Bluhmki, Werner Hacke and Rdiger von Kummer

Christian Berger, Marco Fiorelli, Thorsten Steiner, Wolf-Rdiger Schbitz, Luigi Bozzao, Erich
Hemorrhagic Transformation of Ischemic Brain Tissue : Asymptomatic or Symptomatic?
Print ISSN: 0039-2499. Online ISSN: 1524-4628
Copyright 2001 American Heart Association, Inc. All rights reserved.
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Stroke
doi: 10.1161/01.STR.32.6.1330
2001;32:1330-1335 Stroke.
http://stroke.ahajournals.org/content/32/6/1330
World Wide Web at:
The online version of this article, along with updated information and services, is located on the

http://stroke.ahajournals.org//subscriptions/
is online at: Stroke Information about subscribing to Subscriptions:

http://www.lww.com/reprints
Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer process is available in the
Request Permissions in the middle column of the Web page under Services. Further information about this
Once the online version of the published article for which permission is being requested is located, click
can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Stroke in
Requests for permissions to reproduce figures, tables, or portions of articles originally published Permissions:
by guest on September 18, 2012 http://stroke.ahajournals.org/ Downloaded from
Hemorrhagic Transformation of Ischemic Brain Tissue
Asymptomatic or Symptomatic?
Christian Berger, MD; Marco Fiorelli, MD; Thorsten Steiner, MD; Wolf-Rdiger Schbitz, MD;
Luigi Bozzao, MD; Erich Bluhmki, MD; Werner Hacke, MD; Rdiger von Kummer, MD
Background and PurposeThe term symptomatic hemorrhage secondary to ischemic stroke implies a clear causal
relationship between clinical deterioration and hemorrhagic transformation (HT) regardless of the type of HT. The aim
of this study was to assess which type of HT independently affects clinical outcome.
MethodsWe used the data set of the European Cooperative Acute Stroke Study (ECASS) II for a post hoc analysis. All
patients had a control CT scan after 24 to 96 hours or earlier in case of rapid and severe clinical deterioration. HT was
categorized according to radiological criteria: hemorrhagic infarction type 1 and type 2 and parenchymal hematoma type
1 and type 2. The clinical course was prospectively documented with the National Institutes of Health Stroke Scale and
the modified Rankin Scale. The independent risk of each type of HT was calculated for clinical deterioration at 24 hours
and disability and death at 3 months after stroke onset and adjusted for possible confounding factors such as age, severity
of stroke syndrome at baseline, and extent of the ischemic lesion on the initial CT.
ResultsCompared with absence of HT, only parenchymal hematoma type 2 was associated with an increased risk for
deterioration at 24 hours after stroke onset (adjusted odds ratio, 18; 95% CI, 6 to 56) and for death at 3 months (adjusted
odds ratio, 11; 95% CI, 3.7 to 36). All other types of HT did not independently increase the risk of late deterioration.
ConclusionsOnly parenchymal hematoma type 2 independently causes clinical deterioration and impairs prognosis. It
has a distinct radiological feature: it is a dense homogeneous hematoma 30% of the ischemic lesion volume with
significant space-occupying effect. (Stroke. 2001;32:1330-1335.)
Key Words: hematoma

hemorrhagic stroke

stroke outcome

thrombolysis
S
everal large, randomized, placebo-controlled trials of
thrombolytic therapy in acute ischemic stroke have been
conducted during the past years. Secondary hemorrhagic
transformations (HT) after different types of thrombolytic
therapy are frequently reported as important safety parame-
ters in these studies, but they also occur as natural events in
the evolution of a cerebral infarct.
14
The increase of HT in
the actively treated arm is often used as evidence against the
treatment tested, as reported in the early streptokinase trials,
with a concurrent increase of mortality and HT in the actively
treated groups.
57
Although the large European and American
tissue plasminogen activator (tPA) trials provided evidence of
a benefit in ischemic stroke patients, the fear of hemorrhagic
events frequently precludes the use of thrombolysis in clinical
practice. Unfortunately, previous trials addressing safety and
efficacy of thrombolytic therapy in acute stroke applied
different definitions for the term symptomatic hemorrhage or
symptomatic hemorrhagic transformation. In the tPA trial
sponsored by the National Institutes of Neurological Disor-
ders and Stroke (NINDS),
8,9
symptomatic intracranial hem-
orrhage was defined as any CT-documented hemorrhage
that was temporally related to deterioration in the patients
clinical condition in the judgment of the clinical investiga-
tor, no matter how trivial the hemorrhage on the CT scan
might have been. In addition, symptomatic intracranial hem-
orrhage attributable to study medication was defined as
symptomatic hemorrhage that occurred within 36 hours
from treatment onset. In the Multicenter Acute Stroke Trial
of Italy and Europe,
6,7
symptomatic HT was defined as
clinical deterioration temporally related to HT documented
by CT-scan or autopsy. A CT scan was regularly obtained at
5 days after stroke onset or earlier in the event of clinical
deterioration. Further studies continued to apply the term
symptomatic hemorrhage according to the NINDS proto-
col
911
or without a definition.
12,13
The European Cooperative Acute Stroke Study (ECASS) I
and II
14,15
went in the opposite direction and used a pure
radiological, prospective definition: HT were categorized into
4 different subtypes without taking into consideration
whether or not any hemorrhage was associated with clinical
Received October 26, 2000; final revision received February 21, 2001;accepted February 22, 2001.
From the Department of Neurology, University of Heidelberg (Germany) (C.B., T.S., W-R.S., W.H.); Department of Neuroradiology, University of
Dresden (Germany) (R. von K.); Department of Neuroradiology, University of Rome (Italy) (M.F., L.B.); and Boehringer Ingelheim, Biberach, Germany
(E.B.).
Correspondence to Christian Berger, MD, Department of Neurology, University of Heidelberg, Im Neuenheimer Feld 400, D-69221 Heidelberg,
Germany. E-mail christian_berger@med.uni-heidelberg.de
2001 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org
1330
by guest on September 18, 2012 http://stroke.ahajournals.org/ Downloaded from
deterioration. In addition to the pure radiological definition,
the category of symptomatic hemorrhage was used for pa-
tients with clinical deterioration by 4 points on the National
Institutes of Health Stroke Scale (NIHSS) and with no CT
findings that might have been responsible for this deteriora-
tion other than a hemorrhage.
Fiorelli et al
16
recently demonstrated on the basis of the
ECASS I data that, in both the placebo and the rtPA groups,
only parenchymal hematomas (PH) 30% of the infarcted
area with significant space-occupying effect increased the
risk of early neurological deterioration and of 3-month death.
Hemorrhagic infarctions (HI) or PH with only mild space-
occupying effect did not modify the risk of early neurological
deterioration, death, and disability. The different clinical
outcome after different subtypes of HT illustrates the diffi-
culty in defining symptomatic hemorrhage precisely and
clearly. In fact, the spectrum of HTs differs widely and may
include some trivial hemorrhagic petechiae as well as PH
with space-occupying effect. To determine whether any
neurological deterioration is due to HT itself or due to
massive infarct and ischemic edema with a coincidental HT
remains crucial in thrombolysis or other stroke trials. The aim
of this study on the ECASS II data set was to define specific
HT that are independently associated with an increased risk
of clinical deterioration or worse outcome according to the
hypothesis of Fiorelli et al.
16
We sought to further assess the
impact of possible confounding factors on clinical deteriora-
tion in patients with HT.
Subjects and Methods
The study design and primary results of ECASS II have been
described in detail elsewhere.
15
ECASS II was a double-blind,
placebo-controlled trial evaluating safety and efficacy of 0.9 mg/kg
IV recombinant tPA (rtPA) in patients presenting within 6 hours
from the onset of an acute hemispheric ischemic stroke. Intravenous
heparin or oral anticoagulants within the first 24 hours were not
allowed.
All patients were examined with CT scan before randomization
and after 24 to 96 hours (median, 1 day; range, 0 to 4 days). Nine
patients had their follow-up scan within 24 hours of stroke onset, 643
patients between 24 and 48 hours, 128 between 48 and 72 hours, and
3 between 72 and 96 hours. An additional CT was performed after 1
week. We used this CT for analysis in 6 patients in whom the first
follow-up CT was not available. Following the prospective ECASS
protocol, all CT scans were evaluated twice, first by the local
investigators and then independently by 3 members of the CT
reading panel. The members of the CT reading panel were blinded to
treatment allocation, any clinical events occurring after randomiza-
tion, and the reading of the local investigators and did not see
follow-up scans before evaluating the baseline CT. After the exclu-
sion of 10 patients (3 rtPA, 7 placebo) with CT scans of too poor
quality to allow unequivocal assessment of HT, 790 of 800 random-
ized patients remained for the analysis. HT occurring on the first
follow-up CT scan after randomization were prospectively catego-
rized according to definitions previously described
17,18
: HI and PH.
HI type 1 (HI-1) was defined as small petechiae along the margins of
the infarct, and HI type 2 (HI-2) was defined as more confluent
petechiae within the infarcted area but without space-occupying
effect. PH type 1 (PH-1) was defined as hematoma in 30% of the
infarcted area with some slight space-occupying effect; PH type 2
(PH-2) was defined as dense hematoma 30% of the infarcted area
with substantial space-occupying effect or as any hemorrhagic lesion
outside the infarcted area (examples are shown in Figure 1). In cases
of 1 hemorrhagic lesion on CT scan, the worst possible HT
category was assumed.
Baseline and follow-up CT scans were obtained nonenhanced and
on the same scanner if possible. Windows and center levels were set
to optimally distinguish gray and white matter. The study protocol
recommended a window width of 80 to 100 Hounsfield units and a
center level of 30 to 40 Hounsfield units. The chairman of the CT
reading panel gathered the categorized findings from the panel
members, checked them for disagreements, disclosed disagreements
to the other panel members, and discussed each discrepant CT to
achieve consensus. The final judgments were sent to the data
management center, which then was allowed to send the follow-up
scans to the members of the CT reading panel. We defined
hypoattenuation as a visible decrease in x-ray attenuation of brain
tissue compared with other portions of the same anatomic structure
or its contralateral counterpart. We categorized the extent of hypoat-
tenuation of the middle cerebral artery (MCA) territory as none,
33% (small), and 33% (large).
19,20
The total MCA territory was
considered the brain volume between 2 lines being drawn from the
anterior horn of the ventricle and from the trigonum perpendicular to
the skull involving most parts of the frontal, temporal, and parietal
lobe with the exception of the parasagittal structures. On follow-up
CT scans, we measured the volume of acute ischemic lesions by
multiplying the maximum diameter of hypoattenuation, maximum
diameter perpendicular to it in the same slice, number of slices
affected, slice distance, and a conversion factor of 0.5 using the
formula for irregular volumes. In patients with HT, the entire
ischemic and hemorrhagic lesion was measured.
From the ECASS II database, we retrieved the following vari-
ables: age, sex, allocation to placebo/rtPA treatment, severity of
neurological deficit on admission as quantified with the NIHSS and
the Scandinavian Stroke Scale,
21
NIHSS score 242 hours after
stroke onset, extent of initial ischemic lesion on baseline CT scan
(none, 33% of MCA territory, and 33% of MCA territory), body
temperature, blood glucose, and blood pressure at the inclusion time.
Figure 1. Examples of different subtypes of HT: HI-1 with scat-
tered, heterogeneous petechiae along the margins of the infarct
(A); HI-2 with more conuent but still heterogeneous petechiae
within the infarcted area (B); PH-1 with a homogeneous hema-
toma covering 30% of the infarcted area and only mild space-
occupying effect (C); and PH-2 with a dense hematoma 30%
of the lesion volume with signicant space-occupying effect (D).
Berger et al Hemorrhagic Transformation of Ischemic Brain Tissue 1331
by guest on September 18, 2012 http://stroke.ahajournals.org/ Downloaded from
We calculated the frequencies of the different subtypes of HT for
placebo and rtPA groups. Statistical difference was tested with the
2
test.
To examine which of the collected variables accounted for clinical
deterioration 24 hours after stroke onset (increase in NIHSS score by
4 between baseline and after 242 hours
22
according to the
prospective definition used in ECASS II), for disability and death at
3 months (modified Rankin Scale score 2), and for death up to 3
months after stroke onset, we performed a logistic regression
analysis on each of these variables. Variables with a P0.05 in the
Wald test and the logistic likelihood ratio test were assumed to be
irrelevant for the outcome tested and were excluded from further
analysis. We calculated the odds ratios (ORs) and 95% CIs for each
HT subtype, comparing the ORs of each level of HT with no HT.
Subsequently, we performed a multiple regression analysis including
all variables considered relevant factors from the univariate analysis
and calculated ORs for clinical deterioration at 24 hours, disability
and death after 3 months, and death at 3 months after stroke onset.
In a backward stepwise regression, we excluded all variables from
our model that did not contribute significantly in at least 1 outcome
test. The adjusted ORs are given for all variables finally kept in the
model. Variables leading to a clear change of the ORs for HT were
considered confounding factors.
Finally, we evaluated the relationship between the frequency of
PH-2 and the extent of hypoattenuation on baseline CT scan and
calculated ORs for the development of PH-2. The Newman-Keuls
test was applied to assess the relationship between incidence of
different types of HT and the lesion volume at 24 hours. In all tests,
a statistical significance was assumed for P0.05. Analyses were
performed with StatView statistical software (edition 5.0.1).
Results
Of a total of 790 intent-to-treat patients in ECASS II, 406
patients received rtPA (0.9 mg/kg), and 384 patients were
treated with placebo. In the first follow-up CT scan, signifi-
cantly more HT occurred in the rtPA group: 29.5% versus
18.5% in the placebo group (Table 1). This was mainly due to
a higher incidence of PH-2 in the rtPA group (7.6% versus
0.5%; P0.0001). The incidences of HI-1, HI-2, and PH-1
were statistically not different between the placebo and the
actively treated group, although there was a clear trend for
HI-2 and PH-1 to more likely occur after thrombolysis.
Table 2 presents the unadjusted and adjusted risk ratios for
the 4 different types of HT and for other factors found to
contribute significantly to outcome. Sex, body temperature,
and blood pressure at the inclusion time were irrelevant for
outcome and thus were excluded from analysis. PH-2 signif-
icantly increased the risk of early deterioration, of disability
and death at 3 months, and of death at 3 months alone. PH-1
accounted for an increased risk of early deterioration but not
of disability or death at 3 months. Thus, outcome at 3 months
was modified only by presence of PH-2 on the first follow-up
CT scan. In our search for factors contributing significantly to
outcome, we found a confounding effect of severity of stroke
syndrome at onset and of presence of hypodensity 33% on
baseline CT scan. Both were associated with an increased risk
rate for disability and death. Presence of a large hypodensity
was associated with early deterioration. Other contributing
factors for outcome were age and glucose level. Increase of
age or glucose was associated with an increased risk of
disability and death at 3 months. Treatment with rtPA led to
a significantly decreased risk of disability or death after 3
TABLE 1. Frequencies of Different Types of Hemorrhage in the
Placebo and rtPA Groups
Placebo (n384) rtPA (n406)
No. % No. %
HI-1 53 13.8 50 12.3
HI-2 11 2.9 28 6.9
PH-1 5 1.3 11 2.7
PH-2* 2 0.5 31 7.6
Total* 71 18.5 120 29.5
*Significant difference in frequency between placebo and rtPA group
according to
2
test (P0.05).
TABLE 2. Comparison of Adjusted and Unadjusted ORs
Deterioration After 24 h Disability/Death up to 3 mo Death up to 3 mo
OR (95% CI) P OR (95% CI) P OR (95% CI) P
Nonadjusted HT
HI-1 0.6 (0.21.3) 0.2 1.4 (0.92.2) 0.09 0.9 (0.42.1) 0.9
HI-2 0.8 (0.22.6) 0.7 1.3 (0.72.6) 0.4 0.9 (0.33.2) 0.9
PH-1 4.2 (1.412.4) 0.01 1.2 (0.43.1) 0.8 2.6 (0.79.6) 0.1
PH-2 41 (16104) 0.0001 6.5 (2.517.0) 0.0002 20 (943) 0.0001
Adjusted HT
HI-1 0.2 (0.10.6) 0.003 0.7 (0.41.2) 0.2 0.2 (0.070.6) 0.004
HI-2 0.5 (0.21.9) 0.3 0.9 (0.41.9) 0.8 0.6 (0.22.5) 0.5
PH-1 3.5 (1.111.6) 0.04 0.6 (0.21.9) 0.4 1.0 (0.24.8) 0.9
PH-2 18 (656) 0.0001 1.9 (0.75.7) 0.2 11.4 (3.736) 0.0001
Treatment with rtPA 0.8 (0.51.3) 0.4 0.7 (0.40.9) 0.02 0.6 (0.31.2) 0.2
Hypodensity 33% on initial CT 2.6 (1.06.5) 0.047 2.8 (1.26.4) 0.01 8 (3.319.5) 0.0001
Severe stroke syndrome at
onset
1.1 (0.62.1) 0.7 4.1 (2.56.5) 0.0001 4.3 (2.47.7) 0.0001
Age 1.0 (0.91.0) 0.8 1.04 (1.021.05) 0.0001 1.09 (1.051.13) 0.0001
Glucose 1.0 (0.91.1) 0.6 1.1 (1.041.13) 0.0006 1.1 (1.021.2) 0.02
1332 Stroke June 2001
by guest on September 18, 2012 http://stroke.ahajournals.org/ Downloaded from
months; however, its influence for early deterioration or death
alone at 3 months was insignificant.
Figure 2 demonstrates the positive correlation between
large lesion volumes and large PH in both the actively treated
and the placebo-treated groups. In both treatment groups, the
total lesion volumes in patients with PH-2 were significantly
larger than those of all other groups, including the group
without HT. Thus, the space-occupying effect of PH-2
becomes evident. Table 3 presents the relationship between
incidence of PH-2 and extent of hypoattenuation on baseline
CT scan. Clearly, the incidence of PH-2 is more frequent with
presence of large hypodensity on CT scan. The risk ratio of
developing PH-2 with an initially large extent of the ischemic
lesion was 4.4.
Discussion
HT of a brain infarct is a major concern when thrombolysis is
used in acute stroke. Frequently, HT of any type and extent
found on routine follow-up CT scans is considered a serious
adverse event and is often termed symptomatic hemorrhage
regardless of whether a certain type of HT was really causing
symptoms. In the larger thrombolysis studies, various defini-
tions for symptomatic hemorrhage consisting of radiological
and clinical criteria were applied. The NINDS tPA trial
defined symptomatic hemorrhage as any CT-documented
hemorrhage that was temporally related to deterioration in the
patients clinical condition in the judgment of the clinical
investigator.
8
Symptomatic hemorrhage was considered to
be due to study medication if it occurred within 36 hours of
treatment. It appears inconceivable that a clinical deteriora-
tion under this definition could not be merely coincidental
with any HT appearing on CT scan. In fact, the deterioration
could well have been the natural course of a stroke indepen-
dent of the coinciding HT, especially if the HT was only a
minor part of a large infarct. In a post hoc analysis by the
NINDS Stroke Study Group,
9
it was nevertheless stated that
presence of a severe neurological deficit and clear signs of
brain edema or mass effect on pretreatment CT increased the
risk of symptomatic hemorrhage. Again, pure radiological
criteria of different types of HT were not considered, and
other factors causing clinical deterioration were ignored.
Similar definitions for symptomatic hemorrhage were applied
in the Multicenter Acute Stroke Trial of Italy and Europe
6,7
and in recent thrombolysis surveys.
10,11
In ECASS I and II, prospective radiological definitions of
the 4 different subtypes of HT were applied regardless of
clinical events, and CT scans with or without HT were
analyzed regardless of clinical data. CT was performed at 24
hours and 7 days after symptom onset and treatment and in
case of clinical deterioration. Almost all major hemorrhages
were detected by the first follow-up CT. However, local
investigators collecting data on neurological status might
have created some unavoidable bias in that they knew the
result of the CT scan and might have implemented a precon-
ceived notion of any relationship between HT and neurolog-
ical deterioration. We believe that a prospective protocol for
CT scanning at predetermined dates is necessary to study the
question of whether certain types of hemorrhage are associ-
ated with clinical deterioration. In our view, an early
follow-up CT within 48 hours of treatment is necessary to
detect hemorrhages caused by rtPA.
In this post hoc analysis of the ECASS II data set, we
confirm previous results by Fiorelli et al
16
that only PH
30% of the infarcted area with considerable space-
occupying effect (PH-2) significantly increased the risk of
early clinical deterioration and of a worse long-term outcome,
including death. Other types of HT, particularly HI-1 and
HI-2, were not associated with clinical deterioration. This
was also true for PH-1, which at least did not increase the risk
of disability or death up to 3 months after stroke onset. Thus,
while these latter types of HT would have been termed
symptomatic hemorrhages according to the definition of
many ongoing trials, their presence might not have been
responsible for deterioration. Instead, other factors such as
age, initial glucose level, extent of initial hypodensity on
baseline CT scan, and severity of stroke syndrome at baseline
Figure 2. Comparison between lesion volume and presence of
hemorrhage, showing mean lesion volume (error bars indicate
SD) and corresponding group of patients with a certain type of
HT or without any HT. In both treatment groups, signicant dif-
ference in lesion volume was reached between PH-2 and all
other groups, including the group without any HT. In the rtPA
group, lesion volume in patients without HT was also signi-
cantly lower than in patients with HI-1 and HI-2 (Newman-Keuls
test). See Table 1 for frequencies of different HT subtypes.
TABLE 3. Observed Frequencies of PH-2 With Different Extents of
Hypoattenuation on Baseline CT Scan and Their ORs for the Development of PH-2
Hypoattenuation
at Baseline
Frequency OR
PH-2 % Total OR (95% CI) P
No 12 3.3 356 0.04 (0.020.07) 0.0001
33% 18 4.5 398 1.8 (0.83.7) 0.1
33% 3 8.3 36 4.4 (1.116.8) 0.03
Total 33 4.2 790
Berger et al Hemorrhagic Transformation of Ischemic Brain Tissue 1333
by guest on September 18, 2012 http://stroke.ahajournals.org/ Downloaded from
additionally contributed to clinical outcome, as described in a
previous study.
23
Treatment with rtPA, although associated
with a higher incidence of PH-2, decreased the overall risk of
disability and death at 3 months.
In particular, stroke severity and the presence of large
hypodensity on the initial CT scan were associated with a
significantly increased risk for disability and death up to 3
months after stroke onset. Thus, they appear as confounding
factors for the influence of HT on outcome. After adjustment
for these confounding factors, the risk for disability or death
at 3 months was increased insignificantly after PH-2, but with
respect to a long-term outcome, PH-2 independently and
significantly increased only the risk of death at 3 months.
HI-1 was even associated with some clinical improvement
and may, at least in this early stage, indicate the effect of
successful early reperfusion. Since PH-2 is of major clinical
relevance or may in fact be considered hemorrhage that is
most likely to be symptomatic, emphasis must be placed on
the recognition of stroke patients who are more likely to
develop large PH and subsequently should be excluded from
thrombolysis. PH-2 occurred more frequently in patients with
hypodensity 33% on early CT. These early ischemic
changes identified on the pretreatment CT represent early
cytotoxic edema and possibly the development of irreversible
injury.
24
In a study by Hamann et al,
25
the presence of
microscopic hemorrhages in the ischemic area has been
related to a loss of cerebrovascular basal lamina integrity.
Since these microscopic hemorrhages always occur within
the infarcted area, particularly in the subcortical core, a large
infarct volume a priori offers a larger area of lost basal lamina
integrity, thus possibly leading more frequently to parenchy-
mal hemorrhages. We were able to confirm this relationship
by demonstrating a positive correlation between large lesion
volume and presence of PH-2. Thus, to detect signs of early
infarction on CT remains crucial, although it is questioned
whether these signs can be detected reliably.
26
In a previous
study the interobserver agreement in assessing subtle CT
signs of cerebral infarction was moderate to substantial.
20
Similar results for assessment of the hypoattenuation in thirds
of the MCA territory were obtained by the Alberta Stroke
Programme Early CT Score (ASPECTS) study group.
27
Whether general reading of CT scans can be improved by
quantification and division of the MCA territory into many
regions of interest remains to be confirmed in clinical
practice. To date, presence of hypoattenuation larger than one
third of the MCA territory on baseline CT seems to be the
most simple and readily available predictor of the develop-
ment of large PH.
In summary, only large PH 30% of the infarcted area
with space-occupying effect (PH-2) independently modify
the risk of a worse clinical outcome both early and late after
stroke onset. Smaller but still homogeneous PH (PH-1)
increase the risk of early deterioration but not of a worse
long-term outcome. Heterogeneous, petechial HI are not
associated with worse early or late outcome. Thus, the term
symptomatic hemorrhage should be applied with caution
because it implies a causal relationship between signs of HT
found on CT and clinical deterioration. Additional factors,
such as extent of the lesion, edema formation, and severity of
the stroke syndrome, however, offer an explanation for a
patients deterioration or bad outcome. In future reports
dealing with safety aspects of thrombolysis, a clearer distinc-
tion between nonrelevant forms of HT and PH with risk of
clinical deterioration should be made.
Acknowledgment
ECASS II was supported exclusively by Boehringer Ingelheim,
Biberach, Germany.
References
1. Hornig CR, Dorndorf W, Agnoli AL. Hemorrhagic cerebral infarction: a
prospective study. Stroke. 1986;17:179185.
2. Bozzao L, Angeloni U, Bastianello S, Fantozzi LM, Pierallini A, Fieschi
C. Early angiographic and CT findings in patients with hemorrhagic
infarction in the distribution of the middle cerebral artery. AJNR Am J
Neuroradiol. 1991;12:11151121.
3. Moulin T, Crepin-Leblond T, Chopard JL, Bogousslavsky J. Hemor-
rhagic infarcts. Eur Neurol. 1994;34:6477.
4. Toni D, Fiorelli M, Bastianello S, Sacchetti ML, Sette G, Argentino C,
Montinaro E, Bozzao L. Hemorrhagic transformation of brain infarct:
predictability in the first 5 hours from stroke onset and influence on
clinical outcome. Neurology. 1996;46:341345.
5. Donnan GA, Davis SM, Chambers BR, Gates PC, Hankey GJ, McNeil JJ,
Rosen D, Stewart-Wynne EG, Tuck RR, for the Australian Streptokinase
(ASK) Trial Study Group. Streptokinase for acute ischemic stroke with
relationship to time of administration. JAMA. 1996;276:961966.
6. . The Multicentre Acute Stroke TrialItaly Group (MAST-I). Randomised
controlled trial of streptokinase, aspirin, and combination of both in
treatment of acute ischaemic stroke Lancet. 1995;346:15091514.
7. The Multicenter Acute Stroke TrialEurope (MAST-E). Thrombolytic
therapy with streptokinase in acute ischemic stroke. N Engl J Med.
1996;335:145150.
8. The National Institute of Neurological Disorders and Stroke rt-PA Stroke
Study Group. Tissue plasminogen activator for acute ischemic stroke.
N Engl J Med. 1995;333:15811587.
9. The NINDS t-PA Stroke Study Group. Intracerebral hemorrhage after
intravenous t-PA therapy for ischemic stroke. Stroke. 1997;28:
21092118.
10. Tanne D, Bates VE, Verro P, Kasner SE, Binder JR, Patel SC, Mansbach
HH, Daley S, Schultz LR, Karanjia PN, Scott P, Dayno JM, Vereczkey-
Porter K, Benesch C, Book D, Coplin WM, Dulli D, Levine SR, for the
t-PA Stroke Survey Group. Initial clinical experience with IV tissue
plasminogen activator for acute ischemic stroke: a multicenter survey.
Neurology. 1999;53:424427.
11. Buchan AM, Barber PA, Newcommon N, Karbalai HG, Demchuk AM,
Hoyte KM, Klein GM, Feasby TE. Effectiveness of t-PA in acute ische-
mic stroke: outcome relates to appropriateness. Neurology. 2000;54:
679684.
12. Chiu D, Krieger D, Villar-Cordova C, Kasner SE, Morgenstern LB,
Bratina PL, Yatsu FM, Grotta JC. Intravenous tissue plasminogen acti-
vator for acute ischemic stroke: feasibility, safety, and efficacy in the first
year of clinical practice. Stroke. 1998;29:1822.
13. Gonner F, Remonda L, Mattle H, Sturzenegger M, Ozdoba C, Lovblad
KO, Baumgartner R, Bassetti C, Schroth G. Local intra-arterial
thrombolysis in acute ischemic stroke. Stroke. 1998;29:18941900.
14. Hacke W, Kaste M, Fieschi C, Toni D, Lesaffre E, von Kummer R,
Boysen G, Bluhmki E, Hoxter G, Mahagne MH, et al, for the European
Cooperative Acute Stroke Study (ECASS). Intravenous thrombolysis
with recombinant tissue plasminogen activator for acute hemispheric
stroke. JAMA. 1995;274:10171025.
15. Hacke W, Kaste M, Fieschi C, von Kummer R, Davalos A, Meier D,
Larrue V, Bluhmki E, Davis S, Donnan G, Schneider D, Diez-Tejedor E,
Trouillas P, for the Second European-Australasian Acute Stroke Study
Investigators. Randomised double-blind placebo-controlled trial of
thrombolytic therapy with intravenous alteplase in acute ischaemic stroke
(ECASS II). Lancet. 1998;352:12451251.
16. Fiorelli M, Bastianello S, von Kummer R, del Zoppo GJ, Larrue V,
Lesaffre E, Ringleb AP, Lorenzano S, Manelfe C, Bozzao L. Hemor-
rhagic transformation within 36 hours of a cerebral infarct: relationships
with early clinical deterioration and 3-month outcome in the European
Cooperative Acute Stroke Study I (ECASS I) cohort. Stroke. 1999;30:
22802284.
1334 Stroke June 2001
by guest on September 18, 2012 http://stroke.ahajournals.org/ Downloaded from
17. Pessin MS, del Zoppo GJ, Estol CJ. Thrombolytic agents in the treatment
of stroke. Clin Neuropharmacol. 1990;13:271289.
18. Wolpert SM, Bruckmann H, Greenlee R, Wechsler L, Pessin MS, del
Zoppo GJ, for the rt-PA Acute Stroke Study Group. Neuroradiologic
evaluation of patients with acute stroke treated with recombinant tissue
plasminogen activator. AJNR Am J Neuroradiol. 1993;14:313.
19. von Kummer R, Allen KL, Holle R, Bozzao L, Bastianello S, Manelfe C,
Bluhmki E, Ringleb P, Meier DH, Hacke W. Acute stroke: usefulness of
early CT findings before thrombolytic therapy. Radiology. 1997;205:
327333.
20. von Kummer R, Holle R, Gizyska U, Hofmann E, Jansen O, Petersen D,
Schumacher M, Sartor K. Interobserver agreement in assessing early CT
signs of middle cerebral artery infarction. AJNR Am J Neuroradiol.
1996;17:17431748.
21. The Scandinavian Stroke Study Group. Multicenter trial of hemodilution
in ischemic stroke: background and study protocol. Stroke. 1985;16:
885890.
22. Brott TG, Haley EC Jr, Levy DE, Barsan W, Broderick J, Sheppard GL,
Spilker J, Kongable GL, Massey S, Reed R, Marler JR. Urgent therapy for
stroke, part I: pilot study of tissue plasminogen activator administered
within 90 minutes. Stroke. 1992;23:632640.
23. Larrue V, von Kummer R, del Zoppo G, Bluhmki E. Hemorrhagic
transformation in acute ischemic stroke: potential contributing factors in
the European Cooperative Acute Stroke Study. Stroke. 1997;28:957960.
24. del Zoppo GJ, von Kummer R, Hamann GF. Ischaemic damage of brain
microvessels: inherent risks for thrombolytic treatment in stroke. J Neurol
Neurosurg Psychiatry. 1998;65:19.
25. Hamann GF, Okada Y, del Zoppo GJ. Hemorrhagic transformation and
microvascular integrity during focal cerebral ischemia/reperfusion.
J Cereb Blood Flow Metab. 1996;16:13731378.
26. Wardlaw JM, Dorman PJ, Lewis SC, Sandercock PA. Can stroke phy-
sicians and neuroradiologists identify signs of early cerebral infarction on
CT? J Neurol Neurosurg Psychiatry. 1999;67:651653.
27. Barber PA, Demchuk AM, Zhang J, Buchan AM, for the ASPECTS
Study Group (Alberta Stroke Programme Early CT Score). Validity and
reliability of a quantitative computed tomography score in predicting
outcome of hyperacute stroke before thrombolytic therapy. Lancet. 2000;
355:16701674.
Berger et al Hemorrhagic Transformation of Ischemic Brain Tissue 1335
by guest on September 18, 2012 http://stroke.ahajournals.org/ Downloaded from