For Peer Review Only Journal: BMJ Open

BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

The efficacy and safety of complete pericardial drainage by means of intrapericardial fibrinolysis in the prevention of complications of pericardial effusion: a systematic review protocol

For peer review only Journal: BMJ Open

Manuscript ID: bmjopen-2015-007842

Article Type: Protocol

Date Submitted by the Author: 08-Feb-2015

Complete List of Authors: Kakia, Aloysious; Walter Sisulu University, Family Medicine and Rural Health Wiysonge, Charles; Stellenbosch University, Centre for Evidence-based Health Care Ochodo, Eleanor; Stellenbosch University, Centre for Evidence-based Health Care Awotedu, Abolade; Walter Sisulu University, Internal Medicine Ristic, Arsen; Belgrade University, Department of Cardiology, Clinical Center of Serbia and Belgrade University School of Medicine Mayosi, Bongani; University of Cape Town, Department of Medicine, Groote Schuur Hospital http://bmjopen.bmj.com/ Primary Subject Cardiovascular medicine Heading:

Secondary Subject Heading: Evidence based practice

Adult cardiology < CARDIOLOGY, Paediatric cardiology < CARDIOLOGY, Keywords: PREVENTIVE MEDICINE, Cardiology < INTERNAL MEDICINE

on September 30, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 10 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 4 The efficacy and safety of complete pericardial drainage by means of 5 6 intrapericardial fibrinolysis in the prevention of complications of pericardial 7 8 9 effusion: a systematic review protocol 10 11 12

13 1* 2, 3 2 4 5 14 Aloysious Kakia , Charles S. Wiysonge , Eleanor A. Ochodo , Abolade A. Awotedu , Arsen D. Ristic, 15 Bongani M. MayosiFor6 peer review only 16 17

18 19 20 1 21 Department of Family Medicine and Rural Health, Walter Sisulu University, Mthatha, South 22 23 Africa;2 Centre for Evidencebased Health Care, Stellenbosch University, Cape Town, South 24 25 Africa;3 Division of Community Health, Stellenbosch University, Cape Town, South 26 27 4 28 Africa; Department of Medicine, Nelson Mandela Academic Hospital and Walter Sisulu 29 30 University, Mthatha, South Africa;5Department of Cardiology, Clinical Center of Serbia and 31 32 Belgrade University School of Medicine, Belgrade, Serbia;6Groote Schuur Hospital and 33 34 http://bmjopen.bmj.com/ 35 University of Cape Town, Cape Town, South Africa 36 37 Email addresses: 38 39 40 Aloysious Kakia: [email protected]; Charles Shey Wiysonge: [email protected]; 41 42 Eleanor A. Ochodo: [email protected]; Abolade A. Awotedu: [email protected]; on September 30, 2021 by guest. Protected copyright. 43 44 Arsen Ristic: [email protected]; Bongani M. Mayosi: [email protected] 45 46 47 48 * Corresponding author. 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 10 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 ABSTRACT: 4 5 6 7 Background: Intrapericardial fibrinolysis has been proposed as a means of preventing 8 9 complications of pericardial effusion such as cardiac tamponade, persistent and recurrent 10 11 12 pericardial effusion, as well as pericardial constriction. There is a need to understand the efficacy 13 14 and safety of this procedure because it shows promise. The purpose of this review therefore is to 15 For peer review only 16 conduct a systematic analysis on the efficacy and safety of intrapericardial fibrinolysis in the 17 18 19 prevention of the complications of pericardial effusion. 20 21 22 Methods: We will include studies that evaluate the efficacy and/or safety of complete pericardial 23 24 fluid drainage by intrapericardial fibrinolysis for preventing complications of pericardial effusion 25 26 27 irrespective of study design, geographical location, language, status of publication, age of 28 29 participants, aetiology of pericardial disease, or types of fibrinolytics used up to 31st December 30 31 32 2014. The primary outcomes will be development of cardiac tamponade, persistent or recurrent 33 34 pericarditis without tamponade, constrictive pericarditis, hospitalization and death, as well as http://bmjopen.bmj.com/ 35 36 adverse events related to the use of intrapericardial fibrinolytics. We shall search PubMed, the 37 38 39 Cochrane Library, African Journals online, Cumulative Index to Nursing and Allied Health 40 41 Literature, Trip database, Clinical trials.gov and the WHO International Clinical Trials Registry 42 on September 30, 2021 by guest. Protected copyright. 43 Platform. Two reviewers will do the search independently, screen the search outputs for 44 45 46 potentially eligible studies, and assess whether the studies meet the inclusion criteria. 47 48 Discrepancies between the two reviewers will be resolved through discussion with a third 49 50 reviewer and consensus. Data from the selected studies shall be extracted using a standardised 51 52 53 data collection form which will be piloted before use. The methodological quality of studies will 54 55 be assessed using the Cochrane Collaboration’s tool for assessing risk of bias for experimental 56 57 58 studies and the Scottish Intercollegiate Guidelines Network (SIGN) checklist for other study 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 10 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 designs. The primary metaanalysis will use random effects models due to expected interstudies 4 5 6 heterogeneity. Dichotomous data will be analysed using relative risk and continuous data using 7 8 weighted mean differences (or standardised mean differences), both with 95% CIs. 9 10 11 Discussion: This systematic review will shed light on the evidence to date regarding the efficacy 12 13 14 and safety of intrapericardial fibrinolysis in preventing constrictive pericarditis, and guide future 15 For peer review only 16 research on this theme. 17 18 Study strengths: Unbiased selection of many studies conducted in different settings. This will 19 20 21 strengthen the validity of the review results. 22 23 Study limitations: Heterogeneity of the study settings of the lowincome, lowermiddleincome 24 25 26 and uppermiddleincome countries as well as heterogeneity in study designs. 27 28 29 30 Review registration: This review is registered with PROSPERO, registration number 31 32 33 CRD42014015238 34 http://bmjopen.bmj.com/ 35 36 Key words: ‘Pericarditis’, ‘tuberculous pericarditis,’ ‘purulent pericarditis,’ 37 38 ’pericardiocentesis’, ‘therapeutic pericardiocentesis,’ ‘fibrinolytics,’ ‘intrapericardial 39 40 41 fibrinolytics,’ ‘urokinase,’ ‘streptokinase,’ ‘tissue plasminogen activator.’ 42 on September 30, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 10 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 BACKGROUND 4 5 6 7 The era of human immunedeficiency virus (HIV) has seen an increase in the incidence of 8 9 pericarditis1. The main cause of pericarditis in Africa is tuberculosis2. Pericarditis may 10 11 complicate to tamponade in the short term, and chronic effusive pericarditis and constrictive 12 13 14 pericarditis in the long term. Cardiac tamponade and constrictive pericarditis lead to death if not 15 For peer review only 16 treated. The definitive management for constrictive pericarditis involves pericardiectomy, which 17 18 is associated with a mortality of up to fourteen percent3, and is an expensive procedure4. Imazio 19 20 5 21 et al have shown that tuberculous and purulent pericarditis are more likely to progress to 22 23 constrictive pericarditis than pericarditis due to other causes. Ntsekhe et al6 found a 10.9 % 24 25 26 incidence of constrictive pericarditis over a six month period in patients with pericardial 27 28 effusions presumed to be tuberculous. These findings highlight the importance of efforts to 29 30 prevent progression of pericarditis to constrictive pericarditis. 31 32 33 34 Various strategies have been used to prevent progression of acute pericarditis to constrictive http://bmjopen.bmj.com/ 35 36 pericarditis. Early diagnosis and prompt treatment of pericarditis, including treating the 37 38 underlying cause and draining effusions, are a major step in this direction. The use of colchicine 39 40 41 as adjunctive treatment to prevent recurrent and persistent pericarditis, and thereby reducing the 42 on September 30, 2021 by guest. Protected copyright. 43 risk of constriction, showed promise in a randomized clinical trial conducted by Imazio et al7. 44 45 Corticosteroids have been found to be useful in several trials; however, the findings of Mayosi et 46 47 8 48 al have shown that corticosteroids could increase the risk of cancers in patients coinfected with 49 50 HIV. 51 52 53 Intrapericardial fibrinolysis has been proposed as a way of stemming the development of 54 55 56 cardiac tamponade and constriction in patients with effusive pericarditis. The objective of 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 10 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 fibrinolysis is to target fibrin formation, to optimize evacuation of a thick fluid, and therefore to 4 5 9 6 prevent both persistent purulent pericarditis and constrictive pericarditis . The procedure is also 7 8 minimally invasive. A clinical review conducted by Augustine et al9 concluded that 9 10 10 11 intrapericardial fibrinolysis may be useful for prevention of constrictive pericarditis. Cui et al 12 13 investigated the efficacy of intrapericardial fibrinolysis in preventing constrictive pericarditis in 14 15 patients with infectiveFor pericardial peer effusion, review 60 % of which were onlyof tuberculous origin. They 16 17 18 found that the early employment of fibrinolysis optimized complete evacuation of the pericardial 19 20 effusion, significantly reduced progress to pericardial constriction, and was safe. 21 22 In view of the promise held by intrapericardial fibrinolysis, there is currently a need to better 23 24 25 understand the safety and efficacy of the procedure. We propose therefore to conduct a 26 27 systematic review to assess the efficacy and safety of intrapericardial fibrinolysis in the 28 29 prevention of complications of pericardial effusion, such as cardiac tamponade, recurrent or 30 31 32 persistent effusion, constrictive pericarditis, hospitalisation, and death. 33 34 http://bmjopen.bmj.com/ 35 OBJECTIVES 36 37 38 1. To determine whether complete pericardial drainage by intrapericardial fibrinolysis reduces 39 40 41 the incidence of cardiac tamponade, persistent or recurrent pericardial effusion, constrictive 42 on September 30, 2021 by guest. Protected copyright. 43 pericarditis, hospitalisation and death in patients with pericardial effusion. 44 45 46 2. To determine whether complete pericardial drainage by intrapericardial fibrinolysis can be 47 48 performed safely with respect to the incidence of haemorrhage, procedurerelated cardiac 49 50 tamponade, allergy, serious and nonserious adverse events. 51 52 53 3. To determine the appropriate timing, dose and volume of intrapericardial fibrinolysis. 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 10 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 METHODS 4 5 6 7 Types of studies 8 9 10 We will consider primary studies with the following designs: 11 12 13 14 ▸ Intervention studies: randomised controlled trials (RCTs), and quasiRCTs. 15 For peer review only 16 17 ▸ Observational studies: case reports, cohort studies, case–control studies and crosssectional 18 19 20 studies. 21 22 23 Types of participants 24 25 26 27 People of all ages requiring intrapericardial fibrinolysis for evacuation of pericardial effusion due 28 29 to any cause. 30 31 32 33 Study settings 34 http://bmjopen.bmj.com/ 35 36 We will include studies that evaluate the efficacy and/or safety of intrapericardial fibrinolysis for 37 38 preventing constrictive pericarditis irrespective of geographical location. 39 40 41 42 Types of interventions on September 30, 2021 by guest. Protected copyright. 43 44 45 All types of fibrinolytics will be considered including (but not limited to) urokinase, 46 47 streptokinase, and tissue plasminogen activator. 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 10 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 4 5 6 7 Types of outcome measures 8 9 10 The efficacy outcomes of interest to this review are cardiac tamponade, persistent or recurrent 11 12 13 pericardial effusion, constrictive pericarditis, hospitalisation and death in patients with 14 15 pericardial effusion.For peer review only 16 17 18 The safety outcomes will be the incidence of haemorrhage, procedurerelated cardiac tamponade, 19 20 21 allergy, serious and nonserious adverse events. 22 23 24 Search methods for identification of studies 25 26 27 We will develop a comprehensive strategy to search for all eligible studies available up to 31st 28 29 30 December 2014, regardless of language or publication status. For published literature we will 31 32 search the electronic databases PubMed, Cochrane Library (Cochrane Central Register of 33 34 http://bmjopen.bmj.com/ 35 Controlled Trials, Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews 36 37 of Effects (DARE)), African Journals online (AJOL), Cumulative Index to Nursing and Allied 38 39 Health Literature (CINAHL), and Trip database. We will use a combination of the following 40 41 42 search terms and tailor them appropriately to the different databases: ‘Pericarditis’, ‘tuberculous on September 30, 2021 by guest. Protected copyright. 43 44 pericarditis,’ ‘purulent pericarditis,’ ‘pericardiocentesis’, ‘therapeutic pericardiocentesis,’ 45 46 ‘fibrinolytics,’ ‘intrapericardial fibrinolytics,’ ‘urokinase,’ ‘streptokinase,’ ‘tissue plasminogen 47 48 49 activator.’ To avoid selection bias, two reviewers will do the search independently. To access 50 51 unpublished literature, we will contact experts in the field of therapeutic pericardiocentesis and 52 53 54 search Clinical trials.gov and the WHO International Clinical Trials Registry Platform. 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 10 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 Data collection and analysis 4 5 6 Two authors will independently screen the search outputs for potentially eligible studies, 7 8 compare their results, and resolve disagreements by discussion and consensus. The two authors 9 10 11 will then independently go through the full text of all potentially eligible studies to assess 12 13 whether the studies meet the inclusion criteria defined by the study design, setting, intervention, 14 15 and outcomes. DiscrepanciesFor peer in the list of eligiblereview studies between only the two authors will be 16 17 18 resolved through discussion and consensus. A structured and standardised data collection form 19 20 shall be developed for extracting data from the selected studies. The form will capture key study 21 22 characteristics, including methods, participants, and outcomes. Prior to use, the extraction form 23 24 25 will be piloted on at least three studies identified randomly from the list of included studies. 26 27 The methodological quality of studies will be assessed using the Cochrane Collaboration’s tool 28 29 for assessing risk of bias for experimental studies and the Scottish Intercollegiate Guidelines 30 31 32 Network (SIGN) checklist for other study designs. 33 http://bmjopen.bmj.com/ 34 All eligible studies will be summarized and analyzed using Cochrane Review manager. Two 35 36 37 authors will extract the data; one author will enter the data and the second author will recheck the 38 39 entries. In the event of discrepancy, the authors shall discuss and resolve the disagreement. If the 40 41 studies are sufficiently similar, we will combine the data using the random effects model. We 42 on September 30, 2021 by guest. Protected copyright. 43 44 will stratify the analyses by study design. We will examine statistical heterogeneity between 45 46 study results using the chisquared (χ2) test of homogeneity (with a significance αlevel of 0.1). 47 48 We shall quantify statistical heterogeneity between study results using the inconsistency index 49 50 2 11, 12, 13, 14 51 (I ) .When studies cannot be combined for metaanalysis due to diversity of 52 53 interventions, narrative syntheses will be conducted. 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 10 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 The planned systematic review is registered with the International Prospective Register of 4 5 6 Systematic Reviews (PROSPERO), registration number CRD42014015238. 7 8 9 DISCUSSION 10 11 12 To our knowledge no systematic review on intrapericardial fibrinolysis for the prevention of 13 14 15 constrictive pericarditisFor has peerbeen done to date.review This systematic reviewonly will bring out the current 16 17 evidence regarding the efficacy and safety of intrapericardial fibrinolysis in preventing 18 19 20 constrictive pericarditis. Our discussion of the findings shall be in light of the relevance of this 21 15 22 data in clinical decision making and future research design and direction on this topic. 23 24 25 26 27 28 29 REFERENCES 30 31 32 1. Mayosi BM, Burgess LJ, Doubell AF. Tuberculous pericarditis. Circulation. 2005;112:3608– 33 http://bmjopen.bmj.com/ 34 3616 35 36 37 2. Mayosi BM. Contemporary trends in the epidemiology and management of cardiomyopathy 38 39 and pericarditis in subSaharan Africa. Heart. 2007; 93:1176–1183. 40 41 3. Mutyaba A, Balkaran S, Cloete R, et al. Constrictive pericarditis requiring pericardiectomy at 42 on September 30, 2021 by guest. Protected copyright. 43 44 Groote Schuur Hospital in Cape Town, South Africa: causes and perioperative outcomes in 45 46 the HIV era (1990–2012). J Thorac Cardiovasc Surg (in press). 47 48 4. Mayosi BM, Ntsekhe M, Volmink JA, et al. Interventions for treating tuberculous 49 50 51 pericarditis. Cochrane Database Syst Rev 2002; CD000526. 52 53 5. Imazio M, Brucato A, Maestroni S, et al. Risk of Constrictive Pericarditis after Acute 54 55 56 Pericarditis. Circulation 2011; 124:1270–5. 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 10 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 6. Ntsekhe M, Wiysonge CS, Gumedze F, et al. HIV Infection Is Associated with a Lower 4 5 6 Incidence of Constriction in Presumed Tuberculous Pericarditis: A Prospective Observational 7 8 Study. PLoS ONE:2008;3(6): e2253. 9 10 11 7. Imazio M, Brucato A, Cemin R, et al. Colchicine for Recurrent Pericarditis: A Randomized 12 13 Trial. Ann Intern Med 2011; 155:409414. 14 15 8. Mayosi BM,For Ntsekhe M, peer Bosch J, et al. review prednisolone and Mycobacterium only indicuspranii in 16 17 18 Tuberculous Pericarditis. N Engl J Med 2014; 371: 112130. 19 20 9. Augustin P, Desmard M, Mordant P, et al. Clinical review: intrapericardial fibrinolysis in 21 22 management of purulent pericarditis. Crit Care. 2011 Apr 20;15(2):220. doi: 23 24 25 10.1186/cc10022. 26 27 10. Cui HB, Chen XY, Cui CC, et al. Prevention of pericardial constriction by transcatheter 28 29 intrapericardial fibrinolysis with urokinase. Chin Med Sci J 2005; 20:510. 30 31 32 11. Haidich AB Metaanalysis in medical research. Hippokratia 2010; 14 (Suppl 1): 2937. 33 http://bmjopen.bmj.com/ 34 12. Hardy RJ, Thompson SG. Detecting and describing heterogeneity in metaanalysis. Stat 35 36 37 Med.1998; 17: 841–856. 38 39 13. Higgins JP, Thompsom SG, Deeks JJ, Altman DG. Measuring inconsistency in meta 40 41 analysis. BMJ. 2003; 327: 557–560. 42 on September 30, 2021 by guest. Protected copyright. 43 44 14. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions 45 46 Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, Oxford, UK, 2011. 47 48 15. Maisch B, Ristić AD, Seferović PM, Tsang TSM. Interventional Pericardiology: 49 50 51 Pericardiocentesis, Pericardioscopy, Pericardial Biopsy, Balloon Pericardiotomy, and 52 53 Intrapericardial Therapy. Springer Verlag, Heidelberg 2011. 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

The efficacy and safety of complete pericardial drainage by means of intrapericardial fibrinolysis in the prevention of complications of pericardial effusion: a systematic review protocol

For peer review only Journal: BMJ Open

Manuscript ID bmjopen-2015-007842.R1

Article Type: Protocol

Date Submitted by the Author: 28-Sep-2015

Secondary Subject Heading: Evidence based practice

Adult cardiology < CARDIOLOGY, Paediatric cardiology < CARDIOLOGY, Keywords: PREVENTIVE MEDICINE, Cardiology < INTERNAL MEDICINE

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 14 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 The efficacy and safety of complete pericardial drainage by means of 4 intrapericardial fibrinolysis in the prevention of complications of 5 6 pericardial effusion: a systematic review protocol 7

8 1 2, 3 * 2 4 9 Aloysious Kakia , Charles S. Wiysonge , Eleanor A. Ochodo , Abolade A. Awotedu , 10 11 Arsen Ristic,5 Bongani M. Mayosi6 12 13 14 1 Department of Family Medicine and Rural Health, Walter Sisulu University, Mthatha, South 15 For peer review only 16 Africa; 2 Centre for Evidencebased Health Care, Stellenbosch University, Cape Town, South 17 18 3 19 Africa; Cochrane South Africa, South African Medical Research Council, Cape Town, 20 4 21 South Africa; Department of Medicine, Nelson Mandela Academic Hospital and Walter 22 23 Sisulu University, Mthatha, South Africa; 5Department of Cardiology, Clinical Center of 24 25 Serbia and Belgrade University School of Medicine, Belgrade, Serbia; 6 Department of 26 27 Medicine, Groote Schuur Hospital and University of Cape Town, Cape Town, South Africa 28 29 30 Email addresses: 31 32 Aloysious Kakia: [email protected]; Charles Shey Wiysonge: [email protected]; 33 http://bmjopen.bmj.com/ 34 Eleanor A. Ochodo: [email protected]; Abolade A. Awotedu: [email protected]; 35 36 Arsen Ristic: [email protected]; Bongani M. Mayosi: [email protected] 37 38 39 40 41 *Corresponding author. 42 on September 30, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 14 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 Abstract 4 5 Introduction: 6 Intrapericardial fibrinolysis has been proposed as a means of preventing complications of 7 8 pericardial effusion such as cardiac tamponade, persistent and recurrent pericardial effusion, 9 and pericardial constriction. There is a need to understand the efficacy and safety of this 10 procedure because it shows promise 11 12 Methods and Analysis: 13 We aim to assess the effects of intrapericardial fibrinolysis in the treatment of pericardial 14 effusion. We will search PubMed, the Cochrane Library, African Journals online, 15 Cumulative IndexFor to Nursing peer and Allied reviewHealth Literature, Trip only database, Clinical trials.gov, 16 17 and the WHO International Clinical Trials Registry Platform for studies that evaluate the 18 efficacy and/or safety of complete pericardial fluid drainage by intrapericardial fibrinolysis 19 irrespective of study design, geographical location, language, age of participants, aetiology of 20 pericarditis, or types of fibrinolytics. Two authors will do the search independently, screen 21 the search outputs for potentially eligible studies, and assess whether the studies meet the 22 inclusion criteria. Discrepancies between the two authors will be resolved through discussion 23 and arbitration by a thirdauthor. Data from the selected studies shall be extracted using a 24 standardised data collection form which will be piloted before use. The methodological 25 quality of studies will be assessed using the Cochrane Collaboration’s tools for assessing risk 26 27 of bias for experimental studies and nonrandomised studies respectively. The primary meta 28 analysis will use random effects models due to expected interstudy heterogeneity. 29 Dichotomous data will be analysed using relative risk and continuous data using mean 30 differences, both with 95% confidence intervals. 31 32 Ethics and Dissemination: 33

Approval by an ethics committee is not required for this study as it is a protocol for a http://bmjopen.bmj.com/ 34 systematic review of published studies. The results will be disseminated through a conference 35 presentation and peerreviewed publication. 36 37 38 Review registration: PROSPERO, registration number CRD42014015238 39 40 Key words: Pericarditis, complete intrapericardial drainage, intrapericardial fibrinolysis, 41 urokinase, streptokinase, tissue plasminogen activator. 42 on September 30, 2021 by guest. Protected copyright. 43 44 45 Strengths and limitations 46 • The planned review will shed light on the evidence to date regarding the efficacy and 47 safety of intrapericardial fibrinolysis in preventing complications of pericardial effusion, 48 and guide future research on this theme. 49 • This manuscript is prepared according to the recent Preferred Reporting Items for 50 Systematic review and MetaAnalysis Protocols (PRISMAP) Statement. 51 • Unbiased selection of many studies conducted in different settings will strengthen the 52 validity of the review results. 53 54 • The main limitation of the planned review will be the heterogeneity of the settings and 55 designs of included studies. 56 57 58 59 60 2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 14 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 30, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 14 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 Background 4 5 6 The era of human immunedeficiency virus (HIV) has seen an increase in the incidence of 7 1 2 8 pericarditis . The main cause of pericarditis in Africa is tuberculosis . Pericarditis may 9 10 complicate to tamponade in the short term, and chronic effusive pericarditis and constrictive 11 12 pericarditis in the long term. Cardiac tamponade and constrictive pericarditis lead to death if 13 14 not treated. The definitive management for constrictive pericarditis involves pericardiectomy, 15 For peer review only 16 3 17 which is associated with a mortality of up to fourteen percent , and is an expensive 18 19 procedure4. Imazio et al5 have shown that tuberculous and purulent pericarditis are more 20 21 likely to progress to constrictive pericarditis than pericarditis due to other causes. Ntsekhe et 22 23 al6 found a 10.9 % incidence of constrictive pericarditis over a six month period in patients 24 25 26 with pericardial effusions presumed to be tuberculous. These findings highlight the 27 28 importance of efforts to prevent progression of pericarditis to constrictive pericarditis. 29 30 Various strategies have been used to prevent progression of acute pericarditis to constrictive 31 32 pericarditis. Early diagnosis and prompt treatment of pericarditis, including treating the 33 34 http://bmjopen.bmj.com/ underlying cause and draining effusions, are a major step in this direction. The use of 35 36 37 colchicine as adjunctive treatment to prevent recurrent and persistent pericarditis, and thereby 38 39 reducing the risk of constriction, showed promise in a randomized clinical trial conducted by 40 41 Imazio et al7. Corticosteroids have been found to be useful in several trials; however, the 42 on September 30, 2021 by guest. Protected copyright. 43 findings of Mayosi et al8 have shown that corticosteroids could increase the risk of cancers in 44 45 46 patients coinfected with HIV. 47 48 Intrapericardial fibrinolysis has been proposed as a way of stemming the development of 49 50 cardiac tamponade and constriction in patients with effusive pericarditis. The objective of 51 52 fibrinolysis is to target fibrin formation, to optimize evacuation of a thick fluid, and therefore 53 54 to prevent both persistent purulent pericarditis and constrictive pericarditis9. The procedure is 55 56 9 57 also minimally invasive. A clinical review conducted by Augustin et al concluded that 58 59 60 4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 14 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 intrapericardial ﬁbrinolysis may be useful for prevention of constrictive pericarditis. Cui et 4 5 al10 investigated the efficacy of intrapericardial fibrinolysis in preventing constrictive 6 7 pericarditis in patients with infective pericardial effusion, sixty percent of which were of 8 9 10 tuberculous origin. They found that the early employment of fibrinolysis optimized complete 11 12 evacuation of the pericardial effusion, significantly reduced progress to pericardial 13 14 constriction, and was safe. 15 For peer review only 16 In view of the promise held by intrapericardial fibrinolysis, there is currently a need to better 17 18 19 understand the safety and efficacy of the procedure. We propose therefore to conduct a 20 21 systematic review to assess the efficacy and safety of intrapericardial fibrinolysis in the 22 23 prevention of complications of pericardial effusion, such as cardiac tamponade, recurrent or 24 25 persistent effusion, constrictive pericarditis, hospitalisation, and death. 26 27 28 29 30 Objectives 31 32 33 1. To determine whether complete pericardial drainage by intrapericardial fibrinolysis 34 http://bmjopen.bmj.com/ 35 reduces the incidence of cardiac tamponade, persistent or recurrent pericardial effusion, 36 37 constrictive pericarditis, hospitalisation and death in patients with pericardial effusion. 38 39 2. To determine whether complete pericardial drainage by intrapericardial fibrinolysis can 40 41 42 be performed safely with respect to the incidence of haemorrhage, procedurerelated on September 30, 2021 by guest. Protected copyright. 43 44 cardiac tamponade, allergy, and serious and nonserious adverse events. 45 46 3. To determine the appropriate timing, dose and volume of intrapericardial fibrinolysis. 47 48 49 50 51 52 53 54 55 56 57 58 59 60 5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 14 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 Methods 4 5 6 Types of studies 7 8 We will consider primary studies with the following designs: 9 10 ▸ Intervention studies: randomised controlled trials (RCTs), and quasiRCTs. 11 12 13 ▸ Observational studies: case reports, cohort studies, case–control studies and crosssectional 14 15 studies. For peer review only 16 17 Types of participants 18 19 People of all ages requiring intrapericardial fibrinolysis for evacuation of pericardial effusion 20 21 22 due to any cause. 23 24 Study settings 25 26 We will include studies that evaluate the efficacy and/or safety of intrapericardial fibrinolysis 27 28 for preventing constrictive pericarditis irrespective of geographical location. 29 30 31 Types of interventions 32 33 All types of fibrinolytics will be considered including (but not limited to) urokinase, 34 http://bmjopen.bmj.com/ 35 streptokinase, and tissue plasminogen activator. 36 37 Types of outcome measures 38 39 The efficacy outcomes of interest to this review are cardiac tamponade, persistent or recurrent 40 41 42 pericardial effusion, constrictive pericarditis, hospitalisation and death in patients with on September 30, 2021 by guest. Protected copyright. 43 44 pericardial effusion. 45 46 The safety outcomes will be the incidence of haemorrhage, procedurerelated cardiac 47 48 tamponade, allergy, serious and nonserious adverse events. 49 50 51 Search methods for identification of studies 52 53 We will develop a comprehensive strategy to search for all eligible studies available up to the 54 55 search date, regardless of language or publication status. For published literature we will 56 57 search the electronic databases PubMed, Cochrane Library (Cochrane Central Register of 58 59 60 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 14 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 Controlled Trials, Cochrane Database of Systematic Reviews, Database of Abstracts of 4 5 Reviews of Effects (DARE)), African Journals online (AJOL), Cumulative Index to Nursing 6 7 and Allied Health Literature (CINAHL), and Trip database. We will use a combination of 8 9 10 the following search terms and tailor them appropriately to the different databases: 11 12 ‘Pericarditis’, ‘tuberculous pericarditis,’ ‘purulent pericarditis,’ ‘pericardiocentesis’, 13 14 ‘therapeutic pericardiocentesis,’ ‘fibrinolytics,’ ‘intrapericardial fibrinolytics,’ ‘urokinase,’ 15 For peer review only 16 ‘streptokinase,’ and ‘tissue plasminogen activator.’ Box 1 below gives the provisional search 17 18 19 strategy for PubMed, which will be adapted for each electronic database. To avoid selection 20 21 bias, two authors will do the search independently. To access unpublished literature, we will 22 23 contact experts in the field of therapeutic pericardiocentesis and search Clinical trials.gov and 24 25 the WHO International Clinical Trials Registry Platform. 26 27 28 Box 1. Provisional search strategy for PubMed 29 30 (((((((pericardiocentesis) OR "pericardial drainage") OR "intrapericardial fibrinolysis") 31 32 OR fibrinolysis)) OR (((("tissue plasminogen activator") OR urokinase) OR 33 http://bmjopen.bmj.com/ 34 streptokinase) OR fibrinolytics))) AND ((((((((((((pericarditis) OR "tuberculous 35 36 pericarditis") OR "pericardial effusion") OR "TB pericarditis") OR "purulent 37 38 39 pericarditis") OR "complicated pericarditis") OR "complications of pericarditis") OR 40 41 "complications of pericardial effusions") OR "constrictive pericarditis") OR "recurrent 42 on September 30, 2021 by guest. Protected copyright. 43 pericardial effusions") OR "persistent pericarditis") OR "cardiac tamponade") 44 45 46 47 48 Data collection and analysis 49 50 Two authors will independently screen the search outputs for potentially eligible studies, 51 52 compare their results, and resolve disagreements by discussion and consensus. The two 53 54 authors will then independently go through the full text of all potentially eligible studies to 55 56 assess whether the studies meet the inclusion criteria deﬁned by the study design, setting, 57 58 59 60 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 14 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 intervention and outcomes. Discrepancies in the list of eligible studies between the two 4 5 authors will be resolved through discussion and consensus. A structured and standardised 6 7 data collection form shall be developed for extracting data from the selected studies. The 8 9 10 form will capture key study characteristics, including study design, participants, methods 11 12 used for diagnosis of pericardial effusion (e.g. echocardiography), aetiology of effusions, 13 14 interventions, risk of bias, and outcomes. Prior to use, the extraction form will be piloted on 15 For peer review only 16 at least three studies identified randomly from the list of included studies. 17 18 19 The methodological quality of studies will be assessed using the Cochrane Collaboration’s 20 11 21 tool for assessing risk of bias for experimental studies and the “Cochrane risk of bias 22 23 assessment tool for nonrandomised studies of interventions” for other study designs.12 24 25 All eligible studies will be summarized and analyzed using the Cochrane Review Manager 26 27 software11. Two authors will extract the data, one author will enter the data and the second 28 29 30 author will recheck the entries. In the event of discrepancy, the authors shall discuss and 31 32 resolve the disagreement by discussion and consensus, and if this fails to resolve the 33 http://bmjopen.bmj.com/ 34 disagreement a third author will arbitrate. If the studies are sufficiently similar, we will 35 36 combine the data using the random effects model. We will examine statistical heterogeneity 37 38 2 39 between study results using the chisquared (χ ) test of homogeneity (with a significance α 40 41 level of 0.1). We shall quantify statistical heterogeneity between study results using the 42 on September 30, 2021 by guest. Protected copyright. 43 inconsistency index (I2)13, 14.When studies cannot be combined for metaanalysis due to 44 45 diversity of interventions, narrative syntheses will be conducted. 46 47 We will stratify analysis by aetiology of pericardial effusion (e.g. tuberculous, bacterial), type 48 49 50 of pericarditis (e.g. effusive, effusive constrictive), modality for diagnosis of pericardial 51 52 effusions and constriction (e.g. use of echography, echography not used), and study design 53 54 (e.g. controlled trials, observational studies). For any metaanalysis involving 10 or more 55 56 57 58 59 60 8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 14 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 studies, we will use funnel plots to assess the possibility of publication bias. In addition, we 4 5 will apply the GRADE system to assess the strength of the evidence from the review15. 6 7 Reporting of protocol and systematic review 8 9 10 We plan to report the findings of the review as recommended in the Preferred Reporting 11 16 12 Items for Systematic reviews and MetaAnalyses (PRISMA) guidelines . In addition, we 13 14 prepared the review protocol according to the Preferred Reporting Items for Systematic 15 For peer review only 16 review and MetaAnalysis Protocols (PRISMAP) Statement17. 17 18 19 20 21 Ethics and dissemination 22 23 The planned systematic review is registered with the International Prospective Register of 24 25 26 Systematic Reviews (PROSPERO), registration number CRD42014015238. Systematic 27 28 reviews draw on data available in the public domain, and do not need formal ethical review 29 30 and approval. The findings of this systematic review will be disseminated through peer 31 32 reviewed journal publications and conference presentations. To our knowledge no systematic 33 34 http://bmjopen.bmj.com/ review on intrapericardial fibrinolysis for the prevention of complications of pericardial 35 36 37 effusion has been done to date. Our discussion of the findings shall be in light of the 38 39 relevance of these data in clinical decision making and future research design and direction 40 41 on this topic. 42 on September 30, 2021 by guest. Protected copyright. 43 44 45 46 Acknowledgements 47 48 The authors did not receive any external funding for this manuscript, which was written 49 50 during their routine work in their respective institutions. Neither the authors’ institutions nor 51 52 53 any funder or sponsor played a role in preparing the manuscript. The authors acknowledge 54 55 the Editor and peer reviewers for critical and constructive comments, which helped to 56 57 improve earlier versions of the manuscript. 58 59 60 9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 14 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 Contributors 4 5 6 Bongani Mayosi conceived the study, and Aloysious Kakia and Eleanor Ochodo wrote the 7 8 first draft of the protocol. Aloysious Kakia, Charles Wiysonge, Eleanor Ochodo, Abolade 9 10 Awotedu, Arsen Ristic, and Bongani Mayosi critically revised successive drafts of the 11 12 manuscript and approved the final version for publication. Charles Wiysonge prepared the 13 14 final version and is the guarantor of the manuscript. 15 For peer review only 16 17 Competing interests 18 19 None declaredNo, there are no competing interests. 20 21 22 Provenance and peer review 23 24 Not commissioned; externally peer reviewed. 25 26 27 Open Access 28 29 This is an Open Access article distributed in accordance with the Creative Commons 30 31 Attribution Non Commercial (CC BYNC 4.0) license, which permits others to distribute, 32 33 remix, adapt, build upon this work noncommercially, and license their derivative works on http://bmjopen.bmj.com/ 34 35 36 different terms, provided the original work is properly cited and the use is noncommercial. 37 38 See: http://creativecommons.org/licenses/bync/4.0 39 40 41 42 on September 30, 2021 by guest. Protected copyright. 43 44 45 References 46 47 1. Mayosi BM, Burgess LJ, Doubell AF. Tuberculous pericarditis. Circulation 48 49 2005;112:360816 50 51 52 2. Mayosi BM. Contemporary trends in the epidemiology and management of 53 54 cardiomyopathy and pericarditis in subSaharan Africa. Heart 2007; 93:117683. 55 56 57 58 59 60 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 14 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 3. Mutyaba A, Balkaran S, Cloete R, et al. Constrictive pericarditis requiring 4 5 pericardiectomy at Groote Schuur Hospital in Cape Town, South Africa: causes and peri 6 7 operative outcomes in the HIV era (1990–2012). J Thorac Cardiovasc Surg (in press). 8 9 10 4. Mayosi BM, Ntsekhe M, Volmink JA, Commerford PJ. Interventions for treating 11 12 tuberculous pericarditis. Cochrane Database Syst Rev 2002;4:CD000526. 13 14 5. Imazio M, Brucato A, Maestroni S, et al. Risk of constrictive pericarditis after acute 15 For peer review only 16 pericarditis. Circulation 2011;124:127075. 17 18 19 6. Ntsekhe M, Wiysonge CS, Gumedze F, et al. HIV infection is associated with a lower 20 21 incidence of constriction in presumed tuberculous pericarditis: a prospective 22 23 observational study. PLoS ONE 2008; 3(6): e2253. 24 25 7. Imazio M, Brucato A, Cemin R, et al. Colchicine for recurrent pericarditis: a randomized 26 27 trial. Annals of Internal Medicine 2011; 155:409414. 28 29 30 8. Mayosi BM, Ntsekhe M, Bosch J, et al. Prednisolone and Mycobacterium indicus pranii 31 32 in tuberculous pericarditis. N Engl J Med 2014; 371: 112130. 33 http://bmjopen.bmj.com/ 34 9. Augustin P, Desmard M, Mordant P, et al. Clinical review: intrapericardial fibrinolysis in 35 36 management of purulent pericarditis. Crit Care 2011;15 (2):220. 37 38 39 10. Cui HB, Chen XY, Cui CC, et al: Prevention of pericardial constriction by transcatheter 40 41 intrapericardial fibrinolysis with urokinase. Chin Med Sci J 2005, 20:510 42 on September 30, 2021 by guest. Protected copyright. 43 11. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of 44 45 Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, Oxford, 46 47 UK, 2011. 48 49 50 12. Sterne JAC, Higgins JPT, Reeves BC on behalf of the development group for 51 52 ACROBATNRSI. A Cochrane Risk Of Bias Assessment Tool: for NonRandomized 53 54 Studies of Interventions (ACROBATNRSI), Version 1.0.0, 24 September 2014. 55 56 Available from http://www.riskofbias.info (accessed 28 September 2015). 57 58 59 60 11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 14 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 13. Hardy RJ, Thompson SG. Detecting and describing heterogeneity in metaanalysis. Stat 4 5 Med1998; 17: 841–856. 6 7 14. Higgins JP, Thompsom SG, Deeks JJ, Altman DG. Measuring inconsistency in meta 8 9 10 analysis. BMJ 2003; 327: 557–560. 11 12 15. Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. GRADE guidelines: 1. 13 14 IntroductionGRADE evidence profiles and summary of findings tables. J Clin Epidemiol 15 For peer review only 16 2011;64:383e94. 17 18 19 16. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic 20 21 reviews and metaanalyses of studies that evaluate healthcare interventions: explanation 22 23 and elaboration. BMJ 2009;339:b2700 24 25 17. Shamseer L, Moher D, Clarke M, et al; PRISMAP Group. Preferred reporting items for 26 27 systematic review and metaanalysis protocols (PRISMAP) 2015: elaboration and 28 29 30 explanation. BMJ 2015;349:g7647. 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 30, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 14 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 PRISMA-P (preferred reporting items for systematic review and meta-analysis protocols) 2015 4 checklist: recommended items to address in a systematic review protocol 5 6 Manuscript: BMJOpen-2015-007842.R1 7 The efficacy and safety of complete pericardial drainage by means of intrapericardial fibrinolysis in 8 the prevention of complications of pericardial effusion: a systematic review protocol 9 10 Section and topic Item No Checklist item Page No 11 Administrative information 12 Title: 13 Identification 1a Identify the report as a protocol of a systematic review 1 14 Update 1b If the protocol is for an update of a previous systematic N/A 15 For peerreview, identifyreview as such only 16 Registration 2 If registered, provide the name of the registry (such as 2 & 8 17 PROSPERO) and registration number 18 Authors: 19 20 Contact 3a Provide name, institutional affiliation, email address of all 1 21 protocol authors; provide physical mailing address of 22 corresponding author 23 Contributions 3b Describe contributions of protocol authors and identify the 9 24 guarantor of the review 25 Amendments 4 If the protocol represents an amendment of a previously N/A 26 completed or published protocol, identify as such and list 27 changes; otherwise, state plan for documenting important 28 protocol amendments 29 Support: 30 Sources 5a Indicate sources of financial or other support for the review 8 31 Sponsor 5b Provide name for the review funder and/or sponsor 8 32 Role of sponsor or 5c Describe roles of funder(s), sponsor(s), and/or 8 33 funder institution(s), if any, in developing the protocol http://bmjopen.bmj.com/ 34 Introduction 35 36 Rationale 6 Describe the rationale for the review in the context of what 34 37 is already known 38 Objectives 7 Provide an explicit statement of the question(s) the review 4 39 will address with reference to participants, interventions, 40 comparators, and outcomes (PICO) 41 Methods 42 Eligibility criteria 8 Specify the study characteristics (such as PICO, study 5 on September 30, 2021 by guest. Protected copyright. 43 design, setting, time frame) and report characteristics (such 44 as years considered, language, publication status) to be 45 used as criteria for eligibility for the review 46 Information sources 9 Describe all intended information sources (such as 56 47 electronic databases, contact with study authors, trial 48 registers or other grey literature sources) with planned 49 dates of coverage 50 Search strategy 10 Present draft of search strategy to be used for at least one 6 51 electronic database, including planned limits, such that it 52 could be repeated 53 Study records 54 Data management 11a Describe the mechanism(s) that will be used to manage 67 55 records and data throughout the review 56 57 Selection process 11b State the process that will be used for selecting studies 67 58 (such as two independent reviewers) through each phase 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 14 BMJ Open: first published as 10.1136/bmjopen-2015-007842 on 5 January 2016. Downloaded from

1 2 3 of the review (that is, screening, eligibility and inclusion in 4 metaanalysis) 5 Data collection process 11c Describe planned method of extracting data from reports 67 6 (such as piloting forms, done independently, in duplicate), 7 any processes for obtaining and confirming data from 8 investigators 9 Data items 12 List and define all variables for which data will be sought 7 10 (such as PICO items, funding sources), any preplanned 11 data assumptions and simplifications 12 Outcomes and 13 List and define all outcomes for which data will be sought, 5 and 7 13 prioritization including prioritization of main and additional outcomes, 14 with rationale 15 Risk of bias in individualFor 14 peer Describe review anticipated methods for assessingonly risk of bias of 7 16 studies individual studies, including whether this will be done at the 17 outcome or study level, or both; state how this information 18 will be used in data synthesis 19 Data synthesis 15a Describe criteria under which study data will be 7 20 quantitatively synthesised 21 22 15b If data are appropriate for quantitative synthesis, describe 7 23 planned summary measures, methods of handling data 24 and methods of combining data from studies, including any 2 25 planned exploration of consistency (such as I , Kendall’s τ) 26 15c Describe any proposed additional analyses (such as 7 27 sensitivity or subgroup analyses, metaregression) 28 15d If quantitative synthesis is not appropriate, describe the 7 29 type of summary planned 30 Metabias(es) 16 Specify any planned assessment of metabias(es) (such as 78 31 publication bias across studies, selective reporting within 32 studies) 33 Confidence in 17 Describe how the strength of the body of evidence will be 8 34 cumulative evidence assessed (such as GRADE) http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 30, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml