We have performed a retrospective study to evaluate whether surgical treatment is beneficial in patients with hepatic metastases from breast cancer. Between September 1985 and September 1998, 25 patients with hepatic metastases (14 solitary and 11 multiple), eight of whom had extrahepatic metastases, underwent hepatectomy. All of the detectable liver metastasis were resected in all of the cases. There were no severe postoperative complications. All but one of the patients received adjunctive polychemotherapy after the hepatectomy. After the hepatectomy, recurrent tumors were detected in 18 of the patients, being located in the liver in 12 (67%) of them. Overall, however, hepatectomy ensured that the liver was clinically recurrence-free for a median of 24 months (range 2-132 months). Eleven patients died of recurrent tumors, two died of other causes and the remaining 12 are currently alive. The 2- and 5-year cumulative survival rates after hepatectomy were 71% and 27%, respectively, and the median survival duration was 34.3 +/- 3.2 months, much better than the period of 8.5 months for another series of patients treated with standard or non-surgical therapies at our institution. The number and the size of hepatic metastases, the interval between treatment of the primary lesion and hepatectomy, and the existence of extrahepatic metastasis were not adverse prognostic factors. In conclusion, our data, although limited and highly selective, suggest that surgical treatment of hepatic metastases from breast cancer may prolong survival in certain subgroups of patients to a greater extent than standard or non-surgical therapies.