Abstract

The purpose of this retrospective study was to evaluate the diagnostic accuracy of bronchoscopic biopsy and transthoracic fine needle aspiration (TFNA) specimens in establishing the specific cell type in primary lung cancer and to study the influence of several factors on this accuracy. The study included 177 patients with lung cancer who underwent thoracotomy between January 2001 and June 2002. The initial diagnostic method was bronchoscopic biopsy in 118 patients (group A) and TFNA in 59 patients (group B). The overall concordance rate was 89.8% (Kappa, 0.63) in group A. The concordance rate was 97% in cases with squamous cell carcinoma and 61.5 % in cases with adenocarcinoma (p=0.02). Cell type agreement rate was higher in central lesions than peripheral lesions in group A (p=0.006). In group B, the overall concordance was 74.6% (Kappa, 0.57). The concordance rate was 70.6% in cases with squamous cell carcinoma and 83.3% in cases with adenocarcinoma (p>0.05). Cell type agreement rate was lower in poorly differentiated tumors than in well-differentiated tumors (p=0.043, OR=10.43) in group B. In conclusion, cell type discordance is a more common problem in TFNA specimens than in bronchoscopic biopsy specimens in our series. Several factors such as tumor localization and degree of cell differentiation can affect cell agreement.