The role of liver resections for noncolorectal, nonneuroendocrine metastases: experience with 142 observed cases.

BACKGROUND: To evaluate the role of liver resection for noncolorectal, nonneuroendocrine metastases, indications and results were retrospectively reviewed in 142 observed patients. METHODS: A curative liver resection was performed in 83 cases (58.5%), and the remaining 59 patients received palliative treatments. The primary tumor site was gastrointestinal in 18, breast in 21, genitourinary in 15, leiomyosarcoma in 10, and other in 19. The mean number of metastases was 1.4. The mean diameter of the nodules was 5.7 cm. Liver metastases were synchronous in 11 (13.3%) cases and metachronous in the remaining 72 (86.7%). RESULTS: There was no operative mortality. Postoperative morbidity was 20.5%. The median postoperative stay was 9.5 days. The 3- and 5-year actuarial survival rate was 49.5% and 34.3% in resected cases, respectively, whereas there were almost no survivors 3 years after diagnosis in unresected cases (P < .05). The 3- and 5-year disease-free survival was 41.4% and 23.8%, respectively. Among the 83 resected cases, the 3- and 5-year actuarial survival was 17.3% and 8.6% for metastases from gastrointestinal tumors, 53.9% and 24.6% from breast cancer, 63.7% and 36.4% from leiomyosarcoma, 50.4% and 37.8% from genitourinary neoplasms, and 55.6% and 42.4% from other sites, respectively. Fifteen patients (18.1%) survived longer than 5 years. CONCLUSIONS: Liver resection is an effective treatment for noncolorectal, nonneuroendocrine metastases; it allows satisfactory long-term survival with an acceptable operative risk in selected patients. Hepatic metastases from gastrointestinal carcinoma have the worst prognosis; those from genitourinary tumors show a better outcome. Patient selection is the key to achieving encouraging results.