Background: There is conflicting data on the utility of endobronchial ultrasound (EBUS)-guided transbronchial needle aspiration (TBNA) for the detection and subtyping of lymphomas. Herein, we present our experience with EBUS-TBNA in suspected lymphoma.
Methods: This was a multicenter retrospective study of subjects with suspected lymphoma who underwent EBUS-TBNA (July 2011 to June 2017). The performance characteristics of EBUS-TBNA were calculated separately for suspected new-onset and recurrent lymphoma as well as for subtyping. We also analyzed the factors predicting the yield of EBUS-TBNA in suspected lymphoma.
Results: Among the 4803 EBUS procedures performed, 92 (1.9%) subjects had either suspected or proven lymphoma; 48 were finally diagnosed to have lymphoma. The diagnostic sensitivities of EBUS-TBNA in new-onset and recurrent lymphomas were 72.7% and 73.3%, respectively. Only 24.2% (8/33) subjects with new-onset lymphoma could be appropriately subtyped. This low yield was possibly due to inadequate material for cell block in 10 subjects, and performance of immunophenotyping and flow cytometry in only 5 and 1 subjects, respectively. Among the suspected cases of recurrence, EBUS-TBNA was sufficient for management in 81.8% (18/22). On a multivariate analysis, no factor (rapid onsite evaluation, needle size, number of lymph nodes sampled, passes per node, and size of the largest lymph node sampled) predicted the diagnostic yield.
Conclusion: EBUS-TBNA is a useful investigation in the diagnostic algorithm of suspected lymphoma as it helps avoid other invasive diagnostic procedures. The sensitivity of EBUS-TBNA in subtyping new-onset mediastinal lymphoma depends on the adequacy of cell aspirate and the judicious utilization of pathologic techniques.