Risk of aspirin continuation in spinal surgery: a systematic review and meta-analysis

Accepted Manuscript Title: Risk of aspirin continuation in spinal surgery: a systematic review and meta-analysis Author: R. Goes, I.S. Muskens, T.R. Smith, R.A. Mekary, M.L.D. Broekman, W.A. Moojen PII: DOI: Reference:

S1529-9430(17)30912-9 http://dx.doi.org/doi: 10.1016/j.spinee.2017.08.238 SPINEE 57467

To appear in:

The Spine Journal

Received date: Revised date: Accepted date:

4-2-2017 25-7-2017 9-8-2017

Please cite this article as: R. Goes, I.S. Muskens, T.R. Smith, R.A. Mekary, M.L.D. Broekman, W.A. Moojen, Risk of aspirin continuation in spinal surgery: a systematic review and metaanalysis, The Spine Journal (2017), http://dx.doi.org/doi: 10.1016/j.spinee.2017.08.238. 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.

1

Risk of Aspirin Continuation in spinal surgery: a systematic review

2

and meta-analysis

3 4

R. Goes MD1

5

I. S. Muskens BSc2,3,

6

T.R. Smith MD PhD MPH3

7

R.A. Mekary MSc PhD3,4

8

M.L.D. Broekman MD PhD JD2,3,5

9

W. A. Moojen MD PhD1,6,7

10

1

Department of Neurosurgery, Haaglanden Medical center, The Hague, The Netherlands.

11

2

Department of Neurosurgery, Brain Center Rudolf Magnus, University Medical Center

12

Utrecht, Utrecht, The Netherlands

13

3

14

and Woman’s Hospital, Boston, MA, USA

15

4

16

Boston, MA, USA

17

5

Department of Neurology, Massachusetts General Hospital, Boston, MA, USA

18

6

Department of Neurosurgery, Haga teaching Hospital, The Hague, The Netherlands

19

7

Leiden University Medical Center, Leiden, The Netherlands

Cushing Neurosurgery Outcomes Center (CNOC), Department of Neurosurgery, Brigham

Department of Pharmaceutical Business and Administrative Sciences, MCPHS University,

20 21

Corresponding author:

22

R. Goes, department of neurosurgery, Haaglanden Medical Center, Lijnbaan 32 2512VA The

23

Hague, The Netherlands, [email protected]

24 25 26

1

Page 1 of 20

1

Competing interest:

2

R. Goes: has nothing to disclose

3

I. S. Muskens: has nothing to disclose

4

T.R. Smith: has nothing to disclose

5

R.A. Mekary: has nothing to disclose

6

M.L.D. Broekman: has nothing to disclose

7

W. A. Moojen: has nothing to disclose

8 9

Author contributions:

10

R. Goes: Data collection and drafting of the article

11

I.S. Muskens: Data collection and drafting of the article

12

T.R. Smith: Critical revision of the article

13

R.A. Mekary: Data analysis and interpretation, critical revision of the article

14

M.L.D. Broekman: Conception of the design of the work, critical revision of the article

15

W.A. Moojen: Conception of the design of the work, critical revision of the article

16 17

Abstract:

18

Background context:

19

Aspirin is typically discontinued in spinal surgery because of increased risk of haemorrhagic

20

complications. The risk of peri-operative continuation of aspirin in neurosurgery needed to

21

be evaluated.

22 23

Purpose:

24

To evaluate all available evidence about continuation of aspirin and to compare peri- and

25

post-operative blood loss and complication rates between patients that continued aspirin

26

and those who discontinued aspirin peri-operatively in spinal surgery.

27 28

Study Setting:

29

Systematic review and meta-analysis.

30 31

Method:

2

Page 2 of 20

1

A meta-analysis was conducted according to the PRISMA guidelines. Studies comparing

2

aspirin continuation with discontinuation were included. Studies using a combination of

3

anticlotting agents or non-spinal procedures were excluded. Operative outcomes (blood loss

4

and operative length) and different complications (surgical site infection (SSI), stroke,

5

myocardial infarction within 30 days post-operatively) were extracted. Overall prevalence

6

and means were calculated for the reported outcomes in fixed-effects models with

7

heterogeneity (I2) and effect modification (P-interaction) assessment.

8 9

Results:

10

Out of 1339 studies, 3 case series were included in the meta-analysis. No significant

11

differences in mean operating time were seen between the aspirin-continuing group

12

(mean=201.8 minutes, 95%CI=193.3; 210.3; I2=95.4%; 170 patients) and the aspirin-

13

discontinuing group (mean=178.4 minutes, 95%CI=119.1; 237.6; I2=93.5%; 200 patients); (P-

14

interaction=0.78). No significant differences in mean peri-operative blood loss were seen

15

between the aspirin-continuing group (mean=553.9 millilitres, 95%CI=468.0; 639.9;

16

I2=83.4%; 170 patients) and the aspirin-discontinuing group (mean=538.7 millilitres,

17

95%CI=427.6; 649.8; I2=985.5%; 200 patients); (P-interaction=0.96). Similar non-significant

18

differences between the 2 groups were found for cardiac events, stroke, and surgical site

19

infections.

20 21

Conclusions:

22

This meta-analysis showed an absence of significant differences in peri-operative

23

complications between aspirin continuation and discontinuation. Due to the paucity of

24

included studies, further well-designed prospective trials are imperative to demonstrate

25

potential benefit and safety.

26 27

Keywords: low-dose aspirin, peri-operative complications, neurosurgery, spinal surgery, meta-

28

analysis

3

Page 3 of 20

1 2

Introduction Cardiovascular disease (CVD), including myocardial infraction (MI), coronary heart

3

disease (CHD), stroke, and venous thromboembolism (VTE), is the primary cause of death in

4

developed countries and is expected to be the leading cause of death worldwide by 2020.[1]

5

In the primary and secondary prevention of CVD, antiplatelet drugs play a key role, with

6

aspirin being the most commonly used.[1-3] The anticlotting effect of aspirin is the result of

7

the permanent inactivation of the platelet enzyme cyclooxygenase, resulting in a reduction

8

of platelet aggregation and thus thrombocyte deactivation.[2] With the growing incidence of

9

CVD, the use of antiplatelet drugs will increase. This presents a challenge prior to and during

10 11

(neuro-) surgical interventions.[4, 5] The continuation of aspirin in surgical procedures is still a matter of debate among

12

surgeons. Reviews by Burger et al. and Kiberd et al. showed that the continuation of aspirin

13

should be safe in most surgical interventions; however, they advised an individual decision

14

to be made in every case.[6, 7]

15

For neurosurgical patients, recent surveys showed that the protocol of most German

16

hospitals was to discontinue aspirin prior to surgery, to optimize peri-operative coagulation

17

conditions with the aim to reduce the risk of post-operative hemorrhagic complications.[8,

18

9] However, it is unclear to what extent the discontinuation of aspirin effectively reduced

19

the chance of clinically relevant post-operative bleeding complications and mortality.

20

Currently, there seems to be a lack of consensus about aspirin (dis) continuation before

21

neurosurgical interventions.[8, 9] The aim of this review and meta-analysis was to evaluate

22

all available evidence about continuation of aspirin and to compare peri- and post-operative

23

blood loss and complication rates between patients that continued aspirin and those who

24

discontinued aspirin peri-operatively in spinal surgery. The secondary goal was to compare

25

post-operative cardiovascular events between the two groups.

26 27 28 29 30 31

4

Page 4 of 20

1 2

Methods Literature search

3

A systematic review of the current literature was conducted to identify studies

4

reporting outcomes of continuation versus discontinuation of low-dose aspirin in

5

neurosurgery. With help from a professional librarian, PubMed, Embase, Cochrane, and Web

6

of Science databases were searched. For the search strategy, the keywords “neurosurgery”

7

which included spinal surgery and “aspirin” with synonyms were used. The search strategy

8

was performed on June 21st 2016, and updated on May 5th 2017 and was limited to studies

9

in humans (Supplementary table 1). This review was performed in accordance with the

10

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.

11

[10]

12

Study selection

13

Duplicates were removed in Endnote X7.5. Two authors (RG and IM) screened titles

14

and abstracts independently. Only literature in English and Dutch was reviewed. Case-

15

reports, congress abstracts, commentaries and reviews were excluded. Studies in spinal

16

patients, that investigated outcomes and complications of continuation versus

17

discontinuation of low-dose aspirin were included. Studies were excluded when patients

18

used other anticoagulants or a combination of anticoagulants; a control group was missing;

19

or when patients did not undergo a neurosurgical procedure, cranial or peripheral nerve

20

surgery. Non-clinical studies (laboratory or in-vitro studies) were also excluded.

21

Data extraction

22

The independently selected studies were discussed for final inclusion and

23

disagreements were solved by discussion. If consensus was not reached, two independent

24

experts (WM and MB) were asked for their opinion. The following variables were when

25

possible independently extracted by RG and IM from the full text of each study: study

26

design, number of patients, patient age, indication for aspirin usage, history of cardiac stent

27

placement, fusion versus non-fusion, spinal location, interval between discontinuation and

28

surgery, operation time, blood loss (peri-operative blood loss and post-operative drainage),

29

number of transfusions and post-operative complications in the first 30 days such as surgical

30

site infections, stroke, post-operative significant hematomas (requiring surgery), ischemic

31

events, including myocardial infarction, transient ischemic attacks and venous

32

thromboembolism. Study quality was assessed with the modified Newcastle-Ottawa scale

5

Page 5 of 20

1

(NOS).[11] Disagreements in extracted data and study quality evaluation were solved by

2

consultation of one of the senior authors (WM of MB).

3 4

Data analysis

5

Data analysis was conducted using Comprehensive Meta-Analysis (CMA) software

6

version 3. For continuous variables such as blood loss (combined - peri- and post-operative

7

blood loss- and not combined – peri-operative blood loss) and operating time, fixed and

8

random overall means were calculated with I-squared tests for heterogeneity assessment

9

and p-interaction values. To evaluate the ratio of cardiac events, surgical site infections,

10

stroke, and epidural hematomas, fixed and random overall mean prevalence was calculated.

11

Significant interaction was assessed with calculated P-interaction values emanating from

12

sub-group analyses. Heterogeneity was assessed by I-squared tests. For publication bias

13

analysis, the funnel plot, Egger’s test, and Begg’s tests were performed for each of the

14

outcomes.

15

6

Page 6 of 20

1 2 3

Results

4

titles and abstracts, 1406 articles were excluded and 51 were evaluated full-text. Forty-eight

5

studies were excluded because they did not fulfill the inclusion criteria (Figure 1). Three case

6

series were included, with a total of 370 patients.[12-14]

After removing duplicates, 1457 unique articles were identified. After screening for

7 8

Validation

9

The quality of the three included studies were assessed using the Newcastle-Ottawa scale

10

(NOS) score (Supplemental table 2). The total score averaged 5 out of 7 stars. Although the

11

included articles where regarded as case series, they were analyzed in the original studies

12

using a univariate analysis approach, without adjusting for any potential confounders in a

13

multivariate regression model. For instance, two of the spinal studies did not control for

14

possible relevant surgical confounders.[12] No further case series with correction for

15

confounders, adequate power, and reasonable follow-up were identified. Hence, the

16

comparability component was not accounted for in the NOS scale. As the length of follow-

17

up and loss to follow-up were not specified in the included case series, a limited number of

18

stars were assigned to the outcome group for all studies.

19 20

Descriptive analysis

21

Three case series and no randomized controlled trials were identified. The number of

22

patients in the studies ranged from 102 to 200 and the percentage of males ranged from

23

47% to 77%. The total number of participants continuing aspirin ranged from 30 to 100. The

24

dosage of aspirin ranged from 81 mg to 325 mg in the spinal studies. The mean age of

25

participants ranged from 64 to 70 years (Table 1a). All patients continuing their aspirin used

26

aspirin as secondary prevention, whereas the discontinuation group included patients that

27

were on aspirin as prophylaxis and treatment. All patients in the study by Cuellar et al. had a

28

history of cardiac stent placement.[13] The majority of surgeries involved the thoracic or

29

lumbar spine, with only 27 cervical spine surgeries. The pre-operative discontinuation period

30

ranged from 5 to more than 7 days (Table 1b).[12-14]

31

7

Page 7 of 20

1

All included articles described the peri-operative blood loss, which included both the peri-

2

operative and post-operative blood loss. Post-operative blood loss was measured in the

3

blood drainage system after surgery. All studies compared the operation time between the

4

two groups[12-14]. The mean operation time in both groups showed a non-significant

5

increase in the aspirin group. Peri-operative blood loss also increased in all studies. However,

6

only Park et al. could find a significant increase in blood loss in the aspirin group versus the

7

non-users group after correction for NSAID usage (960.3ml versus 1298.6ml, P-

8

interaction=0.043) [12]. The rate of post-operative complications was the same in the two

9

groups of all reviewed articles. Only two studies reported post-operative transfusion rates,

10

which was not significantly different between the two groups. [13, 14]

11 12

Quantitative Analysis

13

No significant p-interaction values between the discontinuation and continuation groups

14

were found in the peri-operative measurements (Table 2). Mean operating times were

15

similar between the aspirin-continuation group (mean=201.8 minutes, 95%CI=193.3; 210.3;

16

I2=95.4%; 170 patients) and the aspirin-discontinuation group (mean=178.4 minutes,

17

95%CI=119.1; 237.6; I2=93.5%; 200 patients); (P-interaction=0.78). As for the peri-operative

18

blood loss, the mean was higher in the aspirin-continuation group (mean=512.0 milliliters,

19

95%CI=442.3; 581.7; I2=98.0%; 70 patients) than in the aspirin-discontinuation group

20

(mean=345.2 milliliters, 95%CI=265.7; 424.7; I2=98.0%; 100 patients) although there was no

21

indication of a significant difference between the 2 groups (P-interaction=0.96). Additionally,

22

the peri-operative and post-operative mean blood loss was similar between the aspirin-

23

continuation group (mean=553.9 milliliters, 95%CI=468.0; 639.9; I2=83.4%; 170 patients) and

24

the aspirin-discontinuation group (mean=538.7 milliliters, 95%CI=427.6; 649.8; I2=985.5%;

25

200 patients); (P-interaction=0.96).

26 27

There was no significant difference in the occurrence of cardiac events between the aspirin-

28

continuation group (1.40%; 95%CI=0.40%; 5.50%; fixed-effects model; I2=0%; p-

29

heterogeneity=0.64; 170 patients) and the aspirin-discontinuation group (1.20%;

30

95%CI=0.30%; 4.70%; fixed-effects model; I2=0%; p-heterogeneity=0.77; 200 patients) (p-

31

interaction=0.87[12, 14] (Table 3 and Figure 2). Similar non-significant differences were

32

found for stroke between the aspirin-continuation group (0.90%; 95%CI=0.20%; 4.40%; 8

Page 8 of 20

1

I2=0%; p-heterogeneity=0.87; 170 patients) and the aspirin-discontinuation group (0.90%;

2

95%CI=0.20%; 4.10%; fixed-effects model; I2=0%; P-heterogeneity=0.84; 200 patients). (P-

3

interaction=0.95) (Table 3 and Figure 3). [12]Non-significant differences were also found for

4

surgical site infections between the aspirin-continuation group (3.20%; 95%CI=1.30%; 7.40%;

5

I2=0%; P-heterogeneity=0.57; 170 patients) and the aspirin-discontinuation group (3.10%;

6

95%CI=1.40%; 7.00%; fixed-effects model; I2=0%; p-heterogeneity=0.63; 200 patients) (p-

7

interaction=0.98) (Table 3). Similarly, non-significant differences were found for significant

8

epidural hematoma between the aspirin-continuation group (1.40%; 95%CI=0.4%; 5.5%;

9

I2=0%; P-heterogeneity=0.64; 170 patients) and the aspirin-discontinuation group (0.9%;

10

95%CI=0.2%; 4.10%; fixed-effects model; I2=0%; p-heterogeneity=0.84; 200 patients) (p-

11

interaction=0.65).

12 13

There was a considerable heterogeneity in the operation variables; I-squared ranged from

14

83.4% to 98%. Due to the limited amount of included studies and the small number of

15

patients, further subgroup analyses were not possible. The Begg’s test revealed a lack of

16

publication bias for each of these outcomes (0.35 ≤P-value ≤1), except for stroke.

9

Page 9 of 20

1 2 3

Discussion This is the first systematic review and meta-analysis on aspirin continuation in spinal

4

surgery. The results showed that there were no significant differences in peri- and post-

5

operative complications when aspirin was continued during spinal procedures.

6

Previously, a large RCT in non-cardiac surgery showed a significant increase of major

7

post-operative bleeding complications, without a significant reduction of death or nonfatal

8

myocardial infarctions. [16]These results are contrasted by an RCT examining high and

9

intermediate risk non-cardiac surgery (1-5% or more risk of post-operative cardiac

10

complications), in which a significant reduction of major cardiac events was found.[17] These

11

results are supported by two reviews, that included mostly non-RCT studies.[6, 7] The review

12

by Burger et al. reported a greater risk of bleeding complications without more severe

13

complications.[6] In cardiac surgery, the results differ, as a systematic review on the peri-

14

operative usage of aspirin during coronary artery surgery concluded that there was a

15

reduction in peri-operative myocardial infarctions, at the expense of a non-significant

16

increase of bleeding related complications and re-explorations.[18] These results were not

17

reproduced in a recent RCT in coronary surgery.[19]

18

Given these conflicting results, both the European and American Cardiac Association

19

guidelines indicated that the continuation of aspirin should remain at the surgeons’

20

discretion.[20, 21] This is in contrast to the European Society of Vascular Surgery, which

21

advises to perform carotid endarterectomies while continuing aspirin in order to prevent

22

post-operative stroke.[22]

23

The surgeons’ opinions and experience on the matter of peri-operative aspirin

24

continuation, was evaluated by Korinth et al. among German Neurosurgeons concluding that

25

approximately 75% of all neurosurgeons considered aspirin to be a risk factor for

26

hemorrhagic complications. Furthermore, about half of the respondents had experienced

27

complications related to peri-operative continuation of aspirin.[9]

28

Despite evaluation of all available literature on the comparison of peri-operative

29

continuation and discontinuation of aspirin, this meta-analysis was based on only three case

30

series. One study by Akhavan et al. was excluded as Clopidogrel was used for anti-

31

coagulation.[15] Furthermore, a meta-regression on the aspirin dosage, usage indication,

32

anatomical location, with or without cardiac stent, and interval between discontinuation and

10

Page 10 of 20

1

surgery, and type of surgery could not be performed due to the low number of studies and

2

included patients (table 1a and 1b). Types of surgery varied from instrumented spinal

3

surgery and conventional discectomies. With regard to post-operative complications,

4

cardiovascular complications were only investigated in two studies, which limits the

5

conclusions. [13, 14] The same applies to cerebrovascular complications as described by Park

6

et al.[12] Furthermore, every meta-analysis inherently has inter-rater variability as each

7

outcome and the study quality was evaluated by two investigators.

8

Therefore, no recommendation can be given and it is important that future

9

prospective research aims to evaluate the ideal peri-operative anticoagulant scheme and

10

outcomes. This is especially important for an aging population that will require more spinal

11

surgery.

12 13

Conclusion

14

This meta-analysis suggests that continuation of aspirin during neurosurgical procedures

15

does not result in a higher rate of complications. However, due to the limited number of

16

included case series, further prospective evaluation of safety of peri-operative aspirin

17

continuation is required. Therefore, the matter of aspirin continuation during spinal

18

operation remains a subject of debate and should be evaluated in each individual case.

19 20 21

Funding

22 23

No funding

11

Page 11 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

References 1. Cohen A, Imfeld S, Markham J, Granziera S. The use of aspirin for primary and secondary prevention in venous thromboembolism and other cardiovascular disorders. Thrombosis research. 2015;135(2):217-25. 2. Hall R, Mazer CD. Antiplatelet drugs: a review of their pharmacology and management in the perioperative period. Anesthesia & Analgesia. 2011;112(2):292-318. 3. Steiger N, Cifu AS. Primary Prevention of Stroke. JAMA. 2016;316(6):658-9. 4. Biondi-Zoccai GG, Lotrionte M, Agostoni P, et al. A systematic review and metaanalysis on the hazards of discontinuing or not adhering to aspirin among 50 279 patients at risk for coronary artery disease. European heart journal. 2006;27(22):2667-74. 5. Maulaz AB, Bezerra DC, Michel P, Bogousslavsky J. Effect of discontinuing aspirin therapy on the risk of brain ischemic stroke. Archives of neurology. 2005;62(8):1217-20. 6. Burger W, CHEMNITIUS JM, Kneissl G, Rücker G. Low‐dose aspirin for secondary cardiovascular prevention–cardiovascular risks after its perioperative withdrawal versus bleeding risks with its continuation–review and meta‐analysis. Journal of internal medicine. 2005;257(5):399-414. 7. Kiberd MB, Hall RI. Aspirin in the perioperative period: a review of the recent literature. Current Opinion in Anesthesiology. 2015;28(3):349-55. 8. Korinth CM. Low-dose aspirin before intracranial surgery – results of a survey among neurosurgeons in Germany. Acta Neurochirurgica. 2006;148(11):1189-96. 9. Korinth MC, Gilsbach JM, Weinzierl MR. Low-dose aspirin before spinal surgery: results of a survey among neurosurgeons in Germany. European Spine Journal. 2007;16(3):365-72. 10. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264-9, W64. 11. Wells G, Shea B, O’connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 2000. 12. Park HJ, Kwon KY, Woo JH. Comparison of blood loss according to use of aspirin in lumbar fusion patients. Eur Spine J. 2014;23(8):1777-82. 13. Cuellar JM, Petrizzo A, Vaswani R, Goldstein JA, Bendo JA. Does aspirin administration increase perioperative morbidity in patients with cardiac stents undergoing spinal surgery? Spine (Phila Pa 1976). 2015;40(9):629-35. 14. Soleman J, Baumgarten P, Perrig WN, Fandino J, Fathi A-R. Non-instrumented extradural lumbar spine surgery under low-dose acetylsalicylic acid: a comparative risk analysis study. European Spine Journal. 2015:1-8. 15. Akhavan-Sigari R, Rohde V, Abili M. Continuation of medically necessary platelet aggregation inhibitors - acetylsalicylic acid and clopidogrel - during surgery for spinal degenerative disorders: Results in 100 patients. Surg Neurol Int. 2014;5(Suppl 7):S376-S9. 16. Devereaux PJ, Mrkobrada M, Sessler DI, et al. Aspirin in Patients Undergoing Noncardiac Surgery. New England Journal of Medicine. 2014;370(16):1494-503. 17. Oscarsson A, Gupta A, Fredrikson M, et al. To continue or discontinue aspirin in the perioperative period: a randomized, controlled clinical trial. British journal of anaesthesia. 2010;104(3):305-12. 18. Hastings S, Myles P, McIlroy D. Aspirin and coronary artery surgery: a systematic review and meta-analysis. British journal of anaesthesia. 2015:aev164.

12

Page 12 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

19. Myles PS, Smith JA, Forbes A, et al. Stopping vs. Continuing Aspirin before Coronary Artery Surgery. New England Journal of Medicine. 2016;374(8):728-37. 20. Fleisher LA, Fleischmann KE, Auerbach AD, et al. 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. Developed in collaboration with the American College of Surgeons, American Society of Anesthesiologists, American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Anesthesiologists, and Society of Vascular Medicine Endorsed by the Society of Hospital Medicine. Journal of nuclear cardiology: official publication of the American Society of Nuclear Cardiology. 2015;22(1):162-215. 21. Kristensen SD, Knuuti J, Saraste A, et al. 2014 ESC/ESA Guidelines on non-cardiac surgery: cardiovascular assessment and management. European heart journal. 2014:ehu282. 22. Liapis C, Bell P, Mikhailidis D, et al. ESVS guidelines. Invasive treatment for carotid stenosis: indications, techniques. European journal of vascular and endovascular surgery. 2009;37(4):1-19.

20

13

Page 13 of 20

1

Figure 1: Study selection process of the identified articles in our meta-analysis

Records identified through database searching (n = 2052)

Additional records identified through other sources (n = 0)

Records after duplicates removed (n = 1457)

Included

Eligibility

Screening

Identification

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Legends:

Records screened (n = 1457)

Full-text articles assessed for eligibility (n = 51)

Studies included in qualitative synthesis (n =4)

Records excluded (n =1406)

Full-text articles excluded, with reasons (n = 47): -Not a neurosurgical procedure (18) -Not a clinical study (8) -Peripheral nerve surgery (1) -Not specific for only aspirin continuation during surgery (20)

Cranial study (n =1) Studies included in quantitative synthesis (meta-analysis) (n = 3)

14

Page 14 of 20

1

Figure 2: Forest plot represents the prevalence of post-operative cardiac events (95% CI)

2

with continuing or discontinuing of aspirin peri-operative. The horizontal lines denote the

3

95%-CIs whereas the solid squares represent the point estimate of each study and the

4

diamonds represent the pooled estimate of each subgroup. The size of the solid squares is

5

proportional to the weight of the study. Weights are from the fixed-effects analysis using the

6

inverse variance method. The “continue” group represents a heterogeneity of I2 0%; p=0.64

7

and the “discontinue” group I20%; P=0.77, with a p-interaction value of 0.87. Prevalence of cardiac events by aspirin continuation or discontinuation Model

Group by Asp_disc

Study name

Continue

Cuellar B2015

0.005

0.000

0.074

Continue

Park B 2014

0.013

0.001

0.175

Continue

Soleman B 2015

0.025

0.004

0.157

0.014

0.004

0.055

Event rate

Fixed Continue

Lower limit

Upper limit

Discontinue

Cuellar A 2015

0.005

0.000

0.074

Discontinue

Park A 2014

0.016

0.001

0.211

Discontinue

Soleman A 2015

0.016

0.002

0.106

Fixed Discontinue

0.012

0.003

0.047

Fixed Overall

0.013

0.005

0.034

Event rate and 95%CI

-0.50 -0.25 0.00 0.25 0.50

Continue group: I2=0%; P=0.64; Discontinue group: I2=0%; P=0.77; P-interaction=0.87

8 9

15

Page 15 of 20

1

Figure 3: Forest plot represents the prevalence of stroke events (95% CI) with continuing or

2

discontinuing of aspirin peri-operative. The horizontal lines denote the 95%CIs whereas the

3

solid squares represent the point estimate of each study and the diamonds represent the

4

pooled estimate of each subgroup. The size of the solid squares is proportional to the weight

5

of the study. Weights are from the fixed-effects analysis using the inverse variance method.

6

The “continue” group represents a heterogeneity of I2 0%; p=0.87 and the “discontinue”

7

group I2:0%; p=0.84, with a p-interaction value of 0.95.

Prevalence of stroke by aspirin continuation or discontinuation Model

Group by Asp_disc

Study name

Continue

Cuellar B2015

0.005

0.000

0.074

Continue

Park B 2014

0.013

0.001

0.175

Continue

Soleman B 2015

0.012

0.001

0.167

0.009

0.002

0.044

Event rate

Fixed Continue

Lower limit

Upper limit

Discontinue

Cuellar A 2015

0.005

0.000

0.074

Discontinue

Park A 2014

0.016

0.001

0.211

Discontinue

Soleman A 2015

0.008

0.000

0.115

Fixed Discontinue

0.009

0.002

0.041

Fixed Overall

0.009

0.003

0.027

Event rate and 95%CI

-0.50 -0.25 0.00 0.25 0.50

Continue group: I2=0%; P=0.87; Discontinue group: I2=0%; P=0.84; P-interaction=0.95

8

16

Page 16 of 20

1 2

Table 1a: Study characteristics for each of the included studies

First author, Publication year

Study design

Study years

Country of origin

Type of surgery

Modified NOS scale (7)

Cuellar, 201513

Case series

Feb 2009Nov 2013

USA

Spinal

5

Park, 201412

Soleman, 201514

Case series

Case series

Sept 2008Jan 2011

South Korea

Jan 2010April 2012

Switzerland

Spinal

Spinal

5

5

Total N of participants

Mean age (SD)

% Male

Dose of aspirin

Continued group

100

69 (9)

79

N 140: 81mg, N 37: 325 mg

210 (136)

697 (1187)

Discontinued Group

100

68 (9)

75

NA

266 (143)

642 (905)

Continued group

30

64.3 (10.5)

46.7

Unknown

211 (27)

1332.8 (464.8)

Discontinued group Continued Group

38

66.3 (9.4) 70.9 (9.5)

52.6

NA

72.5

100mg

200 (28) 95 (39)

1281.7 (393.6) 373.95 (285.5)

Discontinued Group

62

69.4

NA

109 (45)

312.7 (203.3)

40

70.8 (9.5)

Operation duration in minutes (SD)

Blood loss in milliliter (SD)

3 4

Table 1a legend: abbreviations: SD: standard Deviation, NA: not applicable, NOS: Newcastle-Ottawa Scale with a maximum of 7 stars, N: number of patients.

5

Table 1b: study characteristics continued

First Author, publication year

Number of patients in group

Prophylaxis or treatment

Stent placement

(Non) instrumented

Location C/T/L*

Aspirin discontinuation prior to surgery (days)

17

Page 17 of 20

Cuellar, 201513 Continued group

Park, 201412

Soleman, 201514

1 2 3 4 5 6

100

Treatment

100

Non-instrumented

L

NA

Discontinued Group

100

Treatment

100

Non-instrumented

L

At least 5

Continued group

30

Treatment

Unknown

Instrumented

L

NA

Discontinued group Continued Group

38

Mixed

Unknown

Instrumented

L

7

40

Treatment

21

Instrumented (55) Non-instrumented (45)

C: 7 T/L**: 93

NA

Discontinued Group

62

Mixed

8

Instrumented (70) Non-instrumented (30)

C: 20 T/L**: 80

At least 7

Table 1b legend: *C/T/L: Cervical/ Thoracic/Lumbar, ** T/L: Thoracic and Lumbar, NA: not applicable

18

Page 18 of 20

1 2 3

Table 2: Pooled mean prevalence of blood loss, blood loss combined, and operation time. Operation Group variables

Fixed-Effect model

Mean

95% CI

Blood loss Continuation (ml)* Discontinuation Overall

512 345 440

Blood loss Continuation combined Discontinuation (ml)** Overall Operation Continuation time (min) Discontinuation Overall

Random-Effect Model

Heterogeneity

Mean

95% CI

P- value

I2 %

P-value

442-581 266-424 387-492

Pvalue <0.01 <0.01 <0.01

428 449 438

-119-975 -100-997 51-825

0.13 0.11 0.03

98 98

<0.01 <0.01

554 539 548

468-640 428-650 480-616

<0.01 <0.01 <0.01

478 529 503

212-745 258-800 313-693

<0.01 <0.01 <0.01

83 86

201 204 203

193-201 195-213 196-209

<0.01 <0.01 <0.01

190 178 184

130-250 119-237 142-226

<0.01 <0.01 <0.01

95 94

P-Interaction value

P- value Begg’s test

0.96

1.00

0.80

0.55

0.78

0.57

<0.01 <0.01

<0.01 <0.01

4 5 6 7

Table 2 legend: *Blood loss included peri-operative blood loss, **Blood loss combined included peri- and post-operative blood loss

19

Page 19 of 20

1 2

Table 3: Pooled mean prevalence of postoperative complications significant epidural hematoma, surgical site infection, cardiac events, and stroke.

3 Complication

SEH

SSI

Cardiac events

Stroke

Group

Fixed-effect model

Heterogeneity

Mean

95% CI

I2 %

Continue Discontinue Overall

0.01 0.01 0.01

0.01- 0.06 0.01- 0.04 0.01- 0.03

0 0

Continue Discontinue Overall

0.03 0.03 0.03

0.01- 0.07 0.01- 0.07 0.02- 0.06

0 0

Continue Discontinue Overall

0.01 0.01 0.01

0.01- 0.06 0.01- 0.05 0.01- 0.03

0 0

Continue Discontinue Overall

0.01 0.01 0.01

0.01- 0.04 0.01-0.04 0.01- 0.03

0 0

4 5 6 7 8

Table 3 legend: Abbreviations: SEH: Significant epidural hematoma, SSI: Surgical site infection

9

evaluated as case series. Abbreviations: NA: not evaluated

P-interaction value

P- value Begg’s test

0.65

0.45

0.98

0.70

0.87

0.71

0.95

<0.01

P-value for Cochrane Q test 0.64 0.84

0.57 0.63

0.64 0.77

0.87 0.84

Supplementary table 2 legend: *Maximum number of stars per category. **not evaluated, as comparability cannot be scored as the articles are

10

20

Page 20 of 20