The Usefulness of Thin-Section Spiral CT in the Evaluation of Mediastinal Lymph Node Metastasis from Non-Small Cell Lung Cancer: AProspective Study and Comparison with Thick-Section Spiral CT

ABSTRACT

PURPOSE: To compare the accuracy of thick-and thin-section spiral CT and to determine whether, in diagnosing mediastinal lymph node metastasis from non-small cell lung cancer, the latter is superior to the former. MATERIALS AND METHODS: Between March 1997 and March 1998, 51 patients with pathologically proven non-s-mall cell lung cancer underwent thoracotomy with full nodal dissection. Thick- and thin-section spiral CT were performed in all patients, with a mean interval of 14 days. The former was performed with 10 mm thick-ness and 10 mm interval, and the latter with 3 mm thickness and 3 mm interval. Mediastinal lymph nodes were localized according to the lymph node mapping scheme of the American Thoracic Society and were considered positive for metastasis if they exceeded 10 mm in short-axis diameter. RESULTS: A total of 227 mediastinal nodal stations in 51 patients were obtained. Of these, 188 stations included in thin-section spiral CT were analyzed and the prevalence of ediastinal nodal metastasis was found to be 10%. On a station-by-station basis, and for thick-and thin-section spiral CT, respectively, the overall sensitivi-ties of mediastinal lymph node metastasis were 32% and 53% (p .05). Although there were no statistically significant differences in sensitivity and specificity according to nodal station, thin-section spiral CT tended to be superior to the thick-section type for stations 7 and 10R in terms of sensitivity, and for stations 4L and 5 in terms of specificity. CONCLUSION: Thin-section spiral CT was more sensitive than thick-section spiral CT is the evaluation of medi-astinal lymph node metastasis from non-small cell lung cancer. This may be due to the higher resolution of the former and its ability to discriminate between lymph node and vessel.