Should the number of pulmonary metastases influence the surgical decision?

To assess, using a large homogeneous retrospective series, the prognostic value of the number of resected pulmonary metastases, and thus, to determine to what extent the number of resectable metastases should influence the surgical decision.

The survival analysis of all patients operated on for pulmonary metastases at a single center, the comparisons of 2 'histologic' groups (sarcoma and carcinoma) and, within each histologic group, of three subgroups with different numbers of resected metastases (1, 2-4, and > or = 5) were performed. The log-rank test was used to compare survival curves.

Among 575 adult patients operated on with curative intent before December 1991, the first operation allowed the complete resection of a known number of histologically proven viable pulmonary metastases in 230 and 151 patients with metastases from carcinoma and sarcoma, respectively. The 5- and 10-year probabilities of survival (Kaplan-Meier) were 37 and 23%, respectively in carcinoma patients, and 31 and 28%, respectively in sarcoma patients (log-rank test: ns). Only the difference between patients with 1 versus 2-4 metastases from carcinoma proved statistically significant (P = 0.02), with 5-year survival estimates of 41 and 25%, respectively. Beside survival, the only significant difference between the subgroups of patients with different numbers of resected metastases was the mean interval between the diagnosis of pulmonary metastases and the resection of pulmonary metastases, which was significantly longer in patients with several metastases in both histologic groups.

In patients with resectable pulmonary metastases from sarcoma or carcinoma, the number of metastases should have little influence on the surgical decision, except for delaying this decision in patients with several metastases until a significant interval, with or without treatment, has shown that metastatic disease remains resectable and confined to the lungs.