Automated quantitative image cytometry of bronchial washings in suspected lung cancer: comparison with cytology, histology and clinical diagnosis

Automated image cytometry represents a new method for the quantitative analysis of nuclear structure and DNA-content of exfoliative airway epithelial cells. In the present investigation, we examined the correlation between automated cytometry, conventional cytology and histopathology with the final...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Pneumologie (Stuttgart, Germany) Germany), 1999-12, Vol.53 (12), p.583-595
Hauptverfasser: Marek, W, Krampe, S, Dickgreber, N J, Nielsen, L, Muti, A, Khanavkar, B, Müller, K M, Atay, Z, Topalidis, T, Nakhosteen, J A
Format: Artikel
Sprache:ger
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Automated image cytometry represents a new method for the quantitative analysis of nuclear structure and DNA-content of exfoliative airway epithelial cells. In the present investigation, we examined the correlation between automated cytometry, conventional cytology and histopathology with the final diagnosis as the "gold standard". In 142 patients (100 males and 42 females) with suspected lung cancer and 50 controls (COPD, asthma), bronchial washings (5-10 ml) were obtained during bronchoscopy before taking biopsies for cytological and/or histological examinations. The washings were collected in 20 ml Saccomanno's fixative and centrifuged (500 g, 15 min). The cell pellet was resuspended in Saccomanno's solution. Two specimens were stained according to Papanicolaou and another two using the Feulgen reaction with thionine. Image cytometry was performed by means of a special, trainable classifier for exfoliative cells of the respiratory tract, using the Cyto-Sacant (Oncometrics, Vancouver). In the patients with suspected lung cancer we found numerous abnormal nuclei in 97 samples, 36 samples contained normal cells only, and 9 samples were insufficient. In our control group there was no sample with abnormal nuclei, and all washings were evaluable. Compared to the final diagnosis of lung cancer, we found a sensitivity of 90% (92/102) and a specificity of 84% (26/31). For histology sensitivity was 91% (73/80) and specificity 100%, while we found a sensitivity of 92% (92/100) and specificity of 100% for cytology. For automated cytometry the positive predicted value was 95%, the negative predicted value 71%. In the investigation of patients with suspected lung cancer, automated image cytometry of bronchial washings is a sensitive and reliable method for the detection of malignant changes in the tracheobronchial mucosa. The automated procedure seems well suited not only for analysing bronchial washings, but also for a screening procedure.
ISSN:0934-8387