The value of rapid on-site evaluation during EBUS-TBNA
A.V.. Cardosoa,, , I.. Nevesa, A.. Magalhãesa, M.. Sucenaa, H.. Barrocab, G.. Fernandesa,c
a Pulmonology Department, Centro Hospitalar de São João, Porto, Portugal
b Pathology Department, Centro Hospitalar de São João, Porto, Portugal
c Faculty of Medicine of University of Porto, Porto, Portugal
Abstract
Background

Rapid on-site evaluation (ROSE) has the potential to increase endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) accuracy in the diagnosis of mediastinal lesions and lung cancer staging. However, studies have reported controversial results.

The purpose of our study was to evaluate the influence of ROSE on sample adequacy and diagnostic accuracy of EBUS-TBNA.

Methods

Prospective observational study that enrolled 81 patients who underwent EBUS-TBNA for investigation of hilo-mediastinal lesions or lung cancer staging. The first 41 patients underwent EBUS-TBNA with ROSE (ROSE group) and the last 40 patients without ROSE (non-ROSE group). Sample adequacy and diagnostic accuracy of EBUS-TBNA in both groups were compared.

Results

Adequate samples were obtained in 93% of the patients in the ROSE group and 80% in non-ROSE group (p = 0.10). The diagnostic accuracy of EBUS-TBNA was 91% in ROSE group and 83% in non-ROSE group (p = 0.08). Analyzing the EBUS-TBNA purpose, in the subgroup of patients who underwent EBUS-TBNA for investigation of hilo-mediastinal lesions, these differences between ROSE and non-ROSE group were higher compared to lung cancer staging, 93% of patients with adequate samples in the ROSE group vs. 75% in the non-ROSE group (p = 0.06) and 87% of diagnostic accuracy in ROSE group vs. 77% in non-ROSE group (p = 0.10).

Conclusions

Despite the lack of statistical significance, ROSE appears to be particularly useful in the diagnostic work-up of hilo-mediastinal lesions, increasing the diagnostic yield of EBUS-TBNA.

Keywords
EBUS-TBNA, ROSE, Sample adequacy, Diagnostic accuracy

Metrics

  • Impact Factor: 1.560(2016)
  • 5-years Impact Factor: 1,100
  • SCImago Journal Rank (SJR):0,29
  • Source Normalized Impact per Paper (SNIP):0,685