The Use of Systematic Reviews and Meta-Analyses When Designing and Reporting Surgical Randomized Controlled Trials: A Cross-Sectional Analysis

Article type
Authors
Lu C1, Ke L2, Lu T1, Lan T3, Guo Y4, Ge L1, Shen R2, Hua Y2, Yang K1
1Evidence-Based Medicine Center, School of Basic Medical Sciences, Lanzhou University
2Department of Liver Surgery, The First Affiliated Hospital, Sun Yat-Sen University
3Department of Pancreatobiliary Surgery, The First Affiliated Hospital of Sun Yat-Sen University
4Organ Transplant Center, The First Affiliated Hospital, Sun Yat-Sen University
Abstract
Background: Well designed and conducted surgical randomized controlled trials (RCTs) are essential to surgery clinical practice, and the prerequisite of planning RCTs should be based on an extensive literature review, especially systematic reviews (SRs) and meta-analysis (MAs). However, knowledge integration before conducting RCTs is not uniformly applied.

Objectives: This cross-sectional study is to evaluate the use of SRs and MAs in conducting surgical RCTs using two high-impact factor surgery journals as data sources, including Annal of Surgery (Ann Surg) and British Journal of Surgery (Br J Surg).

Methods: Studies published in Ann Surg and Br J Surg journals were searched on January 15, 2020 through PubMed database from January 1, 2015 to December 31, 2019. The search terms were as follows: “Ann Surg [Journal]” OR “Br J Surg [Journal]”. All records were downloaded from PubMed and imported into Endnote X9 software, then two independent authors read each title, abstract, full-text, and selected RCTs. Data will be extracted separately and cross-checked, using a standard form: name of first author, year of publication, journal, funding, type of trial, type of intervention, sample size and significance of the study (P-values was classified as positive, < 0.05, or negative, ≥0.05). For each RCT, the number of SRs cited in the introduction, methods, and discussions will be documented by reviewing each publication’s reference list for the following key words: meta-analysis, systematic review, and Cochrane review. Each full-text will be reviewed to determine if the SR can be explicitly cited as justification for conducting the RCT, whether the cited SR can be inferred as forming the basis for initiating the RCT or just be cited in other ways unrelated to the RCT development. All discrepancies will be resolved by consensus. Descriptive data will be presented as frequencies and percentages along with 95% confidence interval (95% CI). All proportions will be calculated alongside their 95% CIs using the Wilson binomial proportion method. All data analyses will be performed using Excel 2019 (Microsoft, WA, USA) and Stata version 13.0 (StataCorp, College Station, TX, USA).

Results and Conclusions: A total of 4404 records were retrieved from PubMed database and 371 RCTs were identified after full-text screening. The data extraction and data analysis are ongoing. Eventually, the results of this study will be submitted to a peer-reviewed journal for publication.

Patient or healthcare consumer involvement: Not applicable.