Surgery in the treatment of locally advanced lung carcinoma.

Many differing views exist on the management of locally advanced lung carcinoma. One reason for this is that experience has shown that the most reliable method of determining the extent of the spread outside the lung is by thoracotomy. Having confirmed the extent of invasion by exploration, opinions vary on the subsequent steps in management, from immediate closure of the chest followed by radiation therapy to an attempt at resection in every instance, despite invasion of mediastinal structures. We have followed this latter course, even though we are aware of the advantages of other methods and of the very small return from this form of surgery. In so doing, we emphatically do not advocate a policy of operating on clearly inoperable lesions, using this term in the strictest surgical sense, and, in fact, our high resectability rate suggests that for various reasons many patients are rejected who would be operated upon in surgical centres averaging a more usual resectability rate of about 75% of patients explored. In this contribution I will describe the fate of patients from whom portions of lung carcinoma in inaccessible mediastinal structures could not be removed at the time of resection, suggest indications for resection of this nature, and consider from a small experience the possible benefits from surgery in three conditions arising from extension of tumour beyond the confines of the lung, namely, superior vena caval obstruction, the superior sulcus syndrome, and oesophagobronchial fistula due to penetration of the oesophagus by lung carcinoma. The statistics of this study may be summarized in this way. Of 510 patients, the tumour was removed by some form of resection in 489, and in 21 resection was not possible. The resectability rate is 95-9 per cent. Of the 489 patients whose tumour was resected, 33 died in hospital following operation, an operative mortality of 6-6 per cent. Every patient was operated upon and followed up personally. In the ten-year period under review it is estimated that 4,000 patients were referred for a surgical opinion, although many were clearly inoperable cases. Thus, only one in every eight patients seen was operated upon. From an area in the county of Warwickshire in which the total deaths from lung carcinoma and the number of patients operated upon can be estimated, it is believed that 350 died from lung carcinoma and that 27 patients were explored during the same period from the same area. It appears, therefore, that only one in every 13 cases of lung carcinoma is operable. These 510 patients are a selected group. Concurrently with the practice of resecting locally advanced lung tumours, a number of local excisions and segmental resections have been carried out with satisfactory long-term results. There are strong reasons for treating the individual patient rather than the disease, and, from a consideration of the results, for making definite indications for resection even when total removal of the tumour is not possible. Judged by the extent of its invasion of mediastinal structures, the differences between the tumour which is totally resectable, the tumour which is incompletely resectable, and that which is not considered resectable, may be small. For this reason, in forming indications for partial resection of an advanced tumour, factors other than the extent of mediastinal invasion influence the decision to continue with resection. As suggested by Chamberlain, McNeill, Parnassa, and Edsall (1959), the operation under discussion will be described as a non-curative resection inasmuch as tumour removal is incomplete. They define this operation and promote its use in certain circumstances. It is clear that the higher the resectability rate which is achieved, the greater must be the incidence of resections in which tumour removal is incomplete. Reviewing his experience of both conservative and extensive