Methodological challenges in surgical randomized trials: systematic review of rectal cancer studies
In brief
Most recent rectal cancer surgery trials show bias from protocol deviations and outcome measurement
In a review of 35 randomized trials published since 2020, methodological problems specific to surgery were most common in the Cochrane risk-of-bias domains for deviations from intended interventions and for outcome measurement, and also created indirectness in GRADE assessments. These issues stem from the structural nature of surgical trials rather than poor study conduct, highlighting the need to adapt appraisal tools and trial designs for more reliable evidence.
- Journal
- BJS open (Q1)
- Published
- 4 September 2026
- Study design
- Systematic review / meta-analysis of RCTs
- Evidence level
- Level 1, High (CEBM 1a)
- Authors
- Daichi Kitaguchi, Mariano Giménez, Antonello Forgione, Tatsuya Oda, Jacques Marescaux
- PMID
- 42765562
- DOI
- 10.1093/bjsopen/zrag102
Why clinicians should know about it
- Picked for Breast and Endocrine Surgery (top studies of the week, 27 September 2026): Surgical RCTs face bias in deviations and outcome measurement
Abstract
BACKGROUND: Surgical randomized clinical trials (RCTs) have structural features that differ fundamentally from those of non-surgical trials, which may complicate appraisal using standard assessment frameworks. This methodological systematic review used rectal cancer surgery RCTs as a high-complexity exemplar to identify surgically specific methodological challenges and to examine their interaction with established tools for assessing risk of bias and certainty of evidence. METHODS: A systematic literature search was conducted in PubMed, Embase, and the Cochrane Library on 1 November 2025. Eligible studies were RCTs evaluating surgery-related interventions for rectal cancer in adult patients, published in English from 2020 onward, with sufficient methodological detail. Included trials were categorized into predefined, mutually exclusive surgical RCT types. Methodological features were descriptively summarized and conceptually mapped to the domains of the revised Cochrane Risk of Bias (RoB 2) tool and the GRADE framework using both quantitative tabulation and narrative synthesis. RESULTS: Of 1529 records identified, 45 publications reporting on 35 RCTs were included. Trials were classified into five surgical RCT categories. Surgically specific methodological challenges were most frequently observed in RoB 2 domains D2 (bias due to deviations from intended interventions) and D4 (bias in measurement of the outcome), as well as in the GRADE domain of indirectness. CONCLUSION: This review suggests that several challenges encountered when applying RoB 2 and GRADE to surgical RCTs arise from structural characteristics of surgical trial design rather than obvious methodological shortcomings. Contextualizing standard appraisal frameworks within the realities of surgical research may allow for a more nuanced interpretation of existing evidence, inform future trial design, and support more appropriate evidence synthesis.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.