Diagnostic yield of preoperative computed tomography imaging and the importance of a clinical decision for lung cancer surgery.

PURPOSE: This study aimed to evaluate the diagnostic yield of preoperative computed tomography (CT) imaging and the validity of surgical intervention based on the clinical decision to perform surgery for lung cancer or suspected lung cancer.
METHODS: We retrospectively evaluated 1755 patients who had undergone pulmonary resection for lung cancer or suspected lung cancer. CT scans were performed on all patients. Surgical intervention to diagnose and treat was based on a medical staff conference evaluation for the suspected lung cancer patients who were pathologically undiagnosed. We evaluated the relation between resected specimens and preoperative CT imaging in detail.
RESULTS: A total of 1289 patients were diagnosed with lung cancer by preoperative pathology examination; another 466 were not pathologically diagnosed preoperatively. Among the 1289 patients preoperatively diagnosed with lung cancer, the diagnoses were confirmed postoperatively in 1282. Among the 466 patients preoperatively undiagnosed, 435 were definitively diagnosed with lung cancer, and there were 383 p-stage I disease patients. There were 38 noncancerous patients who underwent surgery with a diagnosis of confirmed or suspected lung cancer. Among the 1755 patients who underwent surgery, 1717 were pathologically confirmed with lung cancer, and the diagnostic yield of preoperative CT imaging was 97.8%. Among the 466 patients who were preoperatively undiagnosed, 435 were compatible with the predicted findings of lung cancer.
CONCLUSION: Diagnostic yields of preoperative CT imaging based on clinical evaluation are sufficiently reliable. Diagnostic surgical intervention was acceptable when the clinical probability of malignancy was high and the malignancy was pathologically undiagnosed.