Conventional transbronchial needle aspiration (TBNA)has a number of limitations. Endobronchial ultrasound (EBUS TBNA) has significantly improved the diagnostic capability of mediastinal lymph node sampling. We present our experience with conventional TBNA with and without rapid onsite evaluation (ROSE) compared with EBUS TBNA with and without ROSE.
We retrospectively reviewed medical records, with IRB approval, on all patients who underwent conventional TBNA between August of 2005 and September 2006 and on all patients who underwent EBUS/TBNA from June of 2007 through mid August of 2008. The collected data included nodal station, the use of onsite pathology (ROSE), and pathologic diagnosis, including presence of lymphocytes. Intially, only those samples that did not contain lymphocytes on cytopathology were sited as nondiagnostic.
One-hundred and sixty patients underwent conventional TBNA or EBUS TBNA. Seventy-seven patients (48%) had EBUS. Sixty-seven (87%) of these patients were diagnostic, compared with 50 (60%) of the conventional TBNA patients, a difference which was statistically significant (p < .001). Twenty-four (31%) of the EBUS patients had ROSE. There was no statistically significant difference in the rate of diagnosis in EBUS patients with and without ROSE (p = .93). Only six (7%) of the patients undergoing conventional TBNA had ROSE and they had a significantly higher rate of diagnosis compared to those without ROSE (p = 0.04). The most frequently sampled lymph nodal station in all patients was 7, followed by 4R. No patients with conventional TBNA had node 11R sampled, whereas 19 EBUS TBNA patients did. Ten EBUS TBNA patients were initially considered nondiagnostic. After nine months of clinical followup, four of these patients were deemed true negatives. The other six patients were diagnosed by another method and considered as false negatives.
Compared to conventional TBNA, early experience with EBUS TBNA achieved a higher diagnostic yield, specifically at level 10 and 11 nodal stations. While ROSE significantly enhanced diagnostic yield in conventional TBNA, this was not the case with EBUS TBNA.
Onsite pathology (ROSE) may not be necessary with EBUS TBNA.
Mutaz Labib, No Financial Disclosure Information; No Product/Research Disclosure Information