Protocol for a Randomized Controlled Trial

BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.

When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.

The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.

BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com).

If you have any questions on BMJ Open’s open peer review process please email [email protected] http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from

Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP): protocol for a randomized controlled trial

Journal: BMJ Open ManuscriptFor ID peerbmjopen-2018-028664 review only Article Type: Protocol

Date Submitted by the 28-Dec-2018 Author:

Complete List of Authors: Liu, Songqiao; Department of Critical Care Medicine, Zhongda Hospital, School of Medicine, Southeast University Yao, Chen ; Peking University First Hospital, Medical Statistics Office Zhang, Junhua; Tianjin University of Traditional Chinese Medicine, evidence based medicine; evidence based medicine Yang, Yi; Department of Critical Care Medicine, Zhongda Hospital, School of Medicine, Southeast University Qiu, Haibo; Department of Critical Care Medicine, Zhongda Hospital, School of Medicine, Southeast University Investigators, The EXIT-SEP; Zhongda Hospital, School of Medicine, Southeast University, Department of Critical Care Medicine

Keywords: Xuebijing injection, Sepsis, randomized, controlled trial, protocol

http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 23 BMJ Open

1 2 3

4 Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP): protocol for a BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 randomized controlled trial 8 9 10 11 Songqiao Liu1, Chen Yao2, Junhua Zhang3，Yi Yang1 and Haibo Qiu1 and the EXIT-SEP Investigators 12 13 14 15 Correspondence to Dr Haibo Qiu; [email protected], Department of Critical Care Medicine, 16 17 Zhongda Hospital, School of Medicine, Southeast University, Nanjing, Jiangsu, 210009, China 18 For peer review only 19 20 21 Author affiliations 22 23 1 Department of Critical Care Medicine, Zhongda Hospital, School of Medicine, Southeast 24 25 University, Nanjing, Jiangsu, 210009, China. 26 27 2Medical Statistics Office, Peking University First Hospital, Beijing, China, 100034. 28 29 3 Centre for Evidence-based Medicine, Tianjin University of Traditional Chinese Medicine, Tianjin, 30 31 China, 301617. 32 33 34 35 Word count: 2406 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 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 23

1 2 3 ABSTRACT

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Introduction Sepsis is a major challenge with high incidence and is associated with high mortality. 7 8 Current management of sepsis remains mainly supportive except for treatment with antibiotics. 9 10 Both basic research and clinical investigation have shown that the Chinese herbal derived 11 therapeutic Xuebijing (XBJ) injection is beneficial for sepsis. However, the quality of evidence 12 13 supporting the therapeutic use of XBJ in the ICU is limited. The aim of this trial is to evaluate the 14 15 efficacy of XBJ, compared with a placebo, on the outcome of patients with sepsis in the ICU. 16 17 Methods and analysis In this multicentre, blinded, randomized, controlled trial, we are recruiting a 18 For peer review only 19 total of 1800 subjects meeting the criteria within The Third International Consensus Definitions for 20 21 Sepsis and Septic Shock (Sepsis 3.0). Subjects will be randomized (1:1) to receive XBJ, q12h for 5 22 23 days or a matching placebo and usual care. The primary outcome is 28 days all-cause mortality. 24 25 Secondary outcomes will be the improvement of Sequential Organ Failure Assessment (SOFA) 26 27 scores, the improvement of the Acute Physiology and Chronic Health Evaluation II (APACHE II) 28 29 score, duration of mechanical ventilation, mortality in ICU and duration of stay in the ICU. 30 31 Investigators, participants and statisticians will be blinded to the allocated treatment. 32 33 Ethics and dissemination This trial has been approved by Ethics Committees of all of the centres that 34 35 will participate in this trial. The findings of the study will be disseminated in peer-reviewed journals and 36 37 at conferences. Once this study is complete, the results of this trial may help to provide evidence-based http://bmjopen.bmj.com/ 38 39 recommendations for complementary therapeutic options for patients with sepsis. 40 41 Trial registration number This clinical trial was registered at Clinicaltrials.gov (NCT03238742) and 42 43 the Chinese Clinical Trial Registry (ChiCTR-IPR-17012713). 44 45 Protocol date: 10 May 2017. on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 KEYWORDS 51 52 Xuebijing injection; Sepsis; randomized, controlled trial; protocol 53 54 55 56 57 58 59 60 2

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 23 BMJ Open

1 2 3 Strengths and limitations of this study

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6  The EXIT-SEP trial is a multicentre, blind, randomized, controlled study including a wide 7 8 representation of the Chinese population, allowing for inferences regarding 9 10 epidemiology, management and outcomes of sepsis patients in the entire country. 11  The study will have a sample size large enough to provide high-quality evidence to 12 13 evaluate the potential benefit of XBJ injection on outcomes in patients with sepsis. 14 15  One limitation of the study is the blinding. The independent drug administrators and a 16 17 minority of the nurses were able to determine which patients were in the treatment 18 For peer review only 19 arm; however, the investigators, patients, and statistician were blinded. Therefore, 20 21 satisfactory levels of blinding in a multicentre context regarding treatment arms was 22 23 achieved. 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 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 23

1 2 3 INTRODUCTION

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Sepsis is a common and devastating condition that is responsible for more than 6 million deaths 7 1 2 8 annually worldwide and is the leading cause of death in critically ill patients . Half of all sepsis 9 10 patients are treated in the intensive care unit (ICU), representing 10% of all ICU admissions. The 11 World Health Organization (WHO) reports that sepsis affects more than 30 million people 12 13 worldwide every year. The burden of sepsis is most likely highest in low- and middle-income 14 15 countries3 4. 16 17 The pathogenesis of sepsis includes an overwhelming inflammatory response, endothelial 18 For peer review only 19 injury, altered coagulation, initiation of innate immunity and sepsis-associated 20 21 immunosuppression, resulting in life-threatening organ dysfunction. Despite the more than 100 22 23 therapeutic clinical trials that have been conducted, no FDA-approved treatment options are 24 25 currently recommended for sepsis5. Although there are improvements in supportive care 26 27 strategies, mortality still remains unacceptably high. During the past 30 years, an understanding 28 29 of the complex biological mechanisms of sepsis has not been fully explored, nor have promising 30 31 therapies been developed6. Some adjuvant treatments for sepsis have been developed at great 32 33 expense but have failed to significantly improve mortality 7 8. Until recently, research on sepsis 34 35 therapies has resulted in failed clinical trials 9. 36 37 There is accumulating evidence to suggest that Xuebijing (XBJ) injection has the potential to http://bmjopen.bmj.com/ 38 39 improve the outcomes for critically ill patients with sepsis. XBJ, a Chinese herb-derived therapeutic, 40 41 has been approved for sepsis treatment in critically ill patients (China Food and Drug 42 43 Administration; Beijing, China, No. Z20040033). The main components of XBJ include amino acids, 44 45 phenolic acids, flavonoid glycoside, terpene glycoside, and phthalides. XBJ has effects on at least on October 1, 2021 by guest. Protected copyright. 46 47 ten sepsis/inflammation pathways and its 21 major active ingredients regulate 550 targets (HRAS,

48 10 49 GSK3B, BTK, AK, et al.) . Basic research of XBJ suggests that XBJ has satisfactory 50 immunomodulatory, anti-coagulation and anti-inflammatory activity. Animal studies and ex vivo 51 52 studies of XBJ on sepsis have suggested that the effects of XBJ inhibit the expression of HMGB1, 53 54 inhibit proinflammatory cytokines secretion and ameliorate sepsis-induced lung injury 11-14. 55 56 Clinical trials have demonstrated that XBJ improved outcomes and a meta-analysis showed 57 58 that XBJ was associated with high survival rate (RR 0.62, 95% CI 0.51-0.76 P < 0.000 01, I2 = 0%) 15 59 60 4

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 23 BMJ Open

1 2 3 16. These clinical trials, however, were conducted with small sample sizes, even when pooled,

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 17 18 19 20 6 limiting the evidence of potential benefit in patients with sepsis . We hypothesize that XBJ 7 8 will reduce all-cause mortality than the placebo among sepsis patients. Thus, we aim to conduct a 9 10 large-scale, blind, randomized controlled trial (RCT) to evaluate the effects of XBJ in adult sepsis 11 patients in the ICU. 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 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 23

1 2 3 METHODS

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 Design 6 7 This study was designed to be a multicentre, blind, randomized and placebo-controlled trial. 8 9 Eligible participants will be randomly assigned to either the XBJ group or the placebo group in a 10 11 ratio of 1:1. The participants will receive XBJ (XBJ group) or the solvent (placebo group) within 24 12 13 h of enrolment for 5 days. Figure 1 shows the design of the study. This study protocol follows the 14 15 Standard Protocol Items: Recommendations for Interventional Trials statement 16 22 17 recommendations . The items from the trial registration data set are described in the online 18 For peer review only 19 supplementary file 1. This study protocol was registered at Clinicaltrials.gov (NCT03238742) and 20 21 at the Clinical Trial Registry (ChiCTR) before randomization (ChiCTR-IPR-17012713). 22 23 24 Population 25 26 Participant recruitment is currently ongoing at Intensive Care Units (ICUs) from the 45 medical 27 28 centres in China. Patients who fulfil the inclusion criteria and who sign informed consent forms 29 30 will enter the screening period. Patients with sepsis who meet the exclusion criteria will be 31 32 excluded before randomization. The recruitment duration will last for 24 months, from October 33 34 2017 to September 2019. 35 36 37 http://bmjopen.bmj.com/ 38 Inclusion criteria 39 40 Patients are eligible for the trial if they are ≥18 and ≤75 years old, meet the criteria of sepsis 3.0 41 42 (The Third International Consensus Definition for Sepsis and Septic Shock) 21, and have a SOFA 43 44 score of 2-13. 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 Exclusion criteria 49 50 Patients will be excluded if they fulfil any of the exclusion criteria: 51 52 1. Diagnosis of sepsis for more than 48 h; 53 54 2. Pregnant and lactating women; 55 56 3. Severe primary disease including unrespectable tumours, blood diseases and Human 57 58 Immunodeficiency Virus (HIV); 59 60 4. Severe liver and kidney dysfunction (single liver or kidney SOFA score ≥ 3 points); 6

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 23 BMJ Open

1 2 3 5. Use of an immunosuppressant or having an organ transplant within the previous 6 months;

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 6. Participating in other clinical trials in the previous 30 days. 7 8 9 Ethics and informed consent 10 11 This trial is approved by the Research Ethics Boards at Zhongda Hospital (File number: 12 13 2017ZDSYLL025-P01). The research ethics committees of all participant centres approved the 14 15 study protocol(supplementary-material-1). The patients’ written informed consent is required at 16 17 all participating centres. Site investigators will be responsible for obtaining informed consent from 18 For peer review only 19 study participants or their families. Subject confidentiality will be assured through data 20 21 anonymization and controlled access to case report forms, the electronic data capture system and 22 23 datasets. Any breaches of confidentiality, study protocol or AEs attributable to this study will be 24 25 reported to the research ethics committees. 26 27 28 29 Randomization and allocation concealment 30 31 Randomization is generated centrally using an interactive web response system (IWRS). When a 32 33 subcentre accepts an eligible participant, the drug administrators will log into the central 34 35 randomization system, where a random number is produced. The independent drug 36 37 administrators receive group information based on the random number, and then they assign the http://bmjopen.bmj.com/ 38 39 study drug to the nurses to administer. 40 41 42 43 Interventions 44 45 The participants will receive XBJ (normal saline 100 ml + XBJ 100 ml, q12h, intravenous infusion on October 1, 2021 by guest. Protected copyright. 46 47 for 80 min) in the XBJ group and the solvent only (normal saline, 200 ml, q12h) in the placebo 48 49 group. XBJ specifications include a 10 ml/ampule, packaged in 5 ampules/container. They were 50 manufactured by a Good Manufacturing Practice (GMP) certified company in China (Tianjin Chase 51 52 Sun Pharmaceutical Co., Ltd., Z20040033). Patients in both groups will receive standard care by 53 54 the attending physician according to the International Guidelines for Management of Sepsis and 55 56 Septic Shock. The patients will receive the study drug within 24 h of enrolment, for a 5 day 57 58 treatment period. According to the 2016 guidelines on sepsis /sepsis shock, the routine care should 59 60 7

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 23

1 2 3 include early fluid resuscitation, antimicrobials anticoagulants, nutritional support and other

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 treatment measures. Ulinastatin is not allowed during the study period. Use of all drugs, if any, 7 8 should be documented in the Case Report Form (CRF). 9 10 11 Blinding 12 13 Photophobic brown colour infusion bags and infusion devices for both groups will be visually 14 15 inspected by the pharmacy to ensure identical appearance. The independent drug administrators 16 17 will receive group information based on a random number; they will then assign the study drug to 18 For peer review only 19 the nurses to administer. The mixture of drugs will be prepared by the nurses in a separate room. 20 21 This mixture of drugs will be the standard procedure for XBJ, and all the nurses will have experience 22 23 in performing this task. The nurses will also sign a confidentiality agreement about patient 24 25 allocation. The participants, as well as all the members of the study and the health care team, 26 27 outcome assessors, will be blinded to the study drug assignment. Data analysis will be performed 28 29 by a researcher who is blinded to patient allocation. 30 31 32 33 Outcomes 34 35 The primary outcome will be 28-day all-cause mortality. Secondary outcomes will be the 36 37 improvement of Sequential Organ Failure Assessment (SOFA) scores, the improvement of the http://bmjopen.bmj.com/ 38 39 Acute Physiology and Chronic Health Evaluation II (APACHE II) score, duration of mechanical 40 41 ventilation, mortality in the Intensive Care Unit (ICU) and duration of stay in the ICU. 42 43 44 45 Follow-up on October 1, 2021 by guest. Protected copyright. 46 47 Data will be recorded during the follow-up period according to the multiple time-points. The 48 49 details are shown in Table 1. 50 1. Screening period (day 0): 24 h before recruitment; 51 52 2. Intervention period (day 1-5): data will be recorded every day during follow-up; 53 54 3. Period after intervention (within 90 days after treatment): follow-up at day 6 and day 28 and 55 56 day 90. If the patient is discharged, we will contact them via telephone or a short messaging service. 57 58 59 60 8

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 23 BMJ Open

1 2 3 Sample size calculation

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 We determined that an enrolment of 1800 participants would provide the trial with 80% power to 7 8 detect an absolute difference of 6 percentage points in 28-day all-cause mortality of 24.3 % with 9 10 a two-sided P value of 0.05. This calculation allowed for a rate of withdrawal and loss to follow-up 11 of 15%. We estimated the 28 days mortality rate of 24.3% in a previous observational CHESS study 12 13 of 2322 sepsis/septic shock patients with SOFA score of 2-13 in China. 14 15 16 17 Study organization 18 For peer review only 19 The design and implementation of a multicentre study requires collaboration among several 20 21 organizations. Quality control will be undertaken by these organizations: ① Expert committee: 22 23 comprises clinical experts, statisticians, and quality control experts who will be responsible for 24 25 clinical research methodology and for resolving the key issues in the implementation of the study. 26 27 ②Executive committee: the main staff from the members of the expert committee. The 28 29 responsibilities of the executive committee include the design of the clinical research trial, 30 31 selecting the cooperative hospitals, and providing a training course including a manual of 32 33 instructions. ③ DSMB: an independent team that evaluates the safety outcomes and AEs and 34 35 submits a review proposal. ④ Data management and statistical analysis: performed by trained 36 37 staff and statisticians. ⑤ Quality control: sites and researchers will be monitored and inspected http://bmjopen.bmj.com/ 38 39 regularly, following the standard protocol throughout the process. This trial will undergo 40 41 normative monitoring and inspection throughout the entire process. 42 43 44 45 Data management on October 1, 2021 by guest. Protected copyright. 46 47 Qualified sites and researchers are crucial factors that ensure the quality of a clinical trial. The sub- 48 49 centres should be qualified by the “Good Clinical Practice (GCP) training” of the State Food and 50 Drug Administration (SFDA) for compliance in the training. The qualified researchers should 51 52 understand the detailed contents of the protocol. The data filled in the CRF by researchers should 53 54 be accurate, complete, timely, and reliable. 55 56 Data management will be performed by trained staff at each participating centre using the 57 58 electronic data system. The quality of the trial data management will be guaranteed by the 59 60 9

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 23

1 2 3 reliability, access control, and traceability of the system. Data collection will be restricted to those

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 variables necessary to define baseline patient characteristics (demographics, sepsis diagnosis, 7 8 concurrent medical conditions and comorbidities, inclusion and exclusion criteria, severity of 9 10 illness and organ dysfunction scores, and laboratory results), the delivery of the study drugs, 11 potential confounding co-interventions, and outcomes. Randomized participants will be followed 12 13 up until either death or 28 days after randomization, whichever comes first. Follow-up will be 14 15 conducted by study staff by either direct contact with the patient or their next of kin. Participants 16 17 who withdraw from the study for any reason will be followed up with according to the study 18 For peer review only 19 follow-up schedule and data will be analysed according to the intention to treat (ITT) principle. 20 21 22 23 Data analysis 24 25 All analyses will be performed according to the ITT principle. Continuous variables will be reported 26 27 as the means and standard deviations or as medians and interquartile ranges according to data 28 29 distribution. Categorical variables will be reported as proportions. We will compare the data on 30 31 primary outcome of 28-day all-cause mortality using an unadjusted chi-square test for 32 33 proportions or a logistic regression model. We will report frequency (percentage) per treatment 34 35 group with a risk difference and 95% confidence interval and a corresponding odds ratio and 95% 36 37 confidence interval. The missing data of the primary outcome are filled by the worst imputation http://bmjopen.bmj.com/ 38 39 method, that is, "death" is used to fill the missing data. The secondary binary and continuous 40 41 outcomes will be analysed with the use of logistic regression and linear regression, respectively. 42 43 The rate of death in a time-to-event analysis will be reported with the use of Kaplan–Meier plots, 44 45 and differences in survival will be tested using a Cox proportional-hazards model that includes the on October 1, 2021 by guest. Protected copyright. 46 47 group variable. All statistical analyses will be performed using SAS version 9.4 (SAS Institute Inc) 48 49 with a 2-sided P value of less than .05 considered significant. No adjustment will be made for 50 multiple comparisons; therefore, secondary outcomes will be interpreted as exploratory. 51 52 53 54 Patient and public involvement 55 56 No patient or public were involved in the present study. Upon the completion of this trial, a journal 57 58 article manuscript will be prepared to present the trial results. The burden of intervention will not 59 60 10

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 23 BMJ Open

1 2 3 be assessed by trial participants.

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 8 DISCUSSION 9 10 This RCT aims to evaluate the efficacy of XBJ to reduce mortality for sepsis patients. XBJ has been 11 evaluated as a promising Chinese herbal derived therapeutic drug in the management of sepsis, 12 13 by reducing the inflammatory mediators such as TNF-α, IL-1, IL-6 and IL-8, improving the immune 14 15 function, and protecting the vascular endothelial cells, which correspond to the major 16 17 pathogenesis of sepsis23 24. A meta-analysis showed that XBJ reduced all-cause mortality15. 18 For peer review only 19 However, current clinical evidence supporting the efficacy of XBJ in the treatment of sepsis, 20 21 including RCTs, is subject to limitations such as a non-representative patient population, small 22 23 sample size, and concomitant use of immunomodulatory agents. As a result, the effect of XBJ on 24 25 the mortality of sepsis warrants validation by large, well conducted RCTs in the ICU. 26 27 This trial uses the most robust outcome of a superiority trial, with 28-day mortality as the 28 29 primary outcome 25-27. At the time of preparation of this study, patients who met the criteria of 30 31 Sepsis 3.0 were reported to have a hospital mortality rate of 20% in developed countries28. 32 33 Previous studies suggest that ICU patients with sepsis in middle- and low- income countries might 34 35 have a much higher mortality rate than those in high-income countries, possibly due to inadequate 36 37 resources and poor quality of care. Using data from our cross section study ‘CHESS’, we estimate http://bmjopen.bmj.com/ 38 39 that 28-day all-cause mortality will be 31.87% in the control group. The ICU mortality rate was 40 41 greater than 60% in patients with sepsis whose SOFA score more than 13. Those patients were too 42 43 severe to detect the benefit of XBJ in ICU. With a sample size of 1800, this large-scale trial will be 44 45 adequately powered to test the study hypothesis. The definition of Sepsis 3.0 focuses on organ on October 1, 2021 by guest. Protected copyright. 46 47 function, so apart from the regular endpoints in clinical trials, we would like to employ more 48 49 outcome measurements for organ function. 50 There are, however, surrogate outcomes that can be influenced by blinding, which can be 51 52 influenced by discharge decisions and safety judgement. Although the independent drug 53 54 administrators and a minority of nurses were able to determine the treatment arm, we 55 56 demonstrated that concealed drug administration achieved satisfactory levels of blinding in a 57 58 multicentre context. In particular, medications in brown infusion bags and tubes to obscure 59 60 11

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 23

1 2 3 content were sufficient to achieve blinding.

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Recruitment will be completed in September 2019 with a final number of 1800 participants 7 8 in 45 medical centres in China. The collection of primary end point data will conclude in October, 9 10 2019. With the results of the EXIT-SEP trial, we hope to be able to make evidence-based 11 recommendations for XBJ use for sepsis. 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 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 23 BMJ Open

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Funding This trial was supported by the Development Center for Medical Science and Technology, 7 8 National Health and Family Planning Commission of the People’s Republic of China [WK-2016-HR- 9 10 03], which had no role in the design of the protocol, the conduction of the trial, or the analyses or 11 reporting of the data. 12 13 Contributors Haibo Qiu designed the study, drafted the manuscript, edited the manuscript, 14 15 supervised the study, and obtained study funding. Songqiao Liu designed the study, drafted and 16 17 edited the manuscript. Yi Yang drafted and edited the manuscript. Chen Yao designed the study 18 For peer review only 19 and reviewed the manuscript. Junhua Zhang edited the manuscript, and provided study 20 21 supervision. All authors read and approved the final manuscript. 22 23 Competing interests None declared. 24 25 Patient consent Obtained. 26 27 Ethics approval This study has been approved by Ethics Committee of Zhongda hospital, Southeast 28 29 University (Reference number: 2017ZDSYLL025-P01) on the 17 May 2017, and all participants 30 31 provided written informed consent. After the study is completed, the analysis results will be 32 33 published; however, no personal information will be included in the findings. 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 13

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 23

1 2 3 REFERENCES

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 1. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, et al. Surviving 7 8 9 Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 10 11 2016. Critical Care Medicine 2017;45(3):486-552. 12 13 14 2. Singer M, Deutschman CS, Seymour CW, Shankarhari M, Annane D, Bauer M, et al. 15 16 The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). 17 18 For peer review only 19 Jama 2016;315(8):775-87. 20 21 22 3. Fleischmann C, Scherag A, Adhikari NK, Hartog CS, Tsaganos T, Schlattmann P, et al. 23 24 Global burden of sepsis: a systematic review. Critical Care 2015;19(Suppl 1):P21-P21. 25 26 27 4. Fleischmann C, Scherag A, Adhikari NK, Hartog CS, Tsaganos T, Schlattmann P, et al. 28 29 Assessment of Global Incidence and Mortality of Hospital-treated Sepsis. Current 30 31 32 Estimates and Limitations. Am J Respir Crit Care Med 2016;193(3):259-72. 33 34 35 5. Delano MJ, Ward PA. Sepsis-induced immune dysfunction: can immune therapies 36 37 reduce mortality? Journal of Clinical Investigation 2016;126(1):23. http://bmjopen.bmj.com/ 38 39 40 6. Gotts JE, Matthay MA. Sepsis: pathophysiology and clinical management. BMJ 41 42 2016;353:i1585. 43 44 45 7. Paul M, Shani V, Muchtar E, Kariv G, Robenshtok E, Leibovici L. Systematic review and on October 1, 2021 by guest. Protected copyright. 46 47 48 meta-analysis of the efficacy of appropriate empiric antibiotic therapy for sepsis. 49 50 Antimicrobial Agents & Chemotherapy 2010;54(11):4851. 51 52 53 8. Annane D. Adjunct therapy for sepsis: how early? Curr Infect Dis Rep 2010;12(5):361- 54 55 67. 56 57 58 9. Poll TVD, Veerdonk FLVD, Scicluna BP, Netea MG. The immunopathology of sepsis 59 60 14

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 23 BMJ Open

1 2 3

4 and potential therapeutic targets. Nature Reviews Immunology 2017;17(7). BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 10. Feng YY, Xie YY, Wang YP, Qi L, Wang YM, Luo GA, et al. Molecular mechanism of 8 9 Xuebijing injection in treatment of sepsis according to “drug-target-pathway” network. Acta 10 11 12 Pharmaceutica Sinica 2017. 13 14 11. Chen S, Dai G, Hu J, Rong A, Lv J, Su L, et al. Discovery of Xuebijing Injection 15 16 17 Exhibiting Protective Efficacy on Sepsis by Inhibiting the Expression of HMGB1 in Septic 18 For peer review only 19 20 Rat Model Designed by Cecal Ligation and Puncture. American Journal of Therapeutics 21 22 2016;23(6). 23 24 25 12. Jiang M, Zhou M, Han Y, Xing L, Zhao H, Dong L, et al. Identification of NF-κB Inhibitors 26 27 in Xuebijing injection for sepsis treatment based on bioactivity-integrated UPLC-Q/TOF. J 28 29 30 Ethnopharmacol 2013;147(2):426-33. 31 32 33 13. He XD, Yan W, Wu Q, Wang HX, Chen ZD, Zheng RS, et al. Xuebijing Protects Rats 34 35 from Sepsis Challenged with Acinetobacter baumannii by Promoting Annexin A1 36 37 http://bmjopen.bmj.com/ 38 Expression and Inhibiting Proinflammatory Cytokines Secretion. Evidence-Based 39 40 Complementray and Alternative Medicine,2013,(2013-11-28) 2013;2013(1):804940. 41 42 43 14. Wang Q, Wu X, Tong X, Zhang Z, Xu B, Zhou W. Xuebijing Ameliorates Sepsis-Induced 44 45 on October 1, 2021 by guest. Protected copyright. 46 Lung Injury by Downregulating HMGB1 and RAGE Expressions in Mice. Evidence-Based 47 48 Complementray and Alternative Medicine,2015,(2015-3-2) 2015;2015(264):860259. 49 50 51 15. Li C, Wang P, Zhang L, Li M, Lei X, Liu S, et al. Efficacy and safety of Xuebijing injection 52 53 (a Chinese patent) for sepsis: a meta-analysis of randomized controlled trials. J 54 55 56 Ethnopharmacol 2018. 57 58 59 16. Shi H, Hong Y, Qian J, Cai X, Chen S. Xuebijing in the treatment of patients with sepsis. 60 15

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 23

1 2 3

4 The American journal of emergency medicine 2017;35(2):285-91. BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 17. Deng LH, Yao HG, Shao YM. Clinic Observation of Procalcitonin Sepsis with Xuebijing. 8 9 Journal of Practical Medical Techniques 2005. 10 11 12 18. National Center for Complementary and Alternative Medicine.Complementary and 13 14 Alternative Medicine. 15 16 17 19. Qin Y, Li C. Treatment Effects of Xuebijing Injection in Severe Septic Patients with 18 For peer review only 19 20 Disseminated Intravascular Coagulation. Evidence-Based Complementray and Alternative 21 22 Medicine,2014,(2014-3-20) 2014;2014(7):949254. 23 24 25 20. Zhang J, Qiu Y. Clinical Study of Ulinastatin Combined with Xuebijing on Sepsis with 26 27 Septic Shock. Chinese & Foreign Medical Research 2015. 28 29 30 21. Shankarhari M, Phillips GS, Levy ML, Seymour CW, Liu VX, Deutschman CS, et al. 31 32 33 Developing a New Definition and Assessing New Clinical Criteria for Septic Shock: For the 34 35 Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 36 37 http://bmjopen.bmj.com/ 38 2016;315(8):775-87. 39 40 22. Moher D, Chan AW. SPIRIT (Standard Protocol Items: Recommendations for 41 42 43 Interventional Trials): John Wiley & Sons, Ltd, 2014. 44 45 on October 1, 2021 by guest. Protected copyright. 46 23. Zhang SW, Sun CD, Yan W, Yin CH. Effect of treatment with Xuebijing injection (血必 47 48 净注射液 ) on serum inflammatory mediators and Th1/2 of spleen in rats with sepsis. 49 50 51 Chinese Critical Care Medicine 2006;18(11):673-76. 52 53 24. Sun C, Fan H, Jianbin GE. Effect of Xuebijing injection on serum concentrations of 54 55 56 TNF-α,IL-1β and IL-6 in patients with acute pancreatitis. Journal of Clinical Medicine in 57 58 59 Practice 2013;13(7):: R59. 60 16

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 23 BMJ Open

1 2 3

4 25. Opal SM, Pierre-Francois L, Bruno F, Larosa SP, Angus DC, Jean-Paul M, et al. Effect BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 of eritoran, an antagonist of MD2-TLR4, on mortality in patients with severe sepsis: the 8 9 ACCESS randomized trial. Jama 2013;309(11):1154-62. 10 11 12 26. Investigators SS, Finfer S, Mcevoy S, Bellomo R, Mcarthur C, Myburgh J, et al. Impact 13 14 of albumin compared to saline on organ function and mortality of patients with severe 15 16 17 sepsis. Intensive Care Med 2011;37(1):86-96. 18 For peer review only 19 20 27. Stevenson EK, Rubenstein AR, Radin GT, Wiener RS, Walkey AJ. Two Decades of 21 22 Mortality Trends among Patients with Severe Sepsis: A Comparative Meta-analysis. 23 24 25 Critical Care Medicine 2014;42(3):625-31. 26 27 28. Raith EP, Udy AA, Bailey M, Mcgloughlin S, Macisaac C, Bellomo R, et al. Prognostic 28 29 30 Accuracy of the SOFA Score, SIRS Criteria, and qSOFA Score for In-Hospital Mortality 31 32 33 Among Adults With Suspected Infection Admitted to the Intensive Care Unit. Journal of the 34 35 American Medical Association 2017;317(3):290. 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 17

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 23

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 Figure Legend 8 9 10 11 12 Figure 1: Flow chart of the EXIT-SEP study. 13 14 15 16 17 We screened patients who were admitted to the intensive care unit (ICU) with sepsis. 18 For peer review only 19 20 Patients were excluded if they met the exclusion criteria. Patients were randomized to 21 22 receive XBJ or the placebo (1:1). 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 18

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 23 BMJ Open

1 2 3 Table 1 Schedule of enrolment, interventions, assessments and data collection 4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Study Period 7 8 Enrollment / 9 Intervention (5 days) Follow-up 10 Baseline 11 12 Time point D0 D1 D2 D3 D4 D5 D6 D28 D90 13 14 Enrollment: 15 16 Eligibility screen Χ 17 18 Informed consent For peer reviewΧ only 19 20 Randomization Χ 21 22 Interventions: 23 24 Xuebijing injection Χ Χ Χ Χ Χ 25 26 Normal saline Χ Χ Χ Χ Χ 27 28 Assessments: 29 30 Demographic data Χ 31 32 Etiological examination Χ 33 34 Primary disease condition Χ 35 36 General condition Χ Χ Χ 37 http://bmjopen.bmj.com/ 38 SOFA score Χ Χ Χ 39 40 APACHE Ⅱ score Χ Χ Χ 41 42 Drugs Χ Χ Χ Χ Χ Χ Χ X 43 44 Safety X Χ 45 on October 1, 2021 by guest. Protected copyright. 46 AEs Χ Χ Χ Χ Χ 47 48 Mechanical ventilation/CRRT Χ 49 50 Duration of stay in the ICU Χ 51 52 Time and cost of hospitalization Χ Χ 53 54 Survival condition Χ Χ 55 56 57 58 59 60 19

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 23 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 We screened patients who were admitted to the intensive care unit (ICU) with sepsis. Patients were 27 excluded if they met the exclusion criteria. Patients were randomized to receive XBJ or the placebo (1:1). 28 29 224x134mm (300 x 300 DPI) 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on October 1, 2021 by guest. Protected copyright. 42 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: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from

Page 21 of 23 BMJ Open

1 2 3 4 5 6 Supplementary-material-1 7 Ethics committees’ names and approval registration number 8 9 NO. Hospital Ethics committee name Approval number 10 Independent Ethics Committee for Clinical Research of Zhongda Hospital, Affiliated 11 1 Zhongda Hospital, Southeast University 2017ZDSYLL025-P01 with Southeast University 12 For peer review only 13 2 Yantai Yuhuangding Hospital Medical Ethics Committee of Yantai Yuhuangding Hospital [2017]No. 17 14 Beijing Friendship Hospital Affiliated with Medical Ethics Committee of Beijing Friendship Hospital Affiliated with Capital 15 3 2017-P2-083-02 Capital Medical University Medical University 16

17 4 Northern Jiangsu People’s Hospital Medical Ethics Committee of Northern Jiangsu People’s Hospital http://bmjopen.bmj.com/ 2017045 18 The First Affiliated Hospital of Chongqing 5 Ethics Committee of The First Affiliated Hospital of Chongqing Medical University 20172501 19 Medical University 20 21 6 Yantaishan Hospital, Yantai The Clinical Trial Ethics Committee of Yantaishan Hospital 2017004 22 The Affiliated Hospital of Henan University of Medical Ethics Committee of The Affiliated Hospital of Henan University of Science 7 2017-0044 23 Science and Technology and Technology 24

25 8 Zhejiang Provincial People’s Hospital Medical Ethics Committee of Zhejiang Provincial People’s Hospital on October 1, 2021 by guest. Protected copyright. 2017KY019 26 The Second Affiliated Hospital of Harbin Medical Ethics Committee of the Second Affiliated Hospital of Harbin Medical 9 KY2017-241 27 Medical University University 28 Shenzhen Hospital of Traditional Chinese 29 10 Medical Ethics Committee of Shenzhen Hospital of Traditional Chinese Medicine (2017)18 30 Medicine 31 11 Tianjin Medical University General Hospital Medical Ethics Committee of Tianjin Medical University General Hospital IRB2017-130-01 32 The First Affiliated Hospital of Nanjing Medical 33 12 Ethics Committee of the First Affiliated Hospital of Nanjing Medical University 2017-SR-215 34 University (Jiangsu Province People’s Hospital) 35 13 Zhongshan Hospital, Fudan University Medical Ethics Committee of Zhongshan Hospital, Fudan University B2017-138 36 The First Affiliated Hospital of Nanchang Ethics Committee for Medical Research of the First Affiliated Hospital of Nanchang 37 14 2017-059 38 University University 39 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from

BMJ Open Page 22 of 23

1 2 3 4 5 NO. Hospital Ethics committee name Approval number 6 Chinese PLA Rocket General Hospital (Second 7 15 Ethics Committee of Second Artillery General Hospital PLA KY2017016

8 Artillery General Hospital PLA) 9 16 Tianjin Third Central Hospital Medical Ethics Committee of Tianjin Third Central Hospital IRB2017-011-01 10 17 Zhongnan Hospital of Wuhan University Medical Ethics Committee of Zhongnan Hospital of Wuhan University 2017020 11 The First Affiliated Hospital of China Medical Ethics Committee for Medical Science Research of the First Affiliated Hospital of 12 18 For peer review only 2017-166-2 13 University China Medical University 14 19 The First People’s Hospital of Foshan Medical Ethics Committee of the First People’s Hospital of Foshan 2017-8 15

16 20 First Hospital of Shanxi Medical University Ethics Committee of First Hospital of Shanxi Medical University 2017-K030

17 Central Hospital Affiliated to Shenyang Medical http://bmjopen.bmj.com/ 21 Medical Department of Central Hospital Affiliated to Shenyang Medical College NO 18 College 19 Xinhua Hospital Affiliated with Shanghai Jiao Xinhua Hospital Ethics Committee Affiliated with Shanghai Jiao Tong University 20 22 XHEC-C-2017-046-2 21 Tong University School of Medicine School of Medicine 22 23 Qilu Hospital of Shandong University (Qingdao) Ethics Committee of Qilu Hospital of Shandong University (Qingdao) NO 23 24 Wuhan General Hospital of Guangzhou Military Medical Ethics Committee of Wuhan General Hospital of Guangzhou Military [2017]023-1 24

25 25 Henan Province People's Hospital Medical Ethics Committee of Henan Province People's Hospital on October 1, 2021 by guest. Protected copyright. 2017(56) 26 Affiliated Hospital of Guizhou Medical Ethics Committee for Drug Clinical Trials of Affiliated Hospital of Guizhou Medical 27 26 2017-135 University University 28 29 27 China-Japan Friendship Hospital Ethics Committee for Clinical Research of China-Japan Friendship Hospital 2017-110 30 First Affiliated Hospital, Heilongjiang University Ethics Committee of First Affiliated Hospital, Heilongjiang University of Chinese 31 28 HZYLLKY201700601 of Chinese Medicine Medicine 32 33 29 The First Hospital of Jilin University Ethics Committee of The First Hospital of Jilin University 2017-327 34 30 Hunan Provincial People’s Hospital Medical Ethics Committee of Hunan Provincial People’s Hospital [2018]-03 35 31 Shanghai General Hospital Shanghai General Hospital Institutional Review Board [2018]02 36 37 32 Shanghai Changzheng Hospital Medical Ethics Committee of Shanghai Changzheng Hospital 2018SL005 38 39 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from

Page 23 of 23 BMJ Open

1 2 3 4 5 NO. Hospital Ethics committee name Approval number 6 The Third Xiangya Hospital of Central South 7 33 Institutional Review Board of The Third Xiangya Hospital of Central South University R17020

8 University 9 The Affiliated Hospital of Inner Mongolia 34 Ethics Committee of The Affiliated Hospital of Inner Mongolia Medical University 2017(016) 10 Medical University 11 The First Affiliated Hospital of Zhengzhou Ethics Committee for Scientific Research and Clinical Trials of the First Affiliated 12 35 For peer review only Drug-2017-85 13 University Hospital of Zhengzhou University 14 First Affiliated Hospital of the Air Force Medical 15 Independent Ethics Committee for Drug Clinical Trials of First Affiliated Hospital of 36 University (First Affiliated Hospital of Fourth KY20172100-1 16 Fourth Military Medical University

17 Military Medical University) http://bmjopen.bmj.com/ 18 Scientific Research Institutional Review Board of Yijishan Hospital of Wannan 37 Yijishan Hospital of Wannan Medical College 2018(04) 19 Medical College 20 21 38 Shanghai Sixth People's Hospital Ethics Committee of Shanghai Sixth People's Hospital 2018-012 22 The North Hospital of Ruijin Hospital Affiliated Ethics Committee of The North Hospital of Ruijin Hospital Affiliated with Shanghai 23 39 with Shanghai Jiao Tong University School of 2018(009)-1 24 Jiao Tong University School of Medicine

25 Medicine on October 1, 2021 by guest. Protected copyright. 26 Peking Union Medical College Hospital of the Institutional Review Board of Peking Union Medical College Hospital of the Chinese 40 B248 27 Chinese Academy of Medical Sciences Academy of Medical Sciences 28 First Affiliated Hospital of Kunming Medical 29 41 Ethics Committee of First Affiliated Hospital of Kunming Medical University 2018-12 30 University 31 42 The First Hospital of Lanzhou University Ethics Committee for Drug Clinical Trials of The First Hospital of Lanzhou University 2018-13 32 33 43 Chinese PLA General Hospital Medical Ethics Committee of Chinese PLA General Hospital S2018-011-01 34 44 Yanbian University Hospital Medical Ethics Committee of Yanbian University Hospital 2018138 35 45 The Affiliated Hospital of Qingdao University Medical Ethics Committee of The Affiliated Hospital of Qingdao University QYFYKYLL 2018-20 36 37 38 39 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from

Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP): protocol for a randomized controlled trial

Journal: BMJ Open ManuscriptFor ID peerbmjopen-2018-028664.R1 review only Article Type: Protocol

Date Submitted by the 06-Jun-2019 Author:

Primary Subject Intensive care Heading: http://bmjopen.bmj.com/ Secondary Subject Heading: Intensive care

Keywords: Xuebijing injection, Sepsis, randomized, controlled trial, protocol

on October 1, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 32 BMJ Open

1 2 3

4 Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP): protocol for a BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 randomized controlled trial 8 9 10 11 Songqiao Liu1, Chen Yao2, Junhua Zhang3，Yi Yang1 and Haibo Qiu1 and the EXIT-SEP Investigators 12 13 14 15 Correspondence to Dr Haibo Qiu; [email protected], Department of Critical Care Medicine, 16 17 Zhongda Hospital, School of Medicine, Southeast University, Nanjing, Jiangsu, 210009, China 18 For peer review only 19 20 21 Author affiliations 22 23 1 Department of Critical Care Medicine, Zhongda Hospital, School of Medicine, Southeast 24 25 University, Nanjing, Jiangsu, 210009, China. 26 27 2Medical Statistics Office, Peking University First Hospital, Beijing, China, 100034. 28 29 3 Centre for Evidence-based Medicine, Tianjin University of Traditional Chinese Medicine, Tianjin, 30 31 China, 301617. 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 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 32

1 2 3 ABSTRACT

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Introduction Sepsis is a major challenge with high incidence and is associated with high mortality 7 8 worldwide. Current management of sepsis remains mainly supportive except for treatment with 9 10 antibiotics. Both basic research and clinical investigation have shown that the Chinese herbal 11 derived therapeutic Xuebijing (XBJ) injection is beneficial for patients with sepsis. However, the 12 13 quality of evidence supporting the therapeutic use of XBJ in sepsis is limited. The aim of this trial 14 15 is to evaluate the Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP), compared with a placebo, 16 17 on the outcome of patients with sepsis in the ICU. 18 For peer review only 19 Methods and analysis In this multicentre, blinded, randomized, controlled trial, we are recruiting a 20 21 total of 1800 subjects who met Sepsis 3.0 criteria. Subjects will be randomized (1:1) to receive XBJ, 22 23 q12h for 5 days or a matching placebo and usual care. The primary outcome is 28 days all-cause 24 25 mortality. Secondary outcomes will be the improvement of Sequential Organ Failure Assessment 26 27 (SOFA) scores, the improvement of the Acute Physiology and Chronic Health Evaluation II (APACHE 28 29 II) score, duration of mechanical ventilation, mortality in ICU and duration of stay in the ICU. 30 31 Investigators, participants, and statisticians will be blinded to the allocated treatment. 32 33 Ethics and dissemination This trial has been approved by all ethics committees of the centers that will 34 35 participate in this trial. The findings of the study will be disseminated in peer-reviewed journals and 36 37 present at conferences. Once this study is complete, the results of this trial may help to provide http://bmjopen.bmj.com/ 38 39 evidence-based recommendations for complementary therapeutic options for patients with sepsis. 40 41 Trial registration number This clinical trial was registered at Clinicaltrials.gov (NCT03238742) and 42 43 the Chinese Clinical Trial Registry (ChiCTR-IPR-17012713). 44 45 Protocol date: 10 May 2017. on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 KEYWORDS 51 52 Xuebijing injection; Sepsis; randomized, controlled trial; protocol 53 54 55 56 57 58 59 60 2

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 32 BMJ Open

1 2 3 Strengths and limitations of this study

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6  The EXIT-SEP trial is a blinded, randomized controlled study comparing the effect of XBJ 7 8 injection with placebo on outcome in patients with sepsis. 9  10 This is a multicentre study including a wide representation of the Chinese population in 11 the entire country. 12 13  The study will have a sample size large enough to provide high-quality evidence to 14 15 evaluate the potential benefit of XBJ injection in sepsis. 16 17  One limitation of the study is blinding. Except of drug administrators and a minority of 18 For peer review only 19 the nurses, the investigators, patients, and statistician were all blinded. 20 21  All participating centers of the EXIT-SEP trial are in China. The result of this study cannot 22 23 be directly applied in other different population. 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 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 32

1 2 3 INTRODUCTION

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Sepsis is a common and devastating condition that is responsible for more than 6 million deaths 7 8 worldwide annually and is the leading cause of death in critically ill patients[1]. Half of all sepsis 9 10 patients are treated in the intensive care unit (ICU), representing more than 10% of all ICU 11 admissions[1, 2]. The World Health Organization (WHO) reports that sepsis affects more than 30 12 13 million people worldwide every year[3]. The burden of sepsis is most likely highest in low- and 14 15 middle-income countries. 16 17 The pathogenesis of sepsis includes an overwhelming inflammatory response, endothelial 18 For peer review only 19 injury, altered coagulation, dysregulated immune response, and sepsis-associated 20 21 immunosuppression, resulting in acute life-threatening organ dysfunction. During the past 30 22 23 years, the understanding of the host immune response has advanced considerably, and the more 24 25 than 100 therapeutic clinical trials that have been conducted, no approved treatment options are 26 27 currently recommended for sepsis [4][5]. Although there are improvements in supportive care 28 29 strategies, mortality still remains unacceptably high. Some adjuvant treatments for sepsis have 30 31 been developed at great expense but failed to significantly improve mortality[6, 7]. 32 33 There is accumulating evidence to suggest that Xuebijing (XBJ) injection has the potential to 34 35 improve the outcomes for critically ill patients with sepsis. XBJ, a Chinese herb-derived therapeutic, 36 37 has been approved for sepsis treatment in critically ill patients (China Food and Drug http://bmjopen.bmj.com/ 38 39 Administration; Beijing, China, No. Z20040033). The main components of XBJ include amino acids, 40 41 phenolic acids, flavonoid glycoside, terpene glycoside, and phthalides. XBJ has effects on at least 42 43 ten sepsis/inflammation pathways and its 21 major active ingredients regulate 550 targets (HRAS, 44 45 GSK3B, BTK, AK, et al.)[8]. Basic research of XBJ suggests that XBJ has significant on October 1, 2021 by guest. Protected copyright. 46 47 immunomodulatory, anti-coagulation and anti-inflammatory activity. Animal studies and ex vivo 48 49 studies have suggested that the effects of XBJ inhibit the expression of HMGB1, inhibit 50 proinflammatory cytokines secretion and ameliorate sepsis-induced lung injury[9-12]. 51 52 Clinical trials have demonstrated that XBJ improved outcomes and a meta-analysis showed 53 54 that XBJ was associated with high survival rate (RR 0.62, 95% CI 0.51-0.76 P < 0.000 01, I2 = 0%)[13, 55 56 14]. These clinical trials, however, were conducted with small sample sizes, even when pooled, 57 58 limiting the evidence of potential benefit in patients with sepsis [13, 15]. Therefore, the large RCT 59 60 4

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 32 BMJ Open

1 2 3 is needed to access the benefit of XBJ on patients with sepsis. We hypothesize that XBJ will reduce

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 all-cause mortality than the placebo among early sepsis patients. Thus, we aim to conduct a large- 7 8 scale, blind, randomized controlled trial (RCT) to evaluate the effects of XBJ in adult sepsis patients 9 10 in the ICU. 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 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 32

1 2 3 METHODS

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 Design 6 7 The Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP) study is a multicentre, blind, randomized 8 9 and placebo-controlled trial. Eligible participants will be randomly assigned to either the XBJ group 10 11 or the placebo group in a ratio of 1:1. The participants will receive XBJ (XBJ group) or the solvent 12 13 (placebo group) within 24 h of enrolment for 5 days. Figure 1 shows the flowchart of the study. 14 15 This study protocol follows the Standard Protocol Items: Recommendations for Interventional 16 17 Trials statement recommendations[16]. The items from the trial registration data set are described 18 For peer review only 19 in the online supplementary file 1. This study protocol was registered at Clinicaltrials.gov 20 21 (NCT03238742) and at the Clinical Trial Registry (ChiCTR) before randomization (ChiCTR-IPR- 22 23 17012713). 24 25 26 Population 27 28 Participant recruitment is currently ongoing at Intensive Care Units (ICUs) from the 45 medical 29 30 centers in China. Patients who fulfill the inclusion criteria and who sign informed consent forms 31 32 will enter the screening period. Patients with sepsis who meet the exclusion criteria will be 33 34 excluded before randomization. The recruitment duration will last for 27 months, from October 35 36 2017 to December 2019. 37 http://bmjopen.bmj.com/ 38 39 40 Inclusion criteria 41 42 Patients are eligible for the trial if they are ≥18 and ≤75 years old, meet the criteria of sepsis 3.0 43 44 (The Third International Consensus Definition for Sepsis and Septic Shock), and SOFA score of 2- 45 on October 1, 2021 by guest. Protected copyright. 46 13. 47 48 49 50 Exclusion criteria 51 52 Patients will be excluded if they fulfill any of the exclusion criteria: 53 54 1. Diagnosis of sepsis for more than 48 h; 55 56 2. Pregnant and lactating women; 57 58 3. Severe primary disease including unrespectable tumors, blood diseases and Human 59 60 Immunodeficiency Virus (HIV); 6

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 32 BMJ Open

1 2 3 4. Severe liver and kidney dysfunction (single liver or kidney SOFA score ≥ 3 points);

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 5. Use of an immunosuppressant or having an organ transplant within the previous 6 months; 7 8 6. Participating in other clinical trials in the previous 30 days. 9 10 11 Ethics and informed consent 12 13 This trial is approved by the Research Ethics Boards at Zhongda Hospital (File number: 14 15 2017ZDSYLL025-P01). The research ethics committees of all participant centers approved the 16 17 study protocol(supplementary-material-2) before randomizing patients. The patients’ written 18 For peer review only 19 informed consent(supplementary-material-3) is required at all participating centers. Site 20 21 investigators will be responsible for obtaining informed consent from study participants or legal 22 23 representative of the family member. Subject confidentiality will be assured through data 24 25 anonymization and controlled access to case report forms, the electronic data capture system, and 26 27 datasets. Any breaches of confidentiality, study protocol or AEs attributable to this study will be 28 29 reported to the research ethics committees. 30 31 32 33 Randomization and allocation concealment 34 35 Randomization is generated centrally using an interactive web response system (IWRS). When an 36 37 eligible participant has enrolled this subcentre, the drug administrators will log into the central http://bmjopen.bmj.com/ 38 39 randomization system, where a random number is produced. The independent drug 40 41 administrators receive group information based on the random number, and then they assign the 42 43 study drug to the nurses for administration. 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 Interventions 48 49 The participants will receive XBJ (normal saline 100 ml + XBJ 100 ml, q12h, intravenous infusion 50 for 80 min) in the XBJ group and the solvent only (normal saline, 200 ml, q12h, also administered 51 52 for 80 minutes) in the placebo group as previous study[17]. XBJ injection includes a 10 ml/ampule, 53 54 packaged in 5 ampules/container. They were manufactured by a Good Manufacturing Practice 55 56 (GMP) certified company in China (Tianjin Chase Sun Pharmaceutical Co., Ltd., Z20040033). 57 58 Patients in both groups will receive standard care by the attending physician according to the 59 60 7

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 32

1 2 3 International Guidelines for Management of Sepsis and Septic Shock[1]. The patients will receive

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 the study drug within 12 h of enrolment, for a 5 day treatment period. According to the 2016 7 8 guidelines on sepsis /sepsis shock, the routine care should include early fluid resuscitation, 9 10 antimicrobials anticoagulants, nutritional support, and other treatment. Ulinastatin and Tanreqing 11 injection are not allowed during the study period. Use of all drugs including antibiotics, if any, 12 13 should be documented in the Case Report Form (CRF). 14 15 16 17 Blinding 18 For peer review only 19 Photophobic brown color infusion bags and infusion devices for both groups will be visually 20 21 inspected by the pharmacy to ensure identical appearance. The independent drug administrators 22 23 will receive group information based on a random number; they will then assign the study drug to 24 25 the nurses to administer. The mixture of drugs will be prepared by the nurses in a separate room. 26 27 This mixture of drugs will be the standard procedure for XBJ, and all the nurses will have experience 28 29 in performing this task. The nurses will also sign a confidentiality agreement about patient 30 31 allocation. The participants, as well as all the members of the study and the health care team, 32 33 outcome assessors, will be blinded to the study drug assignment. Data analysis will be performed 34 35 by a researcher who is blinded to patient allocation. 36 37 http://bmjopen.bmj.com/ 38 39 Outcomes 40 41 The primary outcome will be 28-day all-cause mortality. Secondary outcomes will be the 42 43 improvement of Sequential Organ Failure Assessment (SOFA) scores, the improvement of the 44 45 Acute Physiology and Chronic Health Evaluation II (APACHE II) score (the improvement means the on October 1, 2021 by guest. Protected copyright. 46 47 difference of score between the day enrolled in the study and the 6 days), duration of mechanical 48 49 ventilation, mortality in the Intensive Care Unit (ICU), duration of stay in the ICU. 50 51 52 Follow-up 53 54 Data will be recorded during the follow-up period according to the multiple time-points. The 55 56 details are shown in Table 1. 57 58 1. Screening period (day 0): before recruitment; 59 60 8

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 32 BMJ Open

1 2 3 2. Intervention period (day 1-5): data will be recorded every day during follow-up;

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 3. The period after the intervention (within 28 days after treatment): follow-up at day 28 for 7 8 survival data. If the patient is discharged, we will contact them via telephone or a short messaging 9 10 service. 11 12 13 Sample size calculation 14 15 We determined that enrolment of 1800 participants would provide the trial with 80% power to 16 17 detect an absolute difference of 6 percentage points decrease in 28-day all-cause mortality of 24.3 18 For peer review only 19 % with a two-sided P value of 0.05. This calculation allowed for a rate of withdrawal and loss to 20 21 follow-up of 15%. We estimated the 28 days mortality rate of 24.3% in a previous observational 22 23 CHESS study of 2322 sepsis/septic shock patients with the SOFA score of 2-13 in China. 24 25 26 27 Study organization 28 29 The design and implementation of a multicentre study require collaboration among several 30 31 organizations. Quality control will be undertaken by these organizations: ① Expert committee: 32 33 comprises clinical experts, statisticians, and quality control experts who will be responsible for 34 35 clinical research methodology and for resolving the key issues in the implementation of the study. 36 37 ②Executive committee: the main staff from the members of the expert committee. The http://bmjopen.bmj.com/ 38 39 responsibilities of the executive committee include the design of the clinical research trial, 40 41 selecting the cooperative hospitals, and providing a training course including a manual of 42 43 instructions. ③ DSMB(Data Safety and Monitoring Board): an independent team that evaluates 44 45 the safety outcomes and AEs and submits a review proposal. ④ Data management and statistical on October 1, 2021 by guest. Protected copyright. 46 47 analysis: performed by trained staff and statisticians. ⑤ Quality control: sites and researchers will 48 49 be monitored and inspected regularly by two CRO, following the standard protocol throughout the 50 process. This trial will undergo normative monitoring and inspection throughout the entire process. 51 52 53 54 Data management 55 56 Qualified sites and researchers are crucial factors that ensure the quality of a clinical trial. The sub- 57 58 centers should be qualified by the “Good Clinical Practice (GCP) training” of the State Food and 59 60 9

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 32

1 2 3 Drug Administration (SFDA) for compliance in the training. Qualified researchers should

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 understand the detailed contents of the protocol. The data filled in the CRF by researchers should 7 8 be accurate, complete, timely, and reliable. 9 10 Data management will be performed by trained staff at each participating centre using the 11 electronic data system. The quality of the trial data management will be guaranteed by the 12 13 reliability, access control, and traceability of the system. Data collection will be restricted to those 14 15 variables necessary to define baseline patient characteristics (demographics, sepsis diagnosis, 16 17 concurrent medical conditions and comorbidities, inclusion and exclusion criteria, the severity of 18 For peer review only 19 illness and organ dysfunction scores, and laboratory results), the delivery of the study drugs, 20 21 potential confounding co-interventions, and outcomes. Randomized participants will be followed 22 23 up until either death or 28 days after randomization, whichever comes first. Follow-up will be 24 25 conducted by study staff by either direct contact with the patient or their next of kin. Participants 26 27 who withdraw from the study for any reason will be followed up with according to the study 28 29 follow-up schedule and data will be analyzed according to the intention to treat (ITT) principle. 30 31 32 33 Data analysis 34 35 All analyses will be performed according to the intention to treat principle. Continuous variables 36 37 will be reported as the means and standard deviations or as medians and interquartile ranges http://bmjopen.bmj.com/ 38 39 according to data distribution. Categorical variables will be reported as proportions. We will 40 41 compare the data on the primary outcome of 28-day all-cause mortality using an unadjusted chi- 42 43 square test for proportions or a logistic regression model. We will report frequency (percentage) 44 45 per treatment group with a risk difference and 95% confidence interval and a corresponding odds on October 1, 2021 by guest. Protected copyright. 46 47 ratio and 95% confidence interval. The missing data of the primary outcome are filled by the worst 48 49 imputation method, that is, "death" is used to fill the missing data. The secondary binary and 50 continuous outcomes will be analysed with the use of logistic regression and linear regression, 51 52 respectively. The rate of death in a time-to-event analysis will be reported with the use of Kaplan– 53 54 Meier plots, and differences in survival will be tested using a Cox proportional-hazards model that 55 56 includes the group variable. All statistical analyzes will be performed using SAS version 9.4 (SAS 57 58 Institute Inc) with a 2-sided P value of less than .05 considered significant. No adjustment will be 59 60 10

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 32 BMJ Open

1 2 3 made for multiple comparisons; therefore, secondary outcomes will be interpreted as exploratory.

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 8 Patient and public involvement 9 10 No patient or public was involved in the present study. Upon the completion of this trial, a journal 11 article manuscript will be prepared to present the trial results. 12 13 14 15 DISCUSSION 16 17 This trial is designed as a multicentre blinded RCT. The completion of this trial will provide 18 For peer review only 19 definitive and accurate evidence on the effectiveness of XBJ injection to reduce mortality of 20 21 patients with sepsis. 22 23 XBJ injection, an intravenous preparation approved by the China Food and Drug Administration 24 25 (China FDA) in 2004, has been evaluated as a promising Chinese herbal derived therapeutic drug 26 27 in the management of sepsis, by reducing the inflammatory mediators such as TNF-α, IL-1, IL-6 28 29 and IL-8, improving the immune function, and protecting the vascular endothelial cells, which 30 31 correspond to the major pathogenesis of sepsis[18]. XBJ injection has pharmacological effects on 32 33 antagonizing endotoxin, inhibiting inflammatory mediators, improving coagulation function and 34 35 microcirculation, protecting endothelial cells, and regulating the immune response. A meta- 36 37 analysis showed that XBJ reduced all-cause mortality. However, current clinical evidence http://bmjopen.bmj.com/ 38 39 supporting the efficacy of XBJ in the treatment of sepsis, including RCTs, is subject to limitations 40 41 such as a non-representative patient population, small sample size, and concomitant use of 42 43 immunomodulatory agents. As a result, the efficacy of XBJ on the mortality of patients with sepsis 44 45 warrants validation by large, well-conducted RCTs in the ICU. on October 1, 2021 by guest. Protected copyright. 46 47 This trial uses the 28-day mortality as the primary outcome. At the time of preparation of this study, 48 49 patients who met the criteria of Sepsis 3.0 were reported to have a hospital mortality rate of 20% 50 in developed countries. Previous studies suggest that ICU patients with sepsis in middle- and low- 51 52 income countries might have a much higher mortality rate than those in high-income countries, 53 54 possibly due to inadequate resources and poor quality of care[19]. Using data from our cross 55 56 section epidemiology study called ‘CHESS’, we estimate that 28-day all-cause mortality will be 57 58 31.87% in the control group and the 28 days mortality rate of 24.3% in patients of SOFA score no 59 60 11

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 32

1 2 3 more than 13. The ICU mortality rate was greater than 60% in patients with sepsis whose SOFA

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 score more than 13. Those patients were too severe to detect the benefit of XBJ in ICU. With a 7 8 sample size of 1800, this large-scale trial will be adequately powered to test the study hypothesis. 9 10 The definition of Sepsis 3.0 focuses on organ function, so except the regular endpoints in clinical 11 trials, we would like to employ more secondary endpoint such as organ function. 12 13 There are, however, surrogate outcomes that can be influenced by blinding, which can be 14 15 influenced by discharge decisions and safety judgment. Although the independent drug 16 17 administrators and a minority of nurses were able to determine the treatment arm, we 18 For peer review only 19 demonstrated that concealed drug administration achieved satisfactory levels of blinding in a 20 21 multicentre context. In particular, medications in brown infusion bags and tubes to obscure 22 23 content were sufficient to achieve blinding. 24 25 Recruitment will be completed in December 2019 with a final number of 1800 participants in 45 26 27 medical centers in China. The collection of primary endpoint data will conclude in April 2020. With 28 29 the results of the EXIT-SEP trial, we hope to be able to make evidence-based recommendations for 30 31 XBJ use for sepsis. 32 33 34 35 Ethics and dissemination 36 37 This trial has been approved by Ethics Committees of all of the centers that will participate in this trial. http://bmjopen.bmj.com/ 38 39 The findings of the study will be disseminated at conferences and in peer-reviewed journals. Once this 40 41 study is complete, the results of this trial may help to provide evidence-based recommendations of 42 43 complementary therapeutic options for patients with sepsis. 44 45 on October 1, 2021 by guest. Protected copyright. 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 32 BMJ Open

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Funding This trial was supported by the Development Center for Medical Science and Technology, 7 8 National Health and Family Planning Commission of the People’s Republic of China [WK-2016-HR- 9 10 03], which had no role in the design of the protocol, the conduction of the trial, or the analyses or 11 reporting of the data. 12 13 Contributors Haibo Qiu designed the study, drafted the manuscript, edited the manuscript, 14 15 supervised the study, and obtained study funding. Songqiao Liu designed the study, drafted and 16 17 edited the manuscript. Yi Yang drafted and edited the manuscript. Chen Yao designed the study 18 For peer review only 19 and reviewed the manuscript. Junhua Zhang edited the manuscript, and provided study 20 21 supervision. Investigators of the EXIT-SEP have participated in the discussion of the protocol and 22 23 reviewed the manuscript for important intellectual content. All authors have read and approved 24 25 the final manuscript. 26 27 Competing interests None declared. 28 29 Patient consent Obtained. 30 31 Ethics approval This study has been approved by the Ethics Committee of Zhongda Hospital, 32 33 Southeast University (Reference number: 2017ZDSYLL025-P01) on the 17 May 2017, and all 34 35 participants provided written informed consent. After the study is completed, the analysis results 36 37 will be published; however, no personal information will be included in the findings. http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 13

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 32

1 2 3 Figure Legend 4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Fig. 1. The flow chart of the EXIT-SEP study. 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 14

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 32 BMJ Open

1 2 3 REFERENCES

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 1. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, Kumar A, 7 8 9 Sevransky JE, Sprung CL, Nunnally ME et al: Surviving Sepsis Campaign: International 10 11 Guidelines for Management of Sepsis and Septic Shock: 2016. Critical care medicine 12 13 14 2017, 45(3):486-552. 15 16 2. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, 17 18 For peer review only 19 Bellomo R, Bernard GR, Chiche JD, Coopersmith CM et al: The Third International 20 21 22 Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016, 23 24 315(8):801-810. 25 26 27 3. Fleischmann C, Scherag A, Adhikari NK, Hartog CS, Tsaganos T, Schlattmann P, 28 29 Angus DC, Reinhart K, International Forum of Acute Care T: Assessment of Global 30 31 32 Incidence and Mortality of Hospital-treated Sepsis. Current Estimates and Limitations. 33 34 35 American journal of respiratory and critical care medicine 2016, 193(3):259-272. 36 37

4. Delano MJ, Ward PA: Sepsis-induced immune dysfunction: can immune therapies http://bmjopen.bmj.com/ 38 39 40 reduce mortality? The Journal of clinical investigation 2016, 126(1):23-31. 41 42 5. Gotts JE, Matthay MA: Sepsis: pathophysiology and clinical management. BMJ 43 44 45 (Clinical research ed) 2016, 353:i1585. on October 1, 2021 by guest. Protected copyright. 46 47 48 6. Paul M, Shani V, Muchtar E, Kariv G, Robenshtok E, Leibovici L: Systematic review 49 50 and meta-analysis of the efficacy of appropriate empiric antibiotic therapy for sepsis. 51 52 53 Antimicrob Agents Chemother 2010, 54(11):4851-4863. 54 55 7. Annane D: Adjunct therapy for sepsis: how early? Curr Infect Dis Rep 2010, 12(5):361- 56 57 58 367. 59 60 15

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 32

1 2 3

4 8. Zhang N, Cheng C, Olaleye OE, Sun Y, Li L, Huang Y, Du F, Yang J, Wang F, Shi Y BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 et al: Pharmacokinetics-Based Identification of Potential Therapeutic Phthalides from 8 9 XueBiJing, a Chinese Herbal Injection Used in Sepsis Management. Drug Metab 10 11 12 Dispos 2018, 46(6):823-834. 13 14 9. Chen S, Dai G, Hu J, Rong A, Lv J, Su L, Wu X: Discovery of Xuebijing Injection 15 16 17 Exhibiting Protective Efficacy on Sepsis by Inhibiting the Expression of HMGB1 in 18 For peer review only 19 20 Septic Rat Model Designed by Cecal Ligation and Puncture. Am J Ther 2016, 21 22 23(6):e1819-e1825. 23 24 25 10. Jiang M, Zhou M, Han Y, Xing L, Zhao H, Dong L, Bai G, Luo G: Identification of NF- 26 27 kappaB Inhibitors in Xuebijing injection for sepsis treatment based on bioactivity- 28 29 30 integrated UPLC-Q/TOF. J Ethnopharmacol 2013, 147(2):426-433. 31 32 33 11. He XD, Wang Y, Wu Q, Wang HX, Chen ZD, Zheng RS, Wang ZS, Wang JB, Yang Y: 34 35 Xuebijing Protects Rats from Sepsis Challenged with Acinetobacter baumannii by 36 37 http://bmjopen.bmj.com/ 38 Promoting Annexin A1 Expression and Inhibiting Proinflammatory Cytokines Secretion. 39 40 Evid Based Complement Alternat Med 2013, 2013:804940. 41 42 43 12. Wang Q, Wu X, Tong X, Zhang Z, Xu B, Zhou W: Xuebijing Ameliorates Sepsis-Induced 44 45 on October 1, 2021 by guest. Protected copyright. 46 Lung Injury by Downregulating HMGB1 and RAGE Expressions in Mice. Evid Based 47 48 Complement Alternat Med 2015, 2015:860259. 49 50 51 13. Li C, Wang P, Zhang L, Li M, Lei X, Liu S, Feng Z, Yao Y, Chang B, Liu B et al: Efficacy 52 53 and safety of Xuebijing injection (a Chinese patent) for sepsis: A meta-analysis of 54 55 56 randomized controlled trials. J Ethnopharmacol 2018, 224:512-521. 57 58 59 14. Shi H, Hong Y, Qian J, Cai X, Chen S: Xuebijing in the treatment of patients with sepsis. 60 16

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 32 BMJ Open

1 2 3

4 The American journal of emergency medicine 2017, 35(2):285-291. BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 15. Yin Q, Li C: Treatment effects of xuebijing injection in severe septic patients with 8 9 disseminated intravascular coagulation. Evid Based Complement Alternat Med 2014, 10 11 12 2014:949254. 13 14 16. Chan AW, Tetzlaff JM, Gotzsche PC, Altman DG, Mann H, Berlin JA, Dickersin K, 15 16 17 Hrobjartsson A, Schulz KF, Parulekar WR et al: SPIRIT 2013 explanation and 18 For peer review only 19 20 elaboration: guidance for protocols of clinical trials. BMJ (Clinical research ed) 2013, 21 22 346:e7586. 23 24 25 17. Liu SQ, Zheng RQ, Li MQ, Yan J, Chen HY, Mu XW, Chang RH, Ye ZL, Li XS, Gao YH 26 27 et al: [Effect of Xuebijing injection treatment on acute respiratory distress syndrome: a 28 29 30 multicenter prospective randomized control clinical trial]. Zhonghua Yi Xue Za Zhi 2012, 31 32 33 92(15):1017-1022. 34 35 18. Zhang SW, Sun CD, Wen Y, Yin CH: [Effect of treatment with Xuebijing injection on 36 37 http://bmjopen.bmj.com/ 38 serum inflammatory mediators and Th1/2 of spleen in rats with sepsis]. Zhongguo Wei 39 40 Zhong Bing Ji Jiu Yi Xue 2006, 18(11):673-676. 41 42 43 19. Raith EP, Udy AA, Bailey M, McGloughlin S, MacIsaac C, Bellomo R, Pilcher DV, 44 45 on October 1, 2021 by guest. Protected copyright. 46 Australian, New Zealand Intensive Care Society Centre for O, Resource E: Prognostic 47 48 Accuracy of the SOFA Score, SIRS Criteria, and qSOFA Score for In-Hospital Mortality 49 50 51 Among Adults With Suspected Infection Admitted to the Intensive Care Unit. JAMA 52 53 2017, 317(3):290-300. 54 55 56 57 58 59 60 17

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 32

1 2 3 Table 1 Schedule of enrolment, interventions, assessments and data collection 4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Study Period 7 8 Enrollment / 9 Intervention (5 days) Follow-up 10 Baseline 11 Time point D0 D1 D2 D3 D4 D5 D6 D28 12 13 Enrollment: 14 15 Eligibility screen Χ 16 17 Informed consent Χ 18 For peer review only 19 Randomization 20 Χ 21 Interventions: 22 23 Xuebijing injection Χ Χ Χ Χ Χ 24 25 Normal saline Χ Χ Χ Χ Χ 26 27 Assessments: 28 29 Demographic data Χ 30 31 Etiological examination 32 Χ 33 Primary disease condition Χ 34 35 Χ Χ 36 General condition Χ 37 SOFA score http://bmjopen.bmj.com/ 38 Χ Χ Χ 39 APACHE Ⅱ score 40 Χ Χ Χ 41 42 Drugs Χ Χ Χ Χ Χ Χ Χ X 43 44 Safety X Χ 45 on October 1, 2021 by guest. Protected copyright. 46 AEs Χ Χ Χ Χ Χ 47 48 Mechanical ventilation/CRRT Χ 49 50 Duration of stay in the ICU Χ 51 52 Time and cost of hospitalization Χ 53 54 Survival condition Χ 55 56 57 58 59 60 18

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 Fig 1 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 8 SPIRIT 2013 Checklist: Recommended items to address in a clinical trial protocol and 9 related documents* 10 11 Section/item Item Description Page Number 12 No on which item 13 is reported 14 15 Administrative information 16 17 Title 1 Descriptive title identifying the study design, 1 18 Forpopulation, peer interventions, review and, if applicable, only trial 19 20 acronym 21 22 Trial 2a Trial identifier and registry name. If not yet registered, 6 23 registration name of intended registry 24 25 2b All items from the World Health Organization Trial 6 26 Registration Data Set 27 28 Protocol 3 Date and version identifier 20170510- 29 version vision 2 30 31 Funding 4 Sources and types of financial, material, and other 13 32 33 support 34 35 Roles and 5a Names, affiliations, and roles of protocol contributors 1，13 36 responsibilitie 37 5b Name and contact information for the trial sponsor 13 s http://bmjopen.bmj.com/ 38 39 5c Role of study sponsor and funders, if any, in study 13 40 design; collection, management, analysis, and 41 42 interpretation of data; writing of the report; and the 43 decision to submit the report for publication, including 44 whether they will have ultimate authority over any of 45 these activities on October 1, 2021 by guest. Protected copyright. 46 47 5d Composition, roles, and responsibilities of the 48 9 49 coordinating centre, steering committee, endpoint 50 adjudication committee, data management team, and 51 other individuals or groups overseeing the trial, if 52 applicable (see Item 21a for data monitoring 53 54 committee) 55 56 Introduction 57 58 59 60

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 32 BMJ Open

1 2 Background 6a Description of research question and justification for 5 3 and rationale undertaking the trial, including summary of relevant

4 studies (published and unpublished) examining BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 benefits and harms for each intervention 7 8 6b Explanation for choice of comparators 5 9 10 Objectives 7 Specific objectives or hypotheses 5 11 12 Trial design 8 Description of trial design including type of trial (eg, 6 13 parallel group, crossover, factorial, single group), 14 allocation ratio, and framework (eg, superiority, 15 equivalence, noninferiority, exploratory) 16 17 18 Methods: Participants,For interventions, peer reviewand outcomes only 19 20 Study setting 9 Description of study settings (eg, community clinic, 6 21 academic hospital) and list of countries where data 22 will be collected. Reference to where list of study 23 24 sites can be obtained 25 26 Eligibility 10 Inclusion and exclusion criteria for participants. If 6 27 criteria applicable, eligibility criteria for study centres and 28 individuals who will perform the interventions (eg, 29 surgeons, psychotherapists) 30 31 Interventions 11a Interventions for each group with sufficient detail to ，8 32 7 33 allow replication, including how and when they will be 34 administered 35 36 11b Criteria for discontinuing or modifying allocated 8 37 interventions for a given trial participant (eg, drug http://bmjopen.bmj.com/ 38 dose change in response to harms, participant 39 40 request, or improving/worsening disease) 41 42 11c Strategies to improve adherence to intervention 8 43 protocols, and any procedures for monitoring 44 adherence (eg, drug tablet return, laboratory tests) 45 on October 1, 2021 by guest. Protected copyright. 46 11d Relevant concomitant care and interventions that are 8 47 48 permitted or prohibited during the trial 49 50 Outcomes 12 Primary, secondary, and other outcomes, including 8 51 the specific measurement variable (eg, systolic blood 52 pressure), analysis metric (eg, change from baseline, 53 final value, time to event), method of aggregation (eg, 54 55 median, proportion), and time point for each 56 outcome. Explanation of the clinical relevance of 57 chosen efficacy and harm outcomes is strongly 58 recommended 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 32

1 2 Participant 13 Time schedule of enrolment, interventions (including 8，9 3 timeline any run-ins and washouts), assessments, and visits

4 for participants. A schematic diagram is highly BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 recommended (see Figure) 7 8 Sample size 14 Estimated number of participants needed to achieve 9 9 study objectives and how it was determined, 10 including clinical and statistical assumptions 11 supporting any sample size calculations 12 13 Recruitment 15 Strategies for achieving adequate participant 9 14 15 enrolment to reach target sample size 16 17 Methods: Assignment of interventions (for controlled trials) 18 For peer review only 19 Allocation: 20 21 Sequence 16a Method of generating the allocation sequence (eg, 7 22 generation computer-generated random numbers), and list of 23 any factors for stratification. To reduce predictability 24 of a random sequence, details of any planned 25 26 restriction (eg, blocking) should be provided in a 27 separate document that is unavailable to those who 28 enrol participants or assign interventions 29 30 Allocation 16b Mechanism of implementing the allocation sequence NA 31 concealme (eg, central telephone; sequentially numbered, 32 33 nt opaque, sealed envelopes), describing any steps to 34 mechanis conceal the sequence until interventions are 35 m assigned 36 37 Implement 16c Who will generate the allocation sequence, who will 7 http://bmjopen.bmj.com/ 38 39 ation enrol participants, and who will assign participants to 40 interventions 41 42 Blinding 17a Who will be blinded after assignment to interventions 8 43 (masking) (eg, trial participants, care providers, outcome 44 assessors, data analysts), and how 45 on October 1, 2021 by guest. Protected copyright. 46 17b If blinded, circumstances under which unblinding is 8 47 48 permissible, and procedure for revealing a 49 participant’s allocated intervention during the trial 50 51 Methods: Data collection, management, and analysis 52 53 54 55 56 57 58 59 60

3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 32 BMJ Open

1 2 Data 18a Plans for assessment and collection of outcome, 8, Table 1 3 collection baseline, and other trial data, including any related

4 methods processes to promote data quality (eg, duplicate BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 measurements, training of assessors) and a 7 description of study instruments (eg, questionnaires, 8 laboratory tests) along with their reliability and 9 validity, if known. Reference to where data collection 10 11 forms can be found, if not in the protocol 12 13 18b Plans to promote participant retention and complete 8 14 follow-up, including list of any outcome data to be 15 collected for participants who discontinue or deviate 16 from intervention protocols 17 18 Data 19 ForPlans peerfor data entry, review coding, security, only and storage, 9 19 20 management including any related processes to promote data 21 quality (eg, double data entry; range checks for data 22 values). Reference to where details of data 23 management procedures can be found, if not in the 24 25 protocol 26 27 Statistical 20a Statistical methods for analysing primary and 10 28 methods secondary outcomes. Reference to where other 29 details of the statistical analysis plan can be found, if 30 not in the protocol 31 32 20b Methods for any additional analyses (eg, subgroup 10 33 34 and adjusted analyses) 35 36 20c Definition of analysis population relating to protocol 10 37 non-adherence (eg, as randomised analysis), and http://bmjopen.bmj.com/ 38 any statistical methods to handle missing data (eg, 39 multiple imputation) 40 41 Methods: Monitoring 42 43 44 Data 21a Composition of data monitoring committee (DMC); 9 45 monitoring summary of its role and reporting structure; on October 1, 2021 by guest. Protected copyright. 46 statement of whether it is independent from the 47 sponsor and competing interests; and reference to 48 49 where further details about its charter can be found, if 50 not in the protocol. Alternatively, an explanation of 51 why a DMC is not needed 52 53 21b Description of any interim analyses and stopping 9 54 guidelines, including who will have access to these 55 56 interim results and make the final decision to 57 terminate the trial 58 59 60

4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 32

1 2 Harms 22 Plans for collecting, assessing, reporting, and 9 3 managing solicited and spontaneously reported

4 adverse events and other unintended effects of trial BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 interventions or trial conduct 7 8 Auditing 23 Frequency and procedures for auditing trial conduct, 9 9 if any, and whether the process will be independent 10 from investigators and the sponsor 11 12 13 Ethics and dissemination 14 Research 24 Plans for seeking research ethics 15 7 16 ethics committee/institutional review board (REC/IRB) 17 approval approval 18 For peer review only 19 Protocol 25 Plans for communicating important protocol 9 20 amendments modifications (eg, changes to eligibility criteria, 21 outcomes, analyses) to relevant parties (eg, 22 23 investigators, REC/IRBs, trial participants, trial 24 registries, journals, regulators) 25 26 Consent or 26a Who will obtain informed consent or assent from 7 27 assent potential trial participants or authorised surrogates, 28 29 and how (see Item 32) 30 31 26b Additional consent provisions for collection and use 7 32 of participant data and biological specimens in 33 ancillary studies, if applicable 34 35 Confidentialit 27 How personal information about potential and 9，10 36 y enrolled participants will be collected, shared, and 37 http://bmjopen.bmj.com/ 38 maintained in order to protect confidentiality before, 39 during, and after the trial 40 41 Declaration of 28 Financial and other competing interests for principal 13 42 interests investigators for the overall trial and each study site 43 44 Access to 29 Statement of who will have access to the final trial 10 45 data dataset, and disclosure of contractual agreements on October 1, 2021 by guest. Protected copyright. 46 47 that limit such access for investigators 48 49 Ancillary and 30 Provisions, if any, for ancillary and post-trial care, 9 50 post-trial care and for compensation to those who suffer harm from 51 trial participation 52 53 Disseminatio 31a Plans for investigators and sponsor to communicate 12 54 n policy trial results to participants, healthcare professionals, 55 56 the public, and other relevant groups (eg, via 57 publication, reporting in results databases, or other 58 data sharing arrangements), including any 59 publication restrictions 60

5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 32 BMJ Open

1 2 31b Authorship eligibility guidelines and any intended use 13 3 of professional writers

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 31c Plans, if any, for granting public access to the full 13 6 protocol, participant-level dataset, and statistical 7 8 code 9 10 Appendices 11 12 Informed 32 Model consent form and other related documentation Supplementary 13 consent given to participants and authorised surrogates file-3 14 materials 15 16 Biological 33 Plans for collection, laboratory evaluation, and 17 Not applicable 18 specimens Forstorage peer of biological review specimens for geneticonly or 19 molecular analysis in the current trial and for future 20 use in ancillary studies, if applicable 21 22 *It is strongly recommended that this checklist be read in conjunction with the SPIRIT 2013 23 Explanation & Elaboration for important clarification on the items. Amendments to the 24 protocol should be tracked and dated. The SPIRIT checklist is copyrighted by the SPIRIT 25 26 Group under the Creative Commons “Attribution-NonCommercial-NoDerivs 3.0 Unported” 27 license. 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from

BMJ Open Page 26 of 32

1 2 3 4 5 6 Supplementary-file-1 7 Ethics committees’ names and approval registration number 8 9 NO. Hospital Ethics committee name Approval number 10 Independent Ethics Committee for Clinical Research of Zhongda Hospital, Affiliated 11 1 Zhongda Hospital, Southeast University 2017ZDSYLL025-P01 with Southeast University 12 For peer review only 13 2 Yantai Yuhuangding Hospital Medical Ethics Committee of Yantai Yuhuangding Hospital [2017]No. 17 14 Beijing Friendship Hospital Affiliated with Medical Ethics Committee of Beijing Friendship Hospital Affiliated with Capital 15 3 2017-P2-083-02 Capital Medical University Medical University 16

Page 27 of 32 BMJ Open

1 2 3 4 5 NO. Hospital Ethics committee name Approval number 6 Chinese PLA Rocket General Hospital (Second 7 15 Ethics Committee of Second Artillery General Hospital PLA KY2017016

16 20 First Hospital of Shanxi Medical University Ethics Committee of First Hospital of Shanxi Medical University 2017-K030

BMJ Open Page 28 of 32

1 血必净注射液治疗脓毒症疗效的多中心盲法随机对照临床研究版本号：v.2.0 2 方案编号：hr20170101 版本日期：2017 年 05 月 10 日 3

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 知情同意书 6 7 8 9 尊敬的病友（家属）： 10 / 11 12 您（ /您的家属）当前所患的是一种严重危害生命的疾病---脓毒症（SEPSIS）。 13 14 我们将邀请您参加一项脓毒症临床研究。在您决定是否参加这项研究之前，请尽 15 可能仔细阅读以下内容，它可以帮助您了解该项临床治疗研究以及为何要进行这 16 17 项研究，研究的程序和期限，参加研究后可能给您带来的益处、不适和风险。如 18 For peer review only 19 果您愿意，您也可以和您的亲属、朋友一起讨论，或者请您的医生给予解释，帮 20 21 助您做出决定是否参加此项临床研究。如有任何疑问请您向负责该项研究的医生 22 23 提出。 24 25 一、研究背景与目的 26 27 1、研究背景 28 29 脓毒症是机体对感染的反应失调而导致危及生命的器官功能障碍。脓毒症和 30 31 脓毒性休克是目前导致住院患者死亡的重要原因，目前针对脓毒症的有效治疗药 32 33 物非常有限，治疗策略仍以抗感染和生命支持治疗为主。血必净注射液是由天津 34 35 红日药业股份有限公司生产，于 2004 年经国家食品药品监督管理局批准应用于脓 36 37 毒症临床治疗国药准字的中药二类新药，适应症为：治疗因感染诱发

( Z20040033) http://bmjopen.bmj.com/ 38 39 的全身炎症反应综合征（脓毒症），也可配合治疗多器官功能失常综合征的脏器 40 41 功能受损期。本项研究是药物上市后临床研究，在常规治疗方案下增加血必净的 42 43 加载治疗，进一步评价血必净注射液治疗脓毒症的的提高生存率方面的有效性。 44 45 2、研究目的 on October 1, 2021 by guest. Protected copyright. 46 47 本研究的目的是为了评价血必净注射液治疗脓毒症，提高生存率的的有效性。 48 二、研究方法 49 50 本项研究为多中心、随机、盲法、对照临床研究，临床研究方案已经由东南 51 52 大学附属中大医院伦理委员会审议批准。将在东南大学附属中大医院等40余家三 53 54 级综合性医院进行，计划纳入1800名符合条件的脓毒症患者自愿参加。 55 56 本项研究采用随机入组方式，您（/您的家属）既有可能入选血必净组（脓毒 57 58 症常规治疗联合血必净注射液），也有可能入选对照组（脓毒症常规治疗联合 0.9% 59 60 氯化钠注射液）。入选不同的组别不会影响医生对您（/您的家属）的常规治疗。

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 32

1 血必净注射液治疗脓毒症疗效的多中心盲法随机对照临床研究版本号：v.2.0 2 方案编号：hr20170101 版本日期：2017 年 05 月 10 日 3

4 本研究将记录您的个人情况和与疾病相关的资料：包括病史（如生命体征）； BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 常规医学检查（如血尿常规、便隐血、肝肾功能、凝血功能、生化检查、血气分 7 8 析等）；为了客观评价病情变化，还将分三次（筛选期、治疗第 3 天、停药后第 1 9 10 天）详细询问并记录您的病情变化，以及入组后第 28 天的预后情况。 11 12 以上的常规治疗措施和医学检查项目，均为脓毒症患者的诊断和治疗所必须 13 14 的临床常规项目。本项研究未涉及特殊检查和治疗项目，也未给患者增加额外的 15 16 负担。 17 18 三、受试者责任For peer review only 19 20 参加本项研究期间需要您（/您的家属）遵从研究方案，接受医生对疾病预后 21 22 情况的随访。 23 24 四、受试者权益 25 26 您（/您的家属）参加本研究是自愿的，您不必为了治疗疾病而必须选择参加 27 本项研究。您可以拒绝参加，或者已经参加了本研究也可以在任何时间、无需任 28 29 何理由退出，您的这些决定均不会受到歧视或报复，也不会影响对您的正常治疗。 30 31 您可以随时了解与本研究有关的信息资料和研究进展。如果您有与本研究有 32 33 关的问题，或您在研究过程中发生了任何不适，或本研究涉及到您权益方面的事 34 35 宜，您可以随时向医生进行咨询。如果您在参加本研究过程中有投诉意见，请联 36 37 系东南大学附属中大医院临床研究伦理委员会。 http://bmjopen.bmj.com/ 38 39 五、参加研究可能的受益 40 41 您（/您的家属）和社会将可能从本项研究中受益。此种受益包括您的病情有 42 43 可能获得改善，以及本项研究可能帮助用于患有相似病情的其他病人。 44 45 无论是否参加本研究，治疗与相关检验都会按脓毒症常规方案进行。同时本 on October 1, 2021 by guest. Protected copyright. 46 47 研究为治疗组免费提供研究药物 5 天的用量并为所有患者减免与本临床研究相关 48 49 的常规检查费用。因此，参加研究项目不会额外增加您（/您的家属）的治疗负担； 50 51 并且对病情观察与治疗将会更加全面、有利。 52 53 六、在研究中可能承受的风险和不适，以及拟采取的风险防范措施 54 55 “血必净注射液上市后再评价临床安全性集中监测研究”共监测了 31913 例 56 57 患者，结果显示血必净注射液的不良反应发生率极低，血必净注射液说明书中已 58 59 载明的不良反应及其发生率（以‰表示）如下：1）全身性损害：过敏性休克 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 32 BMJ Open

1 血必净注射液治疗脓毒症疗效的多中心盲法随机对照临床研究版本号：v.2.0 2 方案编号：hr20170101 版本日期：2017 年 05 月 10 日 3

4 （<0.01‰）、寒战（0.06‰）、发热（0.16‰）、面色苍白、乏力、大汗、抽搐（0.03‰）； BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 2）皮肤损害：皮肤变态反应、皮疹（0.38‰）、瘙痒（0.78‰），皮肤潮红（0.13‰）； 7 8 3）心血管系统：心悸（0.06‰）、紫绀、血压升高或下降、心律失常；4）神经系 9 10 统：头晕（0.06‰）、头痛（0.09‰）；5）呼吸系统：呼吸困难（0.06‰）、胸闷 11 12 （0.22‰）、憋气（0.09‰）、气促（0.03‰）、咳嗽、喉头水肿（<0.01‰）；6） 13 14 消化系统：恶心（0.13‰）、呕吐（0.03‰）、腹痛、腹泻（0.06‰）、肝功能异 15 16 常（0.03‰）；7）泌尿系统：尿频、尿急、尿痛、血尿；8）其他：面部水肿、结 17 18 膜充血、流泪异常、静脉炎。For peer review only 19 20 医生将尽力预防和治疗由于本项研究可能带来的伤害。如果在临床试验中出 21 22 现不良事件，医学专家委员会将会鉴定其是否与治疗药物有关，并将按照我国《药 23 24 物临床试验质量管理规范》的规定对与试验相关的损害提供治疗的费用及相应的 25 26 经济补偿。 27 七、个人隐私保护 28 29 在本项研究的各种医学资料记录中，您（您的家属）的姓名会被拼音缩写代 30 / 31 替。医疗记录及资料将保存在医院，研究者、研究主管部门、伦理委员会经过批 32 33 准可以查阅医疗记录。任何有关本项研究结果的公开报告都不会披露您的个人身 34 35 份。除本项研究外，有可能在今后的其他研究中会再次利用您的医疗记录和检查 36 37 标本。您现在可以声明拒绝除本研究外的其他研究利用您的医疗记录和检查标本。 http://bmjopen.bmj.com/ 38 39 40 41 您（/您的家属）可以选择不参加本项研究，或者在任何时候通知研究者后退 42 43 出而不会遭到歧视或报复，并且任何医疗待遇与权益不会因此而受到影响。 44 45 您（您的家属）参加本项研究是自愿的。您（您的家属）可随时了解与本研 / / on October 1, 2021 by guest. Protected copyright. 46 47 究有关的信息资料，如果有与本研究有关的问题，或发生与研究相关的损伤，或 48 49 有关于本项研究参加者权益方面的问题可以通过（电话号码）与医生联系。 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 32 of 32

1 血必净注射液治疗脓毒症疗效的多中心盲法随机对照临床研究版本号：v.2.0 2 方案编号：hr20170101 版本日期：2017 年 05 月 10 日 3

4 同意声明 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 我已经阅读了上述有关本研究的介绍，而且有机会就此项研究与医生讨论并 7 8 提出问题。我提出的所有问题都得到了满意的答复。 9 10 我知道参加本研究可能产生的风险和受益。我知晓参加研究是自愿的。我询 11 12 问有关研究的细节，提出的所有与研究相关的问题均得到答复，同时我与我的家 13 14 人有充足的时间对此进行考虑，而且清楚的了解： 15 16 ●我可以随时向医生咨询更多的信息。 17 18 ●我所有的个人信息是保密的；我的隐私权、医疗和知情权得到了保障。For peer review only 19 20 ●我可以随时退出本研究，而不会受到歧视或报复，医疗待遇与受益不会受 21 22 到影响。 23 24 ●我同意卫生管理监督部门、伦理委员会或专业学术委员会查阅我的研究资 25 26 料。 27 ●我将获得一份经过签名并注明日期的知情同意书副本。 28 29 最后，我决定同意参加本研究，并尽量遵从医嘱。 30 31 32 33 34 受试者或法定代理人签名：日期：年月日 35 36 联系电话： 37 http://bmjopen.bmj.com/ 38 签名者与受试者的关系： 39 40 41 42 我确认已准确向受试者解释了本研究的详细情况，包括其权利以及可能的受 43 44 益和风险，并对其提问进行了解答，受试者自愿参加该研究，已给其一份签署过 45 on October 1, 2021 by guest. Protected copyright. 46 的知情同意书副本。 47 48 研究者签名：日期：年月日 49 50 研究者电话： 51 52 53 54 55 56 57 58 59 60

4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from

Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP): protocol for a randomized controlled trial

Journal: BMJ Open ManuscriptFor ID peerbmjopen-2018-028664.R2 review only Article Type: Protocol

Date Submitted by the 17-Jul-2019 Author:

Primary Subject Intensive care Heading: http://bmjopen.bmj.com/ Secondary Subject Heading: Intensive care, Emergency medicine

Keywords: Xuebijing injection, Sepsis, randomized, controlled trial, protocol

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 32 BMJ Open

1 2 3

4 Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP): protocol for a BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 randomized controlled trial 8 9 10 11 Songqiao Liu1, Chen Yao2, Junhua Zhang3，Yi Yang1 and Haibo Qiu1 and the EXIT-SEP Investigators 12 13 14 15 Correspondence to Dr Haibo Qiu; [email protected], Department of Critical Care Medicine, 16 17 Zhongda Hospital, School of Medicine, Southeast University, Nanjing, Jiangsu, 210009, China 18 For peer review only 19 20 21 Author affiliations 22 23 1 Department of Critical Care Medicine, Zhongda Hospital, School of Medicine, Southeast 24 25 University, Nanjing, Jiangsu, 210009, China. 26 27 2Medical Statistics Office, Peking University First Hospital, Beijing, China, 100034. 28 29 3 Centre for Evidence-based Medicine, Tianjin University of Traditional Chinese Medicine, Tianjin, 30 31 China, 301617. 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 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 32

1 2 3 ABSTRACT

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Introduction Sepsis is a major challenge with high incidence and is associated with high mortality 7 8 worldwide. Current management of sepsis remains mainly supportive except for treatment with 9 10 antibiotics. Both basic research and clinical investigation have shown that the Chinese herbal 11 derived therapeutic Xuebijing (XBJ) injection is beneficial for patients with sepsis. However, the 12 13 quality of evidence supporting the therapeutic use of XBJ in sepsis is limited. The aim of this trial 14 15 is to evaluate the Efficacy of Xuebijing InjecTion for SEPsis (EXIT-SEP), compared with a placebo, 16 17 on the outcome of patients with sepsis in the ICU. 18 For peer review only 19 Methods and analysis In this multicentre, blinded, randomized, controlled trial, we are recruiting a 20 21 total of 1800 subjects who met Sepsis 3.0 criteria. Subjects will be randomized (1:1) to receive XBJ, 22 23 q12h for 5 days or a matching placebo and usual care. The primary outcome is 28 days all-cause 24 25 mortality. Secondary outcomes will be the improvement of Sequential Organ Failure Assessment 26 27 (SOFA) scores, the improvement of the Acute Physiology and Chronic Health Evaluation II (APACHE 28 29 II) score, duration of mechanical ventilation, mortality in ICU and duration of stay in the ICU. 30 31 Investigators, participants, and statisticians will be blinded to the allocated treatment. 32 33 Ethics and dissemination This trial has been approved by all ethics committees of the centers that will 34 35 participate in this trial. The findings of the study will be disseminated in peer-reviewed journals and 36 37 present at conferences. Once this study is complete, the results of this trial may help to provide http://bmjopen.bmj.com/ 38 39 evidence-based recommendations for complementary therapeutic options for patients with sepsis. 40 41 Trial registration number This clinical trial was registered at Clinicaltrials.gov (NCT03238742) and 42 43 the Chinese Clinical Trial Registry (ChiCTR-IPR-17012713). 44 45 Protocol date: 10 May 2017. on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 KEYWORDS 51 52 Xuebijing injection; Sepsis; randomized, controlled trial; protocol 53 54 55 56 57 58 59 60 2

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 32 BMJ Open

1 2 3 Strengths and limitations of this study

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 32

1 2 3 INTRODUCTION

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 Sepsis is a common and devastating condition that is responsible for more than 6 million deaths 7 8 worldwide annually and is the leading cause of death in critically ill patients[1]. Half of all sepsis 9 10 patients are treated in the intensive care unit (ICU), representing more than 10% of all ICU 11 admissions[1, 2]. The World Health Organization (WHO) reports that sepsis affects more than 30 12 13 million people worldwide every year[3]. The burden of sepsis is most likely highest in low- and 14 15 middle-income countries. 16 17 The pathogenesis of sepsis includes an overwhelming inflammatory response, endothelial 18 For peer review only 19 injury, altered coagulation, dysregulated immune response, and sepsis-associated 20 21 immunosuppression, resulting in acute life-threatening organ dysfunction. During the past 30 22 23 years, the understanding of the host immune response has advanced considerably, and the more 24 25 than 100 therapeutic clinical trials that have been conducted, no approved treatment options are 26 27 currently recommended for sepsis [4][5]. Although there are improvements in supportive care 28 29 strategies, mortality still remains unacceptably high. Some adjuvant treatments for sepsis have 30 31 been developed at great expense but failed to significantly improve mortality[6, 7]. 32 33 There is accumulating evidence to suggest that Xuebijing (XBJ) injection has the potential to 34 35 improve the outcomes for critically ill patients with sepsis. XBJ, a Chinese herb-derived therapeutic, 36 37 has been approved for sepsis treatment in critically ill patients (China Food and Drug http://bmjopen.bmj.com/ 38 39 Administration; Beijing, China, No. Z20040033). The main components of XBJ include amino acids, 40 41 phenolic acids, flavonoid glycoside, terpene glycoside, and phthalides. XBJ has effects on at least 42 43 ten sepsis/inflammation pathways and its 21 major active ingredients regulate 550 targets (HRAS, 44 45 GSK3B, BTK, AK, et al.)[8]. Basic research of XBJ suggests that XBJ has significant on October 1, 2021 by guest. Protected copyright. 46 47 immunomodulatory, anti-coagulation and anti-inflammatory activity. Animal studies and ex vivo 48 49 studies have suggested that the effects of XBJ inhibit the expression of HMGB1, inhibit 50 proinflammatory cytokines secretion and ameliorate sepsis-induced lung injury[9-12]. 51 52 Clinical trials have demonstrated that XBJ improved outcomes and a meta-analysis showed 53 54 that XBJ was associated with high survival rate (RR 0.62, 95% CI 0.51-0.76 P < 0.000 01, I2 = 0%)[13, 55 56 14]. These clinical trials, however, were conducted with small sample sizes, even when pooled, 57 58 limiting the evidence of potential benefit in patients with sepsis [13, 15]. Therefore, the large RCT 59 60 is needed to access the benefit of XBJ on patients with sepsis. We hypothesize that XBJ will reduce 4

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 32 BMJ Open

1 2 3 all-cause mortality than the placebo among early sepsis patients. Thus, we aim to conduct a large-

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 scale, blind, randomized controlled trial (RCT) to evaluate the effects of XBJ in adult sepsis patients 7 8 in the ICU. 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 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 32

1 2 3 METHODS

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 32 BMJ Open

1 2 3 4. Severe liver and kidney dysfunction (single liver or kidney SOFA score ≥ 3 points);

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 5. Use of an immunosuppressant or having an organ transplant within the previous 6 months; 7 8 6. Participating in other clinical trials in the previous 30 days. 9 10 11 Randomization and allocation concealment 12 13 Randomization is generated centrally using an interactive web response system (IWRS). When an 14 15 eligible participant has enrolled this subcentre, the drug administrators will log into the central 16 17 randomization system, where a random number is produced. The independent drug 18 For peer review only 19 administrators receive group information based on the random number, and then they assign the 20 21 study drug to the nurses for administration. 22 23 24 25 Interventions 26 27 The participants will receive XBJ (normal saline 100 ml + XBJ 100 ml, q12h, intravenous infusion 28 29 for 80 min) in the XBJ group and the solvent only (normal saline, 200 ml, q12h, also administered 30 31 for 80 minutes) in the placebo group as previous study[17]. XBJ injection includes a 10 ml/ampule, 32 33 packaged in 5 ampules/container. They were manufactured by a Good Manufacturing Practice 34 35 (GMP) certified company in China (Tianjin Chase Sun Pharmaceutical Co., Ltd., Z20040033). 36 37 Patients in both groups will receive standard care by the attending physician according to the http://bmjopen.bmj.com/ 38 39 International Guidelines for Management of Sepsis and Septic Shock[1]. The patients will receive 40 41 the study drug within 12 h of enrolment, for a 5 day treatment period. According to the 2016 42 43 guidelines on sepsis /sepsis shock, the routine care should include early fluid resuscitation, 44 45 antimicrobials anticoagulants, nutritional support, and other treatment. Ulinastatin and Tanreqing on October 1, 2021 by guest. Protected copyright. 46 47 injection are not allowed during the study period. Use of all drugs including antibiotics, if any, 48 49 should be documented in the Case Report Form (CRF). 50 51 52 Blinding 53 54 Patients, families all bedside clinicians (research staff, site investigators and physicians) and 55 56 statistician will be blinded to study randomization and allocation. Opaque brown infusion bags and 57 58 infusion devices for both groups will be visually inspected by the pharmacy to ensure identical 59 60 appearance. The independent drug administrators will receive group information based on 7

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 32

1 2 3 randomization system; they will then assign the study drug to the bedside nurses to administer.

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 The mixture of drugs will be prepared by the nurses in a separate room. This mixture of drugs will 7 8 be the standard procedure for XBJ, and all the nurses will have experience in performing this task. 9 10 The nurses will also sign a confidentiality agreement about patient allocation. Outcome 11 adjudicators and data analysts will be blinded to the patient allocation. 12 13 14 15 Outcomes 16 17 The primary outcome will be 28-day all-cause mortality. Secondary outcomes will include mortality 18 For peer review only 19 in the Intensive Care Unit (ICU), the improvement of Sequential Organ Failure Assessment (SOFA) 20 21 scores, the difference of the Acute Physiology and Chronic Health Evaluation II (APACHE II) score 22 23 (between the day enrolled in the study and the 7th days), duration of mechanical ventilation, 24 25 duration of stay in the ICU, concentration of C-reactive protein, concentration of Procalcitonin, 26 27 percentage of Human Leukocyte Antigen-DR. 28 29 30 31 Follow-up 32 33 Data will be recorded during the follow-up period according to the multiple time-points. The 34 35 details are shown in Table 1. 36 37 1. Screening period (day 0): before recruitment; http://bmjopen.bmj.com/ 38 39 2. Intervention period (day 1-5): data will be recorded every day during follow-up; 40 41 3. The period after the intervention (within 28 days after treatment): follow-up at day 28 for 42 43 survival data. If the patient is discharged, we will contact them via telephone or a short messaging 44 45 service. on October 1, 2021 by guest. Protected copyright. 46 47 48 Sample size calculation 49 50 We determined that enrolment of 1800 participants would provide the trial with 80% power to 51 52 detect an absolute difference of 6 percentage points decrease in 28-day all-cause mortality of 24.3 53 54 % with a two-sided P value of 0.05. This calculation allowed for a rate of withdrawal and loss to 55 56 follow-up of 15%. We estimated the 28 days mortality rate of 24.3% in a previous observational 57 58 CHESS study of 2322 sepsis/septic shock patients with the SOFA score of 2-13 in China. 59 60 8

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 32 BMJ Open

1 2 3 Study organization

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 The design and implementation of a multicentre study require collaboration among several 7 8 organizations. Quality control will be undertaken by these organizations: ① Expert committee: 9 10 comprises clinical experts, statisticians, and quality control experts who will be responsible for 11 clinical research methodology and for resolving the key issues in the implementation of the study. 12 13 ②Executive committee: the main staff from the members of the expert committee. The 14 15 responsibilities of the executive committee include the design of the clinical research trial, 16 17 selecting the cooperative hospitals, and providing a training course including a manual of 18 For peer review only 19 instructions. ③ DSMB(Data Safety and Monitoring Board): an independent team that evaluates 20 21 the safety outcomes and AEs and submits a review proposal. ④ Data management and statistical 22 23 analysis: performed by trained staff and statisticians. ⑤ Quality control: sites and researchers will 24 25 be monitored and inspected regularly by two CRO, following the standard protocol throughout the 26 27 process. This trial will undergo normative monitoring and inspection throughout the entire process. 28 29 30 31 Data management 32 33 Qualified sites and researchers are crucial factors that ensure the quality of a clinical trial. The sub- 34 35 centers should be qualified by the “Good Clinical Practice (GCP) training” of the State Food and 36 37 Drug Administration (SFDA) for compliance in the training. Qualified researchers should http://bmjopen.bmj.com/ 38 39 understand the detailed contents of the protocol. The data filled in the CRF by researchers should 40 41 be accurate, complete, timely, and reliable. 42 43 Data management will be performed by trained staff at each participating centre using the 44 45 electronic data system. The quality of the trial data management will be guaranteed by the on October 1, 2021 by guest. Protected copyright. 46 47 reliability, access control, and traceability of the system. Data collection will be restricted to those 48 49 variables necessary to define baseline patient characteristics (demographics, sepsis diagnosis, 50 concurrent medical conditions and comorbidities, inclusion and exclusion criteria, the severity of 51 52 illness and organ dysfunction scores, and laboratory results), the delivery of the study drugs, 53 54 potential confounding co-interventions, and outcomes. Randomized participants will be followed 55 56 up until either death or 28 days after randomization, whichever comes first. Follow-up will be 57 58 conducted by study staff by either direct contact with the patient or their next of kin. Participants 59 60 who withdraw from the study for any reason will be followed up with according to the study 9

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 32

1 2 3 follow-up schedule and data will be analyzed according to the intention to treat (ITT) principle.

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 8 Data analysis 9 10 All analyses will be performed according to the intention to treat principle. Continuous variables 11 will be reported as the means and standard deviations or as medians and interquartile ranges 12 13 according to data distribution. Categorical variables will be reported as proportions. We will 14 15 compare the data on the primary outcome of 28-day all-cause mortality using an unadjusted chi- 16 17 square test for proportions or a logistic regression model. We will report frequency (percentage) 18 For peer review only 19 per treatment group with a risk difference and 95% confidence interval and a corresponding odds 20 21 ratio and 95% confidence interval. The missing data of the primary outcome are filled by the worst 22 23 imputation method, that is, "death" is used to fill the missing data. The secondary binary and 24 25 continuous outcomes will be analysed with the use of logistic regression and linear regression, 26 27 respectively. The rate of death in a time-to-event analysis will be reported with the use of Kaplan– 28 29 Meier plots, and differences in survival will be tested using a Cox proportional-hazards model that 30 31 includes the group variable. The subgroup analyses in relation to the different pathogen and SOFA 32 33 score will be performed. All statistical analyzes will be performed using SAS version 9.4 (SAS 34 35 Institute Inc) with a 2-sided P value of less than .05 considered significant. No adjustment will be 36 37 made for multiple comparisons; therefore, secondary outcomes will be interpreted as exploratory. http://bmjopen.bmj.com/ 38 39 40 41 Patient and public involvement 42 43 No patient or public was involved in the present study. Upon the completion of this trial, a journal 44 45 article manuscript will be prepared to present the trial results. on October 1, 2021 by guest. Protected copyright. 46 47 48 DISCUSSION 49 50 This trial is designed as a multicentre blinded RCT. The completion of this trial will provide 51 52 definitive and accurate evidence on the effectiveness of XBJ injection to reduce mortality of 53 54 patients with sepsis. 55 56 XBJ injection, an intravenous preparation approved by the China Food and Drug Administration 57 58 (China FDA) in 2004, has been evaluated as a promising Chinese herbal derived therapeutic drug 59 60 in the management of sepsis, by reducing the inflammatory mediators such as TNF-α, IL-1, IL-6 10

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 32 BMJ Open

1 2 3 and IL-8, improving the immune function, and protecting the vascular endothelial cells, which

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 correspond to the major pathogenesis of sepsis[18]. XBJ injection has pharmacological effects on 7 8 antagonizing endotoxin, inhibiting inflammatory mediators, improving coagulation function and 9 10 microcirculation, protecting endothelial cells, and regulating the immune response. A meta- 11 analysis showed that XBJ reduced all-cause mortality. However, current clinical evidence 12 13 supporting the efficacy of XBJ in the treatment of sepsis, including RCTs, is subject to limitations 14 15 such as a non-representative patient population, small sample size, and concomitant use of 16 17 immunomodulatory agents. As a result, the efficacy of XBJ on the mortality of patients with sepsis 18 For peer review only 19 warrants validation by large, well-conducted RCTs in the ICU. 20 21 This trial uses the 28-day mortality as the primary outcome. At the time of preparation of this study, 22 23 patients who met the criteria of Sepsis 3.0 were reported to have a hospital mortality rate of 20% 24 25 in developed countries. Previous studies suggest that ICU patients with sepsis in middle- and low- 26 27 income countries might have a much higher mortality rate than those in high-income countries, 28 29 possibly due to inadequate resources and poor quality of care[19]. Using data from our cross 30 31 section epidemiology study called ‘CHESS’, we estimate that 28-day all-cause mortality will be 32 33 31.87% in the control group and the 28 days mortality rate of 24.3% in patients of SOFA score no 34 35 more than 13. The ICU mortality rate was greater than 60% in patients with sepsis whose SOFA 36 37 score more than 13. Those patients were too severe to detect the benefit of XBJ in ICU. With a http://bmjopen.bmj.com/ 38 39 sample size of 1800, this large-scale trial will be adequately powered to test the study hypothesis. 40 41 The definition of Sepsis 3.0 focuses on organ function, so except the regular endpoints in clinical 42 43 trials, we would like to employ more secondary endpoint such as organ function. 44 45 There are, however, surrogate outcomes that can be influenced by blinding, which can be on October 1, 2021 by guest. Protected copyright. 46 47 influenced by discharge decisions and safety judgment. Although the independent drug 48 49 administrators and a minority of nurses were able to determine the treatment arm, we 50 demonstrated that concealed drug administration achieved satisfactory levels of blinding in a 51 52 multicentre context. In particular, medications in brown infusion bags and tubes to obscure 53 54 content were sufficient to achieve blinding. 55 56 Recruitment will be completed in December 2019 with a final number of 1800 participants in 45 57 58 medical centers in China. The collection of primary endpoint data will conclude in April 2020. With 59 60 the results of the EXIT-SEP trial, we hope to be able to make evidence-based recommendations for 11

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 32

1 2 3 XBJ use for sepsis.

4 BMJ Open: first published as 10.1136/bmjopen-2018-028664 on 28 August 2019. Downloaded from 5 6 7 8 Ethics and dissemination 9 This trial has been approved by the Research Ethics Boards at Zhongda Hospital (File number: 10 11 2017ZDSYLL025-P01). The research ethics committees of all participant centers approved the study 12 13 protocol(supplementary-material-2) before randomizing patients. The patients’ written informed 14 15 consent(supplementary-material-3) is required at all participating centers. Site investigators will be 16 17 responsible for obtaining informed consent from study participants or legal representative of the family 18 For peer review only 19 member. Subject confidentiality will be assured through data anonymization and controlled access to 20 21 case report forms, the electronic data capture system, and datasets. Any breaches of confidentiality, 22 23 study protocol or AEs attributable to this study will be reported to the research ethics committees. The 24 25 findings of the study will be disseminated at conferences and in peer-reviewed journals. Once this study 26 27 is complete, the results of this trial may help to provide evidence-based recommendations of 28 29 complementary therapeutic options for patients with sepsis. 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 12