Editor:

OPEN ACCESS C. David Hunt, M.D.

SNI: General Neurosurgery For entire Editorial Board visit : Marquette General
http://www.surgicalneurologyint.com Neurosurgery, Brooklyn,
NY, USA

Original Article
Treatment of intracerebellar haemorrhage: Poor outcome and
high long‑term mortality
Jarno Satopää1, Atte Meretoja2,6, Riku J. Koivunen2,5, Satu Mustanoja2, Jukka Putaala2, Markku Kaste2,
Daniel Strbian2, Turgut Tatlisumak2,3,4, Mika R. Niemelä1
Departments of 1Neurosurgery, 2Neurology, Helsinki University Hospital and Clinical Neurosciences, Neurology, University of Helsinki, Helsinki,
Finland, 3Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy at University of Gothenburg, Gothenburg,
4
Department of Neurology, Sahlgrenska University Hospital, Gothenburg, Sweden, 5Department of Anesthesiology, Päijät‑Häme Central Hospital, Lahti, Finland,
6
Department of Medicine at the Royal Melbourne Hospital, University of Melbourne, Parkville, VIC, Australia

E‑mail: *Jarno Satopää ‑ [email protected]; Atte Meretoja ‑ [email protected]; Riku J. Koivunen ‑ [email protected];

Satu Mustanoja ‑ [email protected]; Jukka Putaala ‑ [email protected]; Markku Kaste ‑ [email protected]; Daniel Strbian ‑ [email protected];
Turgut Tatlisumak ‑ [email protected]; Mika R. Niemelä ‑ [email protected]
*Corresponding author

Received: 03 May 17   Accepted: 22 July 17   Published: 09 November 17

Abstract
Background: Intracerebellar haemorrhage constitutes around 10% of all
spontaneous, non‑aneurysmal intracerebral haemorrhages (ICHs) and often carries
a grim prognosis. In symptomatic patients, surgical evacuation is usually regarded
the standard treatment. Our objective was to compare the in‑hospital mortality
and functional outcome at hospital discharge in either medically or surgically
treated patients, and the impact of either treatment on long‑term mortality after a
cerebellar ICH.
Methods: An observational, retrospective, single‑centre consecutive series of
114 patients with cerebellar ICH. We assessed the effect of different demographic
factors on functional outcome and in‑hospital mortality using logistic regression.
We also divided the patients in medical and surgical treatment groups based on
how they had been treated and compared the clinical and radiological parameters, Access this article online
in‑hospital, and long‑term mortality in the different groups. Website:
www.surgicalneurologyint.com
Results: In our series, 38 patients (33.3%) underwent haematoma evacuation
DOI:
and 76 (66.7%) received medical treatment. Glasgow coma scale <8, blocked 10.4103/sni.sni_168_17
quadrigeminal cistern, and severe hydrocephalus were associated with in‑hospital Quick Response Code:
death or poor functional outcome at discharge  (modified Rankin scale 4‑6).
Surgically treated patients were younger, had larger haematomas both in
volume and diameter, were in a worse clinical condition, and suffered more from
hydrocephalus and brainstem compression. There were no statistically significant
differences in in‑hospital or long‑term mortality. However, the surgically treated
patients remained in a poor clinical condition.

This is an open access article distributed under the terms of the Creative Commons Attribution‑NonCommercial‑ShareAlike 3.0 License, which allows others to remix, tweak, and
build upon the work non‑commercially, as long as the author is credited and the new creations are licensed under the identical terms.
For reprints contact: [email protected]

How to cite this article: Satopää J, Meretoja A, Koivunen RJ, Mustanoja S, Putaala J, Kaste M, et al. Treatment of intracerebellar haemorrhage: Poor outcome and high long-term
mortality. Surg Neurol Int 2017;8:272..
http://surgicalneurologyint.com/Treatment-of-intracerebellar-haemorrhage:-Poor-outcome-and-high-long‑term-mortality/

© 2017 Surgical Neurology International | Published by Wolters Kluwer - Medknow

Surgical Neurology International 2017, 8:272 http://www.surgicalneurologyint.com/content/8/1/272

Conclusions: Surgical treatment of cerebellar ICH can be life‑saving but often

leads to a poor functional outcome. New studies are needed on long‑term functional
outcome after a cerebellar ICH.

Key Words: Cerebellum, intracerebral haemorrhage, long‑term mortality, outcome

INTRODUCTION patient contact or consent, and thus all consecutive cases

could be included in the study.
Intracerebral hemorrhage (ICH) in the cerebellum
Immediately on arrival, all patients underwent a
constitutes around 10% of all “spontaneous”
computerized tomography (CT) and a CT angiography
nontraumatic, nonaneurysmal ICHs.[8] Since the
to exclude aneurysms, AVMs, dural fistulae, and tumors.
1970s, there has been a wide mutual consensus in
Glasgow coma scale (GCS) was assessed on arrival. Based
the neurological and neurosurgical community that
on how they had been treated, we divided the patients
cerebellar ICHs should be operated on. However, the with a cerebellar ICH in a conservative and a surgical
scientific proof is mainly based on small retrospective treatment group for analyses. As there are no decisive
series with conflicting results.[28] Hemorrhage in the guidelines for the treatment of cerebellar ICH, indications
posterior fossa may cause brainstem compression and for surgery were not guideline‑ or protocol‑based but
lead to impaired consciousness, respiratory failure, determined case‑by‑case on an individual basis, weighing
and lower cranial nerve dysfunction. Compression or the hematoma size, declining level of consciousness,
obstruction of the fourth ventricle may cause acute age, and possible comorbidities. As a general rule,
hydrocephalus.[7] To relieve brainstem compression and patients with a symptomatic cerebellar ICH and
hydrocephalus, surgeons tend to favor posterior fossa declining level of consciousness were operated on, unless
craniectomy or craniotomy with hematoma evacuation the clinical situation deemed unsalvageable. Deeply
in patients with a declining level of consciousness.[27] unconscious patients with signs of midbrain or brainstem
Some regard this counterintuitive as long‑term outcomes dysfunction (unreactive, bilaterally dilated or miotic
after surgical treatment of cerebellar ICH are generally pupils, or absent corneal and oculocephalic reflexes),
pessimistic.[17] very high age, or severe comorbidities were treated
Our aim was to observe the short‑term functional medically. Both patient groups were treated in a specialist
outcome and long‑term mortality in medically and neurosurgical intensive care unit (ICU) or neurological
surgically treated patients in a consecutive single‑centre acute stroke unit, with two exceptions – patients with
series. We also searched for those demographic and a small hematoma and excellent clinical condition were
radiological factors that led to poor functional outcome treated in the neurological or neurosurgical wards. Deeply
or death [modified Rankin Scale (mRS) 4–6] at hospital unconscious patients with an extensive hemorrhage
discharge. were also treated in neurological or neurosurgical wards
after a decision to withhold treatment. Surgically
MATERIALS AND METHODS treated patients underwent a suboccipital paramedian or
midline craniotomy and the hematoma was evacuated
We performed a retrospective analysis of 1075 in a standard microneurosurgical manner. One patient
consecutive patients with spontaneous nonaneurysmal underwent a primary bilateral decompressive craniectomy
ICHs presenting at the neurological and neurosurgical of the posterior fossa and subsequent hematoma
emergency departments of our institute between evacuation. All patients underwent a postoperative
January 2005 and March 2010. The patients were CT scan on the first postoperative day to confirm that
followed until June 2016. The catchment population is brainstem compression and hydrocephalus had relieved.
1.8 million and the hospital has the only neurological Reoperations were recorded.
and neurosurgical emergency rooms on a 24/7 basis If symptomatic hydrocephalus was present, patients
in the province. Patients with ICHs originating from received an external ventricular drain (EVD). Patients
arteriovenous malformations (AVMs), dural fistulae, or with EVD but no surgical evacuation of the haematoma
tumors were excluded from the study. The imaging data were included in the conservative, i.e., medical treatment
was reviewed by a neuroradiologist, and the patients with group. Level of consciousness was recorded on the GCS.
an intracerebellar hemorrhage were identified. All data The GCS verbal score for intubated patients was derived
were collected retrospectively from charts, electronic using the method by Meredith.[19] Unconscious patients
patient records, and imaging archives. Mortality data was were intubated and mechanically ventilated. Propofol
collected from the National Death Registry. Institutional was used for sedating the intubated patients, interrupted
authorities approved the study as a registry study with no every 4 hours to check for the level of consciousness.
Surgical Neurology International 2017, 8:272 http://www.surgicalneurologyint.com/content/8/1/272

Systolic blood pressure (BP) was kept between 120 and cerebellar ICH. They were aged from 30 to 94 years with
160 mmHg, controlled with labetalol or clonidine, or a mean age of 68.3 years. Thirty‑eight patients (33.3%)
noradrenalin in hypotensive patients. Electrolyte levels were operated on and 76 (66.7%) were treated medically,
and blood gases were analyzed every 6 hours. In case i.e., conservatively. After the primary craniotomy and
of paralysis of the lower cranial nerves and dysphagia, hematoma evacuation, two patients had to undergo a
patients received a percutaneous tracheostomy bedside at secondary decompression of the posterior fossa to relieve
the ICU. Length of ICU and stroke unit treatment were brainstem compression. Three patients underwent
recorded, as well as length of hospital stay until death or re‑evacuation because of a clinically significant residual
discharge to a rehabilitation unit or home. hematoma in the postoperative scans. The reoperation
On imaging studies, we assessed the maximal diameter, rate was 13.2%.
volume, location, and vascular territory of the hematoma. Clinical and radiological data
ICH volume was calculated using the ABC/2 method.[15] The surgically treated patients were significantly younger
In addition, we classified hydrocephalus on a four‑grade and suffered significantly less often from atrial fibrillation
scale by Stein et al.,[22] quadrigeminal cistern obliteration and hypertension. There were no other statistically
on a three‑grade scale by Taneda et al.,[24] and IV ventricle significant differences in comorbidities or medications,
compression on a three‑grade scale by Kirollos et al.[13] and the premorbid functional status in mRS was similar
We used “tight posterior fossa” as a sign of brainstem between the treatment groups [Table 1].
compression classified as the effacement of the basal
cisterns combined with obstructive hydrocephalus.[26] The surgically treated patients were in a significantly
worse clinical condition on arrival. Their ICHs were larger
As primary outcomes, we compared the in‑hospital both in volume and diameter. In addition, they had a
mortality, the functional outcome of medically and greater rate of hydrocephalus, quadrigeminal cistern
surgically treated patients at hospital discharge, and the effacement, and fourth ventricle compression. We used
impact of the chosen treatment on long‑term mortality. the criteria for “tight posterior fossa” to indirectly assess
We used the modified Rankin scale (mRS) 4–5 to classify brainstem compression, and the rate was greater in the
poor functional outcome. In addition, we compared the surgically treated patients [Table 2]. The hemorrhages in
demographics and radiological factors in surgically and the medically treated group were more frequently located
medically treated patients to look for group differences. in the anterior cerebellar lobe, in the AICA territory, and
As a secondary outcome, we also assessed the effect extended to the brainstem more frequently than in the
of different demographic and radiological factors on surgically treated patients (Online supplement).
functional outcome and in‑hospital mortality.
Treatment
Statistical analyses were carried out on SPSS 24.0 (IBM
The surgically treated patients stayed longer in the
Corp., Armonk, NY). The ordinal and continuous
intensive care and acute stroke units. Their total
variables were tested for normality. We used the
length of hospital stay was also longer. The rates
Kruskal–Wallis and Mann–Whitney U tests for skewed
of EVD and tracheostomy were also higher in the
and one‑way analysis of variance (ANOVA) for normally
surgically treated patients [Table 3]. The mean interval
distributed data. Pearson’s Chi‑square test was used
to compare categorical variables between the groups.
Differences in long‑term mortality between the groups Table 1: Demographics and comorbidities in the surgically
were assessed with Kaplan–Meier analysis. To assess the and medically treated patients
effect of categorical variables on functional outcome Medical Surgical P
at discharge, we used binomial logistic regression to treatment treatment
calculate the odds ratios. We used the Firth penalized (n=76) (n=38)
binomial logistic regression to discard separation in the Male gender, n (%) 45 (59.2%) 22 (57.9%) 0.893
standard maximum likelihood analyses. All deaths were Age in years, mean (95% CI) 71.6 (68.9‑74.4) 61.7 (57.6‑65.8) <0.001
included regardless of the cause. A two‑sided P < 0.05 Atrial fibrillation 16 (20.8%) 0 0.03
was considered significant. Hypertension 54 (71.1%) 18 (47.4%) 0.013
Diabetes 17 (22.4%) 3 (7.9%) 0.055
RESULTS Liver disease 5 (6.6%) 2 (5.4%) 0.808
Heavy drinking 7 (13.5%) 5 (19.2%) 0.506
There were 1075 consecutive patients with a Warfarin 12 (15.8%) 3 (7.9%) 0.24
spontaneous, nonaneurysmal ICH between January Any antiplatelet drug 20 (26.7%) 7 (18.4%) 0.332
2005 and March 2010 presenting at neurological Statins 12 (16.2%) 5 (13.2%) 0.669
and neurosurgical emergency rooms at our institute. mRS at arrival, median (IQR) 0 (0) 0 (0) 0.296
Of these, 114 (10.9%, 47 females) presented with a CI: Confidence interval, mRS: Modified rankin scale, IQR: interquartile range
Surgical Neurology International 2017, 8:272 http://www.surgicalneurologyint.com/content/8/1/272

Table 2: Clinical and radiological characteristics at arrival in the surgically and medically treated patients. We used
Pearson’s Chi square test for dichotomous variables (*), one‑way analysis of variance (ANOVA) for normally distributed
continuous variables (†), and Mann‑Whitney U test for skewed ordinal and continuous variables (‡)
Medical treatment (n=76) Surgical treatment (n=38) P
GCS on arrival, median (IQR) 15 (8‑15) 6 (3‑14) <0.001‡
Systolic BP on arrival, mean (95% CI) 174 (166‑182) 188 (174‑202) 0.065†
Haematoma volume in ml, mean (95% CI) 12.7 (9.9‑15.5) 30.4 (25.3‑35.4) <0.001‡
Maximal diameter in mm, mean (95% CI) 36.4 (33.4‑39.5) 48.0 (43.9‑52.0) <0.001†
Intraventricular blood, n (%) 37 (48.7%) 27 (71.1%) 0.023*
Hydrocephalus, n (%) <0.001‡
No 36 (47.4%) 0
Beginning 14 (18.4%) 11 (28.9%)
Moderate 16 (21.1%) 16 (42.1%)
Severe 10 (13.2%) 11 (28.9%)
4 ventricle, n (%)
th
<0.001‡
Open 18 (23.7%) 0
Compressed 25 (32.9%) 5 (13.2%)
Blocked 33 (43.4%) 33 (86.8%)
Quadr cistern, n (%) <0.001‡
Open 42 (55.3%) 2 (5.3%)
Compressed 20 (26.3%) 20 (52.6%)
Blocked 14 (18.4%) 16 (42.1%)
Brainstem compression, n (%) 24 (31.6%) 35 (92.1%) <0.001*
Hematoma extension to brainstem 9 (11.8%) 1 (2.6%) 0.101*
GCS: Glasgow coma scale, IQR: Interquartile range, BP: Blood pressure, CI: Confidence interval

Table 3: Treatment in the surgically and medically In further analyses, we found that all patients with arrival
treated patients GCS <13 had a poor outcome (mRS: 4–6) and the
Medical Surgical P in‑hospital mortality in this group was 50% (P < 0.001).
treatment (n=76) treatment (n=38) Twenty patients were aged over 80, and 18 (90%)
External ventricular 5 (6.6%) 24 (63.2%) <0.001
were treated medically (P = 0.019). Only two of these
drainage, n (%) 20 patients had a favorable outcome (10%), both
Tracheostomy, n (%) 2 (2.6%) 24 (63.2%) <0.001 treated medically (P = 1). Ten patients had hemorrhage
Treated in ICU or acute 42 (55.3%) 35 (92.1%) <0.001 extending into the brainstem, and 90% of these died in
stroke unit, n (%) hospital [Table 4].
ICU and acute stroke 1 (0‑4) 8 (5‑11) <0.001 In addition, we used univariate binomial logistic
unit days, median (IQR) regression to assess the effect of individual variables
Hospital days, 6.5 (2‑13) 15.5 (10‑26) <0.001 on poor functional outcome or death (mRS: 4–6)
median (IQR) at hospital discharge. Poor outcome was most
ICU: Intensive care unit, IQR: Interquartile range
strongly associated with (1) GCS <8, (2) blocked
quadrigeminal cistern, and (3) severe hydrocephalus
between hospital admission and surgery was 22.7 hours [Table 4 and Supplementary Tables 1 and 2].
(95% CI: 1.2–44.2 hours). On 11 patients, the EVD
was inserted before surgery, and the mean interval We included the same variables in a multivariable logistic
between hospital admission and EVD insertion was 24.7 regression model and found that only two variables had a
hours (95% CI: 6–43.4 hours). statistically significant effect on poor functional outcome
or death (mRS: 4–6) at hospital discharge: (1) GCS <8
Mortality and factors leading to poor outcome and (2) age group 65–79 years were associated with poor
The in‑hospital mortality of all patients with outcome [Table 5].
cerebellar ICH was 27.2% (31 patients). Long‑term
mortality was 49.1% with a mean survival of 5.5 years Short‑ and long‑term outcomes after surgical or
(95% CI: 4.5–6.6 years). In total, 76.3% (87 patients) conservative treatment
either died in hospital or had a poor functional The in‑hospital mortality was 31.6% (24 patients) in the
outcome (mRS: 4–6) at hospital discharge. medically and 18.4% (7 patients) in the surgically treated
Surgical Neurology International 2017, 8:272 http://www.surgicalneurologyint.com/content/8/1/272

Table 4: Results of univariate logistic regression on Table 5: The effect of demographic and radiological
poor functional outcome and mortality (modified Rankin factors on poor functional outcome or mortality at
Scale 4‑6) discharge (modified Rankin Scale 4‑6) in multivariable
Variable Good Poor Odds 95% CI P logistic regression
outcome, outcome, ratio Variable Odds ratio 95% CI P
n (%) n (%) Age group
Age group Under 50 reference
Under 50 3 (27.3%) 8 (72.7%) Ref. 50‑64 1.647 0.064‑626 0.765
50‑64 13 (39.4%) 20 (60.6%) 0.625 0.134‑2.46 0.511 65‑79 113.2 2.27‑12700000 0.012
65‑79 9 (18.0%) 41 (82.0%) 1.799 0.385‑7.24 0.432 Over 80 48.04 0.812‑6200000 0.067
Over 80 2 (10%) 18 (90%) 3.046 0.498‑21.5 0.225 GCS at arrival
GCS at arrival 13‑15 reference
13‑15 27 (42.2%) 37 (57.8%) Ref. 8‑12 9.767 0.345‑2180 0.175
9‑12 0 6 (100%) 11 1.25‑1450 0.027 3‑7 366.1 1.84‑3270000 0.023
3‑8 0 44 (100%) 63.81 8.41‑8193 <0.001 Ventricular blood 11.31 0.838‑3530 0.073
Comorbidities Brainstem compression 6.586 0.293‑3230 0.236
Hypertension 18 (25.0%) 54 (75.0%) 0.835 0.331‑2.01 0.692 Volume >10 ml 0.399 0.017‑5.6 0.496
Diabetes 8 (40%) 12 (60%) 0.380 0.14‑1.06 0.065 Diameter >3 cm 3.145 0.337‑52.9 0.317
Atrial fibrillation 4 (25.0%) 12 (75.0%) 0.876 0.286‑3.14 0.826 Extension to brainstem 5.094 0.045‑5550 0.51
Liver disease 2 (28.6%) 5 (71.4%) 0.689 0.155‑4.01 0.646 Hydrocephalus
Ventricular blood 5 (7.8%) 59 (92.2%) 8.542 3.23‑26.3 <0.001 No reference
Brainstem 3 (5.1%) 56 (94.9%) 12.55 4.21‑49.7 <0.001 Beginning 5.589 0.401‑2280 0.211
compression Moderate 8.82 0.441‑799 0.167
Volume >10 ml 9 (12.9%) 61 (87.1%) 4.517 1.87‑11.6 0.001 Severe 90.83 0.853‑1580000 0.06
Diameter >3 cm 11 (13.1%) 73 (86.9%) 7.272 2.89‑19.2 <0.001 4th ventricle
Extension to 0 10 (100%) 7.448 0.905‑967 0.066 Open reference
brainstem Compressed 3.939 0.51‑51.5 0.202
Hydrocephalus Blocked 0.3738 0‑17.2 0.631
No 20 (55.6%) 16 (44.4%) Ref. Quadrigeminal cistern
Beginning 5 (20%) 20 (80%) 4.632 1.55‑15.7 0.005 Open reference
Moderate 2 (6.3%) 30 (93.8%) 15.17 4.15‑82.6 <0.001 Compressed 1.068 0.063‑100 0.965
Severe 0 21 (100%) 53.41 6.46‑6974 <0.001 Blocked 1.635 0.008‑2400 0.851
4th ventricle Surgical treatment 1.98 0.098‑1010 0.662
Open 22 (50%) 22 (50%) Ref. CI: Confidence interval, GCS: Glasgow Coma Scale

Compressed 5 (12.5%) 35 (87.5%) 5.557 1.66‑20.9 0.005

Blocked 0 30 (100%) 27.44 7.87‑113 <0.001
Quadrigeminal
cistern
Open 13 (72.2%) 5 (27.8%) Ref.
Compressed 9 (30%) 21 (70%) 6.456 2.34‑20.6 <0.001
Blocked 5 (7.6%) 61 (92.4%) 61.01 7.68‑7903 <0.001
Surgical treatment 4 (10.5%) 34 (89.5%) 3.209 1.16‑10.9 0.024
CI: Confidence interval, GCS: Glasgow coma scale

patients, however, the difference was not statistically

significant (P = 0.137). Long‑term mortality was 50%
in the medically and 47.4% in the surgically treated
patients (P = 0.791). The mean follow‑up length
was 27.3 months (95% CI: 20.7–33.8 months) in the
medically and 45.7 months (95% CI: 31.9–59.6 months)
Figure 1: Kaplan–Meier analysis of the long‑term mortality in
in the surgically treated patients. The difference was
the different treatment groups. The difference was statistically
statistically significant (P = 0.005). In Kaplan–Meier insignificant (P = 0.184)
analysis, we did not find a statistically significant
difference in the long‑term mortality between the However, the clinical outcome of the surviving patients at
treatment groups (P = 0.184) [Figure 1]. hospital discharge was significantly worse in the surgically
Surgical Neurology International 2017, 8:272 http://www.surgicalneurologyint.com/content/8/1/272

surgical treatment group. However, the operated patients

more often remained in a poor clinical condition. Sixty
percent of the surviving surgically treated patients needed
a tracheostomy because of lower cranial nerve palsy.
Unfortunately, we do not have data on the duration of
tracheostomy treatment.
Cerebellar ICH is usually considered an especially
deadly stroke subtype because of frequent brainstem
compression and high risk of obstructive hydrocephalus.
Before CT became widely available, cerebellar ICH
was a silent killer – patients deteriorated rapidly and
the diagnosis was often made postmortem.[7] In 1978,
Little et al. showed in a case series of 10 patients that
patients with cerebellar ICHs over 3 cm in diameter
Figure 2: Functional outcome of the surviving patients (n = 30 in benefited from surgical evacuation, whereas smaller could
operative group, n = 53 in conservative group) at hospital discharge. be treated medically.[16] Since then, surgical hematoma
The error bars represent 95% confidence intervals.The difference evacuation or posterior fossa decompression has been the
was statistically significant (P < 0.001)
standard treatment in cerebellar ICH. Even today, due
to the low incidence, literature on cerebellar ICH is still
treated group – the median mRS with interquartile ranges based on small, retrospective single‑centre cohorts.[28]
were 4 (3–4) in the medically and 5 (4–5) in the surgically
treated group [Figure 2]. The difference was statistically The current recommendations for operative treatment
significant (P < 0.001). The incidence of lower cranial vary in the literature. Numerous different criteria have
nerve palsy and tracheostomy in the surviving patients was been suggested for surgical evacuation or decompression:
63.2% in the surgically and 2.6% in the medically treated hematoma diameter >3 cm,[16] patients with brainstem
group (P < 0.001). In addition, 69.3% (53 patients) of the compression,[5,26] patients with GCS 4–13 and hematoma
conservatively and 89.5% (34 patients) of the surgically diameter >40 mm,[14] patients with a deteriorating level
treated patients died in the hospital or had a poor of consciousness and hematoma volume >10 cm3,[9]
functional outcome (mRS: 4–6) (P = 0.019). and even operating all patients with a cerebellar ICH
has been suggested.[2] Some authors do not recommend
DISCUSSION surgical treatment at all because of poor long‑term
results.[4] It has been shown that clinicians usually tend
In this observational study of 114 consecutive cerebellar to wait until clinical deterioration, and then proceed to
ICH patients, we compared the outcome of medically and operative treatment, especially in patients under 70 years
surgically treated patients. Not surprisingly, the groups old.[27] The current treatment guidelines have contrasting
recommendations regarding treatment of cerebellar
had significant differences. The surgically treated patients
ICH. European Stroke Organisation (ESO) treatment
were younger, had larger hemorrhages both in volume and
guidelines did not find strong evidence on when, how,
diameter, were in a worse clinical condition, and suffered
or to whom surgical evacuation or decompression should
more from hydrocephalus and brainstem compression.
be performed.[23] American Heart Association (AHA)
Majority of the surgically treated patients were first
guidelines recommend evacuation of cerebellar ICH
observed before proceeding to surgery, the mean interval
in patients declining neurologically or suffering from
between hospital admission and surgery being 22.7 hours.
brainstem compression and/or acute hydrocephalus.[10]
In addition, older patients with numerous comorbidities
were more frequently treated medically. Published results on long‑term outcome after cerebellar
ICH are rare. In Uppsala, Sweden, 62% of the patients
Low admission GCS, increasing age, and severe
had either died or had an unfavorable functional
hydrocephalus were associated with in‑hospital mortality
outcome in long‑term (70 months) follow‑up.[25]
or poor functional outcome. Due to the generally
Dolderer et al. suggested that surgical treatment was an
accepted tendency for surgical treatment in symptomatic
independent predictor for poor functional outcome in
patients with a declining level of consciousness, there was
long‑term (49 months) follow‑up and we made the same
a strong selection bias towards patients with larger ICHs
observation.[4] In their series, only 14% of the surgically
and more severe symptoms in the surgical group.
treated patients had a favorable outcome. Pong showed
There were no statistically significant differences in that 56.7% of surviving patients had an unfavorable
neither in‑hospital nor long‑term mortality, although we outcome (modified Rankin scale >2) at 6 months.[21]
saw a nonsignificant trend towards better survival in the However, some authors have observed good long‑term
Surgical Neurology International 2017, 8:272 http://www.surgicalneurologyint.com/content/8/1/272

results after cerebellar ICH.[2,3] In our series, long‑term the anterior and posterior cerebellar lobes. The same
mortality was 50% in the medically and 47.4% in the applied when considering the location of the hematoma
surgically treated patients, with a mean survival time of in the cerebellar hemisphere vs. vermis. These observed
5.5 years. The high long‑term mortality in the surgically effects were probably mainly caused by the greater
treated patients may be caused by the high incidence hematoma volume in both situations.
of lower cranial nerve dysfunction at discharge and
We detected a trend, although not statistically significant,
subsequent risk of aspiration and pneumonia, at least
toward lower in‑hospital mortality in the surgically
partially. treated patients. However, operated patients who survived
We used mRS 4–5 as a criterion for poor functional were more often left in a poor condition. In addition,
outcome at discharge because of the known severe the long‑term mortality did not differ between the
morbidity of the surviving patients.[21] The incidence of medically and surgically treated patients. Unfortunately,
good functional outcome at discharge was very low, 30.7% we do not have long‑term data on functional outcome
in the medically and even lower, 10.5% in the surgically of the surviving patients. There are very few reports on
treated patients. Age was associated with poor functional functional recovery after cerebellar ICH. It has been
outcome or death in the multivariable logistic regression shown that moderately disabled patients with cerebellar
in the age group of 65–80 years, but not in other age ischemic stroke recovered well, but a poor clinical
groups. This may have been caused by referral bias; some condition led to a poor rehabilitation outcome.[12]
elderly patients may not have been referred to our tertiary Our series is the largest consecutive single‑centre
care centre. However, this probably does not severely affect cohort published and it includes all patients with a
the findings, as the majority of patients over 80 years were spontaneous cerebellar ICH treated at our institute
treated conservatively and 90% died in hospital or had a within a 5‑year period. However, due to the observational
poor outcome. In our series, GCS <13 on admission lead and retrospective nature of the study, there may be some
to a uniformly poor outcome, which was in accordance selection bias and inconsistency in the data. As there are
with previous data.[3,4,14,25] ICH volume was correlated with no good guidelines, no protocol could be used to decide
bad outcome in univariate analyses.[1,4,14] We have earlier who should receive surgical treatment and who should
shown a similar association in a prospective randomized not. Instead, the decisions were made case‑by‑case. The
trial of early surgery vs. conservative treatment in patients study was carried out in a tertiary level centre, and some
with a deep hemispheric ICH.[11] In the present study, patients may have been left in secondary hospitals due
prior medication[6] or comorbidities[2] did not affect the to suspected grim prognosis or died before arrival to the
outcome. hospital, especially in the oldest age group. Unfortunately,
As previously suggested, hydrocephalus was associated we do not have long‑term functional outcome data.
with poor functional outcome or death in univariate One of the most important findings in our study
analyses,[20,25] although it did not reach statistical were the major differences between the surgically and
significance in the multivariate model. We used the medically treated patients. By using the current widely
criteria for “tight posterior fossa” (effacement of the acknowledged criteria for surgical treatment (hemorrhage
basal cisterns combined with obstructive hydrocephalus) diameter >30 mm, declining level of consciousness,
as an indirect measure of brainstem compression.[26] It brainstem compression and/or hydrocephalus), all but
was associated with poor outcome in univariate analyses, the unsalvageable patients and the ones with a minor
but not in the multivariable model.[3,5] Degree of fourth hemorrhage will undergo surgery. Consequently, the
ventricle obstruction[13] and quadrigeminal cistern conservatively treated patients are healthier and less
obliteration[24] were also associated with poor outcome or symptomatic, as severe symptoms and large hemorrhage
mortality at discharge in the univariate analyses, but not volume lead to surgical treatment. Alas, it is rather
in the multivariable model. Fourth ventricle obstruction is impossible to gather a study population with two similar
usually considered the root cause of acute hydrocephalus groups of patients that would have undergone surgical
in patients with a cerebellar ICH, and quadrigeminal and conservative treatment.
cistern obliteration can be seen as a measure of mass
Currently, no clinical equipoise exists on the surgical
effect in the posterior fossa.
treatment of cerebellar hemorrhage (i.e., surgical
It has also been suggested that vascular territory (SCA/ treatment is currently considered superior in selected
AICA/PICA) would affect the prognosis.[18] In our series, patients), and it surely prevents patients from dying,
patients with a hematoma in the AICA territory had at least in the short‑term. Due to the lack of clinical
the worst outcome, probably caused by the frequent equipoise, it would be difficult to run a randomized
hematoma extension into the brainstem. Location in controlled trial on operative treatment of cerebellar
the cerebellar anterior vs. posterior lobe had an effect on ICH. However, we are trying to support the previously
patient outcome only, if the hematoma extended to both published notion that the outcomes after surgery may not
Surgical Neurology International 2017, 8:272 http://www.surgicalneurologyint.com/content/8/1/272

be as good as we neurosurgeons tend to think, and that Intracerebral Hemorrhage: A Guideline for Healthcare Professionals From
the American Heart Association/American Stroke Association. Stroke
the long‑term functional outcome needs further studies
2015;46:2032‑60.
because of the rather pessimistic short‑term functional 11. Juvela S, Heiskanen O, Poranen A, Valtonen S, Kuurne T, Kaste M, et al. The
outcome and high long‑term mortality. One possibility to treatment of spontaneous intracerebral hemorrhage. A prospective randomized
assess the futility of the treatment would be to measure trial of surgical and conservative treatment. J Neurosurg 1989;70:755‑8.
the long‑term survivors’ health‑related quality of life in 12. Kelly PJ, Stein J, Shafqat S, Eskey C, Doherty D, Chang Y, et al. Functional
recovery after rehabilitation for cerebellar stroke. Stroke 2001;32:530‑4.
the surgically and medically treated patients. 13. Kirollos RW, Tyagi AK, Ross SA, van Hille PT, Marks PV. Management of
spontaneous cerebellar hematomas: A prospective treatment protocol.
CONCLUSIONS Neurosurgery 2001;49:1378‑86.
14. Kobayashi S, Sato A, Kageyama Y, Nakamura H, Watanabe Y, Yamaura A.
Treatment of hypertensive cerebellar hemorrhage‑‑surgical or conservative
Surgical treatment of cerebellar ICH can be life‑saving
management? Neurosurgery 1994;34:246‑50.
but leads to a poor functional outcome. Lack of clinical 15. Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M,
equipoise makes designing randomized controlled trials et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke
difficult. New studies are needed on long‑term recovery 1996;27:1304‑5.
after a cerebellar ICH. 16. Little JR, Tubman DE, Ethier R. Cerebellar hemorrhage in adults. Diagnosis
by computerized tomography. J Neurosurg 1978;48:575‑9.
Financial support and sponsorship 17. Luney MS, English SW, Longworth A, Simpson J, Gudibande S, Matta B, et al.
Acute Posterior Cranial Fossa Hemorrhage—Is Surgical Decompression
Nil.
Better than Expectant Medical Management? Neurocrit Care 2016;25:365‑70.
Conflicts of interest 18. Matsukawa H, Shinoda M, Fujii M, Takahashi O, Yamamoto D, Murakata A,
et al. Relationships among hematoma diameter, location categorized
There are no conflicts of interest. by vascular territory, and 1‑year outcome in patients with cerebellar
hemorrhage. World Neurosurg 2012;77:507‑11.
REFERENCES 19. Meredith  W, Rutledge  R, Fakhry  SM, Emery  S, Kromhout‑Schiro  S. The
conundrum of the Glasgow Coma Scale in intubated patients: A linear
1. Cohen ZR, Ram Z, Knoller N, Peles E, Hadani M. Management and outcome regression prediction of the Glasgow verbal score from the Glasgow eye
of non‑traumatic cerebellar haemorrhage. Cerebrovasc Dis 2002;14:207‑13. and motor scores. J Trauma 1998;44:839‑44.
2. Dahdaleh NS, Dlouhy BJ, Viljoen SV, Capuano AW, Kung DK, Torner JC, 20. Mezzadri JJM, Otero JM, Ottino CA. Management of 50 spontaneous
et al. Clinical and radiographic predictors of neurological outcome following cerebellar haemorrhages. Importance of obstructive hydrocephalus. Acta
posterior fossa decompression for spontaneous cerebellar hemorrhage. Neurochir 1993;122:39‑44.
J Clin Neurosci 2012;19:1236‑41. 21. Pong  V, Chan K‑H, Chong B‑H, Lui W‑M, Leung GK‑K, Tse H‑F, et al.
3. Dammann  P, Asgari  S, Bassiouni  H, Gasser  T, Panagiotopoulos  V, Long‑term outcome and prognostic factors after spontaneous cerebellar
Gizewski ER, et al. Spontaneous cerebellar hemorrhage-experience with hemorrhage. Cerebellum 2012;11:939‑45.
57 surgically treated patients and review of the literature. Neurosurg Rev 22. Stein  M, Luecke  M, Preuss  M, Boeker D‑K, Joedicke  A, Oertel  MF.
2011;34:77‑86. Spontaneous intracerebral hemorrhage with ventricular extension and the
4. Dolderer S, Kallenberg K, Aschoff A, Schwab S, Schwarz S. Long‑Term grading of obstructive hydrocephalus: The prediction of outcome of a special
Outcome after Spontaneous Cerebellar Haemorrhage. Eur Neurol life‑threatening entity. Neurosurgery 2010;67:1243‑51.
2004;52:112‑9. 23. Steiner T, Al‑Shahi Salman R, Beer R, Christensen H, Cordonnier C, Csiba L,
5. Donauer E, Loew F, Faubert C, Alesch F, Schaan M. Prognostic factors in et al. European Stroke Organisation (ESO) guidelines for the management
the treatment of cerebellar haemorrhage. Acta Neurochir 1994;131:59‑66. of spontaneous intracerebral hemorrhage. Int J Stroke 2014;9:840‑55.
6. Falcone GJ, Brouwers HB, Biffi A, Anderson CD, Battey TWK, Ayres AM, 24. Taneda M, Hayakawa T, Mogami H. Primary cerebellar hemorrhage.
et al. Warfarin and Statins are Associated with Hematoma Volume in Primary Quadrigeminal cistern obliteration on CT scans as a predictor of outcome.
Infratentorial Intracerebral Hemorrhage. Neurocrit Care 2013;21:192‑9. J Neurosurg 1987;67:545‑52.
7. Fisher CM, Picard EH, Polak A, Dalal P, Ojemann RG. Acute hypertensive 25. Tsitsopoulos PP, Tobieson L, Enblad P, Marklund N. Prognostic factors
cerebellar hemorrhage: Diagnosis and surgical treatment. J Nerv Ment Dis and long‑term outcome following surgical treatment of 76 patients with
1965;140:38‑57. spontaneous cerebellar haematoma. Acta Neurochir 2012;154:1189‑95.
8. Flaherty ML, Woo D, Haverbusch M, Sekar P, Khoury J, Sauerbeck L, et al. 26. Weisberg  LA. Acute cerebellar hemorrhage and CT evidence of tight
Racial Variations in Location and Risk of Intracerebral Hemorrhage. Stroke posterior fossa. Neurology 1986;36:858‑60.
2005;36:934‑7. 27. Wijdicks EF, St Louis EK, Atkinson JD, Li H. Clinician’s biases toward surgery
9. Han J, Lee HK, Cho TG, Moon JG, Kim CH. Management and Outcome of in cerebellar hematomas: An analysis of decision‑making in 94 patients.
Spontaneous Cerebellar Hemorrhage. J Cerebrovasc Endovasc Neurosurg Cerebrovasc Dis 2000;10:93‑6.
2015;17:185‑93. 28. Witsch  J, Neugebauer  H, Zweckberger  K, Jüttler E. Primary cerebellar
10. Hemphill  JC, Greenberg  SM, Anderson  CS, Becker  K, Bendok  BR, haemorrhage: Complications, treatment and outcome. Clin Neurol
Cushman M, et al. Guidelines for the Management of Spontaneous Neurosurg 2013;115:863‑9.
Supplementary Table 1: Hematoma location in the
medically, i.e., conservatively and surgically treated
patients. We used Mann‑Whitney U test to test for
statistically significant differences
Medical (n=76) Surgical (n=38) P
Anterior lobe, n (%) 17 (22.4%) 1 (2.6%) 0.002
Posterior lobe 47 (61.8%) 24 (63.2%)
Both 12 (15.8%) 13 (34.2%)
Vermis 14 (18.4%) 2 (5.3%) 0.001
Hemisphere 44 (57.9%) 16 (42.1%)
Both 18 (23.7%) 20 (52.6%)
SCA 47 (61.8%) 26 (68.4%) 0.941
AICA 10 (13.2%) 0
PICA 12 (15.8%) 5 (13.2%)
Multiple 7 (9.2%) 7 (18.4%)
SCA: Superior cerebellar artery, AICA: Anterior inferior cerebellar artery,
PICA: Posterior inferior cerebellar artery

Supplementary Table 2: The effect of hematoma location on

poor functional outcome or mortality at discharge (modified
Rankin Scale 4‑6) in univariate logistic regression
Variable Odds ratio 95% CI P
Anteroposterior
Anterior lobe reference
Posterior lobe 0.64 0.18‑1.94 0.444
Both lobes 15.86 1.52‑2154 0.017
Laterality
Vermis reference
Hemisphere 2.277 0.761‑6.99 0.138
Both 25 4.71‑259 <0.001
Vascular territory
SCA reference
AICA 9.171 1.1‑1198 0.039
PICA 0.993 0.336‑3.26 0.99
Multiple 12.67 1.57‑1644 0.011
CI: Confidence interval, SCA: Superior cerebellar artery, AICA: Anterior inferior
cerebellar artery, PICA: Posterior inferior cerebellar artery