Background: Excisional biopsies are typically used to diagnose lymphoma, but data suggest that endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is equally effective. In this study, we determined whether EBUS-TBNA could accurately diagnose and subtype lymphoma.
Methods: The cases of patients who had undergone EBUS-TBNA for suspected lymphoma were retrospectively reviewed. EBUS-TBNA results were categorized as lymphoma, specific nonlymphoma diagnosis, granulomatous inflammation, or adequate or inadequate lymphocytes with no specific diagnosis. To quantify the ability of EBUS-TBNA to diagnose lymphoma, we used likelihood ratios. To quantify the ability of EBUS-TBNA to diagnose and subtype lymphoma, we calculated sensitivity and specificity. For this analysis, lymphoma that could be subtyped on the basis of EBUS-TBNA was classified as a true positive; lymphoma that could not be subtyped was classified as a false negative.
Results: Of the 181 patients included, 75 (41.5%) were ultimately diagnosed with lymphoma. EBUS-TBNA was able to establish a diagnosis of lymphoma in 63 patients (84%). Granulomatous inflammation diagnosed on the basis of EBUS-TBNA was associated with a low likelihood of lymphoma being present (likelihood ratio, 0.00; 95% confidence interval [CI], 0.00-0.276). Adequate lymphocytes were associated with a low likelihood of lymphoma (LR, 0.25; 95% CI, 0.14-0.49). EBUS-TBNA was able to establish a diagnosis and subtype the lymphoma in 67% (95% CI, 0.45-0.88) of patients with de novo lymphoma and 81% (95% CI, 0.70-0.91) of patients with relapsed lymphoma.
Conclusions: EBUS-TBNA is an effective, minimally invasive diagnostic test for patients with suspected lymphoma and can provide valuable clinical information, even with "negative" results.
Keywords: endobronchial ultrasound–guided fine needle aspiration; lymphoma; mediastinal lymphadenopathy.