Abstract: Objective To investigate the reliability of CT-guided percutaneous lung biopsy pathological results in diagnosing Non- small cell lung cancer (NSCLC). Methods Patients (n=91) were selected whose radiology examination revealed pulmonary masses but were failed to be confirmed with pulmonary lesion through sputum or bronchoscopy. Then, they were diagnosed as non-small cell lung cancer by CT-guided percutaneous lung biopsy. After that, they received surgical operation and biopsies were taken during operations for pathological analysis. Pathological results between from percutaneous lung biopsy (pathological types and using different analysis tools) and from surgical operation were compared. Results The coincidence rate of pathological type is 86.81% (79/91) using pathological results through percutaneous lung biopsy or operation. Compared with the pathological results of operation, the coincidence rate of pathological type using percutaneous lung biopsy show no statistical significance [Squamous cell carcinoma 88.57%(31/35) vs Grandular cell carcinoma 86.27%(44/51)vs adeno-squamous carcinoma 80.00%(4/5)， χ2=0.310， P＞0.05]; the coincidence rate of percutaneous lung biopsy pathological results using different approaches demonstrate statistical significance compared with surgical biopsy pathological results
(χ2=9.698, P＜0.05). The coincidence rate of pathological results obtained by Smear and biopsy [94.4% (51/54)] is higher than that of surgical biopsy pathological results using smear alone [60.0%(6/10)] and by biopsy alone [81.4%(22/27)]. Conclusion Using CT-guided percutaneous lung biopsy, it can produce reliable results on pathological type of non-small cell lung cancer. But the reliability can be affected by different analysis approaches that were hired to exam samples from CTguided percutaneous lung biopsy.

Key words: carcinoma, non-small-cell lung, punctures, biopsy, needle, surgical procedures, operative, specimen, percutaneous lung biopsy, pathological results