Differential cytological diagnosis of typical and atypical lung carcinoids

Background: The potential for diagnosing lung carcinoids using cytological methods has been underexplored, with limited publications available to support evidence-based conclusions. Aim: This study aims to evaluate the of cytological diagnostic capabilities for differentiating typical and atypical lung carcinoids. Materials and Methods: The study included cytological examination results from 18 patients diagnosed with carcinoid tumor at the cytological laboratory of the Altai Regional Oncology Dispensary (Barnaul). The cell material of carcinoid tumors was obtained during diagnostic video bronchoscopy and stained using the Pappenheim method. In all observations, the final diagnosis was the result of histological examination of tumors with immunohistochemical reactions. The obtained results were evaluated using Microsoft Office Excel 2010 and Statistica 10.0. Significant cellular signs (using multivariate analysis) were determined which were assessed to solve diagnostic problems when establishing a cytological diagnosis. Differences between carcinoids were assessed using discriminant analysis with Fisher’s F-test. Results: Carcinoid tumors accounted for 2.72% of the number of neuroendocrine lung neoplasms. Typical carcinoid was diagnosed in 13 patients, atypical carcinoid was observed in 5 cases. The average age at onset of typical carcinoid was 58.46±19.04 years, of atypical carcinoid — 63.2±4.66 years. Cellular features of typical and atypical carcinoids were assessed, which formed the basis for the statistical analysis. To identify the most informative combinations of cellular features, a multivariate analysis was used that determined the main cytological features for carcinoids characterization, using only light microscopy. The values of the first, second, and third factors were determined; the factor value of the cellular feature variants was 96.02% of the sample (variance). In the discriminant analysis, 14 initial tumor cell parameters (discriminant indices) were used to classify patients into one of two groups (typical and atypical carcinoids), on the basis of which the results of the carcinoid study were classified. Step-by-step assessment using Fisher’s F-test revealed differences that amounted to 37.97% only, suggesting that the cellular features are statistically insignificant. Conclusion: The data obtained demonstrate the possibilities of cytological diagnosis of lung carcinoids using only light microscopy at the first stage of patient examination. Cellular features enable to diagnose the carcinoid tumor with an accuracy of more than 96%. However, it is possible to differentiate between typical and atypical carcinoids in less than 40% of cases due to the large number of similar features.