Randomised controlled trials evaluating endometrial scratching: assessment of methodological issues.

Area of Science:

• Reproductive Medicine
• Clinical Trial Methodology
• Evidence-Based Medicine

Background:

• Endometrial scratching is widely adopted for infertility treatment based on favorable findings from numerous small randomized controlled trials (RCTs).
• Systematic reviews incorporating these RCTs have supported the efficacy of endometrial scratching.
• Recent large-scale RCTs have questioned the benefit of endometrial scratching prior to in vitro fertilization (IVF).

Purpose of the Study:

• To critically evaluate the methodological quality of RCTs assessing endometrial scratching.
• To identify common methodological issues in RCTs related to trial registration, statistical analysis, and randomization.
• To determine the potential impact of these issues on the reported effectiveness of endometrial scratching.

Main Methods:

• A systematic assessment of 25 RCTs for IVF and 12 RCTs for intrauterine insemination (IUI)/intercourse was conducted.
• Criteria included adequacy of trial registration, statistical methods, reproducibility of analyses, and baseline characteristic consistency.
• Statistical tests like Monte Carlo simulations and Kolmogorov-Smirnov were used to evaluate randomization.

Main Results:

• A significant majority of RCTs exhibited methodological issues: 32% of IVF RCTs and 17% of IUI/intercourse RCTs had adequate registration.
• Statistical method issues were present in 40% of IVF RCTs and 42% of IUI/intercourse RCTs.
• Reproducibility of statistical results was a problem in 74% of IVF RCTs and 80% of IUI/intercourse RCTs, with IUI/intercourse RCTs showing significant randomization inconsistencies.

Conclusions:

• The majority of RCTs evaluating endometrial scratching suffer from significant methodological deficiencies.
• These flaws, including poor registration and statistical irreproducibility, may compromise the validity of reported findings.
• Further development in critical appraisal methods is needed for reliable evaluation of clinical trials in reproductive medicine.