| pubmed-article:17089063 | pubmed:abstractText | Liver resection has been recognized as the best treatment for patients with colorectal liver metastases, but as a curative resection for multiple and bilobar colorectal liver metastases (MBCLM) it is definitely less effective. We clarify predictors of survival for unresectable MBCLM. Potential predictors of overall survival, and the correlation between tumor marker and survival were evaluated for patients with synchronous unresectable MBCLM, including 6 rectal and 17 colon cancers. In univariate analysis, survival in patients with the following parameters were longer than those without them: number of liver metastases (</=10), without lung metastasis and peritoneal invasion, and with a <1.0 ratio of postoperative CEA/preoperative CEA. In multivariate analysis, the numbers of liver metastases (>10) and a >1.0 ratio of postoperative CEA/preoperative CEA were factors of poor prognosis, and patients with two such factors had an even worse prognosis. There was a tendency for correlation between the ratio of postoperative CEA/pre-operative CEA and survival (R=-0.492, P=0.053; y=17.388-3.733x). Thus, we clarified some of the predictors of survival for MBCLM, and the usefulness of serum CEA. | lld:pubmed |
