Review article

European Stroke Journal

2018, Vol. 3(2) 101–109
Prognostic factors in pontine ! European Stroke Organisation
2018
haemorrhage: A systematic review Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/2396987317752729
journals.sagepub.com/home/eso

Réza Behrouz

Abstract
Background: Pontine haemorrhage comprises approximately 10% of intracerebral haemorrhages. There is a common
presumption that pontine haemorrhage is inherently associated with poor outcome.
Purpose: The aim of the review was to identify chief predictors of prognosis in (pontine haemorrhage) through
systematic review of published literature.
Methods: A query of PubMed/MEDLINE was conducted in search of studies in English language since, 1980 focusing
specifically on outcome in pontine haemorrhage. References for each publication were reviewed for additional studies
not detected by the PubMed/MEDLINE probe. Surgical outcome studies were excluded from the review.
Findings: The query identified 7867 titles, after removal of duplicates and irrelevant studies, 20 titles were included in
the review. In a total of 1437 pontine haemorrhage patients included in the 20 studies, the overall rate for early all-cause
mortality was 48.1%. Level of consciousness on admission and haemorrhage size were the most consistent predictors of
mortality in patients with pontine haemorrhage. Haemorrhage localisation within the pons was also a prognostic factor,
but not consistently. Age and intraventricular extension were not found to be powerful prognostic predictors.
Discussion/Conclusion: Based on this review, level of consciousness on admission and haemorrhage size were the
most influential prognostic factors in pontine haemorrhage, whereas age, haemorrhage localisation, and intraventricular
haemorrhage did not consistently predict prognosis.

Keywords
Pontine haemorrhage, intracerebral haemorrhage, pons, brainstem, prognosis, outcomes
Date received: 18 October 2017; accepted: 15 December 2017

Introduction
Pontine haemorrhage (PH) comprises approximately
10% of intracerebral haemorrhages (ICH).1 With an
estimated mortality rate ranging widely from 30% to
90%, PH is the most pernicious form of ICH.2–4 There
is an inherent presumption of poor outcome associated
with PH to the extent that some therapeutic clinical
trials in ICH have excluded this category of
patients.5–7 Preconceived notion regarding PH progno-
sis, however, could usher into self-fulfilling prophecies.
It is therefore important to identify reliable prognostic
factors that would allow realistic expectation of out-
come and prevent early care limitations.
Over the years, a host of predictors for outcome in PH
have been proposed, yet the utility and reliability of many
have not been substantiated. This is a review of some of
the important factors that have been shown to influence
outcomes in patients with PH.
Methods
Preferred Reporting Items for Systematic reviews
and Meta-Analyses (PRISMA) was utilised as guidance
for this systematic review.8 A query of PubMed/
MEDLINE was conducted between 3 September
2017 and 8 September 2017, in which medical subject
headings ‘pons þ haemorrhage,’ ‘pontine haemorrhage’
and ‘brainstem haemorrhage’ and ‘infratentorial
Department of Neurology, School of Medicine, University of Texas
Health Science Center, San Antonio, TX, USA
Corresponding author:
Réza Behrouz, Medical Arts & Research Center, University of Texas
Health Science Center San Antonio, 8300 Floyd Curl Drive, MC-7883
San Antonio, TX 78229, USA.
Email: behrouz@uthscsa.edu
102
Figure 1. The preferred reporting items for systematic reviews and meta-analyses flowchart.8 *The results of one study was
reported in the review but not used in the analyses.11
haemorrhage’ were used. Studies in English that focused
on outcomes in PH from 1980 onward were isolated and
gathered. References for each publication were reviewed
for additional studies not detected by the PubMed/
MEDLINE probe. Surgical outcome studies were exclud-
ed from the review. Positive predictive values (PPV), neg-
ative predictive values (NPV) and positive likelihood
ratios (PLR) were calculated using the standard formulas.
The query identified 7867 citations. A total of 2462
titles were found to be relevant, from which four non-
English language studies were excluded. Thirteen addi-
tional studies were identified by reviewing the reference
list of each publication. Twenty-four full-text articles
were reviewed. Two post-surgical PH studies were
excluded. Two studies included the same cohort of
patients, one of which (the one with the least number
of patients) was excluded. Twenty-one studies were
used for this review. Cohort size per study ranged
from 10 to 281 patients.9,10 The oldest cohort was
from 1983 and the most recent was from 2017; the
majority were retrospective. The PRISMA flowchart
European Stroke Journal 3(2)
of this query is displayed in Figure 1.8 Of the 21 out-
come studies in PH, one was excluded from the overall
analysis because of high number of PH caused by cav-
ernous and arteriovenous malformations.11 The results
of this study were, however, reported in the review.
Results
Mortality and functional outcome
Based on 20 studies consisting of 1437 patients, the
overall rate of early (within 30 days of PH or shortly
thereafter) death of all causes was 48.1%
(Table 1).1,2,4,9,10,12–26 This figure reflects results of
studies spanning four decades. The individual mortality
rates reported by these studies ranged from 33.7% to
77.5%.14,26 Mortality rates for the oldest and the most
recent studies were 71.7% and 39.8%, respectively.12,26
Twelve studies reported on outcomes in survivors,
totalling 951 patients.1,2,10,12–15,17,19–21,26 The rate of
good outcome at various points of follow-up was
Behrouz 103

Table 1. Studies included in assessment of early mortality in patients with pontine haemorrhage.

Author Year Study type N N died Duration to death N good outcome Average follow-up

1 Nakijama12 1983 Retrospective 60 43 14 days 11 (subjective) NR

2 Kushner and Bressman9 1985 Retrospective 10 6 In-hospital NR NR
3 Masiyama et al.13 1985 Retrospective 26 12 23 days 5 (subjective) 6–42 months
4 Weisberg14 1986 Retrospective 40 31 In-hospital 4 (subjective) NR
5 Chung and Park4 1992 NR 61 37 7 days NR NR
6 Wijdicks and St Louis15 1997 Retrospective 38 21 ‘Within days of 8 (subjective) 3–12 months
admission’
7 Fong16 1999 Retrospective 39 23 10 days NR NR
8 Murata et al.2 1999 Retrospective 80 38 ‘Within few days’ 28 (GOS 4) NR
9 Dziewas et al.17 2003 Retrospective 39 27 Average of 16 days 6 (mRS, 0–2) 2–8 years
10 Wessels et al.1 2004 Retrospective 29 9 5  3 days 12 (GOS 4) NR
11 Balci et al.18 2005 Retrospective 32 18 Average 8 days NR NR
12 Jung et al.19 2007 Retrospective 35 13 In-hospital 12 (subjective) Average
13.9 months
13 Shin et al.20 2007 Retrospective 35 15 In-hospital 13 (GOS 3) 3 months
14 Jang et al.10 2011 Retrospective 281 110 30 days 27 (mRS, 0–3) 90 days
15 Matsukawa et al.22 2014 Retrospective 118 66 Median 51 days NR NR
16 Nishizaki et al.21 2012 Retrospective 19 9 In-hospital 0 (GOS 4) NR
17 Ye et al.24 2015 Prospectivea 76 33 30 days NR NR
18 Meguro et al.23 2015 Retrospective 101 59 30 days NR NR
19 Morotti et al.25 2016 Retrospective 49 30 30 days NR NR
20 Huang et al.26 2017 Retrospective 171 68 30 days 74 (mRS, 0–3) 90 days
Prospectiveb 98 33 30 days 50 (mRS, 0–3) 90 days
Total   1437 691  250 
NR: not reported; GOS: Glasgow Outcome Scale; mRS: modified Rankin Scale.
a
Based on post hoc analysis of prospectively collected data.
b
External validation cohort for the study by Huang et al.

26.3% (Table 1). The point at which functional out-
come was assessed after PH varied widely among the
studies, ranging from 30 days to eight years. There was
also significant subjectivity and heterogeneity among
these studies with regard to the definition of good out-
come, which ranged from ability to perform activities
of daily living to quantification of disability using
Glasgow Outcome or modified Rankin scales (mRS).
Two studies used mRS 3 and one used mRS 2 to
indicate good outcome (Table 1).10,17,26 The combined
90-day rate of mRS 3 was 39.8% and the rate of mRS
2 in a follow-up period of two to eight years was
15.4%.
Clinical factors
Aetiology. A study of 44 consecutive patients with PH
indicated that PH due to cavernous malformations had
a more benign course and was associated with more
favourable long-term outcome than primary PH due
to hypertension.11 In addition, patients with primary
PH tended to have more severe clinical presentation,
based on Glasgow Coma Scale (GCS) scores.11 This
study may have been biased by an unusually high
number of PH cases with cavernous malformations
(45.5%). Furthermore, it is unclear how the authors
confidently diagnosed cavernous malformations using
magnetic resonance imaging.
Differentiating between primary PH and that due to
a cavernous malformation is not always simple in the
acute phase. Some features that may assist in delinea-
tion are absence of history of hypertension, presence of
cavernous malformations elsewhere in the brain on
standard magnetic resonance imaging, especially if
there are 5 lesions present suggesting familial cavern-
ous angiomatosis; prior haemorrhage in the same loca-
tion, or presence of developmental venous anomalies,
which are found in 30% of individuals with cavernous
malformations.27,28 Gradient echo magnetic resonance
imaging may also reveal microhaemorrhages in peri-
ventricular areas suggesting prior hypertensive ICH.29
Level of consciousness. Coma on admission is arguably
the most decisive prognosticating factor in patients
with ICH.30 The 20 outcome studies in PH used
either ‘coma on admission,’ or various GCS score cut-
offs – ranging from <4 to 9 – for prediction of 30-day
mortality.1,22–24,26 Irrespective of an arbitrary GCS
score threshold, the plurality of studies looking at var-
iables influencing outcomes in PH have consistently
identified depressed level of consciousness as the
104 European Stroke Journal 3(2)

Black dot indicates independent predictor of mortality (significant); dash indicates that the variable was not used in the multivariate analysis; NS, variable included in multivariate analysis, but the result was
Respiration
single, most reliable predictor of death and
disability.1,2,10,12,13,15–20,22–24 Eight studies, wherein
multivariate regression analysis was conducted using

NS








either admission coma or GCS score as variables

Pupillary
(a total of 968 patients) identified these factors

Hydrocephalus Hyperthermia Hyperglycaemia pressure reflex

as independent predictors of early mortality

NS









(Table 2).10,17,19,22–26
Combing the data from studies of PH patients that

Blood
stratified outcomes strictly based on “coma on

NS

NS
NS
NS

NS




admission” and not according to absolute GCS thresh-
olds yielded 585 patients.1,2,10,12,13,15,17,18 The PPV and
NPV, and PLR of coma on admission for early death
were 77.2% and 82.9%, and 3.98 (95% confidence

NS
interval (CI): 3.18–5.00), respectively.









Grading scales. The ICH score is a validated prognosti-
cation model in ICH, and its ability to predict 30-day
mortality also applies to PH.31–33 However, ICH score

NS

NS






may not be fully applicable in PH due to the small size

Table 2. Studies that conducted multivariate logistic regression to determine independent predictors of early death.
of the pons vis-à-vis its severe clinical manifestations in
case of haemorrhage, even in small volumes.26
Subsequently, other outcome prediction scales have
been proposed specifically for PH. Based on a recent

NS

NS







study of 171 primary PH patients, a grading scale has
been developed and validated for predicting 30-day Admission Admission Haemorrhage Haemorrhage Parenchymal Intraventricular
mortality and 90-day functional outcome.26 This scale Hemorrhage
used two independent factors: PH volume and GCS NS
NS
NS
NS

NS
NS
NS
score (Table 1).26 Point assignments according to this


scale were as follows: GCS score 3 to 4 (2 points), 5 to 7
(1 point) and 8 to 15 (0 point); PH volume >10 mL
extension

(2 points), 5 to 10 mL (1 point) and <5 mL (0 point).

NS

NS

NS
Estimated 30-day mortality rates for patients with





scores of 0, 1, 2, 3 and 4 were 2.7%, 31.6%, 42.7%,

81.8% and 100%, respectively (Table 3).26 The area
location

under the receiver operating characteristic curve

showed that that this scale was more discriminative
NS
NS

NS






than the ICH Score in predicting 30-day mortality

and 90-day good outcome.26 For prediction of death
in 30 days, the sensitivity of this grading system was
90.9% and the specificity was 63.1%.26 For 90-day
size

NS

NS

good outcome, the sensitivity and specificity were










87.8% and 77.6%, respectively.26

An earlier scale incorporated three factors for pre-
coma

dicting 30-day mortality: admission GCS score 6,











absence of pupillary light reflex, and serum glucose

180 – each assigned 1 point.23 The number of patients
in that study was 101.23 Thirty-day mortality rates were
Age GCS








7.7%, 33.3%, 78.9% and 100% for patients with 0, 1, 2

and 3 points, respectively.23 With regard to this scale,
NS

NS
NS








PH volume was dichotomised into <20 and 20 mL,

Matsukawa et al.22

which may be too high of a volume threshold for the

Dziewas et al.17

Morotti et al.25
Meguro et al.23

not significant.
Huang et al.26

finite size of the pons. This particular scoring system

Jung et al.19
Jang et al.10

Ye et al.24

also lacked external validation.

A study of 75 primary PH patients compared the per-
formances of the Acute Physiology and Chronic Health
Behrouz
Table 3. The primary pontine haemorrhage (PPH) score
developed by Huang et al.26,a
Feature
GCS score
3–4
5–7
8–15
Haemorrhage volume
>10 mL
5–10 mL
<5 mL
Score
0 2.7%
1 31.6%
2 42.7%
3 81.8%
4 100%
GCS: Glasgow Coma Scale.
a
Unit for haemorrhage volume is mL. Mortality is within 30 days.
Evaluation (APACHE) II, the Simplified Acute
Physiology Score (SAPS) II, and the ICH score in pre-
dicting mortality.34 Based on measurements using the
Hosmer-Lemeshow goodness-of-fit test for calibration
and the area under the receiver operating characteristic
curve for discrimination, SAPS II had the highest cali-
bration and APACHE II, the highest discrimination.34
The results of this comparative analysis, however, must
be viewed cautiously since SAPS II and APACHE II have
been designed to predict in-hospital death, while the ICH
Score is used to estimate mortality within 30 days.
Other factors. Despite age being a chief prognostic
factor in ICH, only one study identified it as an inde-
pendent predictor of 30-day mortality in PH.25,31
Several other physiological factors have been reported
in the literature as predictors of poor outcome in PH,
though they have not been examined or replicated in
successive investigations. History of hypertension, core
temperature 39 C, systolic blood pressure <100
mmHg, absent motor response, absent corneal or ocu-
locephalic reflexes, abnormal respiration, and need for
mechanical ventilation have all been reported in vari-
ous studies to independently correlate with death or
poor outcome.10,15,20,22,23 Sex was not a prognostic
factor in any of the studies.
Radiological predictors
Haemorrhage size. Haemorrhage size is an unswerving
outcome-determining factor in ICH, and PH is no
exception to this rule. Eight studies totalling 870
patients with PH demonstrated that haemorrhage
size – ascertained by either volume or trans-axial diam-
eter – is an independent predictor of mortality and
105
outcome (Table 2).10,17,19,22–26 In these studies, size
threshold values above which predicted death or unfav-
Point ourable outcome ranged between 4 and 5 mL for
volume, or 20 and 27 mm for trans-axial diame-
ter.10,17,22,24,26 As previously mentioned, the pons itself
2 constitutes a small dimensionality and therefore, the
1 volume factor on the ICH score inadequately applies
0
to PH. A haemorrhage volume of >30 mL in the pons
is materially impossible and would certainly have cata-
2
1 strophic consequences. Furthermore, the ABC/2 formula
0 used to measure ICH volume on computed tomography
mortality (CT) scan may overestimate the size of small haemor-
rhages in the pons.35,36 These concerns may lead to the
belief that trans-axial diameter is perhaps a more suitable
technique to quantify PH size.2 However, in terms of
prognostication, the longitudinal dimension of a brain-
stem lesion cannot be undervalued. Estimation of
volume using the ABC/2 equation takes both dimensions
into account and is subsequently the most justifiable
method to measure haemorrhage size, despite its pre-
sumed modest accuracy in PH. Three-dimensional
voxel-based magnetic resonance or pixel-based CT
images volumetry is a promising method that may be
used in precise measurement of PH volume.37,38
Haemorrhage localisation. Early small case series of
patients with PH reported relatively higher survival
rates for lateral haemorrhage localisation.8,14,39,40 In
1992, Chung and Park classified axial CT features of
primary PH into four types: basal-tegmental, bilateral
tegmental, massive and small unilateral tegmental.4
The massive type was defined as haematomas occupy-
ing both the basis pontis and the tegmentum bilateral-
ly, the bilateral tegmental type as those occupying the
bilateral tegmentum only, the basal-tegmental type as
those occupying the junction between the basis pontis
and the tegmentum bilaterally, and the small unilateral
tegmental type was defined as haematomas located
exclusively in the unilateral tegmentum (Figure 2).4
Case survival rate was 94.1% for the small unilateral
tegmental type, 26.19% for the basal-tegmental type,
14.3% for the bilateral tegmental type, and 7.1% for
the massive type.4 Thus, Chung and Park concluded
that small haematomas located in the unilateral teg-
mentum that had the best prognosis.4
In a study of 39 consecutive patients with PH,
Dziewas et al. divided trans-axial PH locations into
three categories: large paramedian, unilateral basoteg-
mental and lateral tegmental.17 The large paramedian
type of PH predicted a poor prognosis, whereas the
lateral tegmental type was associated with a favourable
outcome.17 The transverse haematoma diameter
threshold value related to outcome was found to be
20 mm.17 Another study of 29 primary PH patients
reported that dorsally-located small haemorrhages
106
Figure 2. Computed tomography classification of pontine haemorrhage localisations according to Chung and Park.4 From top left,
clockwise: massive, basal tegmental, bilateral tegmental and unilateral tegmental.
(<4 mL) had significantly better outcome than those
with ventrally-situated lesions.1 Survival rates for
dorsal, ventral and massive PH were 75%, 25% and
0%, respectively.1 It is important to note that axial
location of PH is also dependent on its size. The local-
isations described in the above-mentioned studies relate
to small haemorrhages. The largest outcomes study of
primary PH by Jang et al. which involved 281 patients,
reported 30-day mortality rates of 66.9%, 24.7% and
1.5% for massive, ventral and dorsal localisations,
respectively.10 Functional recovery rates at 90 days
were not different between ventral and dorsal haemor-
rhages (14.8% versus 14.9%).10 However, none of the
stated figures were stratified according to PH size.
Haemorrhage extension and growth
Extension of ICH into the ventricles is associated with
increased risk of death and adverse outcome.31,32 In a
total of 1009 PH patients from studies that reported on
intraventricular extension (IVE), the prevalence was
39.5%.1,2,10,15–19,22,25,26 Among seven studies that
included IVE as a prognostic variable in multivariate
logistic analysis, none could establish it as an indepen-
dent predictor of early death in PH.10,17,19,22,23,25,26
One study – the largest by Jang et al. – identified the
absence of IVE as an independent predictor of 90-day
functional recovery (mRS score, 0–3), with an odds
ratio of 3.64 (95% CI: 2.18–7.08; p ¼ 0.001).10 It did
not, however, find an association between IVE and
30-day mortality in PH.10 It is unclear why IVE is a
major prognostic factor in ICH and not in PH. One
theory may be the frequency of extraventricular drain-
age use.10 IVE in PH is more often localised to
the fourth ventricle and cerebral aqueduct and
European Stroke Journal 3(2)
subsequently, carry a higher risk of hydrocephalus.
Development of hydrocephalus will in turn increase
the use of extraventricular drainage. In ICH patients
with IVE, extraventricular drainage is associated
with reduced mortality and improved short-term
outcomes.41
Extrapontine extension into the midbrain with or
without thalamic involvement was shown by only one
study of 118 consecutive PH patients to be
independently associated with death within a median
follow-up period of 51 days (hazard ratio, 2.2; 95% CI:
1.1–4.4; p ¼ 0.033).22 Three other studies that examined
the effect of parenchymal extension on survival could
not confirm this finding.19,23,26
Haemorrhage expansion is an independent determi-
nant of mortality and functional outcome after ICH.42
A 1 mL increase in haematoma growth is associated
with a 5% higher risk of death or dependency.43 Very
few studies to date have specifically investigated the
impact of PH expansion in on outcomes. The CT
angiogram ‘spot sign’ is an independent predictor of
haematoma expansion, in-hospital mortality and poor
long-term outcome in supratentorial ICH.44 A recent
secondary analysis of Antihypertensive Treatment of
Acute Cerebral Hemorrhage II (ATACH-II) rando-
mised clinical trial, however, demonstrated that the
predictive performance of the spot sign for ICH expan-
sion was lower than in prior reports.45 In a retrospec-
tive analysis of 49 consecutive patients with primary
PH, the CT angiogram spot sign showed an accuracy
of 57% for 30-day mortality, and did not independent-
ly predict in-hospital or 30-day mortality.25 Moreover,
no significant difference was observed in haemorrhage
expansion rates between spot sign positive and negative
patients.25 Due to its size and the condensed functional
Behrouz
machinery it contains, it is conceivable that expansion
of the haemorrhage in the pons would have devastating
impact on survival and outcomes.
Hydrocephalus
Among 12 studies with a total of 1063 patients, 30.3%
developed hydrocephalus.1,2,10,15–22,25,26 Only one of
the three studies that included hydrocephalus as a
variable in multivariate analysis found it to be an
independent predictor of mortality and functional out-
come.10,19,26 This study was by Jang et al., which esti-
mated the odds ratios for 30-day mortality and 90-day
functional recovery (absence of hydrocephalus) at
2.98 (95% CI: 1.84–5.02; p ¼ 0.038) and 2.84 (95%
CI: 1.79–5.32; p ¼ 0.03), respectively.10
Discussion
Based on the available evidence, the overall all-cause
mortality rate for PH is roughly 50%. Level of con-
sciousness on admission and haemorrhage size are the
strongest and most consistent predictors of mortality in
patients with PH. Small PH with dorsal and lateral
localisations may have better survival rates, compared
to other localisations; certainly, more favourable prog-
nosis than massive PH. Age, IVE and hydrocephalus
were not consistently found to be strong prognostic
features. No study to date has investigated the effect
of haemorrhage growth specifically in PH.
Over the four decades in which these studies were
conducted, the survival of PH appears to have
improved. This may be partly due to advancement is
the fields of neurology, neurosurgery and intensive care
medicine. Improved public perception of ICH and
stroke in general may have contributed to this progress.
However, it is not clear if functional outcomes have
improved in parallel to survival.
There were a few limitations to this review. The
majority of the studies were retrospective and there
was significant heterogeneity in how data were collect-
ed and interpreted. These factors precluded the ability
to perform a meta-analysis. Studies spanned four dec-
ades and management of ICH has evolved over the
years. Therefore, survival and outcomes were certainly
affected by difference in management strategies
between older and more recent studies. Studies that
involved surgical evacuation and those older before,
1980 were excluded to ensure some level of consistency.
Furthermore, there may have been regional influences
on how PH patients were managed, especially in older
series when standardised guidelines were not available.
The variability in perception of ‘good outcome’ also
limited the ability to determine the frequency of
favourable functional outcome in patients with PH.
107
Other factors that may have limited this review were
heterogeneity in definitions and reporting of conditions
such as hydrocephalus, as well as missing data in indi-
vidual studies.
In conclusion, patients with PH do not seem to
invariably have poor outcomes. Limiting care due to
preconceived unfavourable prognosis in PH is not rea-
sonable. Although in most randomised trials, tailored
sample populations must be used to reasonably per-
form the study, omitting PH patients from clinical
trials will prevent establishing inclusive treatment strat-
egies for ICH. In order to better prognosticate PH and
understand this insidious condition, large, prospective
studies are certainly needed.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
Informed consent
None required; this is a review article.
Ethical approval
None required; this is a review article.
Guarantor
RB.
Contributorship
RB is the sole author. RB researched literature, wrote the
draft of the manuscript.
Acknowledgements
None.
References
1. Wessels T, M€ oller-Hartmann W, Noth J, et al. CT find-
ings and clinical features as markers for patient outcome
in primary pontine hemorrhage. AJNR Am J Neuroradiol
2004; 25: 257–260.
2. Murata Y, Yamaguchi S, Kajikawa H, et al. Relationship
between the clinical manifestations, computed tomo-
graphic findings and the outcome in 80 patients with
primary pontine hemorrhage. J Neurol Sci 1999; 167:
107–111.
3. Jeong JH, Yoon SJ, Kang SJ, et al. Hypertensive pontine
microhemorrhage. Stroke 2002; 33: 925–929.
4. Chung CS and Park CH. Primary pontine hemorrhage: a
new CT classification. Neurology 1992; 42: 830–834.
108
5. Qureshi AI, Palesch YY, Barsan WG, et al. ATACH-2
Trial investigators and the Neurological Emergency
Treatment Trials network. Intensive blood-pressure low-
ering in patients with acute cerebral hemorrhage. N Engl
J Med 2016; 375: 1033–1043.
6. Hanley DF, Lane K, McBee N, et al. CLEAR III
Investigators. Thrombolytic removal of intraventricular
haemorrhage in treatment of severe stroke: results of the
randomised, multicentre, multiregion, placebo-controlled
CLEAR III trial. Lancet 2017; 389: 603–611.
7. Baharoglu MI, Cordonnier C, Al-Shahi Salman R, et al.
Platelet transfusion versus standard care after
acute stroke due to spontaneous cerebral haemorrhage
associated with antiplatelet therapy (PATCH): a rando-
mised, open-label, phase 3 trial. Lancet 2016; 387:
2605–2613.
8. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA
statement for reporting systematic reviews and meta-
analyses of studies that evaluate health care interven-
tions: explanation and elaboration. PLoS Med 2009; 6:
e1000100.
9. Kushner MJ and Bressman SB. The clinical manifesta-
tions of pontine hemorrhage. Neurology 1985; 35:
637–643.
10. Jang JH, Song YG and Kim YZ. Predictors of 30-day
mortality and 90-day functional recovery after primary
pontine hemorrhage. J Kor Med Sci 2011; 26: 100–107.
11. Rabinstein AA, Tisch SH, McClelland RL, et al. Cause is
the main predictor of outcome in patients with pontine
hemorrhage. Cerebrovasc Dis 2004; 17: 66–71.
12. Nakajima K. Clinicopathological study of pontine hem-
orrhage. Stroke 1983; 14: 485–493.
13. Masiyama S, Niizuma H and Suzuki J. Pontine haemor-
rhage: a clinical analysis of 26 cases. J Neurol Neurosurg
Psychiatry 1985; 48: 658–662.
14. Weisberg LA. Primary pontine haemorrhage: clinical and
computed tomographic correlations. J Neurol Neurosurg
Psychiatry 1986; 49: 346–352.
15. Wijdicks EF and St Louis E. Clinical profiles predictive
of outcome in pontine hemorrhage. Neurology 1997; 49:
1342–1346.
16. Fong CS. Factors affecting the outcome in pontine hem-
orrhage. Acta Neurol Taiwan 1999; 8: 6–11.
17. Dziewas R, Kremer M, Lüdemann P, et al. The prognos-
tic impact of clinical and CT parameters in patients with
pontine hemorrhage. Cerebrovasc Dis 2003; 16: 224–229.
18. Balci K, Asil T, Kerimoglu M, et al. Clinical and neuro-
radiological predictors of mortality in patients with pri-
mary pontine hemorrhage. Clin Neurol Neurosurg 2005;
108: 36–39.
19. Jung DS, Jeon BC, Park YS, et al. The predictors of
survival and functional outcome in patients with pontine
hemorrhage. J Korean Neurosurg Soc 2007; 41: 82–87.
20. Shin SC, Lim DJ, Kim SD, et al. Primary pontine hem-
orrhage. An analysis of 35 cases and research in prognos-
tic factors. Kor J Cerebrovasc Surg 2007; 9: 41–45.
21. Nishizaki T, Ikeda N, Nakano S, et al. Factors determin-
ing the outcome of pontine hemorrhage in the absence of
surgical intervention. OJMN 2012; 2: 17–20.
European Stroke Journal 3(2)
22. Matsukawa H, Shinoda M, Fujii M, et al. Risk factors
for mortality in patients with non-traumatic pontine
hemorrhage. Acta Neurol Scand 2015; 131: 240–245.
23. Meguro T, Kuwahara K, Tomita Y, et al. Primary pon-
tine hemorrhage in the acute stage: clinical features and a
proposed new simple scoring system. J Stroke
Cerebrovasc Dis 2015; 24: 860–865.
24. Ye Z, Huang X, Han Z, et al. Three-year prognosis of
first-ever primary pontine hemorrhage in a hospital-
based registry. J Clin Neurosci 2015; 22: 1133–1138.
25. Morotti A, Jessel MJ, Brouwers HB, et al. CT angiogra-
phy spot sign, hematoma expansion, and outcome in pri-
mary pontine intracerebral hemorrhage. Neurocrit Care
2016; 25: 79–85.
26. Huang K, Ji Z, Sun L, et al. Development and validation
of a grading scale for primary pontine hemorrhage.
Stroke 2017; 48: 63–69.
27. Labauge P, Laberge S, Brunereau L, et al. Hereditary
cerebral cavernous angiomas: clinical and genetic features
in 57 French families. Société Française de
Neurochirurgie. Lancet 1998; 352: 1892–1897.
28. Rabinov JD. Diagnostic imaging of angiographically
occult vascular malformations. Neurosurg Clin N Am
1999; 10: 419–432.
29. Campbell PG, Jabbour P, Yadla S, et al. Emerging clin-
ical imaging techniques for cerebral cavernous malforma-
tions: a systematic review. Neurosurg Focus 2010; 29: E6.
30. Parry-Jones AR, Abid KA, Di Napoli M, et al. Accuracy
and clinical usefulness of intracerebral hemorrhage grad-
ing scores: a direct comparison in a UK population.
Stroke 2013; 44: 1840–1845.
31. Hemphill JC, 3rd, Bonovich DC, Besmertis L, et al. The
ICH score: a simple, reliable grading scale for intracere-
bral hemorrhage. Stroke 2001; 32: 891–897.
32. Hemphill JC, 3rd, Farrant M, Neill TA. Jr. Prospective
validation of the ICH Score for 12-month functional out-
come. Neurology 2009; 73: 1088–1094.
33. Del Brutto OH and Campos X. Validation of intracere-
bral hemorrhage scores for patients with pontine hemor-
rhage. Neurology 2004; 62: 515–516.
34. Huang KB, Ji Z, Wu YM, et al. The prediction of 30-day
mortality in patients with primary pontine hemorrhage: a
scoring system comparison. Eur J Neurol 2012; 19:
1245–1250.
35. Kothari RU, Brott T, Broderick JP, et al. The ABCs of
measuring intracerebral hemorrhage volumes. Stroke
1996; 27: 1304–1305.
36. Huttner HB, Steiner T, Hartmann M, et al. Comparison
of ABC/2 estimation technique to computer-assisted pla-
nimetric analysis in warfarin-related intracerebral paren-
chymal hemorrhage. Stroke 2006; 37: 404–408.
37. Strik HM, Borchert H, Fels C, et al. Three-dimensional
reconstruction and volumetry of intracranial haemor-
rhage and its mass effect. Neuroradiology 2005; 47:
417–424.
38. Scherer M, Cordes J, Younsi A, et al. Development and
validation of an automatic segmentation algorithm for
quantification of intracerebral hemorrhage. Stroke
2016; 47: 2776–2782.
Behrouz
39. Caplan LR and Goodwin JA. Lateral tegmental brain-
stem hemorrhages. Neurology 1982; 32: 252–260.
40. Del-Brutto OH, Noboa CA and Barinagarrementerı́a F.
Lateral pontine hemorrhage: reappraisal of benign cases.
Stroke 1987; 18: 954–956.
41. Herrick DB, Ullman N, Nekoovaght-Tak S, et al.
Determinants of external ventricular drain placement
and associated outcomes in patients with spontaneous
intraventricular hemorrhage. Neurocrit Care 2014; 21:
426–434.
42. Davis SM, Broderick J, Hennerici M, et al. Recombinant
activated factor VII intracerebral hemorrhage trial inves-
tigators. Hematoma growth is a determinant of mortality
and poor outcome after intracerebral hemorrhage.
Neurology 2006; 66: 1175–1181.
43. Delcourt C, Huang Y, Arima H, et al. INTERACT1
investigators. Hematoma growth and outcomes in
109
intracerebral hemorrhage: the INTERACT1 study.
Neurology 2012; 79: 314–319.
44. Demchuk AM, Dowlatshahi D, Rodriguez-Luna D, et al.
PREDICT/Sunnybrook ICH CTA study group.
Prediction of haematoma growth and outcome in
patients with intracerebral haemorrhage using the CT-
angiography spot sign (PREDICT): a prospective obser-
vational study. Lancet Neurol 2012; 11: 307–314.
45. Morotti A, Brouwers HB, Romero JM, et al.
Antihypertensive Treatment of Acute Cerebral
Hemorrhage II and Neurological Emergencies
Treatment Trials investigators. Intensive blood pressure
reduction and spot sign in intracerebral hemorrhage: a
secondary analysis of a randomized clinical trial.
JAMA Neurol 2017; 74: 950–960.