- Email: [email protected]
10-Year Institutional Retrospective Case Series of Decompressive Craniectomy for Malignant Middle Cerebral Artery Infarction (mMCAI)
Accepted Manuscript 10-Year Institutional Retrospective Case Series of Decompressive Craniectomy for Malignant Middle Cerebral Artery Infarction (mMCAI) Simon Lammy, MRCS (Ed) PgDip (Oxon), SpR Neurological Surgery (Neurosurgery), Basel Al-Romhain, MRCS (Ed), SpR Neurological Surgery (Neurosurgery), Laura Osborne, MB ChB, FY1 General Medicine, Edward J. St. George, FRCS (SN), Consultant Neurological Surgeon (Neurosurgery) PII:
S1878-8750(16)30817-8
DOI:
10.1016/j.wneu.2016.09.004
Reference:
WNEU 4544
To appear in:
World Neurosurgery
Received Date: 3 June 2016 Revised Date:
31 August 2016
Accepted Date: 1 September 2016
Please cite this article as: Lammy S, Al-Romhain B, Osborne L, St. George EJ, 10-Year Institutional Retrospective Case Series of Decompressive Craniectomy for Malignant Middle Cerebral Artery Infarction (mMCAI), World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.09.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 10-Year Institutional Retrospective Case Series of Decompressive Craniectomy for Malignant Middle Cerebral Artery Infarction (mMCAI)
Simon Lammy MRCS (Ed) PgDip (Oxon) SpR Neurological Surgery (Neurosurgery)
RI PT
Department of Neurosurgery Institute of Neurological Sciences 1345 Govan Road
SC
Glasgow G51 4TF
Corresponding Author
TE D
Email: [email protected]
Basel Al-Romhain MRCS (Ed)
SpR Neurological Surgery (Neurosurgery)
EP
Department of Neurosurgery
Institute of Neurological Sciences
AC C
1345 Govan Road Glasgow
G51 4TF
United Kingdom
M AN U
United Kingdom
ACCEPTED MANUSCRIPT Laura Osborne MB ChB FY1 General Medicine Department of General Medicine Forth Valley Royal Hospital Stirling Road
FK5 4WR United Kingdom
Edward J. St. George FRCS (SN)
Department of Neurosurgery Institute of Neurological Sciences 1345 Govan Road
G51 4TF
AC C
EP
United Kingdom
TE D
Glasgow
M AN U
Consultant Neurological Surgeon (Neurosurgery)
SC
RI PT
Larbert
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Keywords: Malignant; Craniectomy; Decompressive; Mortality; Glasgow Outcome Scale (GOS) Score
ACCEPTED MANUSCRIPT Abstract A 10-year retrospective case series of patients undergoing decompressive craniectomy for malignant middle cerebral artery infarction (mMCAI) was undertaken (2005-2015). Patient demographics, co-morbidities, pre- and post-operative neurological state, operative timescales, craniectomy dimensions and Glasgow Outcome Scale (GOS) scores were analysed. Overall 40 patients underwent a decompressive craniectomy for mMCAI with a 30-
RI PT
day mortality of 17.5% (n=7). 17 patients (42.5%) were male with a mean age of 43yrs (range: 16-64yrs). Patients who survived had a lower mean age of 41yrs (range: 16-59yrs) than those who did not of 50yrs (range: 42-63yrs). The modal ictal and preoperative GCS were 14 (range: 5–15) and 7 (range: 3–12) which corresponded to motor
SC
scores of 6 and 5 respectively. The mean time from ictus to INS admission was 23.5hrs (range: 0.5-66hrs) and from INS admission to decompression 7.5hrs (range: 0.5-46hrs). Approximately 60% of patients had an “early”
craniectomy (under 48hrs from ictus) and 60% of patients had a craniectomy performed less than 24hrs from INS
M AN U
admission. The mean maximum antero-posterior (AP) craniectomy diameter measured 13cm (range: 10.93– 15.12cm) and the mean surface area was 92.68 cm2 (range: 76.14–124.42cm2). Overall 80% of patients had a modal Glasgow Outcome Scale (GOS) score of 3 (range: 2-5) at discharge, 3-months, 6-months, 9-months and 12-months. The median length of stay was 3 days (range: 6hrs - 11 days) for non-survivors and 13 days (range: 1 - 365 days) for survivors. Decompressive craniectomy for mMCAI is suitable in selected patients and the local
AC C
EP
TE D
practice is consistent with current evidence1.
ACCEPTED MANUSCRIPT Introduction Malignant transformation of a middle cerebral artery (MCA) infarct is a life threatening complication that has 80% mortality if conservatively treated1. With infarction of greater than 50% of MCA territory significant mass effect and concomitant herniation of brain tissue may occur. Recent randomised controlled trials (RCTs)2,3,4 and subsequent meta-analysis of the data5 make it clear that surgical intervention, through a decompressive craniectomy,
RI PT
improves survival but at the potential expense of good functional outcome5. In this retrospective case series
outcome data is presented in the light of recent published evidence. Current controversies in the neurosurgical
literature are discussed and factors which might have contributed to an unfavourable outcome are examined and
AC C
EP
TE D
M AN U
SC
compared to published evidence.
ACCEPTED MANUSCRIPT Methods A retrospective analysis was performed of all patients who underwent a decompressive craniectomy for malignant middle cerebral artery infarction (mMCAI) at the Institute of Neurological Sciences (INS) between January 2005 and January 2015 (see Figure 1). The INS is the busiest and biggest neurosurgical centre in Scotland and covers one of the larger geographical areas in the United Kingdom having a catchment population of over 3 million. The
RI PT
study was conducted according to the STROBE Statement on Cohort Studies for a retrospective case series and our historical selection criteria was judged against National Institute for Health and Care Excellence (NICE)
SC
guidelines as published in the United Kingdom in 20096,7.
Current selection criteria are based on these guidelines and includes four parameters: age <60 years, National Institute of Health Stroke Scale (NIHSS) >15, >50% MCA territory hypodensity on CT (or >145cm3 on diffusion
M AN U
weighted [dw] MRI) and urgent referral to a neurosurgical centre <24 hours from symptom onset and subsequent decompression in eligible patients <48 hours from ictus7. If a patient fulfils these guidelines our practice is to consider each patient on a case by case basis in regard to co-morbidities and pre-ictal functional status. Due of this analysis being retrospective and pre-dating NICE guidelines our inclusion criteria for this case series were patients that had both a clinical and radiological diagnosis of malignant MCA infarction (according to the World
TE D
Health Organisation [WHO] criteria)8 and who then subsequently underwent a decompressive craniectomy.
Electronic and hand written case notes, Scottish National Picture Archiving and Communication System (PACS)
EP
and operative theatre log books were interrogated to extract the relevant data. Once an MCA territory infarction and subsequent decompressive craniectomy were confirmed there was no exclusion criteria based on extent of
AC C
MCA territory infarction (if it was >50% of MCA territory), imaging modalities used, adverse radiological signs and other characteristics, e.g. patient demographics (see below).
Patient demographics (including age, sex and co-morbidities), pre- and post-operative neurological state, operative timescales, craniectomy dimensions and Glasgow Outcome Scale (GOS) scores at discharge, 3months, 6-months, 9-months and 12-months were documented. Unless cranioplasty is considered follow up is primarily conducted by stroke physicians and examination of detailed inpatient and outpatient clinic notes (up to 12-months post-ictus) was conducted to extract outcome information regarding GOS over this time period.
ACCEPTED MANUSCRIPT Risk factors for vascular disease, e.g. pre-ictal transient ischaemic attack (TIA), hypertension (HTN), diabetes mellitus (DM), ischaemic heart disease (IHD), smoking history and a history of alcohol excess were noted. Other non-specific co-morbidities recorded included depression, obesity, a history of malignancy, endocrine and other chronic diseases.
RI PT
Focal neurological findings included the presence of hemi-paresis (change from a patient’s pre-ictus baseline),
speech deficits, e.g. receptive or expressive dysphasia’s, seizure activity, confusion, collapse, incontinence and headache. These signs were recorded both pre-operatively at the time of the initial ictus and post-operatively.
SC
Other post-operative domains included the presence of depression, language comprehension and development of common surgical complications including thrombo-embolic phenomena (as defined by the incidence of deep vein
M AN U
thrombosis and pulmonary embolus) up to 30-days post-operatively.
Additionally, the ictal and pre-operative Glasgow Coma Scale (GCS) scores (specifically the motor scores), incidence of thrombolysis, site of the infarct and sub-sequent decompression, timescale from ictus to INS admission and from INS admission to decompression, overall length of stay, physical dimensions of the
TE D
craniectomy and subsequent cranioplasty complications were documented.
To minimise errors in data collection, all craniotomies performed in the West of Scotland over the study period were scrutinised for accuracy of diagnosis. The radiological modality used was computer tomography (CT) which
EP
represents the primary imaging modality used in the diagnosis of stroke across the United Kingdom. IBM’s
AC C
Statistical Package for the Social Sciences (SPSS) was used to perform statistical analysis.
ACCEPTED MANUSCRIPT Results 40 patients underwent a decompressive craniectomy for malignant MCA infarction. The 30-day mortality was 17.5% (n=7). 17 patients (42.5%) were male with a mean age of 43yrs (range: 16-64yrs) (NB: only one patient was over 60yrs) and 18 had a left sided (dominant) hemispheric infarction (45%). Patients who survived had a lower mean age of 41yrs (range: 16-59yrs) then those who did not of 50yrs (range: 42-64yrs) (p=0.04). The most
not statistically significant with regards to mortality (see Table 1 [p=0.68]).
RI PT
frequently recorded risk factor for cerebrovascular disease was smoking in our cohort (n=22 [55%]) but this was
SC
Other risk factors for cerebrovascular disease, e.g. HTN and IHD, were not predictive of mortality (Table 1). Age proved to be a stronger predictor of the likelihood of survival than the presence of co-morbidity5. This can be further demonstrated by analysing the occurrence of multiple co-morbidities in an individual: interestingly non-
M AN U
survivors although being older had less co-morbidities as a group and less co-morbidities as individuals (see Figure 2).
Pre-operative focal neurological signs did not significantly change following decompression, e.g. the incidence of aphasia was unchanged at 55%. Thrombolysis performed at the time of the original ictal event (before mMCAI
TE D
developed) did not appear to influence mortality. The risk of death in the 12 patients thrombolysed was 16% (n=2) and in the 28 patients not thrombolysed was 18% (n=5) (see Table 1). This was statistically insignificant (p=1.00) and thrombolysis had a relative risk (RR) of death of 81% (i.e. 0.81). Overall 30% of survivors had been
EP
thrombolysed compared to 29% of non-survivors. Due to the small numbers of non-survivors it remains uncertain if
AC C
these results are clinically significant one.
The modal ictal and preoperative GCS were 14 (range: 5–15) and 7 (range: 3–12) which corresponded to motor scores of 6 and 5 respectively. Patients who survived had a higher modal ictal GCS (14: range: 10-15) than those who did not at (10: range: 6-15) but the modal pre-operative GCS scores were similar between the two groups (8: range: 3-15 for survivors and 8: range: 7-10 for non-survivors) (see Table 1). The mean time from ictus to INS admission was 23.5hrs (range: 0.5-66hrs) and from INS admission to decompression 7.5hrs (range: 0.5-46hrs). Approximately 60% of patients had an “early” craniectomy (under 48hrs from ictus) and 60% of patients had a craniectomy performed in under 24hrs from INS admission (see Table 2).
ACCEPTED MANUSCRIPT The mean maximum antero-posterior (AP) craniectomy diameter measured 13cm (10.93–15.12cm) and the mean surface area was 92.68 cm2 (76.14–124.42cm2). Post-operative complications following decompression included infection (respiratory [40%], urinary [10%] and wound [13%]), thrombo-embolic phenomena (DVT and PE [5%]) and post-operative depression (20%). Thirty percent (30%) of patients exhibited ongoing poor comprehension
RI PT
following surgery.
Overall 80% of patients had a modal Glasgow Outcome Scale (GOS) score of 3 (range: 2-5) at discharge, 3-
months, 6-months, 9-months and 12-months. Of the survivors 17 patients subsequently had an insertion of a
SC
titanium cranioplasty and 7 of an autologous bone flap re-insertion within 12-months following decompression.
Post-operative complications related to cranioplasty included 4 revisions for infection (titanium) and 1 bone flap
M AN U
being discarded for wound irritation.
Overall the mean Neurosurgical Unit (NSU) length of stay was 29 days (range: 6hrs - 365 days), which was split into a mean length of stay of 3.89 days (range: 6hrs - 11 days) for non-survivors and 23.12 (range: 1 - 365 days) for survivors. This translated into a median length of stay of 3 for non-survivors and 13 for survivors. Survivors
AC C
EP
TE D
were subsequently discharged to rehabilitation units.
ACCEPTED MANUSCRIPT Discussion The practice of decompressive craniectomy for malignant middle cerebral artery infarction (mMCAI) is controversial: it is an emergency neurosurgical intervention for a life threatening condition whose success cannot simply be measured in terms of survival1,9. Quality of survival is an important factor in complex elective neurosurgical decision making but such a luxury is not afforded to the neurosurgeon in a rapidly deteriorating
RI PT
situation. The context of such a controversy is best understood in terms of its epidemiology, pathophysiology and subsequent management options.
SC
In the United Kingdom stroke is the 4th leading cause of death and the largest cause of complex disability9,10. Over one third of stroke survivors are dependent on others. Over 152,000 people in the United Kingdom have a stroke per annum with an associated 30-day mortality of 12.5% increasing to 25% at 12-months10. Ischaemia due to
M AN U
occlusive cerebrovascular disease accounts for approximately 80% of strokes1,9,10. The incidence of mMCAI is 1% of all strokes (approximately 1520 patients per annum)9,10 and carries an 80% 30-day mortality if neurosurgical intervention does not occur1,8.
The systematic review by Vahedi et al5 of the first three published randomised controlled trials (RCTs) of
TE D
decompressive craniectomy for mMCAI (DECIMAL2, DESTINY3 and HAMLET4) demonstrated an absolute risk reduction (ARR) for mortality of 51% (Oxford Centre for Evidence Based Medicine Level 1b)11. This translates into a number needed to treat (NNT) to prevent death of 2. To prevent a modified Rankin Scale score (mRS) of >5 the
EP
NNT is 2 and to prevent an mRS>4 up to 4 patients need to be decompressed (NNT 4)9. Despite showing an improvement in survival there were discrepancies between the three trials regarding what constituted a good
AC C
functional outcome for the patient. Vahedi et al considered a patient with an mMRS <4 as having a good outcome. Of survivors 29% had an mRS>4 in the surgical arm and 23% in the conservative arm indicating that surgery increased survival at the cost of a slightly worsened functional outcome, i.e. surgery produced more dependent survivors5,9.
Nevertheless, this should be judged in context of the actual numbers. If one considers an annual incidence of 1520 cases of mMCAI then conservative management alone affords a 20% survival rate (n=304) whereas craniectomy (due to an ARR of 51%) permits an extra 152 survivors (n=456). If an mRS<4 is considered a good functional outcome then conservative management results in 234 patients (77% [23% mRS>4]) to exist in those
ACCEPTED MANUSCRIPT categories compared to 323 (71% [29% mRS>4] patients in the surgical arm. Simply put decompressive craniectomy produces more patients with a good functional outcome (around 100 extra patients) i.e. an mRS<4 as a result of an ARR of 51%.
Data from three subsequent trials (e.g. HeADDFIRST (2003)12, Zhao J et al (2012)13 and DESTINY II (2014)14,
RI PT
demonstrated mortality rates of 27%, 13% and 43% in the surgical arms and 46%, 61% and 75% in the non-
surgical arms respectively. The HeADDFIRST12 trial did not publish data on functional outcomes but Zhao et al13
and DESTINY II14 reported an mRS<3 in 9% and 10% of survivors in the surgical arm compared to 21% and 10%
SC
in the non-surgical arm. These generally reflect the pooled analysis by Vahedi et al5.
The pathophysiology of mMCAI is not straightforward and its clinical and radiological predictors are broad9. The
M AN U
condition arises when there is >50% MCA territory infarction producing a potentially significant space occupying effect and consequent herniation of brain tissue. Death usually occurs under one week of symptom onset as a result of progressive swelling of ischaemic brain tissue and sub-falcine (as opposed to rostral-caudal) herniation15. There is a focal (as opposed to a global) increase in intra-cranial pressure (ICP) and extension of the ischaemia to
TE D
adjacent vascular territories9,15. Survivors may therefore be highly disabled and have a poor quality of life5.
Signs include nausea, vomiting, headache, decreased GCS and respiratory compromise9. These signs may occur much quicker in younger patients (approximately 3 hours) due to less compensatory space and sometimes much
EP
later in older patients (around 5 days)1,9. Although elderly patients may deteriorate much later due to relatively more compensatory space such clinical deterioration may indicate a focal lesion that is relatively larger and less
AC C
amenable to aggressive medical and surgical intervention. This might account for the lower mean age of survivors in this study and those of others.
High risk patients have a baseline National Institute of Health Stroke Scale (NIHSS) score of >20 in dominant and >15 in non-dominant hemispheric infarcts in conjunction with >50% hypodensity of MCA territory on CT <6 hours from symptom onset9. In the United Kingdom CT remains the emergency radiological modality of choice due its ease of access and has a sensitivity between 61-94% and a specificity of 91-94% in predicting a fatal outcome. This compares favourably to diffusion weighted MRI which has 100% sensitivity and 94% specificity in predicting a poor outcome, e.g. death and mRS>4, when the lesion is >145cm3 in volume1,5,9,16.
ACCEPTED MANUSCRIPT Despite the controversies surrounding decompressive craniectomy in mMCAI it remains the only condition that has randomised controlled trial (RCT) data to support its use to improve survival. Clarke and Harris (1958) originally proposed the hypothesis that mMCAI should be considered a surgical mass lesion due to the similarities with an extra-dural haematoma (EDH) and its ability to produce malignant focal increases in ICP17. Surgical intervention should have clear goals and is primarily intended to prevent death from brainstem herniation and
RI PT
secondarily to permit the swollen infarct to expand away from the midline enabling a permanent decrease in focal ICP.
SC
Understandably concerns remain regarding patient selection. In 2009 the National Institute for Health and Care
Excellence (NICE), on the basis of the pooled analysis by Vahedi et al5, proposed guidelines predictive of survival and good functional outcome7. These included four parameters: age <60 years, NIHSS >15, >50% MCA territory
M AN U
hypodensity on CT (if possible >145cm3 on dwMRI) and urgent referral to a neurosurgical centre <24 hours from symptom onset and subsequent decompression in eligible patients <48 hours from ictus. This guideline as well as data from the six individual RCTs was used to assess this cohort’s outcomes. This study’s results were comparable to published data with respect to patient selection, timescale to craniectomy, craniectomy dimensions
TE D
and subsequent functional outcome and these will be discussed in turn.
In the present series survivors had a lower mean age than non-survivors (41yrs vs. 50yrs) which is consistent to the sub-group pooled analysis which demonstrated improved survival in patients <50yrs5. Patient selection
EP
conformed to the NICE guidelines in terms of age as only one patient was <18yrs and one >60yrs in the cohort. The importance of age as a predictor of outcome was demonstrated as the best functional outcome at discharge
AC C
(GOS 5) was achieved in patients with a mean age of 30yrs (range: 16-45yrs [n=4]), GOS 4 in those with a mean age of 37yrs (range: 33-51yrs [n=7]) and GOS 3 (often considered a good functional outcome and roughly equivalent to mRS 4 in the pooled analysis) in survivors with a mean age of 44yrs (range: 28-59yrs [n=22]). The mean age of patients achieving a GOS 2 was 53yrs (range: 46-59yrs [n=2]) and the mean age for those with GOS 1 was 51yrs (range: 42-63yrs [n=5]).
An almost linear relationship between age and GOS at discharge was demonstrated with a very high negative correlation (Spearman’s Correlation Coefficient r= -0.900 [R2=0.81] p= 0.37). Increased age decreases patients GOS and up to 81% of the variation in GOS is related to the variation in age (Spearman’s Correlation Coefficient2)
ACCEPTED MANUSCRIPT (see Figure 3). Furthermore, the results demonstrate a GOS 3-4 is anticipated in patients with a mean age of 40 (compared to 50 as observed in the Vahedi pooled analysis) which is similar to the age related outcome derived from the traumatic brain injury prognosis calculator developed from the CRASH studies18.
Baseline Glasgow Coma Scale (GCS) scores (particularly the motor component) is often not reported in studies of
RI PT
this nature. In traumatic brain injury (TBI) and the subsequent prognostications from the CRASH studies the motor component and age have been found to be good predictors of outcome 18. In this case series patients who
survived had a higher modal ictal GCS at 14 (range: 10-15) than those who did not at 10 (range: 6-15) but the
SC
modal pre-operative GCS scores were the same for the two groups (8 for survivors [range: 3-15] and 8 for nonsurvivors [range: 7-10]) (see Table 1).
M AN U
Although non-survivors in this study had a higher mean age they also had less co-morbidity burden (in terms of number and complexity) than survivors and this cannot be easily explained (see Table 1). This can be further demonstrated by analysing the frequency of multiple co-morbidities in an individual (see Figure 2). No association was found between co-morbidity and survival in the study presented. Furthermore, our results tentatively suggest that if a patient had been thrombolysed as part of medical management following the initial ictus, acceptance of
TE D
the patient for decompression (once mMCAI becomes apparent) probably should not be delayed (see Results).
Despite the mean time from ictus to INS admission being 23.5hrs (0.5-66hrs) and from INS admission to
EP
subsequent decompression being 7.5hrs (0.5-46hrs) (adhering to NICE guidelines based on Vahedi et al) a higher mortality was seen in the group treated <48hrs (“early”) (n=6 [19.3%]) when compared to those treated >48hrs
AC C
(“late”) (n=1 [11.1%]) (see Table 2). However, the overall numbers are small and not comparable (n=31 <48hrs vs. n=9 >48hrs) and meaningful statistical and clinical correlations cannot be generated but the finding does raise the possibility that decompressing patients faster might not necessarily increase survival in this sub-group and the impact on functional outcome remains undetermined (see Table 2).
Using post-operative 3D reconstruction of CT images the mean AP diameter was 13cm (range: 10.93–15.12cm) resulting in a mean surface area of 92.68 cm2 (range: 76.14–124.42cm2) which is in keeping with published guidelines15,19. As no post-operative CT was performed on five of the seven non-survivors (due to death occurring <24hrs post-operatively) no correlation could be established between craniectomy dimension and survival (or
ACCEPTED MANUSCRIPT craniectomy dimension and functional outcome). The previously recommended anterior posterior (AP) diameter of 12cm has been increased to 14cm as this permits decompression beyond the infarct margins15.
This is usually surgeon dependent variable but a 76% increase in extra volume occurs from when the AP diameter is increased from 12cm to 14cm15. The minimum dorsal ventral (DV) diameter is ~9cm and although being variable
RI PT
it is generally surgeon independent as it is limited by the floor of the middle cranial fossa and superior sagittal
sinus (SSS)15. This dimension is of particular importance in regard to preventing uncal herniation and subsequent
SC
brainstem ischaemia and irreversible infarction (death). Our results are consistent to current practice.
The incidence of cranioplasty complications in this cohort could not be placed in context as these are seldom reported in the literature. But craniectomy specific complications do occur and these suggest that decompressive
M AN U
craniectomy should be perceived as “a two stage operation”, i.e. initially to preserve life (and possibly function) but subsequently one which requires a further procedure to repair the defect (otherwise the risk of sinking flap syndrome and delayed dysautonmic symptoms is increased). Consequently, a multi-centre prospective cohort study that documents this complication is needed in order to gain adequate numbers to properly quantify post-
TE D
operative cranioplasty risks.
Potential sources of bias included selection bias as the overall number of patients referred to our neurosurgical centre for mMCAI from 2005-2015 could not be quantified (due to the retrospective nature of this study). Also
EP
internal validity was not possible as no previous institutional case series according to the Strobe Statement on Cohort Studies had been conducted on this subject matter. The extent of infarction to predict survival could not be
AC C
conducted as volumetric analysis using MRI was not employed as a mode of imaging in a single patient.
ACCEPTED MANUSCRIPT Conclusions The centre’s practice has been and remains consistent with NICE guidelines. However, it demonstrates that early decompression might not influence mortality nor improve functional outcome and that more research is needed regarding functional outcome and GCS (particular the motor component) at ictus.
RI PT
Acknowledgements
The authors wish to thank NHS Greater Glasgow & Clyde and the Department of Neurosurgery at the Institute of
SC
Neurological Sciences (INS) for their support.
Author Contributions
BA prepared the Introduction. SL prepared the Abstract and Methods and Results and Discussion (and formulated
M AN U
the Tables and Figures). LO extracted the data for 2005–2013 which was independently checked by SL. SL extracted further data for 2013-2015. JSt.G and SL revised the whole manuscript and double checked it for accuracy and fluency. SL re-revised the manuscript again in light of editorial comments. JSt.G proposed and supervised the project.
Funding
EP
None declared.
TE D
Declaration of Conflicting Interests
This research received no specific grant from any funding agency in the public, commercial or not-for-profit
AC C
sectors.
ACCEPTED MANUSCRIPT Tables Characteristics of Patients Alive and Dead Following Craniectomy for Malignant MCA Infraction Dead (n=7)
P Value
Male
13 (39.9%)
4 (57.1%)
0.43*
Age
41.52yrs
51.50yrs
0.04^
(16-59yrs [+/- 9.830])
(42-64yrs [+/- 6.979])
16 (48.4%)
2 (28.6%)
0.42*
n=10
n=2
1.00*
Thrombolysis Risk Factors (Vascular) Hypertension
5 (15.2%)
•
Ischaemic Heart Disease
3 (9.1%)
•
Transient Ischaemic Attack
•
Diabetes Mellitus
• •
1 (14.2%)
1.00*
0 (0.0%)
1.00*
M AN U
•
SC
Dominant Hemisphere
RI PT
Alive (n=33)
8 (24.2%)
0 (0.0%)
0.31*
3 (9.1%)
1 (14.2%)
0.55*
Current Smoker
19 (57.6%)
3 (42.9%)
0.68*
Alcohol Excess
10 (30.3%)
1 (14.2%)
0.65*
4 (12.1%)
0 (0.0%)
1.00*
8 (24.2%)
1 (14.2%)
1.00*
19 (57.6%)
2 (28.6%)
0.23*
8 (24.2%)
1 (14.2%)
1.00*
14 (10-15)
10 (6-15)
0.73”
8 (3-15)
8 (7-10)
1.00”
TE D
Risk Factors (Other) Obesity
•
Depression
•
Other 1, e.g. Malignancy
•
Other 2, e.g. Endocrine
AC C
GCS (Modal)
EP
•
•
Ictal
•
Pre-Operative
Table 1: Demographic and clinical characteristics of survivors and non-survivors. Non-survivors were older and had a lower modal ictal GCS but less co-morbidities than survivors. NB: the time to decompression did not influence outcome survival nor function (NB: * Fischer Exact Test; ^Wilcoxon Signed Rank Test, “Chi Squared Test)
ACCEPTED MANUSCRIPT Characteristics of Patients Having Early and Late Craniectomy for Malignant MCA Infraction Late >48hrs (n=9)
P Value
Male
10 (32.3%)
7 (77.8%)
0.02*
Age
31yrs (26-64yrs)
42yrs (16-51yrs)
0.55
Dominant Hemisphere
11 (35.4%)
7 (77.8%)
0.05*
Mortality
6 (19.3%)
1 (11.1%)
1.00*
n=9 (Mortality: 67% [n=6])
n=3 (Mortality: 33%
1.00* (Mortality:
[n=1])
0.53*)
•
Ictal
15 (9-15)
•
Pre-Operative
M AN U
GCS (Modal)
SC
Thrombolysis
RI PT
Early <48hrs (n=31)
GOS (Modal)
14 (10-15)
0.89^
8 (3-15)
6, 8, 9 (6-12)
1.00^
3 (1-5)
3 (3-4)
1.00^
Table. 2: Characteristics of patients having “early” (<48hrs) compared to “late” (>48hrs) decompressive
TE D
craniectomies. Patients decompressed late did not necessarily have higher mortality nor higher modal ictal and
AC C
EP
pre-operative GCS scores despite having a higher mean age. (NB: * Fischer Exact Test; “Chi Squared Test)
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Figures
AC C
EP
TE D
Fig. 1: This shows the number of decompressive craniectomies performed between January 2005 – January 2015
Fig. 2: Number of patients (y-axis) and the frequency of co-morbidities (x-axis). Non-survivors had less comorbidities individually and as a group than survivors
AC C
EP
TE D
Fig. 3: Age (y-axis) vs. GOS (x-axis)
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT References
1.
Treadwell SD, Thanvi B.Malignantmiddle cerebral artery (MCA) infarction: pathophysiology, diagnosisand management. Postgrad Med J. 2010; 86(1014): 235-42
Vahedi K, Vicaut E, Mateo J, Kurtz A, Orabi M, Guichard JP, Boutron C, Couvreur G, Rouanet F, Touzé
RI PT
2.
E, Guillon B, Carpentier A, Yelnik A, George B, Payen D, Bousser MG; DECIMAL Investigators.
Sequential-design, multicenter, randomized, controlled trial of early decompressive craniectomy in
3.
SC
malignant middle cerebral artery infarction (DECIMAL Trial). Stroke. 2007; 38(9): 2506-17
Jüttler E, Schwab S, Schmiedek P, Unterberg A, Hennerici M, Woitzik J, Witte S, Jenetzky E, Hacke W;
M AN U
DESTINY Study Group. Decompressive Surgery for the Treatment of Malignant Infarction of the Middle Cerebral Artery (DESTINY): a randomized, controlled trial. Stroke. 2007; 38(9): 2518-25
4.
Hofmeijer J, Kappelle LJ, Algra A, Amelink GJ, van Gijn J, van der Worp HB; HAMLET investigators. Surgical decompression for space-occupying cerebral infarction (the Hemicraniectomy After Middle
TE D
Cerebral Artery infarction with Life-threatening Edema Trial [HAMLET]): a multicentre, open, randomised trial. Lancet Neurol. 2009; 8(4): 326-33
Vahedi K, Hofmeijer J, Juettler E, Vicaut E, George B, Algra A, Amelink GJ, Schmiedeck P, Schwab S,
EP
5.
Rothwell PM, Bousser MG, van der Worp HB, Hacke W; DECIMAL, DESTINY, and HAMLET
AC C
investigators. Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomised controlled trials. Lancet Neurol. 2007; 6(3): 215-22
6.
STROBE: Strengthening the Reporting of Observational Studies in Epidemiology. Checklist. Case Control Studies 2007 (http://www.strobe–statement.org/index.php?id=available–checklists).
7.
NICE : National institue for Heath and Clinical Excellence. Surgical Referral for DecompressiveHemicraniectomyhttps://www.nice.org.uk/guidance/cg68/resources/13-2-surgicalreferral-for-decompressive-hemicraniectomy
ACCEPTED MANUSCRIPT 8.
WHO: World HealthOrganisation. Topic. Stroke. 2014 (http://www.who.int/topics/cerebrovascular_accident/en/)
9.
Subramaniam S, Hill MD. Decompressive hemicraniectomy for malignant middle cerebral artery
10. Stroke Association. State of The Nation: Stroke Statistics. 2016
SC
(https://www.stroke.org.uk/resources/state-nation-stroke-statistics)
RI PT
infarction: an update. Neurologist. 2009; 15(4):178-84
11. OCEBM: Oxford Centre for Evidence Based Medicine. Levels of Evidence. 2009
M AN U
(http://www.cebm.net/oxford-centre-evidence-based-medicine-levels-evidence-march-2009/)
12. Frank JI, Schumm LP, Wroblewski K, Chyatte D, Rosengart AJ, Kordeck C, Thisted RA; HeADDFIRSTTrialists. Hemicraniectomy and durotomy upon deterioration from infarction-related swelling trial: randomized pilot clinical trial. Stroke. 2014; 45(3): 781-7
TE D
13. Zhao J, Su YY, Zhang Y, Zhang YZ, Zhao R, Wang L, Gao R, Chen W, Gao D. Decompressive hemicraniectomy in malignant middle cerebral artery infarct: a randomized controlled trial enrolling
EP
patients up to 80 years old. Neurocrit Care. 2012; 17(2): 161-71
14. Jüttler E, Bösel J, Amiri H, Schiller P, Limprecht R, Hacke W, Unterberg A; DESTINY II Study Group.
AC C
DESTINY II: DEcompressive Surgery for the Treatment of malignant INfarction of the middle cerebral arterY II. Int J Stroke. 2011; 6(1): 79-86
15. Johnson RD, Maartens NF, Teddy PJ.Technical aspects of decompressive craniectomy for malignant middle cerebral artery infarction. J ClinNeurosci. 2011; 18(8): 1023-7
16. Hatefi D, Hirshman B, Leys D, Lejeune JP, Marshall L, Carter BS, Kasper E, Chen CC.Hemicraniectomy in the management of malignant middle cerebral artery infarction: Lessons from randomized, controlled trials. SurgNeurol Int. 2014; 15(5): 72
ACCEPTED MANUSCRIPT 17. Clarke E, Harris P. Thrombosis of the internal carotid artery simulating an intracranial space-occupying lesion. Lancet. 1958; 1(7030): 1085-9
18. Roberts I, Yates D, Sandercock P, Farrell B, Wasserberg J, Lomas G, Cottingham R, Svoboda P, Brayley N, Mazairac G, Laloë V, Muñoz-Sánchez A, Arango M, Hartzenberg B, Khamis H,
RI PT
Yutthakasemsunt S, Komolafe E, Olldashi F, Yadav Y, Murillo-Cabezas F, Shakur H, Edwards P;
CRASH trial collaborators. Effect of intravenous corticosteroids on death within 14 days in 10008 adults
2004; 364(9442): 1321-8.
SC
with clinically significant head injury (MRC CRASH trial): randomised placebo-controlled trial. Lancet.
19. Ransohoff J, Benjamin V. Hemicraniectomy in the treatment of acute subdural haematoma. J Neurol
AC C
EP
TE D
M AN U
Neurosurg Psychiatry. 1971; 34(1): 106.
ACCEPTED MANUSCRIPT Highlights (For Review) A 10-year retrospective case series of patients undergoing decompressive craniectomy for malignant middle cerebral artery infarction (mMCAI) was undertaken (2005-2015). Patient demographics, co-morbidities, pre- and post-operative neurological state, operative timescales, craniectomy dimensions and Glasgow Outcome Scale (GOS) scores were analysed. Overall 40 patients underwent a decompressive craniectomy for mMCAI with a 30-day mortality of 17.5% (n=7). 17 patients (42.5%) were male with a
RI PT
mean age of 43yrs (range: 16-64yrs). Patients who survived had a lower mean age of 41yrs (range: 16-59yrs) than those who did not of 50yrs (range: 42-63yrs). The modal ictal and preoperative GCS were 14 (range: 5–15) and 7 (range: 3–12) which
corresponded to motor scores of 6 and 5 respectively. The mean time from ictus to INS admission was 23.5hrs (range: 0.5-66hrs)
SC
and from INS admission to decompression 7.5hrs (range: 0.5-46hrs). Approximately 60% of patients had an “early” craniectomy (under 48hrs from ictus) and 60% of patients had a craniectomy performed less than 24hrs from INS admission. The mean maximum antero-posterior (AP) craniectomy diameter measured 13cm (range: 10.93–15.12cm) and the mean surface area was 92.68 cm2
M AN U
(range: 76.14–124.42cm2). Overall 80% of patients had a modal Glasgow Outcome Scale (GOS) score of 3 (range: 2-5) at discharge, 3-months, 6-months, 9-months and 12-months. The median length of stay was 3 days (range: 6hrs - 11 days) for nonsurvivors and 13 days (range: 1 - 365 days) for survivors. Decompressive craniectomy for mMCAI is suitable in selected patients and the local practice is consistent with current evidence1.
TE D
Mr. Simon Lammy MRCS (Ed) SpR Neurological Surgery (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
EP
Mr. Basel Al-Romhain MRCS (Ed) SpR Neurological Surgery (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
AC C
Dr. Laura Osborne MB ChB FY1 General Medicine Department of General Medicine Forth Valley Royal Hospital
Mr. Edward J. St. George FRCS (SN) Consultant Neurological Surgeon (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
ACCEPTED MANUSCRIPT Abbreviations and Acronyms ARR: Absolute Risk Reduction CRASH: Corticosteroid Randomisation After Significant Head Injury CT: computer tomography DM: Diabetes Mellitus
RI PT
EDH: Extra-Dural Haematoma GCS: Glasgow Coma Scale GOS: Glasgow Outcome Scale
SC
HTN: Hypertension ICP: Intracranial Pressure
INS: Institute of Neurological Sciences MCA: Middle Cerebral Artery mMCAI: Middle Cerebral Artery Infarction MRI: magnetic resonance imaging mRS: Modified Rankin Scale
M AN U
IHD: Ischaemic Heart Disease
TE D
NICE: National Institute for Health and Clinical Excellence NIHSS: National Institute of Health Stroke Scale NNT: Number Needed to Treat
EP
NSU: Neurosurgical Unit
PACS: Picture Archiving and Communication System
AC C
RCT: Randomised Controlled Trial
SPSS: Statistical Package for the Social Sciences SSS: Superior Sagittal Sinus
STROBE: Strengthening the Reporting of Observational Studies in Epidemiology TBI: Traumatic Brain Injury TIA: Transient Ischaemic Attack WHO: World Health Organisation Mr. Simon Lammy MRCS (Ed) SpR Neurological Surgery (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
ACCEPTED MANUSCRIPT Mr. Basel Al-Romhain MRCS (Ed) SpR Neurological Surgery (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
RI PT
Dr. Laura Osborne MB ChB FY1 General Medicine Department of General Medicine Forth Valley Royal Hospital
AC C
EP
TE D
M AN U
SC
Mr. Edward J. St. George FRCS (SN) Consultant Neurological Surgeon (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
S1878-8750(16)30817-8
DOI:
10.1016/j.wneu.2016.09.004
Reference:
WNEU 4544
To appear in:
World Neurosurgery
Received Date: 3 June 2016 Revised Date:
31 August 2016
Accepted Date: 1 September 2016
Please cite this article as: Lammy S, Al-Romhain B, Osborne L, St. George EJ, 10-Year Institutional Retrospective Case Series of Decompressive Craniectomy for Malignant Middle Cerebral Artery Infarction (mMCAI), World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.09.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 10-Year Institutional Retrospective Case Series of Decompressive Craniectomy for Malignant Middle Cerebral Artery Infarction (mMCAI)
Simon Lammy MRCS (Ed) PgDip (Oxon) SpR Neurological Surgery (Neurosurgery)
RI PT
Department of Neurosurgery Institute of Neurological Sciences 1345 Govan Road
SC
Glasgow G51 4TF
Corresponding Author
TE D
Email: [email protected]
Basel Al-Romhain MRCS (Ed)
SpR Neurological Surgery (Neurosurgery)
EP
Department of Neurosurgery
Institute of Neurological Sciences
AC C
1345 Govan Road Glasgow
G51 4TF
United Kingdom
M AN U
United Kingdom
ACCEPTED MANUSCRIPT Laura Osborne MB ChB FY1 General Medicine Department of General Medicine Forth Valley Royal Hospital Stirling Road
FK5 4WR United Kingdom
Edward J. St. George FRCS (SN)
Department of Neurosurgery Institute of Neurological Sciences 1345 Govan Road
G51 4TF
AC C
EP
United Kingdom
TE D
Glasgow
M AN U
Consultant Neurological Surgeon (Neurosurgery)
SC
RI PT
Larbert
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Keywords: Malignant; Craniectomy; Decompressive; Mortality; Glasgow Outcome Scale (GOS) Score
ACCEPTED MANUSCRIPT Abstract A 10-year retrospective case series of patients undergoing decompressive craniectomy for malignant middle cerebral artery infarction (mMCAI) was undertaken (2005-2015). Patient demographics, co-morbidities, pre- and post-operative neurological state, operative timescales, craniectomy dimensions and Glasgow Outcome Scale (GOS) scores were analysed. Overall 40 patients underwent a decompressive craniectomy for mMCAI with a 30-
RI PT
day mortality of 17.5% (n=7). 17 patients (42.5%) were male with a mean age of 43yrs (range: 16-64yrs). Patients who survived had a lower mean age of 41yrs (range: 16-59yrs) than those who did not of 50yrs (range: 42-63yrs). The modal ictal and preoperative GCS were 14 (range: 5–15) and 7 (range: 3–12) which corresponded to motor
SC
scores of 6 and 5 respectively. The mean time from ictus to INS admission was 23.5hrs (range: 0.5-66hrs) and from INS admission to decompression 7.5hrs (range: 0.5-46hrs). Approximately 60% of patients had an “early”
craniectomy (under 48hrs from ictus) and 60% of patients had a craniectomy performed less than 24hrs from INS
M AN U
admission. The mean maximum antero-posterior (AP) craniectomy diameter measured 13cm (range: 10.93– 15.12cm) and the mean surface area was 92.68 cm2 (range: 76.14–124.42cm2). Overall 80% of patients had a modal Glasgow Outcome Scale (GOS) score of 3 (range: 2-5) at discharge, 3-months, 6-months, 9-months and 12-months. The median length of stay was 3 days (range: 6hrs - 11 days) for non-survivors and 13 days (range: 1 - 365 days) for survivors. Decompressive craniectomy for mMCAI is suitable in selected patients and the local
AC C
EP
TE D
practice is consistent with current evidence1.
ACCEPTED MANUSCRIPT Introduction Malignant transformation of a middle cerebral artery (MCA) infarct is a life threatening complication that has 80% mortality if conservatively treated1. With infarction of greater than 50% of MCA territory significant mass effect and concomitant herniation of brain tissue may occur. Recent randomised controlled trials (RCTs)2,3,4 and subsequent meta-analysis of the data5 make it clear that surgical intervention, through a decompressive craniectomy,
RI PT
improves survival but at the potential expense of good functional outcome5. In this retrospective case series
outcome data is presented in the light of recent published evidence. Current controversies in the neurosurgical
literature are discussed and factors which might have contributed to an unfavourable outcome are examined and
AC C
EP
TE D
M AN U
SC
compared to published evidence.
ACCEPTED MANUSCRIPT Methods A retrospective analysis was performed of all patients who underwent a decompressive craniectomy for malignant middle cerebral artery infarction (mMCAI) at the Institute of Neurological Sciences (INS) between January 2005 and January 2015 (see Figure 1). The INS is the busiest and biggest neurosurgical centre in Scotland and covers one of the larger geographical areas in the United Kingdom having a catchment population of over 3 million. The
RI PT
study was conducted according to the STROBE Statement on Cohort Studies for a retrospective case series and our historical selection criteria was judged against National Institute for Health and Care Excellence (NICE)
SC
guidelines as published in the United Kingdom in 20096,7.
Current selection criteria are based on these guidelines and includes four parameters: age <60 years, National Institute of Health Stroke Scale (NIHSS) >15, >50% MCA territory hypodensity on CT (or >145cm3 on diffusion
M AN U
weighted [dw] MRI) and urgent referral to a neurosurgical centre <24 hours from symptom onset and subsequent decompression in eligible patients <48 hours from ictus7. If a patient fulfils these guidelines our practice is to consider each patient on a case by case basis in regard to co-morbidities and pre-ictal functional status. Due of this analysis being retrospective and pre-dating NICE guidelines our inclusion criteria for this case series were patients that had both a clinical and radiological diagnosis of malignant MCA infarction (according to the World
TE D
Health Organisation [WHO] criteria)8 and who then subsequently underwent a decompressive craniectomy.
Electronic and hand written case notes, Scottish National Picture Archiving and Communication System (PACS)
EP
and operative theatre log books were interrogated to extract the relevant data. Once an MCA territory infarction and subsequent decompressive craniectomy were confirmed there was no exclusion criteria based on extent of
AC C
MCA territory infarction (if it was >50% of MCA territory), imaging modalities used, adverse radiological signs and other characteristics, e.g. patient demographics (see below).
Patient demographics (including age, sex and co-morbidities), pre- and post-operative neurological state, operative timescales, craniectomy dimensions and Glasgow Outcome Scale (GOS) scores at discharge, 3months, 6-months, 9-months and 12-months were documented. Unless cranioplasty is considered follow up is primarily conducted by stroke physicians and examination of detailed inpatient and outpatient clinic notes (up to 12-months post-ictus) was conducted to extract outcome information regarding GOS over this time period.
ACCEPTED MANUSCRIPT Risk factors for vascular disease, e.g. pre-ictal transient ischaemic attack (TIA), hypertension (HTN), diabetes mellitus (DM), ischaemic heart disease (IHD), smoking history and a history of alcohol excess were noted. Other non-specific co-morbidities recorded included depression, obesity, a history of malignancy, endocrine and other chronic diseases.
RI PT
Focal neurological findings included the presence of hemi-paresis (change from a patient’s pre-ictus baseline),
speech deficits, e.g. receptive or expressive dysphasia’s, seizure activity, confusion, collapse, incontinence and headache. These signs were recorded both pre-operatively at the time of the initial ictus and post-operatively.
SC
Other post-operative domains included the presence of depression, language comprehension and development of common surgical complications including thrombo-embolic phenomena (as defined by the incidence of deep vein
M AN U
thrombosis and pulmonary embolus) up to 30-days post-operatively.
Additionally, the ictal and pre-operative Glasgow Coma Scale (GCS) scores (specifically the motor scores), incidence of thrombolysis, site of the infarct and sub-sequent decompression, timescale from ictus to INS admission and from INS admission to decompression, overall length of stay, physical dimensions of the
TE D
craniectomy and subsequent cranioplasty complications were documented.
To minimise errors in data collection, all craniotomies performed in the West of Scotland over the study period were scrutinised for accuracy of diagnosis. The radiological modality used was computer tomography (CT) which
EP
represents the primary imaging modality used in the diagnosis of stroke across the United Kingdom. IBM’s
AC C
Statistical Package for the Social Sciences (SPSS) was used to perform statistical analysis.
ACCEPTED MANUSCRIPT Results 40 patients underwent a decompressive craniectomy for malignant MCA infarction. The 30-day mortality was 17.5% (n=7). 17 patients (42.5%) were male with a mean age of 43yrs (range: 16-64yrs) (NB: only one patient was over 60yrs) and 18 had a left sided (dominant) hemispheric infarction (45%). Patients who survived had a lower mean age of 41yrs (range: 16-59yrs) then those who did not of 50yrs (range: 42-64yrs) (p=0.04). The most
not statistically significant with regards to mortality (see Table 1 [p=0.68]).
RI PT
frequently recorded risk factor for cerebrovascular disease was smoking in our cohort (n=22 [55%]) but this was
SC
Other risk factors for cerebrovascular disease, e.g. HTN and IHD, were not predictive of mortality (Table 1). Age proved to be a stronger predictor of the likelihood of survival than the presence of co-morbidity5. This can be further demonstrated by analysing the occurrence of multiple co-morbidities in an individual: interestingly non-
M AN U
survivors although being older had less co-morbidities as a group and less co-morbidities as individuals (see Figure 2).
Pre-operative focal neurological signs did not significantly change following decompression, e.g. the incidence of aphasia was unchanged at 55%. Thrombolysis performed at the time of the original ictal event (before mMCAI
TE D
developed) did not appear to influence mortality. The risk of death in the 12 patients thrombolysed was 16% (n=2) and in the 28 patients not thrombolysed was 18% (n=5) (see Table 1). This was statistically insignificant (p=1.00) and thrombolysis had a relative risk (RR) of death of 81% (i.e. 0.81). Overall 30% of survivors had been
EP
thrombolysed compared to 29% of non-survivors. Due to the small numbers of non-survivors it remains uncertain if
AC C
these results are clinically significant one.
The modal ictal and preoperative GCS were 14 (range: 5–15) and 7 (range: 3–12) which corresponded to motor scores of 6 and 5 respectively. Patients who survived had a higher modal ictal GCS (14: range: 10-15) than those who did not at (10: range: 6-15) but the modal pre-operative GCS scores were similar between the two groups (8: range: 3-15 for survivors and 8: range: 7-10 for non-survivors) (see Table 1). The mean time from ictus to INS admission was 23.5hrs (range: 0.5-66hrs) and from INS admission to decompression 7.5hrs (range: 0.5-46hrs). Approximately 60% of patients had an “early” craniectomy (under 48hrs from ictus) and 60% of patients had a craniectomy performed in under 24hrs from INS admission (see Table 2).
ACCEPTED MANUSCRIPT The mean maximum antero-posterior (AP) craniectomy diameter measured 13cm (10.93–15.12cm) and the mean surface area was 92.68 cm2 (76.14–124.42cm2). Post-operative complications following decompression included infection (respiratory [40%], urinary [10%] and wound [13%]), thrombo-embolic phenomena (DVT and PE [5%]) and post-operative depression (20%). Thirty percent (30%) of patients exhibited ongoing poor comprehension
RI PT
following surgery.
Overall 80% of patients had a modal Glasgow Outcome Scale (GOS) score of 3 (range: 2-5) at discharge, 3-
months, 6-months, 9-months and 12-months. Of the survivors 17 patients subsequently had an insertion of a
SC
titanium cranioplasty and 7 of an autologous bone flap re-insertion within 12-months following decompression.
Post-operative complications related to cranioplasty included 4 revisions for infection (titanium) and 1 bone flap
M AN U
being discarded for wound irritation.
Overall the mean Neurosurgical Unit (NSU) length of stay was 29 days (range: 6hrs - 365 days), which was split into a mean length of stay of 3.89 days (range: 6hrs - 11 days) for non-survivors and 23.12 (range: 1 - 365 days) for survivors. This translated into a median length of stay of 3 for non-survivors and 13 for survivors. Survivors
AC C
EP
TE D
were subsequently discharged to rehabilitation units.
ACCEPTED MANUSCRIPT Discussion The practice of decompressive craniectomy for malignant middle cerebral artery infarction (mMCAI) is controversial: it is an emergency neurosurgical intervention for a life threatening condition whose success cannot simply be measured in terms of survival1,9. Quality of survival is an important factor in complex elective neurosurgical decision making but such a luxury is not afforded to the neurosurgeon in a rapidly deteriorating
RI PT
situation. The context of such a controversy is best understood in terms of its epidemiology, pathophysiology and subsequent management options.
SC
In the United Kingdom stroke is the 4th leading cause of death and the largest cause of complex disability9,10. Over one third of stroke survivors are dependent on others. Over 152,000 people in the United Kingdom have a stroke per annum with an associated 30-day mortality of 12.5% increasing to 25% at 12-months10. Ischaemia due to
M AN U
occlusive cerebrovascular disease accounts for approximately 80% of strokes1,9,10. The incidence of mMCAI is 1% of all strokes (approximately 1520 patients per annum)9,10 and carries an 80% 30-day mortality if neurosurgical intervention does not occur1,8.
The systematic review by Vahedi et al5 of the first three published randomised controlled trials (RCTs) of
TE D
decompressive craniectomy for mMCAI (DECIMAL2, DESTINY3 and HAMLET4) demonstrated an absolute risk reduction (ARR) for mortality of 51% (Oxford Centre for Evidence Based Medicine Level 1b)11. This translates into a number needed to treat (NNT) to prevent death of 2. To prevent a modified Rankin Scale score (mRS) of >5 the
EP
NNT is 2 and to prevent an mRS>4 up to 4 patients need to be decompressed (NNT 4)9. Despite showing an improvement in survival there were discrepancies between the three trials regarding what constituted a good
AC C
functional outcome for the patient. Vahedi et al considered a patient with an mMRS <4 as having a good outcome. Of survivors 29% had an mRS>4 in the surgical arm and 23% in the conservative arm indicating that surgery increased survival at the cost of a slightly worsened functional outcome, i.e. surgery produced more dependent survivors5,9.
Nevertheless, this should be judged in context of the actual numbers. If one considers an annual incidence of 1520 cases of mMCAI then conservative management alone affords a 20% survival rate (n=304) whereas craniectomy (due to an ARR of 51%) permits an extra 152 survivors (n=456). If an mRS<4 is considered a good functional outcome then conservative management results in 234 patients (77% [23% mRS>4]) to exist in those
ACCEPTED MANUSCRIPT categories compared to 323 (71% [29% mRS>4] patients in the surgical arm. Simply put decompressive craniectomy produces more patients with a good functional outcome (around 100 extra patients) i.e. an mRS<4 as a result of an ARR of 51%.
Data from three subsequent trials (e.g. HeADDFIRST (2003)12, Zhao J et al (2012)13 and DESTINY II (2014)14,
RI PT
demonstrated mortality rates of 27%, 13% and 43% in the surgical arms and 46%, 61% and 75% in the non-
surgical arms respectively. The HeADDFIRST12 trial did not publish data on functional outcomes but Zhao et al13
and DESTINY II14 reported an mRS<3 in 9% and 10% of survivors in the surgical arm compared to 21% and 10%
SC
in the non-surgical arm. These generally reflect the pooled analysis by Vahedi et al5.
The pathophysiology of mMCAI is not straightforward and its clinical and radiological predictors are broad9. The
M AN U
condition arises when there is >50% MCA territory infarction producing a potentially significant space occupying effect and consequent herniation of brain tissue. Death usually occurs under one week of symptom onset as a result of progressive swelling of ischaemic brain tissue and sub-falcine (as opposed to rostral-caudal) herniation15. There is a focal (as opposed to a global) increase in intra-cranial pressure (ICP) and extension of the ischaemia to
TE D
adjacent vascular territories9,15. Survivors may therefore be highly disabled and have a poor quality of life5.
Signs include nausea, vomiting, headache, decreased GCS and respiratory compromise9. These signs may occur much quicker in younger patients (approximately 3 hours) due to less compensatory space and sometimes much
EP
later in older patients (around 5 days)1,9. Although elderly patients may deteriorate much later due to relatively more compensatory space such clinical deterioration may indicate a focal lesion that is relatively larger and less
AC C
amenable to aggressive medical and surgical intervention. This might account for the lower mean age of survivors in this study and those of others.
High risk patients have a baseline National Institute of Health Stroke Scale (NIHSS) score of >20 in dominant and >15 in non-dominant hemispheric infarcts in conjunction with >50% hypodensity of MCA territory on CT <6 hours from symptom onset9. In the United Kingdom CT remains the emergency radiological modality of choice due its ease of access and has a sensitivity between 61-94% and a specificity of 91-94% in predicting a fatal outcome. This compares favourably to diffusion weighted MRI which has 100% sensitivity and 94% specificity in predicting a poor outcome, e.g. death and mRS>4, when the lesion is >145cm3 in volume1,5,9,16.
ACCEPTED MANUSCRIPT Despite the controversies surrounding decompressive craniectomy in mMCAI it remains the only condition that has randomised controlled trial (RCT) data to support its use to improve survival. Clarke and Harris (1958) originally proposed the hypothesis that mMCAI should be considered a surgical mass lesion due to the similarities with an extra-dural haematoma (EDH) and its ability to produce malignant focal increases in ICP17. Surgical intervention should have clear goals and is primarily intended to prevent death from brainstem herniation and
RI PT
secondarily to permit the swollen infarct to expand away from the midline enabling a permanent decrease in focal ICP.
SC
Understandably concerns remain regarding patient selection. In 2009 the National Institute for Health and Care
Excellence (NICE), on the basis of the pooled analysis by Vahedi et al5, proposed guidelines predictive of survival and good functional outcome7. These included four parameters: age <60 years, NIHSS >15, >50% MCA territory
M AN U
hypodensity on CT (if possible >145cm3 on dwMRI) and urgent referral to a neurosurgical centre <24 hours from symptom onset and subsequent decompression in eligible patients <48 hours from ictus. This guideline as well as data from the six individual RCTs was used to assess this cohort’s outcomes. This study’s results were comparable to published data with respect to patient selection, timescale to craniectomy, craniectomy dimensions
TE D
and subsequent functional outcome and these will be discussed in turn.
In the present series survivors had a lower mean age than non-survivors (41yrs vs. 50yrs) which is consistent to the sub-group pooled analysis which demonstrated improved survival in patients <50yrs5. Patient selection
EP
conformed to the NICE guidelines in terms of age as only one patient was <18yrs and one >60yrs in the cohort. The importance of age as a predictor of outcome was demonstrated as the best functional outcome at discharge
AC C
(GOS 5) was achieved in patients with a mean age of 30yrs (range: 16-45yrs [n=4]), GOS 4 in those with a mean age of 37yrs (range: 33-51yrs [n=7]) and GOS 3 (often considered a good functional outcome and roughly equivalent to mRS 4 in the pooled analysis) in survivors with a mean age of 44yrs (range: 28-59yrs [n=22]). The mean age of patients achieving a GOS 2 was 53yrs (range: 46-59yrs [n=2]) and the mean age for those with GOS 1 was 51yrs (range: 42-63yrs [n=5]).
An almost linear relationship between age and GOS at discharge was demonstrated with a very high negative correlation (Spearman’s Correlation Coefficient r= -0.900 [R2=0.81] p= 0.37). Increased age decreases patients GOS and up to 81% of the variation in GOS is related to the variation in age (Spearman’s Correlation Coefficient2)
ACCEPTED MANUSCRIPT (see Figure 3). Furthermore, the results demonstrate a GOS 3-4 is anticipated in patients with a mean age of 40 (compared to 50 as observed in the Vahedi pooled analysis) which is similar to the age related outcome derived from the traumatic brain injury prognosis calculator developed from the CRASH studies18.
Baseline Glasgow Coma Scale (GCS) scores (particularly the motor component) is often not reported in studies of
RI PT
this nature. In traumatic brain injury (TBI) and the subsequent prognostications from the CRASH studies the motor component and age have been found to be good predictors of outcome 18. In this case series patients who
survived had a higher modal ictal GCS at 14 (range: 10-15) than those who did not at 10 (range: 6-15) but the
SC
modal pre-operative GCS scores were the same for the two groups (8 for survivors [range: 3-15] and 8 for nonsurvivors [range: 7-10]) (see Table 1).
M AN U
Although non-survivors in this study had a higher mean age they also had less co-morbidity burden (in terms of number and complexity) than survivors and this cannot be easily explained (see Table 1). This can be further demonstrated by analysing the frequency of multiple co-morbidities in an individual (see Figure 2). No association was found between co-morbidity and survival in the study presented. Furthermore, our results tentatively suggest that if a patient had been thrombolysed as part of medical management following the initial ictus, acceptance of
TE D
the patient for decompression (once mMCAI becomes apparent) probably should not be delayed (see Results).
Despite the mean time from ictus to INS admission being 23.5hrs (0.5-66hrs) and from INS admission to
EP
subsequent decompression being 7.5hrs (0.5-46hrs) (adhering to NICE guidelines based on Vahedi et al) a higher mortality was seen in the group treated <48hrs (“early”) (n=6 [19.3%]) when compared to those treated >48hrs
AC C
(“late”) (n=1 [11.1%]) (see Table 2). However, the overall numbers are small and not comparable (n=31 <48hrs vs. n=9 >48hrs) and meaningful statistical and clinical correlations cannot be generated but the finding does raise the possibility that decompressing patients faster might not necessarily increase survival in this sub-group and the impact on functional outcome remains undetermined (see Table 2).
Using post-operative 3D reconstruction of CT images the mean AP diameter was 13cm (range: 10.93–15.12cm) resulting in a mean surface area of 92.68 cm2 (range: 76.14–124.42cm2) which is in keeping with published guidelines15,19. As no post-operative CT was performed on five of the seven non-survivors (due to death occurring <24hrs post-operatively) no correlation could be established between craniectomy dimension and survival (or
ACCEPTED MANUSCRIPT craniectomy dimension and functional outcome). The previously recommended anterior posterior (AP) diameter of 12cm has been increased to 14cm as this permits decompression beyond the infarct margins15.
This is usually surgeon dependent variable but a 76% increase in extra volume occurs from when the AP diameter is increased from 12cm to 14cm15. The minimum dorsal ventral (DV) diameter is ~9cm and although being variable
RI PT
it is generally surgeon independent as it is limited by the floor of the middle cranial fossa and superior sagittal
sinus (SSS)15. This dimension is of particular importance in regard to preventing uncal herniation and subsequent
SC
brainstem ischaemia and irreversible infarction (death). Our results are consistent to current practice.
The incidence of cranioplasty complications in this cohort could not be placed in context as these are seldom reported in the literature. But craniectomy specific complications do occur and these suggest that decompressive
M AN U
craniectomy should be perceived as “a two stage operation”, i.e. initially to preserve life (and possibly function) but subsequently one which requires a further procedure to repair the defect (otherwise the risk of sinking flap syndrome and delayed dysautonmic symptoms is increased). Consequently, a multi-centre prospective cohort study that documents this complication is needed in order to gain adequate numbers to properly quantify post-
TE D
operative cranioplasty risks.
Potential sources of bias included selection bias as the overall number of patients referred to our neurosurgical centre for mMCAI from 2005-2015 could not be quantified (due to the retrospective nature of this study). Also
EP
internal validity was not possible as no previous institutional case series according to the Strobe Statement on Cohort Studies had been conducted on this subject matter. The extent of infarction to predict survival could not be
AC C
conducted as volumetric analysis using MRI was not employed as a mode of imaging in a single patient.
ACCEPTED MANUSCRIPT Conclusions The centre’s practice has been and remains consistent with NICE guidelines. However, it demonstrates that early decompression might not influence mortality nor improve functional outcome and that more research is needed regarding functional outcome and GCS (particular the motor component) at ictus.
RI PT
Acknowledgements
The authors wish to thank NHS Greater Glasgow & Clyde and the Department of Neurosurgery at the Institute of
SC
Neurological Sciences (INS) for their support.
Author Contributions
BA prepared the Introduction. SL prepared the Abstract and Methods and Results and Discussion (and formulated
M AN U
the Tables and Figures). LO extracted the data for 2005–2013 which was independently checked by SL. SL extracted further data for 2013-2015. JSt.G and SL revised the whole manuscript and double checked it for accuracy and fluency. SL re-revised the manuscript again in light of editorial comments. JSt.G proposed and supervised the project.
Funding
EP
None declared.
TE D
Declaration of Conflicting Interests
This research received no specific grant from any funding agency in the public, commercial or not-for-profit
AC C
sectors.
ACCEPTED MANUSCRIPT Tables Characteristics of Patients Alive and Dead Following Craniectomy for Malignant MCA Infraction Dead (n=7)
P Value
Male
13 (39.9%)
4 (57.1%)
0.43*
Age
41.52yrs
51.50yrs
0.04^
(16-59yrs [+/- 9.830])
(42-64yrs [+/- 6.979])
16 (48.4%)
2 (28.6%)
0.42*
n=10
n=2
1.00*
Thrombolysis Risk Factors (Vascular) Hypertension
5 (15.2%)
•
Ischaemic Heart Disease
3 (9.1%)
•
Transient Ischaemic Attack
•
Diabetes Mellitus
• •
1 (14.2%)
1.00*
0 (0.0%)
1.00*
M AN U
•
SC
Dominant Hemisphere
RI PT
Alive (n=33)
8 (24.2%)
0 (0.0%)
0.31*
3 (9.1%)
1 (14.2%)
0.55*
Current Smoker
19 (57.6%)
3 (42.9%)
0.68*
Alcohol Excess
10 (30.3%)
1 (14.2%)
0.65*
4 (12.1%)
0 (0.0%)
1.00*
8 (24.2%)
1 (14.2%)
1.00*
19 (57.6%)
2 (28.6%)
0.23*
8 (24.2%)
1 (14.2%)
1.00*
14 (10-15)
10 (6-15)
0.73”
8 (3-15)
8 (7-10)
1.00”
TE D
Risk Factors (Other) Obesity
•
Depression
•
Other 1, e.g. Malignancy
•
Other 2, e.g. Endocrine
AC C
GCS (Modal)
EP
•
•
Ictal
•
Pre-Operative
Table 1: Demographic and clinical characteristics of survivors and non-survivors. Non-survivors were older and had a lower modal ictal GCS but less co-morbidities than survivors. NB: the time to decompression did not influence outcome survival nor function (NB: * Fischer Exact Test; ^Wilcoxon Signed Rank Test, “Chi Squared Test)
ACCEPTED MANUSCRIPT Characteristics of Patients Having Early and Late Craniectomy for Malignant MCA Infraction Late >48hrs (n=9)
P Value
Male
10 (32.3%)
7 (77.8%)
0.02*
Age
31yrs (26-64yrs)
42yrs (16-51yrs)
0.55
Dominant Hemisphere
11 (35.4%)
7 (77.8%)
0.05*
Mortality
6 (19.3%)
1 (11.1%)
1.00*
n=9 (Mortality: 67% [n=6])
n=3 (Mortality: 33%
1.00* (Mortality:
[n=1])
0.53*)
•
Ictal
15 (9-15)
•
Pre-Operative
M AN U
GCS (Modal)
SC
Thrombolysis
RI PT
Early <48hrs (n=31)
GOS (Modal)
14 (10-15)
0.89^
8 (3-15)
6, 8, 9 (6-12)
1.00^
3 (1-5)
3 (3-4)
1.00^
Table. 2: Characteristics of patients having “early” (<48hrs) compared to “late” (>48hrs) decompressive
TE D
craniectomies. Patients decompressed late did not necessarily have higher mortality nor higher modal ictal and
AC C
EP
pre-operative GCS scores despite having a higher mean age. (NB: * Fischer Exact Test; “Chi Squared Test)
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Figures
AC C
EP
TE D
Fig. 1: This shows the number of decompressive craniectomies performed between January 2005 – January 2015
Fig. 2: Number of patients (y-axis) and the frequency of co-morbidities (x-axis). Non-survivors had less comorbidities individually and as a group than survivors
AC C
EP
TE D
Fig. 3: Age (y-axis) vs. GOS (x-axis)
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT References
1.
Treadwell SD, Thanvi B.Malignantmiddle cerebral artery (MCA) infarction: pathophysiology, diagnosisand management. Postgrad Med J. 2010; 86(1014): 235-42
Vahedi K, Vicaut E, Mateo J, Kurtz A, Orabi M, Guichard JP, Boutron C, Couvreur G, Rouanet F, Touzé
RI PT
2.
E, Guillon B, Carpentier A, Yelnik A, George B, Payen D, Bousser MG; DECIMAL Investigators.
Sequential-design, multicenter, randomized, controlled trial of early decompressive craniectomy in
3.
SC
malignant middle cerebral artery infarction (DECIMAL Trial). Stroke. 2007; 38(9): 2506-17
Jüttler E, Schwab S, Schmiedek P, Unterberg A, Hennerici M, Woitzik J, Witte S, Jenetzky E, Hacke W;
M AN U
DESTINY Study Group. Decompressive Surgery for the Treatment of Malignant Infarction of the Middle Cerebral Artery (DESTINY): a randomized, controlled trial. Stroke. 2007; 38(9): 2518-25
4.
Hofmeijer J, Kappelle LJ, Algra A, Amelink GJ, van Gijn J, van der Worp HB; HAMLET investigators. Surgical decompression for space-occupying cerebral infarction (the Hemicraniectomy After Middle
TE D
Cerebral Artery infarction with Life-threatening Edema Trial [HAMLET]): a multicentre, open, randomised trial. Lancet Neurol. 2009; 8(4): 326-33
Vahedi K, Hofmeijer J, Juettler E, Vicaut E, George B, Algra A, Amelink GJ, Schmiedeck P, Schwab S,
EP
5.
Rothwell PM, Bousser MG, van der Worp HB, Hacke W; DECIMAL, DESTINY, and HAMLET
AC C
investigators. Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomised controlled trials. Lancet Neurol. 2007; 6(3): 215-22
6.
STROBE: Strengthening the Reporting of Observational Studies in Epidemiology. Checklist. Case Control Studies 2007 (http://www.strobe–statement.org/index.php?id=available–checklists).
7.
NICE : National institue for Heath and Clinical Excellence. Surgical Referral for DecompressiveHemicraniectomyhttps://www.nice.org.uk/guidance/cg68/resources/13-2-surgicalreferral-for-decompressive-hemicraniectomy
ACCEPTED MANUSCRIPT 8.
WHO: World HealthOrganisation. Topic. Stroke. 2014 (http://www.who.int/topics/cerebrovascular_accident/en/)
9.
Subramaniam S, Hill MD. Decompressive hemicraniectomy for malignant middle cerebral artery
10. Stroke Association. State of The Nation: Stroke Statistics. 2016
SC
(https://www.stroke.org.uk/resources/state-nation-stroke-statistics)
RI PT
infarction: an update. Neurologist. 2009; 15(4):178-84
11. OCEBM: Oxford Centre for Evidence Based Medicine. Levels of Evidence. 2009
M AN U
(http://www.cebm.net/oxford-centre-evidence-based-medicine-levels-evidence-march-2009/)
12. Frank JI, Schumm LP, Wroblewski K, Chyatte D, Rosengart AJ, Kordeck C, Thisted RA; HeADDFIRSTTrialists. Hemicraniectomy and durotomy upon deterioration from infarction-related swelling trial: randomized pilot clinical trial. Stroke. 2014; 45(3): 781-7
TE D
13. Zhao J, Su YY, Zhang Y, Zhang YZ, Zhao R, Wang L, Gao R, Chen W, Gao D. Decompressive hemicraniectomy in malignant middle cerebral artery infarct: a randomized controlled trial enrolling
EP
patients up to 80 years old. Neurocrit Care. 2012; 17(2): 161-71
14. Jüttler E, Bösel J, Amiri H, Schiller P, Limprecht R, Hacke W, Unterberg A; DESTINY II Study Group.
AC C
DESTINY II: DEcompressive Surgery for the Treatment of malignant INfarction of the middle cerebral arterY II. Int J Stroke. 2011; 6(1): 79-86
15. Johnson RD, Maartens NF, Teddy PJ.Technical aspects of decompressive craniectomy for malignant middle cerebral artery infarction. J ClinNeurosci. 2011; 18(8): 1023-7
16. Hatefi D, Hirshman B, Leys D, Lejeune JP, Marshall L, Carter BS, Kasper E, Chen CC.Hemicraniectomy in the management of malignant middle cerebral artery infarction: Lessons from randomized, controlled trials. SurgNeurol Int. 2014; 15(5): 72
ACCEPTED MANUSCRIPT 17. Clarke E, Harris P. Thrombosis of the internal carotid artery simulating an intracranial space-occupying lesion. Lancet. 1958; 1(7030): 1085-9
18. Roberts I, Yates D, Sandercock P, Farrell B, Wasserberg J, Lomas G, Cottingham R, Svoboda P, Brayley N, Mazairac G, Laloë V, Muñoz-Sánchez A, Arango M, Hartzenberg B, Khamis H,
RI PT
Yutthakasemsunt S, Komolafe E, Olldashi F, Yadav Y, Murillo-Cabezas F, Shakur H, Edwards P;
CRASH trial collaborators. Effect of intravenous corticosteroids on death within 14 days in 10008 adults
2004; 364(9442): 1321-8.
SC
with clinically significant head injury (MRC CRASH trial): randomised placebo-controlled trial. Lancet.
19. Ransohoff J, Benjamin V. Hemicraniectomy in the treatment of acute subdural haematoma. J Neurol
AC C
EP
TE D
M AN U
Neurosurg Psychiatry. 1971; 34(1): 106.
ACCEPTED MANUSCRIPT Highlights (For Review) A 10-year retrospective case series of patients undergoing decompressive craniectomy for malignant middle cerebral artery infarction (mMCAI) was undertaken (2005-2015). Patient demographics, co-morbidities, pre- and post-operative neurological state, operative timescales, craniectomy dimensions and Glasgow Outcome Scale (GOS) scores were analysed. Overall 40 patients underwent a decompressive craniectomy for mMCAI with a 30-day mortality of 17.5% (n=7). 17 patients (42.5%) were male with a
RI PT
mean age of 43yrs (range: 16-64yrs). Patients who survived had a lower mean age of 41yrs (range: 16-59yrs) than those who did not of 50yrs (range: 42-63yrs). The modal ictal and preoperative GCS were 14 (range: 5–15) and 7 (range: 3–12) which
corresponded to motor scores of 6 and 5 respectively. The mean time from ictus to INS admission was 23.5hrs (range: 0.5-66hrs)
SC
and from INS admission to decompression 7.5hrs (range: 0.5-46hrs). Approximately 60% of patients had an “early” craniectomy (under 48hrs from ictus) and 60% of patients had a craniectomy performed less than 24hrs from INS admission. The mean maximum antero-posterior (AP) craniectomy diameter measured 13cm (range: 10.93–15.12cm) and the mean surface area was 92.68 cm2
M AN U
(range: 76.14–124.42cm2). Overall 80% of patients had a modal Glasgow Outcome Scale (GOS) score of 3 (range: 2-5) at discharge, 3-months, 6-months, 9-months and 12-months. The median length of stay was 3 days (range: 6hrs - 11 days) for nonsurvivors and 13 days (range: 1 - 365 days) for survivors. Decompressive craniectomy for mMCAI is suitable in selected patients and the local practice is consistent with current evidence1.
TE D
Mr. Simon Lammy MRCS (Ed) SpR Neurological Surgery (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
EP
Mr. Basel Al-Romhain MRCS (Ed) SpR Neurological Surgery (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
AC C
Dr. Laura Osborne MB ChB FY1 General Medicine Department of General Medicine Forth Valley Royal Hospital
Mr. Edward J. St. George FRCS (SN) Consultant Neurological Surgeon (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
ACCEPTED MANUSCRIPT Abbreviations and Acronyms ARR: Absolute Risk Reduction CRASH: Corticosteroid Randomisation After Significant Head Injury CT: computer tomography DM: Diabetes Mellitus
RI PT
EDH: Extra-Dural Haematoma GCS: Glasgow Coma Scale GOS: Glasgow Outcome Scale
SC
HTN: Hypertension ICP: Intracranial Pressure
INS: Institute of Neurological Sciences MCA: Middle Cerebral Artery mMCAI: Middle Cerebral Artery Infarction MRI: magnetic resonance imaging mRS: Modified Rankin Scale
M AN U
IHD: Ischaemic Heart Disease
TE D
NICE: National Institute for Health and Clinical Excellence NIHSS: National Institute of Health Stroke Scale NNT: Number Needed to Treat
EP
NSU: Neurosurgical Unit
PACS: Picture Archiving and Communication System
AC C
RCT: Randomised Controlled Trial
SPSS: Statistical Package for the Social Sciences SSS: Superior Sagittal Sinus
STROBE: Strengthening the Reporting of Observational Studies in Epidemiology TBI: Traumatic Brain Injury TIA: Transient Ischaemic Attack WHO: World Health Organisation Mr. Simon Lammy MRCS (Ed) SpR Neurological Surgery (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
ACCEPTED MANUSCRIPT Mr. Basel Al-Romhain MRCS (Ed) SpR Neurological Surgery (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences
RI PT
Dr. Laura Osborne MB ChB FY1 General Medicine Department of General Medicine Forth Valley Royal Hospital
AC C
EP
TE D
M AN U
SC
Mr. Edward J. St. George FRCS (SN) Consultant Neurological Surgeon (Neurosurgery) Department of Neurosurgery Institute of Neurological Sciences