Mediastinal lymph node evaluation by computed tomography in lung cancer: An analysis of 345 patients grouped by TNM staging, tumor size, and tumor location

Abstract

To more clearly characterize the role of computed tomography in staging the mediastinal lymph nodes of patients with lung cancer, we analyzed computed tomographic and surgical findings in the chest in 345 consecutive patients with lung cancer who underwent operative staging. Patients were grouped according to the TNM staging system of the American Joint Commission, central or peripheral location of the primary tumor, lobar location of the tumor, and maximum tumor diameter as determined by computed tomography or gross pathology. One third of patients with abnormal findings on the computed tomographic scan did not have mediastinal lymph node metastases. Mediastinal metastases occurred frequently in patients with central cancers (38%). The predictive value of a negative scan in all patients was high (greater than or equal to 90%) except for patients with central T3 lesions (72%), left upper lobe lesions (83%), and central adenocarcinomas (75%). However, only the differences between central T3 and central T2 or T1 lesions, and between central adenocarcinomas and central squamous cell carcinomas, were unlikely to be due to chance alone (p less than 0.05). None of the lobar differences were statistically significant. The frequency of mediastinal metastases in patients with peripheral lesions was 15% (28 of 192 patients); computed tomography correctly identified enlarged mediastinal lymph nodes in all but seven patients. However, there were no true-positive computed tomographic scans in 59 patients with peripheral lesions 2 cm in diameter or smaller; accordingly, we suggest that computed tomography is not indicated for the sole purpose of mediastinal staging in this group. Ninety-four percent of patients in this series undergoing thoracotomy with a curative intent had a curative resection. Only 4% had unresectable lesions; palliative resections were done in 2%.