PURPOSE
The survival impact of secondary cytoreductive surgery in patients with platinum-sensitive recurrent ovarian cancer was studied.

METHODS
We identified published studies from 1983 to 2021 following our inclusion criteria from MEDLINE, EMBASE, and Cochrane library. To integrate the effect size of single-arm studies, meta-analysis was performed using death rate as a primary outcome. The effect of complete cytoreduction and optimal cytoreduction on survival was evaluated using meta-regression. The pooled death rate was presented with a 95% CI. The publication bias was evaluated with the funnel plot and Egger's test, and sensitivity analysis was performed. To overcome missing death rates, the linear regression model was performed on log-transformed median overall survival (OS) time using study size as a weight.

RESULTS
Thirty-six studies with 2,805 patients reporting death rates were used for this meta-analysis of the 80 eligible studies. There was strong heterogeneity, with the P value of the Cochrane Q test of < 0.0001 and Higgins's I2 statistics of 86%; thus, we considered a random effect model. The pooled death rate was 44.2% (95% CI, 39.0 to 49.5), and both the complete and optimal cytoreductions were associated with better survival outcomes as significant moderators in the meta-regression model (P < .001 and P = .005, respectively). Although 14 studies were located outside the funnel plot, Egger's test indicated no publication bias (P = .327). A sensitivity analysis excluding 14 studies showed similar results. In the linear regression model on the basis of 57 studies, the median OS time increased by 8.97% and 7.04% when the complete and optimal cytoreduction proportion increased by 10%, respectively, after adjusting other variables.

CONCLUSION
Secondary cytoreductive surgery, resulting in maximal tumor resection, significantly prolongs OS in platinum-sensitive recurrent ovarian cancer.