The accuracy of clinical evaluative intrathoracic staging in lung cancer as assessed by postsurgical pathologic staging.

The records of 103 patients undergoing thoracotomy for carcinoma of the lung between 1985 and 1988 were reviewed. All patients underwent a uniform staging protocol in the construction of a clinical evaluative stage (cTNM). Using information obtained at thoracotomy supplemented by pathological examination a more accurate stage was constructed (pTNM). We have evaluated the accuracy of cTNM staging using the pTNM staging constructed following thoracotomy. In 46.6% patients cTN and pTN concurred. When comparing T subsets alone 81.6% patients remained unchanged. On comparing nodal staging alone 55.3% patients remained unchanged. Pre-operative evaluation underestimated far more commonly than it overestimated. Mediastinal node involvement was not overestimated since any suggestion of such involvement was confirmed by mediastinal exploration. Construction of a cTNM stage remains a crude evaluation, but we remain convinced that the major aspect of pre-operative evaluation is the exclusion of gross mediastinal gland involvement by mediastinal exploration.