Evaluation of survival and mortality in pelvic exenteration for gynecologic malignancies: a systematic review, meta-analyses, and meta-regression study.

OBJECTIVE

Pelvic exenteration is a radical surgery for advanced or recurrent pelvic tumors, requiring careful patient selection and a multi-disciplinary approach. Despite advancements, it remains high-risk, with limited data on outcomes. The present meta-analysis evaluates survival, mortality, and trends to clarify its role in gynecologic oncology.

METHODS

A systematic search was conducted in January 2025 to identify studies on pelvic exenteration outcomes for gynecologic malignancies. Studies with at least 10 patients reporting 5-year overall survival or 30-day mortality were included. Data extracted included patient and surgical characteristics, and a scoring system based on study design, sample size, and center volume was used to include high-quality studies (score ≥3). Poisson regression models were used to analyze the associations between predictors and outcomes, with results expressed as incidence rate ratios and a 95% CI.

RESULTS

A total of 46 studies involving 4417 patients met the inclusion criteria. Most patients underwent pelvic exenteration for cervical cancer (N = 3183). Positive pelvic and aortic nodal involvement were key predictors of reduced 5-year overall survival, decreasing by 3.9% and 5.9% per 1% increase in nodal positivity, respectively. Pelvic wall involvement also significantly reduced survival by 15.9%. The 30-day mortality rate was 5.1%, with sepsis (27.2%) being the leading cause of death. Peri-operative mortality decreased significantly over time, with each year of publication associated with a 2.6% decrease in incidence rate. However, pelvic sidewall involvement and total exenteration increased 30-day mortality by 11.5% and 0.7%, respectively.

CONCLUSIONS

Pelvic exenteration remains a viable but high-risk option for select patients with advanced gynecologic malignancies. Pre-operative assessment and multi-disciplinary planning are essential for optimizing outcomes.