, Brott TG, Duldner JE, et al: Volume of intracere-
bral hemorrhage. A powerful and easy-to-use predictor of 30-
day mortality. Stroke 24:987–993, 1993
19. Brott T, Broderick J, Kothari R, et al: Early hemorrhage growth
in patients with intracerebral hemorrhage. Stroke 28:1–5, 1997
20. Brott T, Thalinger K, Hertzberg V: Hypertension as a risk
factor for spontaneous intracerebral hemorrhage. Stroke 17:
1078–1083, 1986
21. Cantu C, Arauz A, Murillo-Bonilla LM, et al: Stroke associated
with sympathomimetics contained in over-the-counter cough
and cold drugs. Stroke 34:1667–1672, 2003
22. Caplan LR: General symptoms and signs, in Kase CS, Caplan
LR (eds): Intracerebral Hemorrhage. Boston: Butterworth-
Heinemann, pp 31–43, 1994
23. Cervoni L, Artico M, Salvati M, et al: Epileptic seizures in in-
tracerebral hemorrhage: a clinical and prognostic study of 55
cases. Neurosurg Rev 17:185–188, 1994
24. Challa VR, Moody DM, Bell MA: The Charcot-Bouchard an-
eurysm controversy: impact of a new histologic technique. J
Neuropathol Exp Neurol 51:264–271, 1992
25. Chaney RK, Taber KH, Orrison WW Jr, et al: Magnetic reso-
nance imaging of intracerebral hemorrhage at different field
strengths. Neuroimaging Clin N Am 2:25–51, 1992
26. Charcot JM, Bouchard C: 642 Nouvelles recerches sur la path-
ogenie de l’hemorrhagie cerebrale. Arch Physiol Norm Path-
ol:110–127, 1868
27. Chen X, Yang H, Czherig Z: A prospective randomized trial of
surgical and conservative treatment of hypertensive intracranial
haemorrhage. Acta Acad Med Shanghai 19:237–240, 1992
28. Cohen WA, Hayman LA: Computed tomography of intracra-
nial hemorrhage. Neuroimaging Clin N Am 2:75–87, 1992
29. Cushing H: Some experimental and clinical observations con-
cerning states of increased intracranial tension. Am J Med Sci
124:375–400, 1902
30. Dandapani BK, Suzuki S, Kelley RE, et al: Relation between
blood pressure and outcome in intracerebral hemorrhage.
Stroke 26:21–24, 1995
31. Dennis MS, Burn JP, Sandercock PA, et al: Long-term survival
after first-ever stroke: the Oxfordshire Community Stroke Pro-
ject. Stroke 24:796–800, 1993
32. Doi E, Moriwaki H, Komai N, et al: [Stereotactic evacuation
of intracerebral hematomas.] Neurol Med Chir 22:461–467,
1982 (Jpn)
33. Felber S, Auer A, Wolf C, et al: [MRI characteristics of sponta-
neous intracerebral hemorrhage.] Radiologe 39:838–846, 1999
(Ger)
34. Fernandes HM, Gregson B, Siddique MS, et al: Surgery in in-
tracerebral hemorrhage. The uncertainty continues. Stroke 31:
2511–2516, 2000
35. Fernandes HM, Mendelow AD: Spontaneous intracerebral
haemorrhage: a surgical dilemma. Br J Neurosurg 13:
389–394, 1999
36. Findlay JM, Grace MG, Weir BK: Treatment of intraventricu-
lar hemorrhage with tissue plasminogen activator. Neurosur-
gery 32:941–947, 1993
37. Fisher CM: Pathological observations in hypertensive cerebral
hemorrhage. J Neuropathol Exp Neurol 30:536–550, 1971
38. Fogelholm R, Eskola K, Kiminkinen T, et al: Anticoagulant
treatment as a risk factor for primary intracerebral haemor-
rhage. J Neurol Neurosurg Psychiatry 55:1121–1124, 1992
39. Fujii Y, Takeuchi S, Sasaki O, et al: Multivariate analysis of
14
M. E. Fewel, B. G. Thompson, Jr., and J. T. Hoff
predictors of hematoma enlargement in spontaneous intracere-
bral hemorrhage. Stroke 29:1160–1166, 1998
40. Gallerani M, Trappella G, Manfredini R, et al: Acute intracere-
bral haemorrhage: circadian and circannual patterns of onset.
Acta Neurol Scand 89:280–286, 1994
41. Guridi J, Gallego J, Mornzon F, et al: Intracerebral hemorrhage
caused by transmural dissection of the anterior cerebral artery.
Stroke 24:1400–1402, 1993
42. Garner TB, Curling OD Jr, Kelly DL Jr, et al: The natural his-
tory of intracranial venous angiomas. J Neurosurg 75:
715–722, 1991
43. Griffiths PD, Beveridge CJ, Gholkar A: Angiography in non-
traumatic brain haematoma. Acta Radiol 38:797–802, 1997
44. Hacke W, Brott T, Caplan L, et al: Thrombolysis in acute ische-
mic stroke: controlled trials and clinical experience. Neurology
53 (7 Suppl 4):S3–S14, 1999
45. Halpin SF, Britton JA, Byrne JV, et al: Prospective evaluation
of cerebral angiography and computed tomography in cerebral
haematoma. J Neurol Neurosurg Psychiatry 57:1180–1186,
1994
46. Hankey GJ, Hon C: Surgery for primary intracerebral hemor-
rhage: is it safe and effective? A systematic review of case
series and randomized trials. Stroke 28:2126–2132, 1997
47. Hart RG, Boop BS, Anderson DC: Oral anticoagulants and
intracranial hemorrhage. Facts and hypotheses. Stroke 26:
1471–1477, 1995
48. He J, Whelton PK, Vu B, et al: Aspirin and risk of hemorrhagic
stroke: a meta-analysis of randomized controlled trials. JAMA
280:1930–1935, 1998
49. Hickenbottom SL, Grotta JC, Strong R, et al: Nuclear factor-
kappaB and cell death after experimental intracerebral hemor-
rhage in rats. Stroke 30:2472–2478, 1999
50. Hinton DR, Dolan E, Sima AA: The value of histopathological
examination of surgically removed blood clot in determining
the etiology of spontaneous intracerebral hemorrhage. Stroke
15:517–520, 1984
51. Hirano T, Read SJ, Abbott DF, et al: No evidence of hypoxic
tissue on 18F-fluoromisonidazole PET after intracerebral hem-
orrhage. Neurology 53:2179–2182, 1999
52. Hua Y, Xi G, Keep RF, et al: Complement activation in the
brain after experimental intracerebral hemorrhage. J Neuro-
surg 92:1016–1022, 2000
53. Huang FP, Xi G, Keep RF, et al: Brain edema after experimen-
tal intracerebral hemorrhage: role of hemoglobin degradation
products. J Neurosurg 96:287–293, 2002
54. Iso H, Jacobs DR Jr, Wentworth D, et al: Serum cholesterol lev-
els and six-year mortality from stroke in 350, 977 men screened
for the multiple risk factor intervention trial. N Engl J Med
320:904–910, 1989
55. Italian Acute Stroke Study Group: Haemodilution in acute
stroke: results of the Italian haemodilution trial. Lancet 1:
318–321, 1988
56. Iwama T, Ohkuma A, Miwa Y, et al: Brain tumors manifesting
as intracranial hemorrhage. Neurol Med Chir 32:130–135,
1992
57. Izumihara A, Ishihara T, Iwamoto N, et al: Postoperative out-
come of 37 patients with lobar intracerebral hemorrhage relat-
ed to cerebral amyloid angiopathy. Stroke 30:29–33, 1999
58. Johnsen SP, Pedersen L, Friis S, et al: Nonaspirin nonsteroidal
anti-inflammatory drugs and risk of hospitalization for intrac-
erebral hemorrhage: a population-based case-control study.
Stroke 34:387–391, 2003
59. Juvela S: Prevalence of risk factors in spontaneous intracerebral
hemorrhage and aneurysmal subarachnoid hemorrhage. Arch
Neurol 53:734–740, 1996
60. Juvela S: Risk factors for impaired outcome after spontaneous
intracerebral hemorrhage. Arch Neurol 52:1193–1200, 1995
61. Juvela S, Heiskanen O, Poranen A, et al: The treatment of spon-
taneous intracerebral hemorrhage. A prospective randomized
Neurosurg. Focus / Volume 15 / October, 2003
Spontaneous intracerebral hemorrhage
trial of surgical and conservative treatment. J Neurosurg 70:
755–758, 1989
62. Juvela S, Hillbom M, Palomaki H: Risk factors for spontaneous
intracerebral hemorrhage. Stroke 26:1558–1564, 1995
63. Kanaya H, Kuroda K: Development in neurosurgical approach-
es to hypertensive intracerebral hemorrhage in Japan, in Kauf-
man HH (ed): Intracerebral Hematomas. New York: Raven
Press, 1992, pp 197–210
64. Kaneko M, Tanaka K, Shimada T, et al: Long-term evaluation
of ultra-early operation for hypertensive intracerebral hemor-
rhage in 100 cases. J Neurosurg 58:838–842, 1983
65. Kaneko T, Sawada T, Niimi T, et al: Lower limit of blood pres-
sure treatment of acute hypertensive intracranial hemorrhage. J
Cereb Blood Flow Metab 3 (Suppl 1):S51–S52, 1983
66. Karibe H, Niizuma H, Ohyama H, et al: Hepatitis C virus
(HCV) infection as a risk factor for spontaneous intracerebral
hemorrhage: hospital based case-control study. J Clin Neuro-
sci 8:423–425, 2001
67. Kazui S, Minematsu K, Yamamoto H, et al: Predisposing fac-
tors to enlargement of spontaneous intracerebral hematoma.
Stroke 28:2370–2375, 1997
68. Kazui S, Naritomi H, Yamamoto H, et al: Enlargement of spon-
taneous intracerebral hemorrhage. Incidence and time course.
Stroke 27:1783–1787, 1996
69. Kirkpatrick JB, Hayman LA: Pathophysiology of intracranial
hemorrhage. Neuroimaging Clin N Am 2:11–23, 1992
70. Kirollos RW, Tyagi AK, Ross SA, et al: Management of spon-
taneous cerebellar hematomas: a prospective treatment proto-
col. Neurosurgery 49:1378–1387, 2001
71. Kothari RU, Brott T, Broderick JP, et al: The ABCs of measur-
ing intracerebral hemorrhage volumes. Stroke 27:1304–1305,
1996
72. Kuwata N, Kuroda K, Funayama M, et al: Dysautoregulation in
patients with hypertensive intracerebral hemorrhage. A SPECT
study. Neurosurg Rev 18:237–245, 1995
73. Lee KR, Betz AL, Keep RF, et al: Intracerebral infusion of
thrombin as a cause of brain edema. J Neurosurg 83:
1045–1050, 1995
74. Lee KR, Kawai N, Kim S, et al: Mechanisms of edema forma-
tion after intracerebral hemorrhage: effects of thrombin on cere-
bral blood flow, blood-brain barrier permeability, and cell sur-
vival in a rat model. J Neurosurg 86:272–278, 1997
75. Linfante I, Llinas RH, Caplan LR, et al: MRI features of intrac-
erebral hemorrhage within 2 hours from symptom onset.
Stroke 30:2263–2267, 1999
76. Malik GM, Morgan JK, Boulos RS, et al: Venous angiomas: an
underestimated cause of intracranial hemorrhage. Surg Neurol
30:350–358, 1988
77. Matsumoto K, Hondo H: CT-guided stereotaxic evacuation
of hypertensive intracerebral hematomas. J Neurosurg 61:
440–448, 1984
78. Matthews MK Jr: Association of Ginkgo biloba with intracere-
bral hemorrhage. Neurology 50:1933–1934, 1998
79. Mayer SA: Ultra-early hemostatic therapy for intracerebral hem-
orrhage. Stroke 34:224–229, 2003
80. Mayer SA, Lignelli A, Fink ME, et al: Perilesional blood flow
and edema formation in acute intracerebral hemorrhage. A
SPECT study. Stroke 29:1791–1798, 1998
81. Mayer SA, Sacco RL, Shi T, et al: Neurologic deterioration in
noncomatose patients with supratentorial intracerebral hemor-
rhage. Neurology 44:1379–1384, 1994
82. Maymon R, Fejgin M: Intracranial hemorrhage during pregnan-
cy and puerperium. Obstet Gynecol Surv 45:157–159, 1990
83. McEvoy AW, Kitchen ND, Thomas DG: Intracerebral haemor-
rhage caused by drug abuse. Lancet 351:1029, 1998
84. McKissock W, Richardson A, Taylor J: Primary intracerebral
hemorrhage. A controlled trial of surgical and conservative
treatment in 180 unselected cases. Lancet 2:221–226, 1961
Neurosurg. Focus / Volume 15 / October, 2003
85. Mendelow AD: Surgical trial in intracerebral haemorrhage
(S.T.I.C.H.). Acta Neurochir Suppl 76:521–522, 2000
86. Meyer JS, Bauer RB: Medical treatment of spontaneous in-
tracranial hemorrhage by the use of hypotensive drugs. Neu-
rology 12:36–47, 1962
87. Mohadjer M, Braus DF, Myers A, et al: CT-stereotactic fibri-
nolysis of spontaneous intracerebral hematomas. Neurosurg
Rev 15:105–110, 1992
88. Montes J, Wong JH, Fayad PB, et al: Stereotactic computed
tomographic-guided aspiration and thrombolysis of intracere-
bral hematoma: protocol and preliminary experience. Stroke
31:834–840, 2000
89. Morgenstern LB, Demchuk AM, Kim DH, et al: Rebleeding
leads to poor outcome in ultra-early craniotomy for intracere-
bral hemorrhage. Neurology 56:1294–1299, 2001
90. Morgenstern LB, Frankowski RF, Shedden P, et al: Surgical
treatment for intracerebral hemorrhage (STICH): a single-cen-
ter, randomized clinical trial. Neurology 51:1359–1363, 1998
91. Naff NJ, Wemmer J, Hoenig-Rigamonti K, et al: A longitudinal
study of patients with venous malformations: documentation
of a negligible hemorrhage risk and benign natural history.
Neurology 50:1709–1714, 1998
92. Nath FP, Jenkins A, Mendelow AD, et al: Early hemodynamic
changes in experimental intracerebral hemorrhage. J Neuro-
surg 65:697–703, 1986
93. Niizuma H, Shimizu Y, Yonemitsu T, et al: Results of stereo-
tactic aspiration in 175 cases of putaminal hemorrhage. Neu-
rosurgery 24:814–819, 1989
94. Niizuma H, Suzuki J, Yonemitsu T, et al: Spontaneous intrac-
erebral hemorrhage and liver dysfunction. Stroke
19:852–856, 1988
95. O’Donnell HC, Rosand J, Knudsen KA, et al: Apolipoprotein E
genotype and the risk of recurrent lobar intracerebral hemor-
rhage. N Engl J Med 342:240–245, 2000
96. Park P, Fewel ME, Thompson BG, et al: Recombinant activat-
ed factor VII for the rapid correction of coagulopathy in non-
hemophilic neurosurgical patients. Neurosurgery 53:34–39,
2003
97. Passero S, Rocchi R, Rossi S, et al: Seizures after spontane-
ous supratentorial intracerebral hemorrhage. Epilepsia 43:
1175–1180, 2002
98. Patel MR, Edelman RR, Warach S: Detection of hyperacute pri-
mary intraparenchymal hemorrhage by magnetic resonance
imaging. Stroke 27:2321–2324, 1996
99. Poungvarin N, Bhoopat W, Viriyavejakul A, et al: Effects of
dexamethasone in primary supratentorial intracerebral hemor-
rhage. N Engl J Med 316:1229–1233, 1987
100. Powers WJ, Adams RE, Yundt KD, et al: Acute pharmacolog-
ical hypotension after intracerebral hemorrhage does not
change cerebral blood flow. Stroke 30:242, 1999 (Abstract)
101. Prasad K, Shrivastava A: Surgery for primary supratentorial
intracerebral haemorrhage. Cochrane Database Syst Rev 2:
2000
102. Qureshi AI, Bliwise DL, Bliwise NG, et al: Rate of 24-hour
blood pressure decline and mortality after spontaneous intrac-
erebral hemorrhage: a retrospective analysis with a random
effects regression model. Crit Care Med 27:480–485, 1999
103. Qureshi AI, Hanel RA, Kirmani JF, et al: Cerebral blood flow
changes associated with intracerebral hemorrhage. Neurosurg
Clin N Am 13:355–370, 2002
104. Qureshi AI, Suri MF, Ostrow PT, et al: Apoptosis as a form
of cell death in intracerebral hemorrhage. Neurosurgery 52:
1041–1048, 2003
105. Qureshi AI, Wilson DA, Hanley DF, et al: Pharmacological
reduction of mean arterial pressure does not adversely effect
regional cerebral blood flow and intracranial pressure in
experimental intracerebral hemorrhage. Crit Care Med
27:965–971, 1999
15

106. Roda JM, Carceller F, Perez-Higueras A, et al: Encapsulated
intracerebral hematomas: a defined entity. Case report. J Neu-
rosurg 78:829–833, 1993
107. Rohde V, Rohde I, Reinges MH, et al: Frameless stereotacti-
cally guided catheter placement and fibrinolytic therapy for
spontaneous intracerebral hematomas: technical aspects and
initial clinical results. Minim Invas Neurosurg 43:9–17,
2000
108. Rosner MJ, Becker DP: Origin and evolution of plateau waves.
Experimental observations and a theoretical model. J Neuro-
surg 60:312–324, 1984
109. Ruiz-Sandoval JL, Cantu C, Barinagarrementeria F: Intrace-
rebral hemorrhage in young people: analysis of risk factors,
location, causes, and prognosis. Stroke 30:537–541, 1999
110. Schaller C, Rohde V, Meyer B, et al: Stereotactic puncture
and lysis of spontaneous intracerebral hemorrhage using re-
combinant tissue-plasminogen activator. Neurosurgery 36:
328–335, 1995
111. Schellinger PD, Fiebach JB, Hoffmann K, et al: Stroke MRI in
intracerebral hemorrhage: is there a perihemorrhagic penum-
bra? Stroke 34:1674–1679, 2003
112. Schellinger PD, Jansen O, Fiebach JB, et al: A standardized
MRI stroke protocol: comparison with CT in hyperacute
intracerebral hemorrhage. Stroke 30:765–768, 1999
113. SHEP Cooperative Research Group: Prevention of stroke by
antihypertensive drug treatment in older persons with isola-
ted systolic hypertension. Final results of the Systolic Hy-
pertension in the Elderly Program (SHEP). JAMA 265:
3255–3264, 1991
114. Siddique MS, Fernandes HM, Wooldridge TD, et al: Re-
versible ischemia around intracerebral hemorrhage: a single-
photon emission computerized tomography study. J Neuro-
surg 96:736–741, 2002
115. Strandgaard S, Olesen J, Skinhoj E, et al: Autoregulation of
brain circulation in severe arterial hypertension. Br Med J
1:507–510, 1973
116. Sung CY, Chu NS: Epileptic seizures in intracerebral haemor-
rhage. J Neurol Neurosurg Psychiatry 52:1273–1276, 1989
117. Suzuki J, Kodama N: Moyamoya disease—a review. Stroke
14:104–109, 1983
118. Suzuki K, Kutsuzawa T, Takita K, et al: Clinico-epidemiolog-
ic study of stroke in Akita, Japan. Stroke 18:402–406, 1987
119. Taylor TN, Davis PH, Torner JC: Projected number of strokes
by subtype in the year 2050 in the United States. Stroke
29:322 1998 (Abstract)
120. Teernstra OP, Evers SM, Lodder J, et al: Stereotactic treatment
of intracerebral hematoma by means of a plasminogen acti-
vator: a multicenter randomized controlled trial (SICHPA).
Stroke 34:968–974, 2003
121. Tellez H, Bauer RB: Dexamethasone as treatment in cerebro-
vascular disease. 1. A controlled study in intracerebral hemor-
rhage. Stroke 4:541–546, 1973
122. Thrift AG, McNeil JJ, Forbes A, et al: Risk factors for cerebral
hemorrhage in the era of well-controlled hypertension. Mel-
bourne Risk Factor Study (MERFS) Group. Stroke 27:
2020–2025, 1996
123. Tuhrim S, Horowitz DR, Sacher M, et al: Volume of ventricu-
lar blood is an important determinant of outcome in supraten-
torial intracerebral hemorrhage. Crit Care Med 27:617–621,
1999
16
M. E. Fewel, B. G. Thompson, Jr., and J. T. Hoff
124. Uemura K, Higano S, Shishido F, et al: Regional cerebral blood
flow and oxygen metabolism in patients with hypertensive
intracerebral hemorrhage. A PET study in 23 patients. J Cereb
Blood Flow Metab 7 (Suppl 1):S32, 1987 (Abstract)
125. van Loon J, Van Calenbergh F, Goffin J, et al: Controversies in
the management of spontaneous cerebellar haemorrhage. A
consecutive series of 49 cases and review of the literature.
Acta Neurochir 122:187–193, 1993
126. Vinters HV: Cerebral amyloid angiopathy. A critical review.
Stroke 18:311–324, 1987
127. Wagner KR, Xi G, Hua Y, et al: Ultra-early clot aspiration after
lysis with tissue plasminogen activator in a porcine model of
intracerebral hemorrhage: edema and blood—brain barrier
protection. J Neurosurg 90:491–498, 1999
128. Weir CJ, Murray GD, Adams FG, et al: Poor accuracy of scor-
ing systems for differential clinical diagnosis of intracranial
haemorrhage and infarction. Lancet 344:999–1002, 1994
129. Xi G, Keep RF, Hoff JT: Pathophysiology of brain edema for-
mation. Neurosurg Clin N Am 13:371–383, 2002
130. Xi G, Reiser G, Keep RF: The role of thrombin and thrombin
receptors in ischemic, hemorrhagic, and traumatic brain in-
jury: deleterious or protective? J Neurochem 84:3–9, 2003
131. Xi G, Wagner KR, Keep RF, et al: Role of blood clot forma-
tion on early edema development after experimental intracere-
bral hemorrhage. Stroke 29:2580–2586, 1998
132. Yanaka K, Meguro K, Fujita K, et al: Immediate surgery re-
duces mortality in deeply comatose patients with spontaneous
cerebellar hemorrhage. Neurol Med Chir 40:295–300, 2000
133. Yang GY, Betz AL, Chenevert TL, et al: Experimental intrac-
erebral hemorrhage: relationship between brain edema, blood
flow, and blood-brain barrier permeability in rats. J Neuro-
surg 81:93–102, 1994
134. Yoshida Y, Yoshimoto T, Shirane R, et al: Clinical course, sur-
gical management, and long-term outcome of moyamoya pa-
tients with rebleeding after an episode of intracerebral hemor-
rhage: an extensive follow-up study. Stroke 30:2272–2276,
1999
135. Yu YL, Kumana CR, Lauder IJ, et al: Treatment of acute cere-
bral hemorrhage with intravenous glycerol. A double-blind,
placebo-controlled, randomized trial. Stroke 23:967–971,
1992
136. Zazulia AR, Diringer MN, Videen TO, et al: Hypoperfusion
without ischemia surrounding acute intracerebral hemorrhage.
J Cereb Blood Flow Metab 21:804–810, 2001
137. Zhu XL, Chan MS, Poon WS: Spontaneous intracranial hem-
orrhage: which patients need diagnostic cerebral angiography?
A prospective study of 206 cases and review of the literature.
Stroke 28:1406–1409, 1997
138. Zuccarello M, Brott T, Derex L, et al: Early surgical treatment
for supratentorial intracerebral hemorrhage: a randomized fea-
sibility study. Stroke 30:1833–1839, 1999
Manuscript received August 18, 2003.
Accepted in final form September 16, 2003.
Address reprint requests to: Matthew E. Fewel, M.D., Depart-
ment of Neurosurgery, University of Michigan Health System, 1500
East Medical Center Drive, Room 2128 TC, Ann Arbor, Michigan
48109-0338. email: fewelm@umich.edu.
Neurosurg. Focus / Volume 15 / October, 2003