Articles
Safety and efficacy of minimally invasive surgery plus
Lancet Neurol 2016; 15: 1228–37
See Comment page 1197
†Contributed equally
‡Investigators listed in the
appendix
Department of Neurology,
Brain Injury Outcomes
Coordinating Center
(Prof D F Hanley MD,
N McBee MPH, K Lane CMA,
A J Bistran-Hall BS, P Keyl PhD,
T C Morgan MPH,
N Ullman MPH, W A Mould BA,
J R Carhuapoma MD, W Ziai MD)
and Department of
Neurosurgery (J Huang MD) and
Department of Biostatistics,
School of Public Health
(R E Thompson PhD,
J Muschelli PhD,
M Rosenblum PhD,
C B Thompson MS,
G Yenokyan PhD, E Huang PhD),
Johns Hopkins University,
Baltimore, MD, USA; Emissary
International, Austin, TX, USA
(S W Mayo PD); Department of
Neuroradiology (D Gandhi MD)
and Department of
Neurosurgery (E F Aldrich MD),
University of Maryland,
Baltimore, MD, USA;
Department of Neurology,
Boston University, Boston, MA,
USA (C Kase MD); Department
of Neurology
(W C Broaddus MD) and
Department of Neurosurgery
(R S Graham MD), Virginia
Commonwealth University,
Richmond, VA, USA;
Department of Neurology
(C Wijman MD) and Department
of Neurosurgery (R Dodd MD),
Stanford University, Palo Alto,
CA, USA; Department of
Neurosurgery, University of
Texas, San Antonio, TX, USA
(J-L Caron MD); Department of
Neurosurgery, University of
Kansas, Kansas City, KS, USA
(P Camarata MD); National
Institute of Neurological
1228
alteplase in intracerebral haemorrhage evacuation (MISTIE):
a randomised, controlled, open-label, phase 2 trial
Daniel F Hanley, Richard E Thompson, John Muschelli, Michael Rosenblum, Nichol McBee, Karen Lane, Amanda J Bistran-Hall, Steven W Mayo,
Penelope Keyl, Dheeraj Gandhi, Tim C Morgan, Natalie Ullman, W Andrew Mould, J Ricardo Carhuapoma, Carlos Kase, Wendy Ziai,
Carol B Thompson, Gayane Yenokyan, Emily Huang, William C Broaddus, R Scott Graham, E Francois Aldrich, Robert Dodd, Cristanne Wijman*,
Jean-Louis Caron, Judy Huang, Paul Camarata, A David Mendelow, Barbara Gregson, Scott Janis, Paul Vespa, Neil Martin, Issam Awad†,
Mario Zuccarello†, for the MISTIE Investigators‡
Summary
Background Craniotomy, according to the results from trials, does not improve functional outcome after intracerebral
haemorrhage. Whether minimally invasive catheter evacuation followed by thrombolysis for clot removal is safe and can
achieve a good functional outcome is not known. We investigated the safety and efficacy of alteplase, a recombinant tissue
plasminogen activator, in combination with minimally invasive surgery (MIS) in patients with intracerebral haemorrhage.
Methods MISTIE was an open-label, phase 2 trial that was done in 26 hospitals in the USA, Canada, the UK, and
Germany. We used a computer-generated allocation sequence with a block size of four to centrally randomise patients
aged 18–80 years with a non-traumatic (spontaneous) intracerebral haemorrhage of 20 mL or higher to standard
medical care or image-guided MIS plus alteplase (0·3 mg or 1·0 mg every 8 h for up to nine doses) to remove clots
using surgical aspiration followed by alteplase clot irrigation. Primary outcomes were all safety outcomes: 30 day
mortality, 7 day procedure-related mortality, 72 h symptomatic bleeding, and 30 day brain infections. This trial is
registered with ClinicalTrials.gov, number NCT00224770.
Findings Between Feb 2, 2006, and April 8, 2013, 96 patients were randomly allocated and completed follow-up: 54 (56%)
in the MIS plus alteplase group and 42 (44%) in the standard medical care group. The primary outcomes did not differ
between the standard medical care and MIS plus alteplase groups: 30 day mortality (four [9·5%, 95% CI 2·7–22.6] vs
eight [14·8%, 6·6–27·1], p=0·542), 7 day mortality (zero [0%, 0–8·4] vs one [1·9%, 0·1–9·9], p=0·562), symptomatic
bleeding (one [2·4%, 0·1–12·6] vs five [9·3%, 3·1–20·3], p=0·226), and brain bacterial infections (one [2·4%, 0·1–12·6]
vs zero [0%, 0–6·6], p=0·438). Asymptomatic haemorrhages were more common in the MIS plus alteplase group than
in the standard medical care group (12 [22·2%; 95% CI 12·0–35·6] vs three [7·1%; 1·5–19·5]; p=0·051).
Interpretation MIS plus alteplase seems to be safe in patients with intracerebral haemorrhage, but increased
asymptomatic bleeding is a major cautionary finding. These results, if replicable, could lead to the addition of surgical
management as a therapeutic strategy for intracerebral haemorrhage.
Funding National Institute of Neurological Disorders and Stroke, Genentech, and Codman.
Introduction Haemorrhage Trial (INTERACT) II trial, early blood
The incidence of brain haemorrhage is more than 5 million pressure lowering reduced clot growth by a small amount
cases per year.1 It is the most severe cause of stroke, but (3 mL difference) and led to a non-significant 3·7% gain in
there is no evidence-based primary treatment.2–4 Findings functional outcome.8 We designed MISTIE to assess
from pragmatic trials of therapy used in routine practice whether minimally invasive surgery (MIS) followed by
have not shown that any treatment substantially reduces thrombolytic treatment is safe and reduces or reverses the
haematoma size and brain tissue damage or improves burden of clot on tissue,9,10 and to provide preliminary
functional outcome. In the Surgical Treatment for functional outcome data.10–15 Our aim in using the MIS
Intracerebral Hemorrhage (STICH) I5 and II6 trials, plus alteplase treatment was to achieve almost complete
routine craniotomy did not alter functional outcomes. clot dissolution, without procedure-related safety events
Findings from STICH I suggested that clot removal might that would endanger the lives of the patients beyond the
simplify and shorten overall medical care,5 and findings risks associated with intensive medical treatment.
from STICH II showed a non-significant 5·6% decrease in
mortality. Findings from non-surgical trials of aggressive, Methods
early haemostasis7 showed some stabilisation of haema- Study design and patients
toma growth, but no change in functional outcome. In the MISTIE was a randomised, controlled, open-label,
Intensive Blood Pressure Reduction in Acute Cerebral phase 2 trial16 of image-guided,14 catheter-based13 removal
www.thelancet.com/neurology Vol 15 November 2016

Research in context
Evidence before this study
We used Zhou and colleagues’ published minimally invasive
surgery meta-analysis because it is an up-to-date review. They
searched international databases, websites, and conference
summaries, such as PubMed and the International Clinical Trials
Registry, up to December, 2011. Their inclusion criteria were non-
traumatic (spontaneous) intracerebral haemorrhage diagnosed
with CT and randomised controlled trials with minimally invasive
surgery (MIS) compared with a control group. Exclusion criteria
were traumatic brain injuries, infratentorial intracerebral
haemorrhage, and studies with quality assessments of less than 2
on the Cochrane criteria scale (0–8).
Mounting evidence shows that clinical injury from intracerebral
haemorrhage is directly related to the size of the clot. However,
benefit from reduction of the clot size by mechanical means has
been difficult to show. Now, with completion of the STICH I and
II studies, we have a strong indication that pragmatic use of
open craniotomy does not produce the presumed benefits.
Added value of this study
To the best of our knowledge, MISTIE is the first rigorous study
of MIS in patients with intracerebral haemorrhage, because our
of intracerebral haemorrhage of 20 mL or more,
measured with the ABC/2 method,17 done at 26 intensive
care units in the USA, Canada, the UK, and Germany.16
Each hospital obtained local institutional review board
or ethics committee approval. Patients were eligible if
they were aged 18–80 years and had a spontaneous, non-
traumatic, supratentorial intracerebral haemorrhage of
20 mL or higher as a result of cerebral small-vessel
disease but not a macrovascular cause, a Glasgow Coma
Scale (GCS) score of 14 or less or a National Institutes of
Health Stroke Scale (NIHSS) score of 6 or higher, and a
history of a modified Rankin Scale (mRS) score of 0 or 1,
an intracerebral haemorrhage that remained the same
size for 6 h or more, and who satisfied all other inclusion
and exclusion criteria listed in the appendix (pp 4–5).
All patients provided written informed consent. The
study protocol is available online.
Randomisation and masking
Patients were randomly allocated by local site personnel
using a central web-based enrolment system. The trial
statistician (PK) used a computer-generated number
sequence to randomly allocate patients to the MIS plus
alteplase or standard medical care groups. Patient
allocation was stratified by diagnostic CT-measured clot
size (intracerebral haemorrhage 20–40 mL or >40 mL).
The trial had two pre-planned stages. In the dose-finding
stage (stage one), two doses of alteplase were used (0·3 mg
and 1·0 mg). In stage one, block sizes of four were used to
ensure patients in completed blocks were randomised in a
ratio of 3:1 to MIS plus alteplase or medical management.
www.thelancet.com/neurology Vol 15 November 2016
Articles
Disorders and Stroke, National
Institutes of Health, Bethesda,
MD, USA (S Janis PhD);
trial was a multisite study with a standardised surgical task Department of Neurology
designed to eliminate cortical incision, electrocautery, toxic (P Vespa MD) and Department
exposure to thrombin, and additional loss of deep-brain tissue. of Neurosurgery (N Martin MD),
University of California, Los
The results provide evidence of safety of the MIS plus alteplase Angeles, Los Angeles, CA, USA;
procedure and identify the best dose of thrombolytic drug to be Department of Neurosurgery,
used for clot volume reduction. MIS and alteplase cannot be University of Chicago, Chicago,
combined without increased incidence of asymptomatic IL, USA (Prof I Awad MD);
Department of Neurosurgery,
bleeding. The treatment effect for mRS less than or equal to 3 at University of Cincinnati,
365 days was substantial (>10% absolute benefit) and long Cincinnati, OH, USA
term (1 year). These data show the safety and potential efficacy (Prof M Zuccarello MD); and
of MIS, and overall proof of concept for the hypothesis that it Neurosurgery, Newcastle
University, Newcastle upon
has robust potential as a unique intervention for intracerebral Tyne, UK (A D Mendelow
haemorrhage. FRCS[SN], B Gregson PhD)
Implications of all the available evidence *Dr Wijman died in February, 2013
The data provide a sound basis to estimate a treatment effect Correspondence to:
Prof Daniel F Hanley,
and inform clinical goals for the surgical task of clot size Department of Neurology, Brain
reduction. This study highlights the possibility that MIS has Injury Outcomes Coordinating
promise to mechanically reduce lesion size and has the Center, Johns Hopkins University,
MD 21231, USA
potential to directly alter events caused by blood clot within
dhanley@jhmi.edu
brain tissue and treat a disease for which no therapy exists.
See Online for appendix
A planned protocol amendment occurred after completion
of stage one enrolment and data safety monitoring board
(DSMB) review (amendment occurred on Dec 22, 2009).
This amendment specified the use of alteplase at 1 mg
dose (based on safety profile and clot removal efficiency),
use of a surgical oversight centre (University of Cincinnati,
Cincinnati, OH, USA; based on initial surgical perfor-
mance), and addition of outcome assessments at 365 days.
This amendment was reviewed and approved by the
DSMB and the trial executive committee. In the safety
assessment stage (stage two), patients were randomised in
a 1:1 ratio to receive MIS plus alteplase or medical
management . The first patient at every site was assigned
to surgical management and served as a credentialling
patient to document the surgeon’s ability to do the surgical
protocol. After a surgeon was credentialled, subsequent
eligible patients were randomised to MIS plus alteplase or For the study protocol see
medical management. Site examiners doing outcome http://vistacollaboration.org/
index.php/vista-ich/protocols
assessments were masked to treatment assignment.
Procedures
We managed the risks of initial haematoma growth or
instability by use of a stability protocol that combined
normalisation of coagulation variables, blood pressure
management, and repeat CT assessment of clot size,
measured with the ABC/2 method. 6 h or more after the
diagnostic CT, we did a stability CT to ensure that the
intracerebral haemorrhage clot had not expanded by more
than 5 mL, providing image demonstration of a safe
starting point for clot reduction therapy, defined as the
absence of active bleeding before treatment. The CT could
1229
 Articles
be repeated every 6 h until the clot stabilised or just before
the 48 h eligibility window closed, whichever came first.
Additionally, an MRI or a CT angiography was done to
rule out underlying pathology as the bleeding source; an
angiogram was encouraged in cases with equivocal
findings on vascular pathology screening.17 Patients had to
have an international normalised ratio of 1·3 or less, a
normal activated partial thromboplastin time, and blood
pressure stability before randomisation.18,19 After a protocol
amendment, planned catheter insertion trajectories
describing the skull entry site and planned linear path to
the haematoma target were shared by the site with the
trial’s surgical centre for joint review (stage two only).
Under general anaesthesia in the operating room, we
used image guidance to place an introducer cannula
within the middle two-thirds of the overall haematoma
diameter, through a burr hole or twist drill opening, in
individuals randomised to receive MIS plus alteplase.
The introducer portion of the cannula was then removed.
Clot aspiration was done with a 10 mL handheld syringe
until there was no longer any fluid component of the clot
in the aspirate or until first resistance. We passed a soft
catheter through a rigid, peel-away cannula into the
residual haematoma, tunnelled subcutaneously, and
connected it to a three-way stopcock and closed drainage
system. Postoperative CT was done to confirm
positioning of the soft catheter and stability of the
residual haematoma and catheter tract.
3 h or more after catheter placement, we gave patients
intraclot alteplase (ie, administered directly into the clot
through the catheter) at 0·3 mg in 1 mL or 1·0 mg in 1 mL
every 8 h, for up to nine doses, or until a trial-defined
surgical performance requirement was reached. All doses
were followed by a 3 mL flush of preservative-free normal
saline. After each assigned dose, we closed the system for
1 h to allow drug–clot interaction and then reopened it to
allow for gravitational drainage. Trial-defined surgical
performance requirements were reduction of clot to 20% of
the volume measured on stability CT scan before
randomisation, to 15 mL or less, or occurrence of a clinically
significant rebleeding event, defined as a sustained
decrease of more than 2 points on the GCS motor score,
with CT-demonstrated enlargement of the intracerebral
haemorrhage. Subsequent CT scans were done every 24 h
until dosing was complete to assess safety and drainage.
Use of the American Heart Association recommendations
for treatment of non-traumatic spontaneous intracerebral
haemorrhage18,19 enabled a standard approach to monitoring
patients’ airways, ventilation, intracranial pressure, sed-
ation, and pharmacological treatment of intracranial mass
effect. Patients allocated to the standard medical care group
had follow-up CT scans and other monitoring assessments
on the same schedule as those in the MIS plus alteplase
group.
Patients had an MRI scan at day 7 after randomisation.
Patients returned to clinic on days 30 (stages one and
two), 180 (stages one and two), and 365 (stage two), and
1230
we contacted them by telephone on days 90 (stages one
and two) and 270 (stage two). A certified examiner
assessed the mRS, Barthel Index, Stroke Impact Scale,
Glasgow Outcome Scale, extended Glasgow Outcome
Scale, NIHSS (clinic visits only), and a repeat CT (days 30
and 180 only).
To optimise accuracy and minimise investigator bias, a
core laboratory used semi-automated segmentation and
Hounsfield thresholds to measure clot volumes.9
Measurement of clot volumes was done with OsiriX
software (version 4.1) on DICOM images of each patient’s
stability and treatment scans. This approach has been
validated for accuracy and inter-rater reliability.20 We used
the core laboratory measurements in all analyses. The core
laboratory defined location as either lobar or deep
(putamen or thalamus).
Outcomes
The primary outcomes were all safety outcomes: 30 day
mortality, 7 day procedure-related mortality, 30 day
bacterial brain infection, and symptomatic bleeding
within 72 h after the last dose. Secondary outcomes
were difference in clot-size reduction at the end of the
treatment for each group and its effect on functional
outcome. The mRS was selected as the single functional
outcome to assess for preliminary efficacy in view of its
universal use in stroke and the 5% effect reported in
STICH.5 Efficacy was further explored using the
adjusted 180 day dichotomised mRS score (0–3 vs 4–6),
expressed as the proportion of all patients with an mRS
score of less than or equal to 3, 180 day ordinal mRS,
and 365 day ordinal mRS. Outcomes were not centrally
assessed. All adverse and serious adverse events were
centrally assessed during the acute treatment and all
serious adverse events were assessed until the end of
follow-up.
An independent DSMB managed the two pre-planned
stages of the trial. A third dose, 3 mg, was not investigated
in stage one after a planned DSMB review (protocol
amendment occurred on Oct 2, 2008).
Statistical analysis
We designed and powered the study to explore safety of
MIS plus alteplase, and alteplase dosing. It was not
powered to detect effect of treatment on functional
outcome. The study had 90% power to detect a doubling
in the proportion of any rebleeding from 8% to 16%
between dose groups of 15 patients who received MIS
plus alteplase (stage one). The study had 80% power to
detect a difference in clot dissolution of 3% per day or
greater between groups of 15 patients on different doses
of alteplase (stage one). It had 91–99% power to detect
one or more symptomatic bleeds for 15 patients if the true
proportion of bleeding was 15%. For stages one and two
combined, assuming a total of 80 patients, the study had
a 52–62% power to detect an absolute difference of 25%
in mortality, assuming a 50% mortality. The target sample
www.thelancet.com/neurology Vol 15 November 2016

size for stage one was 20 patients per dose group and, for
stages one and two combined, the final target sample size
was a total of 96 patients (this sample size was reduced
from 110 patients as a result of the DSMB’s decision to
not include the 3 mg dose of alteplase in stage one).
In the safety analysis, we tested the hypothesis that
there was no difference between the treatments. We
defined and prespecified thresholds for safe use of MIS
plus alteplase for treatment of intracerebral haem-
orrhage relative to standard medical care: 70% of
participants for 30 day mortality, 35% for symptomatic
brain bleeding, and 15% for bacterial brain infection. We
tested proportions of events across groups with Fisher’s
exact test and calculated exact binomial 95% CIs. We
used the Kaplan-Meier method to estimate the survival
functions (with 95% CIs) for patients in both groups.
In the modified intention-to-treat analysis, we
estimated the mean benefit of MIS plus alteplase versus
standard medical care. Specifically, we estimated the
difference between the probability of having a
180 day mRS score of 3 or less, referred to as a good
outcome, with MIS plus alteplase versus standard
medical care (secondary outcome). We estimated this
average treatment effect using a simple difference in
proportions of people with 180 day outcomes
(unadjusted estimator) and an estimator that adjusted
(adjusted estimator) for missing outcome data using
the double-robust method for censored data of
Rotnitzky and Colantuoni and colleagues.21,22 The
adjusted estimator is fully described in the
appendix (pp 10–12); it has greater precision than the
unadjusted estimator because of adjustments for
potential imbalances between study groups in
prognostic baseline covariates: NIHSS score, GCS
score, intracerebral haemorrhage volume, and intra-
ventricular haemorrhage volume.21
For the exploratory analyses, patient characteristics,
safety, and outcome measures were reported by each
site. To describe medical events, we used terms and
definitions from the Medical Dictionary for Regulatory
Activities that were centrally adjudicated. We used
longitudinal plots to depict the percentage of
intracerebral haemorrhage clot reduction from stability
over time for each participant. We applied locally
weighted scatter-plot smoothing to calculate mean clot
reduction over time by treatment group.23 To investigate
the role of clot volume reduction by MIS and alteplase
on outcome, we used logistic regression models to
estimate the association between the secondary
outcome measure dichotomised 180 day mRS scores of
3 or less and treatment group, clot removal, and the
baseline variables identified in the study protocol:24
intracerebral haemorrhage clot size at randomisation,
age, enrolment NIHSS score, presence of intra-
ventricular haemorrhage, and location of intracerebral
haemorrhage. We created the multivariable logistic
regression models by first considering univariable
www.thelancet.com/neurology Vol 15 November 2016
Articles
analyses to establish the independent associations
between each covariate and the outcome. For parsimony,
the variables chosen for the multivariable regression
analyses, in addition to treatment, were those with
p<0·1 in the univariable analyses; these variables are
similar to those from previous studies of intracerebral
haemorrhage.6,8,25 For the purpose of hypothesis
generation, we systematically considered other pre-
specified baseline variables that were non-significant in
the univariate analyses as candidate explanatory
variables in the multivariate analysis. As in any variable
selection method for model building, our approach has
limitations and should be regarded as exploratory.26 We
did sensitivity analyses with all possible good and bad
outcomes for patients with missing 180 day mRS scores
to establish a possible tipping point that would change
the statistical association between the outcome and clot
removal.27,28 We used logistic regression for sensitivity
analyses of subgroups.
To manage the effect of time on patients, we defined
volume of clot removed at the specific times as follows:
for patients in the MIS and alteplase group, we defined
4103 patients assessed for eligibility
3980 ineligible
3973 inclusion criteria not met
2 refused consent
5 other
123 enrolled
27 run-in MIS patients*
96 randomised
54 assigned MIS plus 42 assigned standard
alteplase† medical care‡
2 lost to follow-up 4 lost to follow-up
52 included in modified 38 included in modified
intention-to-treat intention-to-treat
analysis analysis
Figure 1: Trial profile
MIS=minimally invasive surgery. rtPA=recombinant tissue plasminogen
activator. *The first one or two patients at each site who were used to credential
surgeons at each study centre. †Eight patients received MIS only. Three patients
received craniotomy. ‡Includes six patients randomly assigned to the medical
care group in stage one that was planned as a 3 mg dose group, but which was
not investigated after the planned data safety monitoring board review and was
used for investigation of endoscopic removal of intracerebral hemorrhage (not
shown). Four patients received craniotomy.
1231
 Articles

the end of treatment as 24 h after the last dose of alteplase; by NINDS provided input during active recruitment. The
we defined the end-of-treatment scan for patients in the DSMB and Genentech approved the decision to submit
medical management group as the scan closest to the for publication. The funders of the study had no role in
median end-of-treatment scan time for patients in the data collection, data analysis, data interpretation, or
MIS plus alteplase group (4 days after randomisation).9 writing of the report. The corresponding author had full
We defined the end of treatment scan for patients who access to all the data in the study and had final
received delayed craniotomy in the medical management responsibility for the decision to submit for publication.
group as the scan before craniotomy. We did all analyses
using the statistical packages Stata (version 12.0) and R Results
(version 3.2). This trial is registered on ClinicalTrials.gov, Between Feb 2, 2006, and Aug 2, 2009, we enrolled
number NCT00224770. 29 patients to MIS plus alteplase and 17 to medical
management in stage one. The slight imbalance of the
Role of the funding source 3:1 randomisation in stage one was due to not all of the
The National Institute of Neurological Diseases and within-site randomisation blocks being filled. The
Stroke (NINDS) provided input into the study design number of subjects randomised and receiving 0·3 mg
during the grant review process and the DSMB appointed and 1·0 mg in each stage or tier is summarised in the
appendix (p 24). Between Dec 21, 2009, and April 3, 2012,
we enrolled 25 patients to MIS plus alteplase and 25 to
Standard care MIS plus alteplase p value
(n=42) (n=54)
medical management in stage two (appendix p 17). We
randomly allocated 54 (56%) patients (stages one and
Demographic variables
two combined) to MIS plus alteplase and 42 (44%) to
Age (years)
standard medical care (including six patients from the
Mean 61·1 (12·3) 60·7 (11) ·· stage one endoscopy group [ie, the 3 mg group not
Median 62 (49·5–73) 60 (54–69) ·· investigated after the planned DSMB review used for
Male sex 28 (67%) 35 (65%) ·· investigation of endoscopic removal of intracerebral
Race hemorrhage]; figure 1). Delayed clinical deterioration
White 23 (55%) 30 (56%) ·· led to craniotomy in four patients in the standard
African American 11 (26%) 18 (33%) ·· medical care group and two in the MIS plus alteplase
Hispanic 5 (12%) 4 (7%) ·· (after the MIS). The final follow-up visit occurred on
Other 3 (7%) 2 (4%) ··
Baseline variables Standard care MIS plus alteplase p value
Diabetes 11 (26%) 14 (26%) ·· (n=42) (n=54)
History of 34 (81%) 49 (91%) ·· (Continued from previous column)
hypertension
Treatment variables
Other cardiovascular 14 (33%) 22 (41%) ··
disease ICP monitoring 10 (24%) 9 (17%) 0·444

Alcohol abuse 7 (17%) 17 (31%) ·· Ventilated 16 (38%) 25 (46%) 0·533

Presentation BP (mm Hg) Time from ictus to 1·3 (0·6) 1·2 (0·5) 0·174
randomisation
Systolic 186·7 (34·1) 186·4 (33·0) ·· (days)
Diastolic 101·9 (20·4) 106·8 (27·7) ·· Systolic BP (mm Hg) 145·3 (20·7) 143·9 (21·1) 0·741
Enrolment GCS score Diastolic BP 73 (14·9) 71·2 (13·1) 0·534
3–8 13 (31%) 17 (31%) ·· (mm Hg)
9–12 12 (29%) 20 (37%) ·· Time from NA 6·6 (7·8) ··
13–15 17 (40%) 17 (31%) ·· randomisation to
surgery (h)
Enrolment NIHSS score
Surgery (elapsed time from symptom onset)
Mean 21·6 (8·9) 22·8 (8·5) ··
≤36 h NA 31 (57%) ··
Median 21 (17–27) 22 (18–29) ··
>36 h NA 23 (43%) ··
Stability CT (last CT before enrolment)
Number of doses of NA 3·5 (2–5·8) ··
ICH volume (mL) 43·1 (15·3) 48·2 (19·6) ·· alteplase
ICH volume (mL) 41·4 (33·2–50) 43·4 (31·6–59·3) ·· Days in ICU 8 (5–13) 8 (6–15) 0·839
IVH volume (mL) 2·4 (3·9) 4·6 (7·7) ·· Days to return home 89 (54–146) 51 (36–89) 0·031
IVH volume (mL) 0·7 (0–3·1) 0·8 (0–4·4) ··
Data are mean (SD), median (IQR), or n (%). MIS=minimally invasive surgery.
Clot location
BP=blood pressure. GCS=Glasgow Coma Scale. NIHSS=National Institutes of Health
Lobar 15 (36%) 18 (33%) ·· Stroke Scale. ICH=intracerebral haemorrhage. IVH=intraventricular haemorrhage.
Deep 27 (64%) 36 (67%) ·· ICP=intracranial pressure. ICU=intensive care unit. NA=not applicable.
(Table 1 continues in next column)
Table 1: Baseline and treatment characteristics

1232 www.thelancet.com/neurology Vol 15 November 2016

 Articles

April 8, 2013, after approval of the planned protocol
amendment. Total follow-up was 180 days in stage one
and 365 days in stage two. Baseline characteristics are
shown in table 1. Eight (15%) of 54 patients in the MIS
plus alteplase group achieved the surgical goal (ie,
reduction in clot volume [with surgery alone and
without the need for alteplase]) at the end of the surgical
procedure (ie, after aspiration and catheter placement)
and 46 (85%) received alteplase through the catheter to
further reduce the haematoma size.
Safety event rates were below prespecified safety
thresholds and the primary safety outcomes were not
significantly different in the two groups (table 2). The
MIS plus alteplase group had non-significantly more
asymptomatic and significantly more bleeding within
72 h after the last dose than did the standard medical
care group (table 2). Kaplan-Meier analysis of survival
over 365 days showed no adverse effect of MIS plus
alteplase on survival (hazard ratio 1·32 [95% CI
0·618–2·82]; p=0·473; appendix p 13). Patients in the
MIS plus alteplase group achieved more rapid volume
reduction of intracerebral haemorrhage than did those
in the standard medical care group (figure 2). 31 (89%)
of 35 neurosurgeons acquired the technical skills. We
noted no differences in volume of clot removed between
a surgeon’s first and fourth procedure or between new
and experienced surgeons (data not shown).
Both the 0·3 mg and 1·0 mg alteplase doses increased
clot removal when compared with the standard medical
care group (appendix p 23), with no differences in
symptomatic bleeding between the two dose groups
(0·3 mg 0 of 14 patients; 1·0 mg two [13%] of 15 patients;
difference 13∙3% [95% CI –3·9 to 30·5]; p=0·483), and
one patient (2% of 36 patients in stages one and two)
with symptomatic bleeding in the standard medical care
group. This finding led to selection of the 1·0 mg dose
for stage two of the trial (appendix p 23). Overall, for both
stages one and two, symptomatic bleeding occurred in
one (2%) patient in the standard medical care group
versus five (9%) in the MIS plus alteplase group
(difference 6·9% [95% CI 4·6–18·1]; p=0·226). Reduction
in clot volume over time and for individual patients in
the modified intention-to-treat analysis at about day 4 is
–0·088 to 0·294; p=0·260) on the basis of the 96 patients
in the intention-to-treat analysis. The adjusted estimate
based on the method by Rotnitzky21 and Colantuoni22
and colleagues was 0·162 (95% CI 0·003–0·323;
p=0·049).
Additional exploration of efficacy was an mRS score of
less than or equal to 3 at 365 days, which we only
measured for stage two patients (56 patients, who had
extended follow-up; table 3). The unadjusted estimate of
the absolute benefit at 365 days was 0·117 (95% CI
–0·146 to 0·370; p=0·376) and the adjusted estimate was
0·115 (–0·171 to 0·306; p=0·330). Additional analyses
focused on ordinal (rather than dichotomous) mRS scores.
The proportion in each mRS category at 180 days (stage
one and two participants) and 365 days (stage two
participants only) is shown in table 3.
Results of the logistic regression analysis on the binary
indicator good versus unfavourable mRS scores at
day 180 are presented in the appendix (p 18). When we
considered univariable (unadjusted) models, we found
that prerandomised intracerebral haemorrhage volume,
age, enrolment NIHSS score, the presence of any
prerandomisation intraventricular haemorrhage, and the
absolute clot volume remaining at the end of treatment
were significant; assignment to MIS plus alteplase
resulted in a non-significant benefit. In the multivariable
model, after controlling for age, enrolment NIHSS,
prerandomised intracerebral haemorrhage volume and
presence of intraventricular haemorrhage, and
assignment to MIS plus alteplase, the absolute volume of
clot remaining at the end of treatment was significant.
Specifically, the model predicted that, with all other
variables held constant, each 10 mL of additional clot
remaining at the end of treatment is associated with a
relative reduction in the odds of a good 180 day outcome
Study stop Standard care MIS plus alteplase p value
threshold (n=42) (n=54)
Died within 0–7 days 10% 0 (0% [0–8·4]) 1 (1·9% [0·1–9·9]) 0·562
Died within 0–30 days 70% 4 (9·5% [2·7–22·6]) 8 (14·8% [6·6–27·1]) 0·542
Bacterial brain 15% 1 (2·4% [0·1–12·6]) 0 (0% [0–6·6]) 0·438
infection 0–30 days*

shown in figure 2. The mean reduction of haematoma Symptomatic bleed 35% 1 (2·4% [0·1–12·6]) 5 (9·3% [3·1–20·3]) 0·226
72 h after last dose†
size (from randomisation to end of treatment) in the MIS
Asymptomatic bleed NA 3 (7·1% [1·5–19·5]) 12 (22·2% [12–35·6]) 0·051
plus alteplase group was 57% (SD 25) versus 5% (10) in 72 h after last dose‡
the standard medical care group. The mean end-of- Symptomatic or NA 4 (9·5% [2·7–22·6]) 15 (27·8% [16·4–41·6]) 0·038
treatment intracerebral haemorrhage was 20 mL (SD 14) asymptomatic bleed
in the MIS plus alteplase group and 41 mL (15) in the 72 h after last dose§
medical management group (mean difference 21 mL Data are n (% [95% CI]). MIS=minimally invasive surgery. NA=not applicable. *Criteria included cultured organism
[95% CI 15–27]; p<0·0001). identification in the presence of fever, relevant laboratory values, and associated clinical symptoms. †Criteria included
The proportion of all patients with an mRS score of 3 radiographic evidence of an increase in clot volume (>5 mL increase) associated with a decrease in the Glasgow Coma
Scale motor scale score of more than 2 points sustained for a minimum of 8 h or associated clinical symptoms in the
or less at 180 days (secondary outcome for the
opinion of the site investigator. ‡Included events for which a clot size increase (>5 mL increase) was confirmed by the
assessment of efficacy) was 21% in the standard medical core laboratory with volumetric measurement of the CT scan by comparison with the previous CT scan, but for which
care and 33% for the MIS plus alteplase groups. The no alteration of Glasgow Coma Scale was noted. §The total of all adjudicated bleeding events.
unadjusted difference between the MIS plus alteplase
Table 2: Safety outcomes
and standard medical care groups was 0·109 (95% CI

www.thelancet.com/neurology Vol 15 November 2016 1233

 Articles

160 Medical care –20

MIS plus rtPA

0
120
ICH volume remaining (%)

ICH volume reduced (mL)

20

80

40

40
60

0 80
0 1 2 3 4 0 20 40 60 80 100
Time from start of treatment (days) Percentile of patients

Figure 2: Intracerebral haemorrhage removal

Data from all 96 randomised patients were included. (A) Percentage of clot remaining as measured with a daily CT scan after achievement of clot size stability and
initiation of MIS plus rtPA. Thin lines represent data from individual patients and thick lines are the fitted average response. The grey shaded areas are the 95% CIs.
The black line marks the average occurrence of the 24 h post-last treatment timepoint. (B) The distribution of each patient’s clot removal, expressed as absolute
volume reduction according to the day 4 end-of-treatment CT scan. The blue dashed line indicates the 50th-percentile patient and respective ICH volume reduction for
the standard care cohort. The green dashed line indicates the 50th percentile patient and respective volume reduction for this patient in the MIS plus rtPA group.
All volumes were established by the core laboratory. ICH=intracerebral haemorrhage. MIS=minimally invasive surgery. rtPA=recombinant tissue plasminogen activator.

potential tissue injury that is inevitable with craniotomy.

Day 180* Day 365†
We used image guidance and a new combination of
MIS plus Standard MIS plus surgery and drug treatment. This approach was reliably
Standard
alteplase care alteplase
care (n=42)
(n=54) (n=31) (n=25)
reproduced by surgeons new to the treatment concept, but
trained in the general principles. The surgical technique is
mRS score of 0 0 0 0 1 (4%)
a simple and logical extension of routine image guidance
mRS score of 1 0 1 (2%) 1 (3%) 3 (12%)
and intracerebral catheter placement. Although both
mRS score of 2 4 (10%) 6 (11%) 2 (6%) 2 (8%)
pragmatic (craniotomy) and simple translational
mRS score of 3 5 (12%) 11 (20%) 3 (10%) 2 (8%)
(factor VIIa) approaches have not worked in trials, our
mRS score of 4 12 (29%) 13 (24%) 6 (19%) 3 (12%)
results are promising and contrary to the belief that
mRS score of 5 6 (14%) 7 (13%) 3 (10%) 2 (8%)
surgical manipulations to remove blood might damage
mRS score of 6 11 (26%) 14 (26%) 11 (35%) 10 (40%)
brain tissue and impair long-term function.5,29,30 Because
mRS score missing 4 (10%) 2 (4%) 5 (16%) 2 (8%)
supportive medical care is the only universally accepted
MIS=minimally invasive surgery. mRS=modified Rankin Scale. *Includes randomly treatment for intracerebral haemorrhage, these results are
allocated patients from stages one and two. A graphical display is presented in the promising19,30 and consistent with trends from studies of
appendix (p 16). †Includes only randomly allocated participants enrolled in stage
convenience samples or at single sites.31,32 We have shown
two—ie, those enrolled after the protocol amendment extending follow-up to
365 days. The six patients randomly assigned to the medical care group in stage that MIS can be done safely at several sites, alteplase can
one that was used for investigation of endoscopic removal of intracerebral be used safely with MIS, and the treatment used in this
hemorrhage were included in the standard medical group. trial, although different from current practice, can be
Table 3: Functional outcomes adopted without difficulty at interested sites. Thus, the
treatment used in MISTIE could be a promising approach
to a worldwide health problem. If the apparent benefits of
by almost 50% (adjusted odds ratio 0·496 [95% CI MIS plus alteplase are replicable, the findings could lead to
0·259–0·949]; p=0·034). Unadjusted analyses in different a change in treatment of intracerebral haemorrhage.
subgroups are shown in the appendix (p 14). The results Our study had several limitations. The trial size was
by subgroup were not significant. small and the screening yield was low (123 patients
enrolled of 4103 screened). The trial was powered to
Discussion observe high safety thresholds (15% bleeding), not
MIS plus alteplase was safe in our phase 2 trial, with a efficacy; thus, the range of estimated benefit in the
possible advantage of better functional outcome at 180 days modified intention-to-treat analysis is wide and the true
than with standard medical care. However, increased benefit could be different. However, all known baseline
asymptomatic bleeding is a major cautionary finding. severity factors (intracerebral haemorrhage size,
MISTIE is an important test of a very gentle approach to intraventricular haemorrhage size, age, NIHSS score,
intracerebral haemorrhage evacuation, minimising GCS score, and stability) were nearly balanced between

1234 www.thelancet.com/neurology Vol 15 November 2016


groups. Point estimates in both adjusted efficacy analyses
suggest a treatment benefit. Safety conclusions are
similarly limited by sample size; the large difference in
asymptomatic bleeding shows that the combination of
MIS and alteplase still has important risks.
We hypothesise that the effect of MIS plus alteplase on
mRS outcome is mediated through clot volume reduction,
although such a relation cannot be assessed without a
second trial. We considered potential confounders of the
relation between clot reduction and mRS outcomes, such
as length of intensive care unit stay and use of intracranial
pressure monitoring; these confounders were similar in
both treatment groups. Surgical bias could inadvertently
account for good outcomes; however, the usual source of
this problem—selection of patients with less severe
intracerebral haemorrhage for surgery—did not occur, as
the patients randomly allocated to surgery were slightly,
but non-significantly, more impaired at baseline, and all
patients consented without knowledge of surgical
allocation, further limiting the possibility of surgical
selection bias. A bias could also be that the population
was not fully representative of the general population
with intracerebral haemorrhage and that the routine care
that the standard medical care group received might
account for the surgical benefit by having selected a
subgroup that had worse illness. This bias is not likely, as
the distribution of good and poor outcomes for the
patients in the standard medical care group is similar to
that in other intracerebral haemorrhage populations. A
bias in the amount of rehabilitation care that a particular
group received is also possible. Finally, specialised skills,
such as surgical skill or stroke centre organisational
resources, could have possibly produced the benefits,
rather than MIS plus alteplase, rendering the results not
reproducible by a wider set of surgeons or centres than
those in this study. This explanation seems unlikely, as
most of the surgeons, although trained in image guidance
and catheter placement, had not previously combined
these skills to treat intracerebral haemorrhage. Similarly,
attention to blood pressure control is already a well
developed and standard approach in established stroke
centre protocols.8,18,23,25,33
Our data suggest that MIS plus alteplase has the potential
to be efficacious where routine craniotomy has not worked.
Comparisons with existing trial31 and meta-analysis32,34 data
are reassuring for the possible generalisability of these
findings. Mortality was low and not different for patients
in the standard medical care or MIS plus alteplase groups
and was similar to STICH I5 and II6 mortalities, suggesting
that the surgical procedure itself is low risk. This finding is
in agreement with other observations of MIS.32 The low
mortality might relate to the infrequent discontinuation of
care. Patients with deep intracerebral haemorrhage
responded poorly to invasive craniotomy in STICH I, but
might benefit from MIS plus alteplase.
Administration of alteplase after MIS also appears safe
in this small sample using mortality and symptomatic
www.thelancet.com/neurology Vol 15 November 2016
Articles
bleeding as the main safety measures, but only if done
under the conditions of the protocol of this study. Some
rebleeding should be expected in patients with an
intracerebral haemorrhage and exposure to surgery and a
thrombolytic drug. The significantly increased occurrence
of asymptomatic bleeding provides evidence for further
safety assessments. The combined use of imaging to
define clot stability and blood pressure control appears to
have limited the frequency of symptomatic bleeding
events in both groups in MISTIE, and the absence of a
symptomatic rebleeding difference between the MIS plus
alteplase group and the standard medical care group is a
reassuring safety profile for a thrombolytic treatment
approach; external monitoring and core imaging
laboratory assessments for enlargement reinforce our
confidence in this safety profile. However, the presence of
higher frequencies for all bleeding categories in the MIS
plus alteplase group than in the standard medical care
group shows that the possible benefits of MIS plus
alteplase come with a clear potential bleeding risk that
must be managed with clinical vigilance. A larger sample
than that used in this study is needed to confirm that the
current clinical vigilance (intracerebral haemorrhage
stability, alteplase dose, and blood pressure control) is
reproducible in the general population. Estimation of
safety and, particularly, bleeding events, would be better
in a phase 3 trial than in a phase 2 trial. MISTIE III is
underway to test the generalisability of the
MISTIE surgical task and medical stabilisation protocol.
The approach used in this trial targets two major
sources of intracerebral haemorrhage morbidity: mass
effect and inflammation. The functional outcomes in the
MIS plus alteplase group are consistent with better tissue
preservation than in the medical management group9,10
and, when compared with the medical management
group equally treated with intracerebral haemorrhage
stabilisation and blood pressure control, are potentially
generalisable to other populations. The trial design was
not able to differentiate whether better tissue preservation
in the MIS plus alteplase group than in the medical
management group was attributable to mass-effect
reduction or removal of inflammation-provoking blood
products; both are most likely to be important effects of
clot reduction. Because removal took place over 3–4 days,
some degree of secondary injury from mass effect and
inflammation is likely to occur over days, not hours. If the
approach of minimal mechanical manipulation is to be
used, most patients will require alteplase irrigation to
achieve large reductions of clot volume. Analysis of
subgroups suggests that the potential for benefit occurs
in a window of at least 48 h. If the time window is this
long, then MIS plus alteplase could possibly be scheduled
urgently rather than in an emergent manner. The model
and subgroup findings represent hypothesis-generating
information, such as time dependence of treatment, and
require independent confirmation. For example, if much
damage occurs by chemical means in the initial hours,
1235
 Articles
additional benefit could exist for very early removal or
biochemical blockade of such events.
To test the reproducibility of our results will require at
least 500 patients in a randomised trial, possibly more if
the estimate of therapeutic effect is inaccurate or the
symptomatic bleeding risk was underestimated. A rigorous
study should include stringent reproduction of the
standard elements of the surgical and drug administration
tasks used in this trial in a cohort of patients recruited
from the widest possible set of stroke hospitals using
image guidance, CT imaging, and blood pressure control
to treat intracerebral haemorrhage. Safety and preliminary
outcome data indicate that MIS plus alteplase is a realistic
approach to treat intracerebral haemorrhage, possibly
improving long-term function and the ability to live at
home, and perhaps, decreasing cost, even for patients with
large intracerebral haemorrhage volumes.
Contributors
DFH and MZ organised the trial hypotheses, designed the trial, and
provided guidance about data analysis and interpretation and
presentation of the data. DFH drafted most of the sections of the
manuscript. EFA, CW, WCB, RSG, RD, J-LC, JH, PC, PV, NMa, IA,
ADM, BG, PK, KL, and NMc were involved in the design of the study
and wrote and revised the manuscript. KL, NMc, SWM, and AJB-H
organised and managed the trial, including trial start-up, data collection,
quality assurance, and trial close-out. DG, TCM, NU, and WAM
provided the region of interest calculations for all volumetric
measurement results. WZ, CK, and JRC provided independent review
and adjudication of all safety events. RET, JM, MR, CBT, GY, and EH
were involved in statistical analysis and data interpretation and
developed and revised the manuscript. SJ provided critical review of the
manuscript. The MISTIE investigators contributed equally to
identification and, when eligible, randomisation of trial participants.
Declaration of interests
IA, DFH, SWM, NU, KL, NMc, WAM, MR (R01NS046309 and
U01NS062851), CBT, and PV report grants from the National Institute of
Neurological Disorders and Stroke (NINDS) during the conduct of the
study. DFH reports non-financial support from Genentech and Johnson
& Johnson (Codman) during the conduct of the study, grants from
NINDS outside the submitted work, and expert testimony. SWM reports
personal fees from Johns Hopkins University outside the submitted
work. JM reports grants from the National Institutes of Health during the
conduct of the study and has a patent (C13388—primary intracerebral
haemorrhage prediction employing logistic regression and features
extracted from CT) pending for Johns Hopkins. ADM reports grants
from the National Institutes of Health during the conduct of the study,
non-financial support as the Director of the Newcastle Neurosurgery
Foundation, and personal fees from Advisor to Stryker and Draeger
outside the submitted work. BG reports grants from Johns Hopkins
University (MISTIE National Institutes of Health grant) during the
conduct of the study and grants from the National Institutes of Health
Research (UK) Health Technology Assessment Programme outside the
submitted work. All other authors declare no competing interests.
Acknowledgments
This study was supported by grant R01NS046309 awarded to DFH by the
National Institute of Neurological Disorders and Stroke. Alteplase was
donated by Genentech. Catheters and introducers were donated by
Codman. We thank the patients and families who volunteered for this
study, Rachel Dlugash for her assistance with data summarisation, and
Pat Reilly for her guidance.
References
1 Krishnamurthi RV, Feigin VL, Forouzanfar MH, et al. Global and
regional burden of first-ever ischaemic and haemorrhagic stroke
during 1990–2010: findings from the Global Burden of Disease
Study 2010. Lancet Glob Health 2013; 1: e259–81.
1236
2 Hachinski V, Donnan GA, Gorelick PB, et al. Stroke: working
toward a prioritized world agenda. Stroke 2010; 41: 1084–99.
3 Cadilhac DA, Dewey HM, Vos T, Carter R, Thrift AG. The health
loss from ischemic stroke and intracerebral hemorrhage:
evidence from the North East Melbourne Stroke Incidence Study
(NEMESIS). Health Qual Life Outcomes 2010; 8: 49.
4 Christensen MC, Mayer S, Ferran JM. Quality of life after
intracerebral hemorrhage: results of the Factor Seven for Acute
Hemorrhagic Stroke (FAST) trial. Stroke 2009; 40: 1677–82.
5 Mendelow AD, Gregson BA, Fernandes HM, et al. Early surgery
versus initial conservative treatment in patients with spontaneous
supratentorial intracerebral haematomas in the International
Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised
trial. Lancet 2005; 365: 387–97.
6 Mendelow AD, Gregson BA, Rowan EN, Murray GD, Gholkar A,
Mitchell PM. Early surgery versus initial conservative treatment in
patients with spontaneous supratentorial lobar intracerebral
haematomas (STICH II): a randomised trial. Lancet 2013;
382: 397–408.
7 Mayer SA, Brun NC, Begtrup K, et al. Efficacy and safety of
recombinant activated factor VII for acute intracerebral
hemorrhage. N Engl J Med 2008; 358: 2127–37.
8 Anderson CS, Heeley E, Huang Y, et al. Rapid blood-pressure
lowering in patients with acute intracerebral hemorrhage.
N Engl J Med 2013; 368: 2355–65.
9 Mould WA, Carhuapoma JR, Muschelli J, et al. Minimally invasive
surgery plus recombinant tissue-type plasminogen activator for
intracerebral hemorrhage evacuation decreases perihematomal
edema. Stroke 2013; 44: 627–34.
10 Wagner KR, Xi G, Hua Y, et al. Ultra-early clot aspiration after lysis
with tissue plasminogen activator in a porcine model of
intracerebral hemorrhage: edema reduction and blood-brain barrier
protection. J Neurosurg 1999; 90: 491–98.
11 Morgan T, Zuccarello M, Narayan R, Keyl P, Lane K, Hanley D.
Preliminary findings of the minimally-invasive surgery plus rtPA
for intracerebral hemorrhage evacuation (MISTIE) clinical trial.
Acta Neurochir Suppl 2008; 105: 147–51.
12 Hanley DF, Lane K, Broaddus WC, et al. MISTIE trial: 365-day
results demonstrate improved outcomes and cost benefit.
International Stroke Conference; Honolulu, Hawaii, USA;
Feb 6–8, 2013 (abstr L21).
13 Montes JM, Wong JH, Fayad PB, Awad IA. Stereotactic computed
tomographic-guided aspiration and thrombolysis of intracerebral
hematoma: protocol and preliminary experience. Stroke 2000;
31: 834–40.
14 Rohde V, Rohde M, Reinges L, Mayfrank L, Gilsbach JM.
Frameless stereotactically guided catheter placement and
fibrinolytic therapy for spontaneous intracerebral hematomas:
technical aspects and initial clinical results.
Minim Invasive Neurosurg 2000; 43: 9–17.
15 Zuccarello M, Brott T, Derex L. Early surgical treatment for
supratentorial intracerebral hemorrhage: a randomized feasibility
study. Stroke 1999; 30: 1833–39.
16 Minimally invasive surgery and rtPA for intracerebral hemorrhage
evacuation (MISTIE). https://clinicaltrials.gov/show/NCT00224770
(accessed June 17, 2015).
17 Josephson CB, White PM, Krishan A, Al-Shahi Salman R.
Computed tomography angiography or magnetic resonance
angiography for detection of intracranial vascular malformations in
patients with intracerebral haemorrhage.
Cochrane Database Syst Rev 2014; 9: CD009372.
18 Morgenstern LB, Hemphill JC 3rd, Anderson C, et al. Guidelines
for the management of spontaneous intracerebral hemorrhage: a
guideline for healthcare professionals from the American Heart
Association/American Stroke Association. Stroke 2010;
41: 2108–29.
19 Broderick J, Connolly S, Feldmann E, et al. Guidelines for the
management of spontaneous intracerebral hemorrhage in adults:
2007 update: a guideline from the American Heart Association/
American Stroke Association Stroke Council, High Blood Pressure
Research Council, and the Quality of Care and Outcomes in
Research Interdisciplinary Working Group. Circulation 2007;
116: e391–413.
www.thelancet.com/neurology Vol 15 November 2016

20 Morgan TC, Dawson J, Spengler D, et al. The Modified Graeb
Score: an enhanced tool for intraventricular hemorrhage
measurement and prediction of functional outcome. Stroke 2013;
44: 635–41.
21 Rotnitzky A, Lei Q, Sued M, Robins JM. Improved double-robust
estimation in missing data and causal inference models. Biometrika
2012; 99: 439–56.
22 Colantuoni E, Rosenblum M. Leveraging prognostic baseline
variables to gain precision in randomized trials. Stat Med 2015;
2602–17.
23 Cleveland WS, Grosse E, Shyu WM. Local regression models.
In: Chambers JM, Hastie TJ, eds. Statistical models. Pacific Grove:
Wadsworth & Brooks/Cole, 1992: 309–76.
24 Tuhrim S, Horowitz DR, Sacher M, et al. Volume of ventricular
blood is an important determinant of outcome in supratentorial
intracerebral hemorrhage. Crit Care Med 1999; 27: 617–21.
25 Davis SM, Broderick J, Hennerici M, et al. Hematoma growth is a
determinant of mortality and poor outcome after intracerebral
hemorrhage. Neurology 2006; 66: 1175–81.
26 Sauerbrei W, Royston P, Binder H. Selection of important
variables and determination of functional form for continuous
predictors in multivariable model building. Stat Med 2007;
26: 5512–28.
27 Yan X, Lee S, Li N. Missing data handling methods in medical
device clinical trials. J Biopharm Stat 2009; 19: 1085–98.
www.thelancet.com/neurology Vol 15 November 2016
Articles
28 Liublinska V, Rubin DB. Sensitivity analysis for a partially missing
binary outcome in a two-arm randomized clinical trial. Stat Med
2014; 33: 4170–85.
29 Steiner T, Kaste M, Forsting M, et al. Recommendations for the
management of intracranial haemorrhage—part I: spontaneous
intracerebral haemorrhage. The European Stroke Initiative Writing
Committee and the Writing Committee for the EUSI Executive
Committee. Cerebrovasc Dis 2006; 22: 294–316.
30 Steiner T, Al-Shahi Salman R, Beer R, et al. European Stroke
Organisation (ESO) guidelines for the management of spontaneous
intracerebral hemorrhage. Int J Stroke 2014; 9: 840–55.
31 Zhou H, Zhang Y, Liu L, et al. Minimally invasive stereotactic
puncture and thrombolysis therapy improves long-term outcome
after acute intracerebral hemorrhage. J Neurol 2011; 258: 661–69.
32 Zhou X, Chen J, Li Q, et al. Minimally invasive surgery for
spontaneous supratentorial intracerebral hemorrhage:
a meta-analysis of randomized controlled trials. Stroke 2012;
43: 2923–30.
33 Ziai W, Stadnik A, Zhang L, et al. Stabilizing bleeding prior to acute
therapies for spontaneous intracerebral hemorrhage. Stroke 2014;
45 (suppl 1): A35.
34 Gregson BA, Broderick JP, Auer LM, et al. Individual patient data
subgroup meta-analysis of surgery for spontaneous supratentorial
intracerebral hemorrhage. Stroke 2012; 43: 1496–504.
1237