Study objective: Fiberoptic bronchoscopy is the most common modality used to diagnose endobronchial carcinoma. Collection of brushing and washing specimens for cytology is common during bronchoscopy for endobronchial abnormality, but it is unknown if collection of these specimens is cost-effective.
Design: Retrospective review of a computerized database with cost-effectiveness analysis.
Setting: Tertiary care medical center.
Patients: Two hundred one patients undergoing bronchoscopy for endobronchial lung tumor.
Intervention: All patients in the study underwent fiberoptic bronchoscopy that included forceps biopsies, washings, and brushings. In addition to analyzing the sensitivity of forceps-biopsy, washings, and brushings at diagnosing malignancy, we analyzed the cost-effectiveness of three potential specimen collection strategies. These strategies were (1) collection of both washings and brushings in addition to forceps biopsy specimen, (2) collection of either washings or brushings in addition to forceps biopsy specimen, and (3) collection of forceps biopsy specimen only.
Measurements and results: The sensitivity of bronchoscopy, including biopsy, washing, and brushing is 85.3% (95% confidence interval [CI], 80.1 to 90.5%). The sensitivity of forceps biopsy is 80.8% (95% CI, 75.0 to 86.6%). The addition of washings and brushings increases the sensitivity of bronchoscopy from 80.8 to 85.3% (McNemar's p = 0.01). Cost-effectiveness analysis reveals that forceps biopsy plus washing or brushing has a marginal cost-effectiveness ratio of $308 per reduced-quality day avoided compared with forceps alone. Adding an additional cytology specimen has a marginal cost-effectiveness ratio of $5,500 per reduced-quality day avoided.
Conclusions: There is a modest but definite increase in the sensitivity of bronchoscopy in diagnosing endobronchial cancer with the addition of washings and brushings for cytology. Cost-effectiveness analysis reveals that collection of either washings or brushings is probably the best strategy.