Background: Routine collection of cytology specimens from bronchial aspirate or washing is thought to increase the sensitivity of bronchoscopy for diagnosing malignant lung lesions. However, the added value of this practice has not been reappraised in a context of changing epidemiology.
Patients and methods: In a retrospective monocenter study, all cytology specimens from bronchial aspirate or washing collected between May 2011 and December 2014 and the corresponding patients' files were reviewed. The final diagnosis of malignancy was based on all available pathologic information.
Results: Bronchoscopy was performed in 2750 patients, and bronchial cytology specimens were collected during 667 procedures, including 474 aspirate or washing cytology specimens collected during conventional (n = 366) or ultrasound-guided (EBUS) (n = 108) bronchoscopy in 455 patients with malignant lung lesion(s). The predominant histologic types were lung adenocarcinoma (43.9%) and squamous cell carcinoma (25.2%), and 271 tumors (59.6%) were endoscopically visible. At least 1 endoscopic sample (bronchial cytology and/or biopsies and/or endobronchial ultrasound-guided samples) was positive for malignancy during 329 (69.4%) of the 474 endoscopies, including 79 samples obtained in nonvisible lesions. Only 67 bronchial cytology specimens proved positive (sensitivity, 14.7%; 95% confidence interval, 11.8%-18.3%), and only 1 specimen (0.2%) produced a diagnosis not made by other samples during the same procedure.
Conclusion: In contrast with older studies, the added value of collecting cytology specimens from bronchial aspirate or washing during bronchoscopy in this series proved negligible, reflecting mainly the increasing prevalence of adenocarcinomas. Abandoning this technique could be considered in centers with similar expertise and patient populations.
Keywords: Diagnostic yield; Futility of diagnostic test; Lung cancer; Lung metastases; Pathology.
Copyright © 2016 Elsevier Inc. All rights reserved.