Objective. An optimal treatment strategy for ovarian metastases of ga stric can cer has not been clearly established. The aim of this study was to examine the r ole of a metastasectomy in the management of metachronous Krukenberg tumors afte r curative surgery for gastric cancer. Methods. Among 1235 female patients who h ad undergone a curative gastric resection for stomach cancer between 1987 and 19 98, 54 (4.4%) developed Krukenberg tumors as a first recurrence without evidenc e of a distant metastasis. Of these 54 patients, 33 underwent a metastasectomy w hile 21 did not. The survival duration between the two groups was analyzed and c ompared. Results. The clinicopathological features of Krukenberg tumors as well as those of the primary cancers in the two groups were similar. All 33 patients in the resection group underwent subsequent adjuvant chemotherapy, including the 7 who received intraperitoneal chemotherapy. The 21 patients in the non-resect ion groupwere managed by either systemic chemotherapy (n = 16)-or supportive ca re (n = 5) alone. The median survival duration of all the patients was 9 months (95%confidence interval, 3-15 months). The median survival time in the resecti on group was 17months (95%confidence interval, 10-24 months),which was signifi cantly longer than that in the non-resection group, 3 months (95%confidence in terval, 2-4 months) (P < 0.001). Conclusion. Our results suggest that a metasta sectomy was associated with an improved survival in patients with metachronous K rukenberg tumors from gastric cancer. These data offer a strong argument in favo r of performing metastasectomy for Krukenberg tumors in the absence of an obviou s distant metastasis. Objective of this study was to examine the r ole of a metastasectomy in the management of metachronous Krukenberg tumors afte r curative surgery for gastric cancer. Methods . Among 1254 female patients who undergone a curative gastric resection for stomach cancer between 1987 and 19 98, 54 (4.4%) developed Krukenberg tumors as a first recurrence without evidenc e of a distant metastasis. Of these 54 patients, 33 underwent a The survival duration between the two groups was analyzed and c ompared. Results. The clinicopathological features of Krukenberg tumors as well as those of the primary cancers in the two groups were similar. All 33 patients in the resection group underwent subsequent adjuvant chemotherapy, including the 7 who received intraperitoneal chemotherapy. The 21 patients in the non-resect ion group managed by either sy The median survival duration of all the patients was 9 months (95% confidence interval, 3-15 months). The median survival time in the resection on (n = 16) -or supportive ca re group was 17months (95% confidence interval, 10-24 months), which was signifi cantly longer than that in the non-resection group, 3 months (95% confidence in terval, 2-4 months) (P <0.001) These results suggest that a metasta sectomy was associated with an improved survival in patients with metachronous K rukenberg tumors from gastric cancer. These data offer a strong argument in favor of r of performing metastasectomy for the absence of tumors of gastric cancer.