PubMed1991
Lung metastases from malignant melanoma are frequent and they often inaugurate the metastatic stage. Exceptionally, they present as one or a few nodules, and in the absence of any other secondary lesion these cases raise the problem of surgical eradication. A retrospective multicentre study was carried out in a series of 38 patients and its results were compared to the data obtained from a review of 435 published cases in order to assess the value of surgery in terms of survival and to delimit its indications as closely as possible. Our series of 38 patients comprised 20 men and 18 women aged from 22 to 93 years (mean 51 years, median 55 years). The primary tumour was located in the trunk in 47 p. 100 of the cases; it was nodular in 33 p. 100 and superficial but extensive in 37.5 p. 100. The time elapsed before the metastases appeared varied from 0 to 108 months (median 40 months). Surgery had been radical in 70 p. 100 of the patients and usually limited, tumorectomies and segmentectomies accounting for 51 p. 100 of the operations. RESULTS. In this series the duration of survival varied between 2 and 144 months (mean 26 months, median close to 15 months), with a 20 p. 100 probability of survival at 5 years (fig. 1). Disease free survival varied from 0 to 144 months (mean 22.5 months, median 10.5 months) (fig. 2, curve 1). The parameters of response as regards patients, primary tumour, metastases and treatment were analysed. Response was uninfluenced by sex and slightly influenced by age, with a difference of borderline significance between subjects under and over 50. The primary tumour characteristics did not affect survival, and the features of metastases were of extremely varied importance. The number of operable metastases was not determinant. On the other hand, the presence of mediastinal lesions, either isolated or associated with lung lesions, worsened the prognosis of terms of survival and much more significantly so in terms of remission (fig. 3 and 4). The evaluation of evolutive characteristics, such as date of appearance and tumour doubling time, was inconclusive. Survival was of the same duration after wide and limited surgery, so that tumorectomy or segmentectomy should preferably be performed. The results of surgical treatment were determinant, with a highly significant difference in survival between radical and incomplete surgery (fig. 5 and fig. 2, curve 2). DISCUSSION. The median survival of patients operated upon for lung metastases is diversely evaluated in the literature as 8 to 29 months (table V), the mean figure of 16 months being virtually the same as that of our series. In this, as in most of the previously published series, the maximum duration of survival was beyond 8 to 10 years. The mean survival rate at 5 years is very close to the one we have recorded (20 p. 100) (table V). Compared with other treatments of lung metastases, surgery may be considered as capable of prolonging survival by 6 months; this is not much unless we add the possibility of a 5-year survival in 1 out of 5 operated patients and the possibility of a survival exceeding 8 or 10 years in 2 to 5 p. 100 of the cases. Some prognostic factors seem to constitute positive or negative criteria of operability. This is the case with mediastinal lesions which may consist of a metastasis of metastasis or of a lymph node invasion associated or not with the lung lesion, but in any case correspond to the involvement of more than one site. Mediastinal lesions must be systematically looked for and treated as contraindications of surgery, as shown by the differences in survival recorded in our series. Opinions differ as regards the value of evolutive parameters of the metastasis. For some authors, a more than 5 years interval before the metastasis appears is associated with a good chance of prolonged survival, whereas a less than 6 months or 1 year interval reflects a steadily high progressiveness and in practice precludes surgery. The value of the