The optimal sequence of bronchial brushing and washing for diagnosing peripheral lung cancer using non-guided flexible bronchoscopy.

The optimum sequence of bronchial brushing and washing for diagnosing peripheral lung cancer, defined as an invisible endobronchial tumour, is not clear and requires further study. We prospectively obtained washing samples after brushing in patients with peripheral lung tumours during non-guided flexible bronchoscopy (FB) to investigate the diagnostic yield of these samples and conducted a retrospective review of the prospectively collected data. The study included 166 patients who met the inclusion criteria. The overall diagnostic yield of bronchial brushing and washing for peripheral lung cancer was 52.4%. The diagnostic yields of brushing and washing were 37.3% and 46.4%, respectively, and that of washing was superior according to McNemar's test (p = 0.017, κ = 0.570). Furthermore, washing was diagnostic, whereas brushing was not, in 15.1% of all cases. Comparison of positive washing cytology (brushing) with the respective pathological diagnosis yielded a concordance rate of 88.3% (90.3%), with κ = 0.769 (0.801) (p < 0.001). Performing washing after brushing during non-guided FB is a very safe, cost-effective procedure that may help improve the diagnostic yield in patients with suspected peripheral lung cancer. Our information will also benefit clinicians performing diagnostic bronchoscopy in patients with suspected peripheral lung cancer when fluoroscopic guidance or advanced bronchoscopy techniques are not available.