Introduction Isolated mediastinal and/or hilar lymphadenopathy (IMHL) is a common indication for EBUS-TBNA. Causes of IHML include granulomatous and malignant disorders or reactive lymphadenopathy due to associated comorbidity (eg emphysema, cardiac failure). A previous study developed a risk stratification model following a negative EBUS-TBNA for IHML to guide future sampling/surveillance0^.1 We conducted a prospective validation of this model.

Methods Consecutive patients undergoing EBUS-TBNA for IMHL at two large EBUS centres in Greater Manchester underwent prospective risk stratification immediately prior to EBUS. Low risk was defined as the presence of at least one comorbidity known to be associated with IMHL AND largest lymph node diameter less than 20 mm, short axis. EBUS-TBNA pathology, Results of any subsequent lymph node sampling and a minimum of six months clinical-radiological follow-up were used to define the final diagnosis in each case.

Results 298 patients (mean age 58.4 years) with IHML underwent EBUS-TBNA between September 2013 and December 2016 (Table 1). Pathological diagnosis of malignancy or granulomatous disease was established by EBUS-TBNA in 98 patients. Of the 200 patients with negative EBUS-TBNA, 143 were ultimately diagnosed with reactive lymphadenopathy, and 57 patients categorised as false negative (46 with sarcoidosis). In the 200 patients with a negative EBUS, all 84 patients prospectively classified as low risk were subsequently diagnosed with reactive lymphadenopathy (NPV 100%). All patients with false negative EBUS were initially classified as high risk (PPV 48%). Only 2/86 patients classified as low risk pre EBUS had a pathological diagnosis at EBUS-TBNA (lung cancer and TB).

Conclusions The risk stratification model following negative EBUS in IMHL has been validated across two EBUS centres demonstrating an excellent NPV. It may provide a simple tool to aid decision making following negative EBUS for IMHL, questioning the role of further sampling or surveillance in such cases. The use of this model as a pre-test decision aid to possibly avoid EBUS-TBNA in low risk patients is a topic for debate and further research.

Reference

1. Evison M, Crosbie PAJ, Morris J, et al. A study of patients with isolated mediastinal and hilar lymphadenopathy undergoing EBUS-TBNA. BMJ Open Resp Res 2014;1:e00040. doi:10.1136/bmjresp-2014-000040