OBJECTIVE: The aim of this study was to evaluate share of each category of cytological assessments in cytological specimens which were taken during bronchoscopy of respiratory tract, and also evaluate sensibility and positive predictive value of lung cancer cytological assessments. STUDY DESIGN: A cross-sectional study. PARTICIPANS AND METHODS: Specimens of participants who were included in the study were taken during bronchoscopy and perthoracalpunctionod lung tissue from January 1st,2015 until December 31st, 2015 at Clinical institute for clinical cytology at Clinical Hospital Centre Osijek. Most common staining in cytology is May Grünwald Giemsa, but when there is suspicion for squamous cell carcinoma, then it is better to use Papanicolaou staining. RESULTS:The study included 608 participants, 427 (70.2%) men and 181 (29.8%) women who were diagnosed with bronchopulmonary lesion by clinical and radiology methods, and thenrequested for further diagnostic analysis. In 728 samples malignant cells were not found, in 164 samples malignant cells werefound, in 32 samples, because of insufficient number of well preserved cells, malignant diagnosis could not be set. Then, samples were marked as suspect. 43 samples were not appropriate for cytological diagnosis. For 41 (7.2%) participants pathological verification was made as well as correlation of cytological and pathohistological assessment. By comparison of cytological and pathohistological assessments, there were 23 (56%) false negative assessmentsfound. For 18 (44%) participants malignant diagnosis was accurately set. Sensitivity of cytological assessment of lung cancer is 44% and positive predictive value is 100%. CONCLUSIONS: In cytological diagnostics multidisciplinary approach is needed and includes clinicians, molecular biologists, radiologists, surgeons and pathologists.