Primary pulmonary cavitating carcinomas.

A primary lung cancer can produce a cavity in three ways. The first is cavitary necrosis' due to breakdown of the growth itself. The second is stenotic abscess' due to infection and breakdown of the lung parenchyma distal to bronchial obstruction caused by the growth. The third type is `spill-over abscess'. In the present series, necrosis and cavitation were observed in 100 cases out of a total of 632 primary bronchial carcinomas seen at the London Chest Hospital from July 1967 to June 1970. There were 91 males and nine females with an average age of 58·45 years. All except one smoked very heavily and had considerable symptoms. The size of the cavities ranged from 1 to 10 cm and their wall thickness from 0·5 to 3 cm. They were single in 92 cases and multiple (up to four) in eight. In 42 cases, the cancerous cavitation was central, in 38 intermediate, and in 20 peripheral. The segments most frequently affected were the apicoposterior segment of the left upper lobe and the superior segment of the left lower lobe. For descriptive purposes, these cavitating carcinomas were also divided into six broad groups on the basis of radiological and pathological correlations. Neoplastic cells in the sputum were found in 64 cases. Bronchoscopy revealed growth in 42 cases and biopsy was positive in 48. The main microscopic feature was vascular invasion of medium-sized muscular arteries and veins found in the vicinity of every cavitating bronchial carcinoma. Invasion along with tumour plugging of the vessels was observed in 75 cases and thrombosis alone in 55 cases. There were 82 squamous-cell carcinomas, 11 undifferentiated carcinomas of large polygonal-cell type, and seven adeno-alveolar cell carcinomas. The single most important and noteworthy feature in the present series was that oat-cell carcinoma hardly ever undergoes necrosis. Out of a total of 95 cases observed, only three showed necrosis, and this was minimal and characteristically devoid of cavitation. In oat-cell carcinoma vascular invasion and tumour plugging was not observed, though all showed rapid growth and most of them blocked the lobar bronchi completely. In the light of the present study, the main factors responsible for tumour necrosis were found to be gradual bronchial obstruction and associated vascular involvement, though in many cases an inherent propensity of the tumour played a major role.