Introduction/Background To investigate if minimally invasive surgical techniques lead to higher disease-specific mortality and all-cause mortality at 4.5 years for patients with early-stage cervical and endometrial cancer.

Methodology PubMed/Medline and EMBASE were searched for results from inception to 2021. Prospective randomised controlled trials reporting disease-specific mortality and all-cause mortality at 4.5 years for patients who had minimally invasive or open procedures for early-stage cervical cancer (< II) or endometrial cancer (< III) were selected. Stata 17 was used to conduct a random-effects meta-analysis generating relative risk estimates, odds ratios and 95% CIs. Heterogeneity was examined, small-study effects (Egger’s test), publication bias and study quality (RoB2) assessments were performed.

Results Seven randomised clinical trials between 2001 and 2020 including 4320 patients from 7 countries were included. Two RCTs for cervical cancer and five RCTs for endometrial cancer were selected. Of these, 2584 (60%) patients had minimally invasive surgery, and 1736 (40%) patients had open abdominal surgery. The non-statistically significant risk of all-cause mortality was 18% higher (RR 1.18, 95% CI 0.80, 1.76, I²50.5%) and of disease-specific mortality was 26% higher for patients who underwent minimally invasive surgery compared to open abdominal surgery (RR 1.26, 95% CI 0.83, 1.89, I²21.4%). However, p = 0.403 (all-cause mortality) and p = 0.265 (disease-specific mortality) indicated little evidence against the null hypothesis. There were no small study effects, little evidence of publication bias and study quality was generally high.

Conclusion Based on a systematic review of the literature and meta-analysis of prospective randomised-controlled trials for patients with early-stage cervical and endometrial cancer, minimally invasive surgery could be associated with a non-significant higher risk of all-cause mortality (18%) and disease-specific mortality (26%) at 4.5 years compared to open abdominal surgery. However, as p > 0.05 and the CI included 1, this meta-analysis was inconclusive.