Survival after Lung Resection for Bronchial Carcinoma

There are many reports of the experience of groups of patients with a bronchial carcinoma treated by lung resection. In this coun'try, for instance, there are series recorded by Sellors, Cruickshank, and Billimoria (1947), Brock (1948), Mason (1949), Taylor and Waterhouse (1950), Price Thomas (1952), and Sellors (1955). But gross yearly survival rates may be a poor guide to prognosis in the individual patient if the duration of survival is greatly influenced by factors not common to all. We have investigated the effects on survival of some of these factors in a relatively large group of patients. The patients were operated on at the Hospitals for Diseases of the Chest by the surgeons of the two hospitals and their assistants between 1940 and 1951. All with a histological diagnosis and a record of the cell type have been included. Records of 531 patients were available; all but 13 (2%) were followed for at least two years or until death. Three were lost sight of during the first two months, two in the third month, and four more in the remaining nine months of the first year. The experience of the patients is expressed as the chances of surviving for each of the first five years after operation. The survival rates have been calculated by the life-table method, which makes full use of the available data where groups of patients have been treated over a period of years and observed for different lengths of time. The validity of the procedure depends on the mortality remaining unchanged throughout the period. This was not true of the immediate mortality after operation. The " operative mortality," arbitrarily defined as death within two months of operation, was high in the early years. Between 1940 and 1946 it was 19%, but in 1947-49 it fell to 16% and in the next two-year period to 10%. The opzrative mortality must therefore be removed from the life-tables, both for investigating the