Accepted Manuscript

Resumption of antithrombotic agents in chronic subdural hematoma (CSDH): a

systematic review and meta-analysis

Kevin Phan, MD BSc MSc MPhil, David Abi-Hanna, BMedSci, Jack Kerferd,
BMedSci, Victor M. Lu, MPhil, Adam A. Dmytriw, MD MSc, Yam-Ting Ho, BS, Jacob
Fairhall, MBBS FRACS, Rajesh Reddy, MBBS FRACS, Peter Wilson, MBBS FRACS
PII: S1878-8750(17)31819-3
DOI: 10.1016/j.wneu.2017.10.091
Reference: WNEU 6744

To appear in: World Neurosurgery

Received Date: 12 August 2017

Revised Date: 15 October 2017
Accepted Date: 17 October 2017

Please cite this article as: Phan K, Abi-Hanna D, Kerferd J, Lu VM, Dmytriw AA, Ho Y-T, Fairhall J,
Reddy R, Wilson P, Rate of thromboembolism however was statistically lower in those who resumed
antithrombotics Resumption of antithrombotic agents in chronic subdural hematoma (CSDH): a
systematic review and meta-analysis, World Neurosurgery (2017), doi: 10.1016/j.wneu.2017.10.091.

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
Resumption of antithrombotic agents in chronic subdural hematoma (CSDH): a

systematic review and meta-analysis

Authors:

Kevin Phan1,2 MD BSc MSc MPhil, David Abi-Hanna1 BMedSci, Jack Kerferd1 BMedSci,

PT
Victor M Lu2 MPhil, Adam A Dmytriw3 MD MSc, Yam-Ting Ho2 BS, Jacob Fairhall4

MBBS FRACS, Rajesh Reddy4 MBBS FRACS, Peter Wilson4 MBBS FRACS

RI
1. NeuroSpine Surgery Research Group (NSURG), Prince of Wales Private Hospital,

SC
Sydney, Australia

2. Sydney Medical School, University of Sydney, Sydney, Australia

U
3. Department of Medical Imaging, University of Toronto, Toronto, Canada
AN
4. Department of Neurosurgery, Prince of Wales Hospital, Randwick, Sydney, Australia
M

Corresponding Author:

Kevin Phan, NeuroSpine Surgery Research Group (NSURG), Prince of Wales Private
D

Hospital, Sydney, Australia. Email: kphan.vc@gmail.com

TE

Cover Title: Resumption of antithrombotics in CSDH

EP

Keywords: Chronic Subdural Hematoma, Chronic Subdural Hemorrhage, Anticoagulant

C

Antiplatelet, Subdural.
AC

Subject Terms: Intracranial Hemorrhage, Anticoagulants, Complications, Meta Analysis

Word Count: 3798

0
 ACCEPTED MANUSCRIPT
ABSTRACT

Background: The clinical decision whether and when to resume antithrombotics in patients

with chronic subdural hematomas (CSDH) postoperatively is limited by a lack of quality

evidence exploring this topic. Our study aims to assess the available evidence of patient

complication outcomes, specifically hemorrhagic and thromboembolic events, following the

PT
resumption or non-resumption of antithrombotic agents postoperatively in CSDH patients

RI
already on these agents before CSDH.

SC
Methods: We followed recommended PRISMA guidelines for systematic reviews. Electronic

database searches were performed to identify included studies. Data were extracted and

U
analysed using meta-analysis.
AN
Results: Eight studies were included for analysis. The most common indication for

antithrombotic treatment before onset of CSDH was atrial fibrillation (29.6%), followed by
M

prosthetic heart valve (16.6%), recent myocardial infarction (14.1%), prior stroke or TIA
D

(11.6%), and finally venous thromboembolism (8.3%). The overall hemorrhagic complication
TE

rate was 14.8% in the resumption group vs 18.6% in the no resumption group (P=0.591). This

did not differ between early (<2 weeks) vs late (>1 month) resumption (15% vs 18.6%,
EP

P=0.97). The rate of thromboembolism however was statistically lower in those who resumed

antithrombotics (2.9% vs 6.8%, P<0.001). There was a non-significant trend towards higher
C

thromboembolic rates with early resumption (5.3% vs 2.1%, P=0.23)

AC

Conclusions: The decision to resume antithrombotics postoperatively in the clinical

management of CSDH patients is a complex one and should therefore be a highly

individualised process. Our meta-analysis demonstrates that in selected cases, it is feasible to

resume early antithrombotic treatment without additional hemorrhagic or thromboembolic

risk.

1
 ACCEPTED MANUSCRIPT
INTRODUCTION

Chronic subdural hematoma (CSDH) involves the abnormal collection of blood in the

cleavage plane between the arachnoid mater and the meningeal layer of the dura mater. It is

predicted to become the most common cranial neurosurgical condition among adults by 2030,
1
with recent reported incidence rates as high as 79.4 cases per 100,000 persons. In healthy

PT
individuals, there is no space between the arachnoid and dura mater meningeal layers as they

RI
are tethered together by a layer of dural border cells. 2 However these layers can be separated,

most commonly due to a traumatic hemorrhagic event, such as the tearing of bridging veins

SC
draining the brain surface into the dural sinuses or the rupturing of cortical arteries,

particularly in the elderly population. 3

U
AN
CSDH development is complex and a different process than acute subdural hematoma

accumulation (ASDH). The etiology of CSDH is varied, but can include maturation of a prior
M

asymptomatic ASDH due to the rupturing of fragile neo-capillaries that develop following the
D

organisation of the primary hematoma. 4 Another mechanism is repeated micro-hemorrhaging

TE

of fragile neo-vessels which have grown into a previous subdural hygroma, accumulations of

cerebrospinal fluid between the meningeal layers, which is common in patients with severe
EP

4,5
baseline atrophy. Patients diagnosed with CSDH have a variety of surgical treatment

options available, all of which are relatively simple, safe and effective; examples include
C

percutaneous twist-drill and burr-hole craniostomies, as well as various sizes of craniotomy

AC

and craniectomy.6

Systemic risk factors associated with the development of CSDH include long-term
7,8
alcohol use, indirect trauma, hemodialysis, low intracranial pressure and epilepsy. Of

additional importance is the use of antithrombotic agents, such as anticoagulant and

antiplatelet medications, which have been indicated to increase the risk of developing CSDH

2
 ACCEPTED MANUSCRIPT
by at least 42.5 times. 9 This provides a particular burden on the elderly community due to the

progressive occurrence with age of microbleeds in the cerebrum, which cannot properly clot

in patients using antithrombotic medication, and then progress to CSDH.10 Patients may be

on these medications for a number of reasons, such as atrial fibrillation or the use of

prosthetic cardiac valves, and studies have quoted that 41-52% of CSDH patients are using

PT
antithrombotic agents at presentation. 11,12 There are, however, ongoing concerns with regard

RI
to the optimal postoperative management of CSDH patients who had antithrombotic

treatment prior, with no clear evidence base to guide clinical decisions. The resumption of

SC
antithrombotic treatment postoperatively requires balancing the risks of re-bleeding with the

risks of thromboembolic complications if antithrombosis is not resumed. Currently, the

U
clinical management of CSDH patients who have either resumed or not resumed the use of
AN
antithrombotic agents postoperatively has been based upon anecdotal evidence and surgical
M

experience due to a lack of quality studies exploring this topic. Whether or not a difference in

outcomes exists between early (<2 weeks) or late (>1 month) resumption also remains
D

uncertain. Furthermore, the majority of studies reporting data regarding restarting

TE

antithrombotic therapies in patients with CSDH are retrospective and not from well-

controlled prospective studies or randomised trials.

EP

To address this issue, our study aims to assess the available evidence of patient
C

complication outcomes, specifically hemorrhagic and thromboembolic events, following the

AC

resumption or non-resumption of antithrombotic agents postoperatively in CSDH patients

who were already on these agents before CSDH.

3
 ACCEPTED MANUSCRIPT
METHODS

Search Strategy and study selection

The present systematic review and meta-analysis followed recommended guidelines

and protocols13-15. Electronic searches were performed using Ovid Medline, PubMed,

PT
Cochrane Central Register of Controlled Trials (CCTR), Cochrane Database of Systematic

Reviews (CDSR), ACP Journal Club, and Database of Abstracts of Review of Effectiveness

RI
(DARE) from their dates of inception to May 2017. To achieve the maximum sensitivity of

SC
the search strategy, we combined the terms: chronic subdural hematoma, subdural

hemorrhage, anticoagulants, anticoagulation, antiplatelet, warfarin, aspirin, dabigatran,

U
rivaroxaban, fondaparinux, heparin, vitamin K, as either key words or MeSH terms. The
AN
reference lists of all retrieved articles were reviewed for further identification of potentially

relevant studies, assessed using the inclusion and exclusion criteria.

M

Eligible studies for the present systematic review and meta-analysis included those in
D

which patients ceased antithrombotic therapy prior to surgical treatment of CSDH. Studies
TE

that did not include complications as endpoints were excluded. When institutions published

duplicate studies with accumulating numbers of patients or increased lengths of follow-up,

EP

only the most complete reports were included for quantitative assessment at each time

interval. All publications were limited to those involving human subjects. Abstracts, case
C

reports, conference presentations, editorials, reviews and expert opinions were excluded.
AC

Data extraction, critical appraisal and treatment effect

All data were extracted from article texts, tables and figures. Two investigators

independently reviewed each retrieved article (K.P., D.A.). Discrepancies between the two

reviewers were resolved by discussion and consensus with a third reviewer (J.K.). If the study

provided medians and interquartile ranges instead of means and standard deviations (SD), we

4
 ACCEPTED MANUSCRIPT
16
inputed the means and SDs as described by Hozo et al. Since quality scoring is

controversial in meta-analyses of observational studies, two reviewers (K.P., V.M.L.)

independently appraised each article included in our analysis according to a critical review
17
checklist of the Dutch Cochrane Centre proposed by MOOSE. The key points of this

checklist include: (I) clear definition of study population; (II) clear definition of outcomes

PT
and outcome assessment; (III) independent assessment of outcome parameters; (IV) sufficient

RI
duration of follow-up; (V) no selective loss during follow-up; and (VI) important

confounders and prognostic factors identified. The final results were reviewed by the senior

SC
investigators (J.F., R.R., P.W.)

U
Statistical analysis AN
A meta-analysis of proportions was conducted for the available main perioperative
18
and postoperative variables, using logit transformations. To incorporate heterogeneity
M

(anticipated among the included studies), transformed proportions were combined using
D

19
DerSimonian-Laird random effects (RE) models. Heterogeneity was evaluated using

Cochran’s Q and I2 test. Subgroup analysis based on resumption status of antithrombotic

TE

therapy was conducted via meta-regression analysis. All analyses were performed using the
EP

metafor package for R version 3.01. P values <0.05 were considered statistically significant.
C
AC

5
 ACCEPTED MANUSCRIPT
RESULTS

Selection of studies

We screened a total of 458 titles and abstracts, from which 8 studies20-27 met the

inclusion criteria. The included studies were summarized in Table 1 and antithrombotic

PT
indications and characteristics in Table 2. The PRISMA search strategy in shown in Figure 1.

All included studies except one27 were retrospective cohort studies. The average follow-up

RI
ranged from 1.4 months to 96 months. Three studies were from Japan20,24,26, one study from

SC
the UK21, one study from Canada22, one study from Switzerland23, one study from Germany25

and one study from Korea27. For evaluation of hemorrhagic complications, we were able to

U
include 7 studies. For assessment of thromboembolism complications, we were able to
AN
include 5 studies in the analysis. Risk of bias was assessed with MOOSE criteria in Table 3.

Oral antithrombotic therapy indications and reinitiation

M

The most common indication for antithrombotic treatment before onset of CSDH was
D

atrial fibrillation (29.6%), followed by prosthetic heart valve (16.6%), recent myocardial
TE

infarction (14.1%), prior stroke or TIA (11.6%), and finally venous thromboembolism

(8.3%). Re-initiation of antithrombotic therapy varied amongst studies. An anticoagulant was

EP

used by 41.7% of cases, antiplatelet in 60.9% of cases, with a small fraction of patients
C

receiving both anticoagulants and antiplatelets. For early resumption ranging from 3 days - 2
AC

weeks, this approach was adopted by 5 studies20,21,24,26,27. The remaining three studies had

median resumption times >1 month22,23,25.

Assessment of hemorrhagic complications

The overall rate of hemorrhagic complications, including recurrence, in the

antithrombotic-resumed group is 14.8% (95% CI, 6.9-28.7%; I2=79.86%; p<0.001) (Figure

1.) In the non-resumed group, the overall rate of hemorrhagic complications was 18.6% (95%

6
 ACCEPTED MANUSCRIPT
CI, 7.7-38.5%; I^2=84.82%; p=0.000) (Figure 2). There was no significant difference

between the two groups (omnibus p-value = 0.591).

The hemorrhagic complication rate in the antithrombotic resumption group were

further subgrouped into studies which employed an early resumption (3 days – 2 weeks)

protocol versus a late resumption protocol. The hemorrhagic rate in the early resumption

PT
group was 15% (95% CI 5.3-35.6%, I2=66.9%) compared with 18.6% (95% CI 7.7.-38.5%,

RI
I2=90.1%) in the late resumption group (Figure 3). This was not statistically significant

(p=0.97)

SC
Assessment of thromboembolism

U
The rate of thromboembolism however was statistically higher in the group which did
AN
not resume antithrombotic therapy (omnibus p-value = 0.01). The antithrombotic resumption

group had a thromboembolism incidence of 2.9% (95% CI, 1.4-5.9%; I2=0, p=0.60), while
M

the non-resumption group had an incidence rate of 6.8% (95% CI, 2.7-15.7%; I2=73.7%;
D

p<0.001) (Figure 4)
TE

The thromboembolic complication rate for the resumption cohort was further

subgrouped according to early versus late resumption. In the early resumption subgroup, the
EP

pooled thromboembolic rate was 5.3% (95% CI 1.5-16.6, I2=0%, p=0.52). This was
C

compared with 2.1% (95% CI 0.8-5.1%, I2=0%, p=0.65). Although there was a trend for the
AC

thromboembolic rate to be higher in the early resumption group, this did not reach statistical

significance (p=0.23) (Figure 5).

Sensitivity Analysis

Given the significant heterogeneity in the random-effects meta-analysis, we

performed a sensitivity analysis to identify whether results were driven by certain studies.

7
 ACCEPTED MANUSCRIPT
Following leave-one out meta-analysis, no significant differences in the trends of the

presented results were found.

DISCUSSION

The results of our meta-analysis emphasize the lack of a consensus in the literature in

PT
regards to restarting antithrombotic therapy in patients who have been treated for CSDH.

However, our review adds to the growing number of studies exploring this issue in order to

RI
help guide clinical decision-making. Overall, no significant difference was found between the

SC
two cohorts in terms of the rate of hemorrhagic complications, including recurrence. The

rates for the CSDH patient groups that resumed and did not resume antithrombotic treatment

U
postoperatively were 14.6% and 16.9% respectively. However, patients who did resume
AN
antithrombotic treatment were found to have a significantly lower thromboembolism rate

(2.9% vs 15.3% for non-resumed patients). We also found that early versus late resumption
M

of antithrombotics did not significantly influence hemorrhagic and thromboembolic

D

complication rates.
TE

The reluctance to resume antithrombotics following CSDH is the theoretical risk of

bleeding and recurrence of hemorrhage. Our current study results support the notion that this
EP

may not be the case. Similar recommendations were made in a recent study exploring the

effect of anticoagulation resumption in the context of acute intracranial hemorrhages, with

C

similar conclusions of no additional heightened risk of rebleeding with anticoagulation

AC

28
resumption. For patients with CSDH, our results are supported by a recent analysis by

Murthy et al., on the effect of restarting antithrombotic agents after the intracranial
29
hemorrhage. They also reported that the resumption of anticoagulants was associated with

a significantly lower thromboembolism rate (6.7% vs 17.6% for non-resumed patients),

however no evidence was found to suggest a significant impact of resumption on recurrence

8
 ACCEPTED MANUSCRIPT
rates (8.7% vs 7.8% for non-resumed patients). 29 Our results are in contrast to those obtained

from a 2012 meta-analysis by Chari et al., on the recommencement of anticoagulants in

30
CSDH patients. This study, while finding a similar result in terms of thromboembolic

complications, reported significant differences in recurrence rates between cohorts resuming

30
and not resuming antithrombotic therapy (11.1% vs 22.2% for non-resumed patients). It

PT
should be noted, however, that only one study used in their meta-analysis had a control group

RI
which contained 18 patients, and it therefore may be difficult to draw valid conclusions for

such a small sample size.

SC
There are a number of additional factors to consider with regards to the re-initiation of

U
antithrombotic therapy in CSDH patients. One of primary importance is the timing at which
AN
the antithrombotic treatment reaches therapeutic levels postoperatively. Our analysis

demonstrated that early resumption <2 weeks had comparable complication rates versus late
M

resumption >1 month. Studies have shown that the majority of recurrences occur less than

one month after the initial surgery, potentially indicating the option for safely restarting
D

30
treatment after this time period. The time taken for antithrombotic treatments to reach
TE

therapeutic levels should be factored into the timepoint at which a clinician may choose to
EP

resume this therapy so as to avoid this initial apparent high recurrence window. This may

explain the non-signficant trend toward increased thromboembolic events among those
C

resuming antithrombotic therapy in the early period. We acknowledge that our meta-analysis
AC

may be limited by statistical power and unadjusted baseline characteristics which can

confound the presented results. Future studies are required to further elucidate the effect of

varying time period at which treatment is resumed, rendering this a potential area for future

research.

Additionally, age-specific trends in CSDH pathophysiology and medication history

should be considered. This includes the rise of spontaneous microbleed-enhanced CSDH and

9
 ACCEPTED MANUSCRIPT
pre-existing medication regimes that are continued before and after surgery in the elderly.

Microbleeds are augmenting contributors to hematoma that are believed to trigger an

inflammatory process which extend the hemorrhagic region31. Achieving hemostasis at the

skin level in addition to that of the subdural level may separate the effectiveness of

antithrombotic by administration mode. Potential anti-inflammatory medication may then

PT
dampen the severity of this CSDH contributor. In addition, recent evidence suggests that

RI
atherosclerosis and other vascular risk factors may be linked to increased risk of cerebral

microbleeds32. This indicates that anti-atherosclerotic and even anti-hypertensive medications

SC
may also reduce CSDH severity. Furthermore, selective serotonin reuptake inhibitors in the

presence of antiplatelet agents can induce a synergistic inhibitory effect on platelets, effecting

bleeding risk in CSDH patients. 33

U
It is thus clear, concomitant medication regimes in
AN
patients, particularly the elderly who are more prone to complex regimes, can affect
M

outcomes, and ideally will be controlled for in future studies. However, it should be clarified

that these potential benefits do not derive from drug-drug interactions with antithrombotics,
D

for their pharmacokinetics do not involve cytochrome P450 enzymes or efflux transporters. 34
TE

Rather, there is potential for medication to affect outcomes that directly affect

pathophysiological contributors to CSDH. Clinicians should be wary of these effects when

EP

evaluating a patient for surgery, however what recommendation, if any, remains unclear in
C

this setting until studies investigating medication regimes are reported.

AC

Another aspect for clinical consideration is the utility of postoperative CT imaging in

the decision matrix for resumption of antithrombotics. While this was not consistently

reported in the included studies, the potential to further optimize patient management is

significant. Indeed, the Nomura classification system of five categories was developed to

indicate cSDH severity based on image densities.35 Theoretically then, bleeding progression

and cessation could be further objectified for clinicians without being invasive to the patient.

10
 ACCEPTED MANUSCRIPT
What imaging threshold to resume/withhold antithrombotics, and at which time point(s) after

the event, with respect to the clinical picture, remain two aspects that will benefit from future

study.

This meta-analysis has limitations that need to be taken into account. Firstly, all of the

studies that were included were retrospective in nature, largely due to a lack of well-

PT
controlled prospective studies and randomised trials in the literature. Studies will need to

RI
overcome a number of biases, including inherent patient bias in medication restrictions,

selection bias in the directives from different clinicans, and attrition bias in age and age-

SC
related comorbitdies of the patients. For example, the overall hemorrhagic complication rate

trended towards higher incidence among those who did not resume antithrombotic treatment

U
(p=0.591), which may be a result of patients already deemed to be at high risk of bleeding
AN
experiencing bleeding regardless if antithrombotic treatment was not resumed. Furthermore,
M

several confounding factors that are likely to affect results, such as the original reason for

antithrombotic treatment, the resumption time of antithrombotic agents postoperatively and

D

pre-existing medication regimes, were not reported in the studies included and as such could
TE

not be analysed. While the use of risk stratification systems, such as CHA2DS2-VASc,

HEMORR2HAGES, HAS-BLED and ATRIA, has been useful to clinicians in assessing

EP

patient risk status for both those that are receviving and no receiving antithrombotic therapy,
C

more information is clearly needed in regards the decision to restart or not to restart
AC

antithrombotic therapy following surgical treatment of CSDH patients. Unfortunately this is

lacking in the currently available evidence. There is also very limited evidence on the role of

newer oral anticoagulants (NOACs) such as factor Xa inihibitors including apixaban and

rivaroxaban, thrombin inhibitors such as dabigatran, as well as other lesser used antiplatelet

agents such as clopidogrel. It is unclear whether the present results apply to this class of

anticoagulants.

11
 ACCEPTED MANUSCRIPT
CONCLUSIONS

The decision to resume antithrombotics postoperatively in the clinical management of

CSDH patients is a complex one and should therefore be a highly individualised process. Our

meta-analysis demonstrates that in selected cases, it is feasible to resume early antithrombotic

treatment without additional hemorrhagic or thromboembolic risk.

PT
RI
Funding: None

SC
Conflicts of interest: None

Acknowledgements: None

U
AN
M
D
TE
C EP
AC

12
 ACCEPTED MANUSCRIPT
REFERENCES

• 1. Balser D, Farooq S, Mehmood T, et al: Actual and projected incidence rates

for chronic subdural hematomas in United States Veterans Administration and civilian populations.
Journal of neurosurgery 123:1209-1215, 2015
• 2. Haines D: On the question of a subdural space. The Anatomical Record
230:3-21, 1991
• 3. Maxeiner H, Wolff M: Pure subdural hematomas: a postmortem analysis of
their form and bleeding points. Neurosurgery 50:503-509, 2002

PT
• 4. Lee KS, Doh JW, Bae HG, et al: Relations among traumatic subdural lesions. J
Korean Med Sci 11:55-63, 1996
• 5. Lee K: The pathogenesis and clinical significance of traumatic subdural

RI
hygroma. Brain Injury 12:595-603, 1998
• 6. Almenawer SA, Farrokhyar F, Hong C, et al: Chronic subdural hematoma
management: a systematic review and meta-analysis of 34,829 patients. Ann Surg 259:449-57, 2014

SC
• 7. Adhiyaman V, Asghar M, Ganeshram K, et al: Chronic subdural haematoma
in the elderly. Postgraduate medical journal 78:71-75, 2002
• 8. Ducruet AF, Grobelny BT, Zacharia BE, et al: The surgical management of

U
chronic subdural hematoma. Neurosurgical review 35:155-169, 2012
• 9. Rust T, Kiemer N, Erasmus A: Chronic subdural haematomas and
AN
anticoagulation or anti-thrombotic therapy. Journal of clinical neuroscience 13:823-827, 2006
• 10. Roob G, Schmidt R, Kapeller P, et al: MRI evidence of past cerebral
microbleeds in a healthy elderly population. Neurology 52:991-991, 1999
• 11. Santarius T, Kirkpatrick PJ, Ganesan D, et al: Use of drains versus no drains
M

after burr-hole evacuation of chronic subdural haematoma: a randomised controlled trial. The
Lancet 374:1067-1073, 2009
• 12. Baechli H, Nordmann A, Bucher H, et al: Demographics and prevalent risk
D

factors of chronic subdural haematoma: results of a large single-center cohort study. Neurosurgical
review 27:263-266, 2004
TE

• 13. Moher D, Liberati A, Tetzlaff J, et al: Preferred reporting items for systematic
reviews and meta-analyses: the PRISMA statement. PLoS Med 6:e1000097, 2009
• 14. Phan K, Tian DH, Cao C, et al: Systematic review and meta-analysis:
techniques and a guide for the academic surgeon. Annals of Cardiothoracic Surgery 4:112-122, 2015
EP

• 15. Phan K, Mobbs RJ: Systematic reviews and meta-analyses in spine surgery,
neurosurgery and orthopedics: guidelines for the surgeon scientist. Journal of Spine Surgery 1:19-27,
2015
C

• 16. Hozo SP, Djulbegovic B, Hozo I: Estimating the mean and variance from the
median, range, and the size of a sample. BMC medical research methodology 5:13, 2005
AC

• 17. Stroup DF, Berlin JA, Morton SC, et al: Meta-analysis of observational studies
in epidemiology: a proposal for reporting. Jama 283:2008-2012, 2000
• 18. Freeman MF, Tukey JW: Transformations related to the angular and the
square root. The Annals of Mathematical Statistics:607-611, 1950
• 19. DerSimonian R, Laird N: Meta-analysis in clinical trials. Controlled clinical
trials 7:177-188, 1986
• 20. Amano T, Takahara K, Maehara N, et al: Optimal perioperative management
of antithrombotic agents in patients with chronic subdural hematoma. Clin Neurol Neurosurg
151:43-50, 2016
• 21. Gonugunta V, Buxton N: Warfarin and chronic subdural haematomas. Br J
Neurosurg 15:514-7, 2001

13
 ACCEPTED MANUSCRIPT
• 22. Guha D, Coyne S, Macdonald RL: Timing of the resumption of antithrombotic
agents following surgical evacuation of chronic subdural hematomas: a retrospective cohort study. J
Neurosurg 124:750-9, 2016
• 23. Kamenova M, Lutz K, Schaedelin S, et al: Does Early Resumption of Low-
Dose Aspirin After Evacuation of Chronic Subdural Hematoma With Burr-Hole Drainage Lead to
Higher Recurrence Rates? Neurosurgery 79:715-721, 2016
• 24. Kawamata T, Takeshita M, Kubo O, et al: Management of intracranial
hemorrhage associated with anticoagulant therapy. Surg Neurol 44:438-42; discussion 443, 1995
• 25. Mirzayan MJ, Calvelli K, Capelle HH, et al: Subdural Hematoma and Oral

PT
Anticoagulation: A Therapeutic Dilemma from the Neurosurgical Point of View. J Neurol Surg A Cent
Eur Neurosurg 77:31-5, 2016
• 26. Tsushima S, Komeichi T, Niwa J: Resumption of Antithrombotic Therapy and

RI
the Operative Procedure in Recurrent Chronic Subdural Hematoma. Japanese Journal of
Neurosurgery 22:625-630, 2013
• 27. Yeon JY, Kong DS, Hong SC: Safety of early warfarin resumption following

SC
burr hole drainage for warfarin-associated subacute or chronic subdural hemorrhage. J Neurotrauma
29:1334-41, 2012
• 28. Murthy SB, Gupta A, Merkler AE, et al: Restarting Anticoagulant Therapy
After Intracranial Hemorrhage: A Systematic Review and Meta-Analysis. Stroke 48:1594-1600, 2017

U
• 29. Murthy SB, Gupta A, Merkler AE, et al: Restarting Anticoagulant Therapy
After Intracranial Hemorrhage. Stroke 48:1594-1600, 2017
AN
• 30. Chari A, Clemente Morgado T, Rigamonti D: Recommencement of
anticoagulation in chronic subdural haematoma: a systematic review and meta-analysis. British
journal of neurosurgery 28:2-7, 2014
• 31. Tang J, Ai J, Macdonald RL: Developing a model of chronic subdural
M

hematoma. Acta Neurochir Suppl 111:25-9, 2011

• 32. Romero JR, Preis SR, Beiser A, et al: Carotid Atherosclerosis and Cerebral
Microbleeds: The Framingham Heart Study. J Am Heart Assoc 5:e002377, 2016
D

• 33. Labos C, Dasgupta K, Nedjar H, et al: Risk of bleeding associated with

combined use of selective serotonin reuptake inhibitors and antiplatelet therapy following acute
TE

myocardial infarction. CMAJ 183:1835-43, 2011

• 34. Russo NW, Petrucci G, Rocca B: Aspirin, stroke and drug-drug interactions.
Vascul Pharmacol 87:14-22, 2016
•
EP

35. Nomura S, Kashiwagi S, Fujisawa H, et al: Characterization of local

hyperfibrinolysis in chronic subdural hematomas by SDS-PAGE and immunoblot. J Neurosurg 81:910-
3, 1994
C
AC

14
 ACCEPTED MANUSCRIPT
FIGURE LEGENDS

Figure 1. PRISMA search strategy flow chart of the present systematic review and meta-

analysis

Figure 2. Rate of hemorrhagic complications including recurrence in patients resuming and

PT
not resuming anti-thrombotic treatment after surgical treatment of chronic subdural

hemorrhage.

RI
Figure 3. Rate of hemorrhagic complications including recurrence in patients resuming anti-

SC
thrombotic treatment after surgical treatment of chronic subdural hemorrhage, subgrouped

according to early vs late resumption.

U
AN
Figure 4. Rate of thromboembolism between patients resuming and not resuming

antithrombotic therapy following surgical treatment of chronic subdural hematoma.

M

Figure 5. Rate of thromboembolism between patients resuming and not resuming

D

antithrombotic therapy following surgical treatment of chronic subdural hematoma,

TE

subgrouped according to early vs late resumption.

C EP
AC

15
 ACCEPTED MANUSCRIPT

Author Year Study Period Country Retro vs Randomized or Average FU

Pro Observational (months)
Amano et al 2016 2011-2015 Japan retrospective observational >3
Gonuguta & 2001 1990-1992; 1995- UK retrospective observational 6 to 96

PT
Buxton 1997
Guha et al 2016 2007-2012 Canada retrospective observational 3.1

RI
Kamenova et al 2016 2009-2014 Switzerland retrospective observational 1.4 (43.5 days)
Kawamata et al* 1995 1983-1993 Japan retrospective observational -

SC
Mirzayan et al* 2016 1995-2003 Germany retrospective observational 39 (1-126)
Tsushima 2013 2007-2012 Japan retrospective observational -
Yeon et al 2012 2008-2010 Korea Prospective observational 1 to 6

U
Table 1. Study characteristics of included articles in the present systematic review and meta-analysis. *patients population included some acute

AN
subdural hemorrhage cases but this made up the minority of the sample size.

M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Author Indications for AT AT AT type Time to

Atrial Prostheti VTE Previou Recen Other prior Anticoagula Antiplatele Anticoagulatio restart
fibrillatio c Heart s stroke t MI to nt t n+ AT (d)
n Valve surger antiplatelet

PT
y (n)
Amano et NR NR NR NR NR NR 44 17 (38.6) 22 (50) 5 (11.4) <7 days

RI
al (all)
Gonuguta 8 (23.5) 8 (23.5) 4 6 (17.6) 0 (0) 8 (23.5) 34 34 (100) 0 0 3-5
& Buxton (11.8) days

SC
(all)
Guha et 72 (31.2) 8 (3.5) 16 34 45 56 231 96* (41.6) 170* (73.6) NR 52 days

U
al (6.9) (14.7) (19.5) (24.2) (median
)

AN
Kamenov NR NR NR NR NR NR 140 0 140 (100) 0 5-90
a et al days
(range)

M
Kawamat 3 (11.1) 20 (74.1) 0 (0) 0 (0) 2 (7.4) 2 (7.4) 27 22 (81.5) 0 5 (18.5) <7 days
a et al (16/22)

D
Mirzayan 19 (38.8) 12 (24.5) 10 1 (2.0) 4 (8.2) 3 (6.1) 49 38 (100) 0 0 35-263

TE
et al (20.4. days
) (range)
Tsushima NR NR NR NR NR NR 38 NR NR NR <14
EP
days
Yeon et 5 (25) 12 (60) 0 (0) 1 (5) 0 (0) 2 (10) 20 20 (100) 0 0 3 days
al
C

Total 107/361 60/361 30/36 42/361 51/361 71/361 583 227/545 332/545 10/545 (1.8) -
AC

(29.6) (16.6) 1 (11.6) (14.1) (19.7) (41.7) (60.9)

(8.3)

Table 2. Overview of antithrombotic indications and characteristics. NR, not reported; AT, antithrombotic; VTE, venous thromboembolism.
*mixed patients with those who received both anticoagulation and antiplatelets. Data expressed as n(%).
 ACCEPTED MANUSCRIPT

Amano Gonugunta Guha Kamenova Mirzayan Tsushima Yeon

(2016) (2001) (2016) (2016) (2016) (2013) (2012)

PT
Clear definition of study
population? Y Y Y Y Y Y Y
Clear definition of outcomes and

RI
outcome assessment? Y Y Y Y Y Y Y
Independent assessment of

SC
outcome parameters? U U U U U U U
N, N, Y,
Sufficient duration of follow-up? N, >3mo Y, >6mo mean=3mo mean=1mo median=32mo U Y, 6 mo

U
No selective loss during follow-

AN
up? Y Y Y Y Y U Y
Important confounders and
prognostic factors identified? Y Y Y Y Y Y Y

M
D
Table 3. MOOSE risk of bias assessment. Y, yes; U, unclear; N, no.

TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
Highlights
We included 8 studies in our meta-analysis of antithrombotic resumption after cSDH
No difference in overall hemorrhagic complication rate in the resumption group vs no resumption
Rate of thromboembolism however was statistically lower in those who resumed antithrombotics

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Abbreviations
chronic subdural hematomas (CSDH)
acute subdural hematoma accumulation (ASDH)
Cochrane Central Register of Controlled Trials (CCTR)
Cochrane Database of Systematic Reviews (CDSR)
Database of Abstracts of Review of Effectiveness (DARE)

PT
Transient ischemic attack (TIA)
Confidence interval (CI)

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
All authors on this paper have no conflicts of interest

PT
RI
U SC
AN
M
D
TE
C EP
AC