Objective: To compare the diagnostic sensitivity and cytologic findings of the sputum-processing method with those of the traditional sputum smear method.
Study design: From May to December 2001, 300 consecutive sputum samples were collected from 168 patients suspected of having lung cancer in the Chest Department, Taipei Veterans General Hospital. After the sputum smear method, the remaining sputum material was processed by homogenization, filtration and fixation. All the slides were stained with Papanicolaou stain and reviewed by 2 cytologists.
Results: Of the 300 sputum samples, 141 from 79 patients were finally diagnosed as lung malignancies. The mean number of sputum samples was 1.78 per patient. Among the 79 patients, 46 had peripheral lung malignancies (58.2%). The overall diagnostic sensitivities of sputum smear and sputum-processing methods were 29% and 31%, respectively (P = .776). Tumor location and cell types did not change the difference significantly; however, among patients with small cell lung cancer, there was a higher detection rate with the sputum-processing method (50% vs. 20%, P = .350). Eight patients with negative results with the sputum smear had positive results with the sputum-processing method (8 of 79 = 10.1%). Microscopic morphologic differences between the 2 methods were described.
Conclusion: There was no significant difference in diagnostic sensitivity between the sputum smear and sputum-processing methods. However, the sputum-processing method may play a role in patients with small cell lung carcinoma.