==== Front BMJ OpenBMJ OpenbmjopenbmjopenBMJ Open2044-6055BMJ Publishing Group BMA House, Tavistock Square, London, WC1H 9JR bmjopen-2018-02168410.1136/bmjopen-2018-021684Evidence Based PracticeResearch15061694Publication status of completed registered studies in paediatric appendicitis: a cross-sectional analysis Breil Thomas 1Boettcher Michael 2Hoffmann Georg F 13http://orcid.org/0000-0002-5054-5741Ries Markus 3 1 Pediatric Gastroenterology and Hepatology, Center for Pediatric and Adolescent Medicine, Heidelberg University Hospital, Heidelberg, Germany 2 Pediatric Surgery, University Medical Center Hamburg-Eppendorf (UKE), Hamburg, Germany 3 Pediatric Neurology and Metabolic Medicine, Center for Pediatric and Adolescent Medicine, Heidelberg University Hospital, Heidelberg, Germany Correspondence to Professor Markus Ries; markus.ries@uni-heidelberg.de2018 16 7 2018 8 7 e02168411 1 2018 11 6 2018 12 6 2018 © Author(s) (or their employer(s)) 2018. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.2018This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.Objective Appendicitis is considered the most frequent surgical emergency in children. While the management of paediatric appendicitis is evolving, the precise amount of unpublished completed trials, potentially introducing bias into meta-analyses, is unknown. Controversial issues include the appropriate choice of surgical procedures, criteria for diagnosis of appendicitis, the role of antibiotic treatment and pain management. Selective reporting may introduce bias into evidence-based clinical decision-making, and the current, precise extent of unpublished results in paediatric appendicitis is unknown. We therefore assessed the publication status of completed clinical studies involving children registered on ClinicalTrials.gov. Design Cross sectional analysis. STrengthening the Reporting of OBservational studies in Epidemiology criteria were applied for design and analysis. Setting and participants ClinicalTrials.gov was queried for completed studies which were matched to publications on ClinicalTrials.gov, PubMed or Google Scholar. If no publication could be identified, principal investigators were contacted. Interventions/exposure Observational analysis. Primary and secondary outcome measures The proportion of published and unpublished studies was calculated. Subgroup analysis included studies on surgical procedures, diagnosis, antibiotic treatment and pain management. Results Out of n=52 completed clinical studies involving children with appendicitis, n=33 (63%) were published and n=19 (37%) were unpublished. Eighty-three per cent (n=43/52) of clinical trials assessed the above-listed controversial issues. Diagnostic studies were most rigorously published (91% of trials reported), data on surgical procedures, antibiotic and pain management were less transparent. Sixty-six per cent of interventional studies and 60% of randomised studies were published. Median time-to-publication, for example, the delay between completion of the trial until public availability of the results was 24 (IQR 12–36), range 2–92 months. Conclusion Despite the importance of appendicitis in clinical practice for the paediatric surgeon, there remains scientific uncertainty due to unpublished clinical trial results with room for improvement in the future. These data are helpful in framing the shifting paradigms in paediatric appendicitis because it adds transparency to the debate. paediatric surgerypain managementethics (see medical ethics)quality in health carespecial-featureunlocked ==== Body Strengths and limitations of this study This is the first study analysing reporting transparency in clinical research of paediatric appendicitis. Clinical trial registration databases other than ClinicalTrials.gov were not analysed. Unregistered clinical studies were not captured by the present study method. Introduction Appendicitis is considered the most frequent surgical emergency in children with an incidence of 86 cases per 100 000 people.1 2 Efforts are increasing to standardise diagnosis and management, nevertheless controversies continue to exist and challenges remain.3 Although a variety of scoring systems have been developed,4–6 there is still no unequivocal consensus on clinical, laboratory and imaging criteria for diagnosing appendicitis. After the diagnosis is made, usually surgical intervention follows. Recent studies have demonstrated that non-operative management for carefully selected children with acute appendicitis is possible.7–9 Different surgical approaches exist: over the years, laparoscopic appendectomies have widely replaced open traditional procedure.10 Many surveys compare different minimally invasive techniques finding no relevant differences in outcome between three-port or single-incision appendectomies.11 12Optimisation of pain management in children with appendicitis has recently become the centre of several investigations.13 14 Selective reporting of clinical trial results introduces bias into evidence-based clinical decision-making.15–17 The precise extent of bias in paediatric appendicitis is unknown. We therefore assessed the public availability of study results of completed clinical studies involving children with appendicitis registered in the major clinical trial database. We drew particular attention on studies focusing on important controversial issues, that is, surgical procedures, diagnosis of appendicitis, antibiotic treatment and pain management. The aim of this study is to render the current publication status of completed, registered, clinical trials in appendicitis involving children transparent. Methods We determined the proportion of published and unpublished results of studies on paediatric appendicitis that were registered and reported as ‘completed’ on ClinicalTrials.gov. ClinicalTrials.gov database query www.clinicaltrials.gov was accessed through the internet. The detailed search criteria were: keyword ‘appendicitis’ and ClinicalTrials.gov query selection parameters ‘completed studies’ and ‘child’. Data were downloaded. Search for publications of completed studies ClinicalTrials.gov, PubMed and Google Scholar were searched for publications related to the completed registered paediatric studies on appendicitis identified as described above. Keywords for literature research included the NCT number, study title as listed in ClinicalTrials.gov or semantic keywords generated from study title as listed in ClinicalTrials.gov, place of study or principal investigator. If no publication could be found on PubMed or Google Scholar as a next step principal investigators or sponsors were contacted directly and asked to provide the publication of the study to make sure that no published study is missed (see the flow sheet in figure 1). All investigators of unpublished studies (n=19) listed in ClinicalTrials.gov were contacted by email, n=5 replied, none provided published study results. Close of database for the search in each repository was 3 May 2016. Figure 1 Study flow diagram: identification of published and unpublished clinical trials registered on ClinicalTrials.gov involving children with appendicitis. Statistical analysis The following continuous or categorical variables were considered: NCT number, study title, gender and age of participants, study type, study design, condition, intervention, recruitment status, completion date, availability of study results, publication date, sponsor/collaborator and country of sponsor/collaborator. The purpose of clinical studies or the intervention was analysed and trials were categorised into five groups according to their major research topic: (1) surgical procedures, (2) diagnostic criteria, (3) antibiotic treatment, (4) pain management and (5) other. Time-to-publication was calculated as the difference in months between publication date and completion date. Standard methods of descriptive statistics were applied. Missing data were not imputed. All calculations were performed with SAS Enterprise Guide V.5.1. STrengthening the Reporting of OBservational studies in Epidemiology criteria were applied for design and analysis of this cross sectional study.18 Close of database was 3 May 2016. A study flow sheet is provided in figure 1. Patient involvement Patients were not involved in this research project. Results Publication status of studies and trial participants Overall, we identified n=52 completed clinical studies on appendicitis involving children registered on ClinicalTrials.gov. Out of those, n=33 (63%) studies were published and n=19 (37%) studies were unpublished (figure 2, table 1, tables 2A,B). Published trials contained data from n=11 997 study participants. The unpublished trials embody information from n=98 673 patients (figure 3). Median size of published trials was 150 (IQR 73–360), range 21–4000 patients whereas median size of unpublished studies was n=184 (IQR 82–500), range 2–40 000 participants. Three unpublished studies were outliers and had 15 000 (one study) and 40 000 participants each (two studies). Year of completion ranges from 2005 to 2016. Out of n=23 observational studies, n=14 (39%) were published and out of n=29 interventional studies, n=19 (66%) were published. n=25 studies were randomised trials. Out of those, n=15 (60%) were published and n=10 (40%) remained unpublished. The published randomised studies contained data of n=2461 patients, the unpublished studies recruited 1411 patients. All studies involved both genders. The difference in publication rates by country of sponsor/collaborator is shown in table 3. Figure 2 Published and unpublished paediatric appendicitis studies: number of trials by year of completion. Figure 3 Published and unpublished paediatric appendicitis studies: number of patients by year of completion (log scale). Three studies were outliers (see tables 1 and 2A). Table 1 Publication status of studies registered as completed on ClinicalTrials.gov involving children with appendicitis Issue Overall number of studies Number and percentage of published studies Number and percentage of published randomised studies Number of patients enrolled in unpublished studies Number of patients enrolled in unpublished randomised studies Surgical procedure 16 9 (56%) 5 (56%) 1479 786 Diagnostic criteria 11 10 (91%) 2 (100%) 500 0 Antibiotic treatment 11 7 (64%) 4 (57%) 15 275* 275 Pain management 5 2 (40%) 2 (66%) 1119 50 Other 9 5 (56%) 2 (59%) 80 300† 300 Outliers: *Study NCT02311452 was registered as completed on ClinicalTrials.gov and having enrolled 15 000 patients. †Studies NCT02018016 and NCT02017951 were reported as completed and having enrolled 40 000 patients each. Table 2A Characteristics of unpublished clinical trials (n=19) Study identifier Study issue Study design Enrolment Completion date NCT01054417 Surgical procedure Observational 142 July 2011 NCT01678365 Antibiotic treatment Interventional 43 October 2009 NCT01115153 Antibiotic treatment Interventional 150 February 2010 NCT02730585 Diagnostic criteria Observational 500 December 2013 NCT01067937 Pain management Interventional 891 April 2011 NCT02724410 Antibiotic treatment Interventional 82 November 2013 NCT02673528 Surgical procedure Observational 451 January 2016 NCT01967745 Surgical procedure Observational 100 September 2013 NCT02625987 Surgical procedure Interventional 200 September 2015 NCT02580487 Pain management Observational 178 October 2015 NCT00554008 Surgical procedure Interventional 400 January 2011 NCT02311452 Antibiotic treatment Observational 15 000 Not specified NCT02352519 Pain management Interventional 50 June 2015 NCT02714023 Other Interventional 240 September 2015 NCT01424631 Surgical procedure Interventional 2 Not specified NCT02018016 Other Observational 40 000 August 2013 NCT02017951 Other Observational 40 000 August 2013 NCT02687217 Other Interventional 60 May 2013 NCT01515293 Surgical procedure Interventional 184 Not specified Close of database 3 May 2016. Table 2B Characteristics of published clinical trials (n=33) Study identifier Study issue Study design Enrolment Completion date NCT01718171 Other Observational 183 July 2012 NCT00528138 Diagnostic criteria Observational 132 September 2008 NCT01697059 Antibiotic treatment Interventional 73 May 2015 NCT00716703 Diagnostic criteria Interventional 250 October 2005 NCT00888888 Surgical procedure Observational 87 June 2010 NCT02137603 Surgical procedure Interventional 36 December 2014 NCT01096927 Antibiotic treatment Interventional 160 February 2011 NCT00435032 Antibiotic treatment Interventional 128 Not specified NCT00630071 Diagnostic criteria Observational 103 August 2008 NCT00723788 Diagnostic criteria Interventional 21 April 2010 NCT00854815 Diagnostic criteria Interventional 220 June 2012 NCT00195923 Antibiotic treatment Observational 100 January 2007 NCT00462020 Antibiotic treatment Interventional 150 November 2008 NCT00783016 Pain management Interventional 234 May 2011 NCT01698099 Diagnostic criteria Observational 500 September 2012 NCT00677417 Diagnostic criteria Observational 538 May 2008 NCT00477061 Pain management Interventional 71 March 2005 NCT01652170 Diagnostic criteria Observational 2201 February 2014 NCT00414375 Antibiotic treatment Interventional 30 March 2009 NCT00908804 Surgical procedure Interventional Not specified Not specified NCT02304653 Diagnostic criteria Observational 226 November 2013 NCT01572558 Antibiotic treatment Interventional 51 October 2012 NCT00913380 Diagnostic criteria Interventional 891 April 2011 NCT00981136 Surgical procedure Interventional 360 December 2011 NCT01002365 Other Interventional Not specified Not specified NCT01738750 Surgical procedure Interventional 100 December 2013 NCT00413855 Other Interventional 40 June 2005 NCT01260064 Surgical procedure Interventional 150 May 2011 NCT01348464 Surgical procedure Observational 150 December 2011 NCT01734837 Surgical procedure Observational 390 August 2013 NCT02047786 Other Observational 4000 August 2013 NCT00925145 Surgical procedure Observational 32 December 2010 NCT01657565 Other Observational 390 January 2011 Close of database 3 May 2016. Table 3 Published (n=33) and unpublished (n=19) completed studies on paediatric appendicitis by country Countries Published studies (N) Unpublished studies (N) Chile 3 0 Croatia 0 1 Denmark 2 0 Egypt 0 1 Finland 1 1 France 0 1 Germany 3 0 India 0 1 Iran 1 0 Israel 1 1 Italy 1 0 Mexico 0 1 Netherlands 1 1 Scotland 1 2 South Korea 3 0 Spain 0 2 Sweden 1 0 Tunisia 0 1 Turkey 1 1 USA 13 5 USA/Germany 1 0 Time to public availability of results Median time-to-publication, that is, the delay from completion of the trial until public availability of the data was 24 (IQR 12–36) range 2–92 months. More recent studies tended to be published faster than older studies (figure 4). Figure 4 Time-to-publication of paediatric appendicitis studies. ‘FDAAA’ indicates the time line mandated by the Food and Drug Administration Amendments Act of 2007.20 Six studies were completed less than 1 year before close of database. Of those, only one study was published (tables 2A,B). This study, a comparison of surgical versus antibiotic therapy, for appendicitis had positive results and was published within 8 months after completion. Study sponsors Three studies were sponsored or cosponsored by the industry. All these studies were published. All other studies were sponsored by academia. Discussion In order to render clinical research transparency, the AllTrials initiative (www.alltrials.net) called for registration and publication of all results of all clinical trials. In addition, publication of clinical research data is considered an ethical imperative.19 In 2007, the prospective registration and mandatory publication of applicable clinical trials within 1 year of completion became federal law in the USA with the Food and Drug Administration Amendments Act (FDAAA).20 Our analysis demonstrated that 37% of registered completed studies in appendicitis involving children remain unpublished. Results for the majority of interventional studies (66%) and randomised studies (60%) were publicly available. In contrast, most observational studies remained unpublished. Unpublished studies account for data of 98 673 patients overall. However, three large retrospective studies were outliers in this group: two studies were designed to compare outcomes of appendectomies between hospitals in 40 000 patients each (studies NCT02018016 and NCT02017951). The third study included data from 15 000 children to compare oral and intravenous treatment in appendicitis and other inflammatory conditions (study NCT02311452). Without these three studies, unpublished studies contained data from 3673 patients. While the majority of trials, that is, n=43/52 (83%) assessed the four important controversial key issues, the availability of answers to these questions varied: diagnostic studies were most rigorously published, with 91% of all trials published, data on surgical procedures, antibiotic and pain management were less transparent. The lack of transparency may lead to an ongoing ambiguity in the field. All industry-sponsored or cosponsored studies were published which may indicate that industry compliance with FDAAA—which mandates publication within 1 year after trial completion— is high.20 Publication rates varied between countries of sponsor/collaborator (table 3). Without overwhelming evidence for or against an intervention, most surgeons may remain faithful to their successfully proven routine; especially, regarding new surgical techniques or alternative treatment options like antibiotic treatment. New concepts in diagnostic criteria and pain management may be more easily adopted. We speculate that professional mentality may play a role, too, and that it might be possible that less traditional attendings may be more open for new developments and try to adopt new surgical techniques, diagnostic criteria and treatment options like antibiotics instead of surgery in selected patients. In general, scientific uncertainty and ambiguity may explain different approaches in surgery. In order to adopt new evidence, it may be appropriate to update and disseminate internal guidelines regularly. Median time-to-publication in the present study was 24 months. This was twice as long as the deadline of 12 months after completion mandated by FDAAA. In 2007, Hopewell et al reviewed time-to-publication as time between start of trial and time of publication and found that results of positive studies were available after 4–5 years and negative studies after 6–8 years.21 There was a trend towards improvement in terms of timely public availability of results. Similar issues exist in other challenging areas of paediatric medicine, such as epilepsy, autism and liver transplantation.15–17 22 Limitations and directions for future research This study has several limitations. Clinical trial registration databases other than ClinicalTrials.gov were not analysed. Unregistered clinical studies were not captured by the present study method. This present analysis relies on accurate data entry into ClinicalTrials.gov.20 We made all efforts to avoid a study being classified as unpublished by searching the two major medical literature databases, PubMed and Google Scholar, and by contacting investigators directly. Reasons why studies remain unpublished were reviewed by Song et al who identified non-submission of study results due to lack of time or low priority and fear of being rejected by journals as the predominant issue.23 In particular, researchers may encounter difficulties to publish their results, such as lack of interest of the journal or multiple rejections by different journals. While multiple unsuccessful attempts to publish a manuscript in high-impact journals may lead to publication delay,24 the ClinicalTrials.gov webpage allowing timely posting of topline clinical trial results is easily accessible and should be helpful in early dissemination of research findings. Six studies were completed less than a year before close of database, which may be too short to publish in a peer-reviewed journal, however posting the results on ClinicalTrials.gov would have been possible. The present data emphasise the necessity to sensitise paediatric and paediatric surgery residents towards awareness of possible publication bias in paediatric appendicitis. In addition, these data serve as baseline for future publication monitoring.25 We strongly encourage publication of all trial results. In particular, negative data are important in order to prevent subjects being exposed to unnecessary, repeated research. In addition, negative data can be helpful in identifying and correcting perpetual scientific misconception in medical practice, exemplified by the case of fluid replacement with hydroxyethyl starch in critical care patients where data in meta-analyses of clinical trials were unable to further support a positive risk-benefit-ratio for this previously widely used intervention.26 27 The present data serve as quantitative baseline for data transparency in paediatric appendicitis, and it would be of high interest to analyse progress on this issue in the future. Conclusion These data raise awareness that despite the importance of appendicitis in clinical practice for the paediatric surgeon, there remains a certain degree of scientific uncertainty due to unpublished clinical trial results with room for improvement in the future. Therefore, biases may exist in the current literature. These data are helpful in framing the shifting paradigms in paediatric appendicitis because they add transparency to the debate. Supplementary Material Reviewer comments Author's manuscript Contributors: Substantial contributions to conception or design of the work, or the acquisition, analysis, or interpretation of data for the work: TB, MB, GFH, MR. Drafting of the work or revising it critically for important intellectual content: TB, MB, GFH, MR. Final approval of the version to be published: TB, MB, GFH, MR. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: TB, MB, GFH, MR. Funding: We acknowledge financial support by Deutsche Forschungsgemeinschaft within the funding programme Open Access Publishing, by the Baden-Württemberg Ministry of Science, Research and the Arts and by Ruprecht-Karls-Universität Heidelberg. Competing interests: MR received consultancy fees or research grants from Alexion, GSK, Oxyrane and Shire unrelated to the subject of this project. Patient consent: Not required. Provenance and peer review: Not commissioned; externally peer reviewed. Data sharing statement: All data are in the manuscript. ==== Refs References 1. St Peter SD , Sharp SW , Holcomb GW , et al An evidence-based definition for perforated appendicitis derived from a prospective randomized trial . J Pediatr Surg 2008 ;43 :2242 –5 . 10.1016/j.jpedsurg.2008.08.051 19040944 2. Addiss DG , Shaffer N , Fowler BS , et al The epidemiology of appendicitis and appendectomy in the United States . Am J Epidemiol 1990 ;132 :910 –25 . 10.1093/oxfordjournals.aje.a115734 2239906 3. Rentea RM , Peter SDS , Snyder CL Pediatric appendicitis: state of the art review . Pediatr Surg Int 2017 ;33 :269 –83 . 10.1007/s00383-016-3990-2 27743024 4. Alvarado A A practical score for the early diagnosis of acute appendicitis . Ann Emerg Med 1986 ;15 :557 –64 . 10.1016/S0196-0644(86)80993-3 3963537 5. Samuel M Pediatric appendicitis score . J Pediatr Surg 2002 ;37 :877 –81 . 10.1053/jpsu.2002.32893 12037754 6. Boettcher M , Breil T , Günther P The Heidelberg appendicitis score simplifies identification of pediatric appendicitis . Indian J Pediatr 2016 ;83 :1093 –7 . 10.1007/s12098-016-2106-2 27115891 7. Svensson JF , Patkova B , Almström M , et al Nonoperative treatment with antibiotics versus surgery for acute nonperforated appendicitis in children: a pilot randomized controlled trial . Ann Surg 2015 ;261 :67 –71 . 10.1097/SLA.0000000000000835 25072441 8. Tanaka Y , Uchida H , Kawashima H , et al Long-term outcomes of operative versus nonoperative treatment for uncomplicated appendicitis . J Pediatr Surg 2015 ;50 :1893 –7 . 10.1016/j.jpedsurg.2015.07.008 26259556 9. Huang L , Yin Y , Yang L , et al Comparison of antibiotic therapy and appendectomy for acute uncomplicated appendicitis in children . JAMA Pediatr 2017 ;171 :426 10.1001/jamapediatrics.2017.0057 28346589 10. Jen HC , Shew SB Laparoscopic versus open appendectomy in children: outcomes comparison based on a statewide analysis . J Surg Res 2010 ;161 :13 –17 . 10.1016/j.jss.2009.06.033 20031168 11. St Peter SD , Adibe OO , Juang D , et al Single incision versus standard 3-port laparoscopic appendectomy: a prospective randomized trial . Ann Surg 2011 ;254 :586 –90 . 10.1097/SLA.0b013e31823003b5 21946218 12. Zhang Z , Wang Y , Liu R , et al Systematic review and meta-analysis of single-incision versus conventional laparoscopic appendectomy in children . J Pediatr Surg 2015 ;50 :1600 –9 . 10.1016/j.jpedsurg.2015.05.018 26095165 13. Robb AL , Ali S , Poonai N , et al Pain management of acute appendicitis in Canadian pediatric emergency departments . CJEM 2017 ;19 :417 –23 . 10.1017/cem.2016.391 27899166 14. Goyal MK , Kuppermann N , Cleary SD , et al Racial disparities in pain management of children with appendicitis in emergency departments . JAMA Pediatr 2015 ;169 :996 –1002 . 10.1001/jamapediatrics.2015.1915 26366984 15. Breil T , Wenning D , Teufel U , et al An assessment of publication status of pediatric liver transplantation studies . PLoS One 2016 ;11 :e0168251 10.1371/journal.pone.0168251 27992485 16. Lampert A , Hoffmann GF , Ries M Ten Years after the International Committee of Medical Journal Editors' Clinical Trial Registration Initiative, One Quarter of Phase 3 Pediatric Epilepsy Clinical Trials Still Remain Unpublished: A Cross Sectional Analysis . PLoS One 2016 ;11 :e0144973 10.1371/journal.pone.0144973 26735955 17. Mechler K , Hoffmann GF , Dittmann RW , et al Defining the hidden evidence in autism research. Forty per cent of rigorously designed clinical trials remain unpublished - a cross-sectional analysis . Int J Methods Psychiatr Res 2017 ;26 10.1002/mpr.1546 18. Vandenbroucke JP , von Elm E , Altman DG , et al Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration . PLoS Med 2007 ;4 :e297 10.1371/journal.pmed.0040297 17941715 19. Pearn J Publication: an ethical imperative . BMJ 1995 ;310 :1313 –5 . 10.1136/bmj.310.6990.1313 7773049 20. FDAAA . Sec. 801. Expanded clinical trial registry data bank 2007 . 2007 http://www.gpo.gov/fdsys/pkg/PLAW-110publ85/html/PLAW-110publ85.htm 21. Hopewell S , Clarke MJ , Stewart L , et al Time to publication for results of clinical trials . Cochrane Database Syst Rev 2007 ;279 :MR000011 10.1002/14651858.MR000011.pub2 22. Anderson ML , Chiswell K , Peterson ED , et al Compliance with results reporting at ClinicalTrials.gov . N Engl J Med 2015 ;372 :1031 –9 . 10.1056/NEJMsa1409364 25760355 23. Song F , Loke Y , Hooper L Why are medical and health-related studies not being published? A systematic review of reasons given by investigators . PLoS One 2014 ;9 :e110418 10.1371/journal.pone.0110418 25335091 24. Evoniuk G , Mansi B , DeCastro B , et al Impact of study outcome on submission and acceptance metrics for peer reviewed medical journals: six year retrospective review of all completed GlaxoSmithKline human drug research studies . BMJ 2017 ;357 :j1726 10.1136/bmj.j1726 28432051 25. Casadevall A , Ellis LM , Davies EW , et al A framework for improving the quality of research in the biological sciences . MBio 2016 ;7 :e01256-16 10.1128/mBio.01256-16 27578756 26. Hartog CS , Welte T , Schlattmann P , et al Fluid replacement with hydroxyethyl starch in critical care--a reassessment . Dtsch Arztebl Int 2013 ;110 :443 –50 . 10.3238/arztebl.2013.0443 23885279 27. Serpa Neto A , Veelo DP , Peireira VG , et al Fluid resuscitation with hydroxyethyl starches in patients with sepsis is associated with an increased incidence of acute kidney injury and use of renal replacement therapy: a systematic review and meta-analysis of the literature . J Crit Care 2014 ;29 :185.e1 –185.e7 . 10.1016/j.jcrc.2013.09.031