The accuracy and adequacy of rapid on-site evaluation conducted by trained pulmonologist on transbronchial needle aspiration samples: A systemic review and meta-analysis

Abstract

Introduction: The effective deployment of rapid on-site cytologic evaluation (ROSE) requires the availability of a cytologist/pathologist. Several studies investigated the capability of trained pulmonologists to assess the adequacy and validity of transbronchial needle aspiration (TBNA) samples. This study aimed to determine the validity and adequacy of ROSE by pulmonologists (P-ROSE).

Methods: A systematic search was conducted in Embase, Cochrane, Scopus, PubMed, and Web of Science. The validity meta-analysis was conducted using Meta-Disc software V 2.0. A bivariate model was employed to perform the analysis.

Results: 14 studies were included in the review, and the collective sensitivity, specificity, negative likelihood ratio, and positive likelihood ratio with 95 % CI were 89.6 % (95 % CI: 0.827-0.94), 95.9 % (95 % CI: 0.845-0.999), 0.108 (95 % CI: 0.064-0.183), and 21.598 (95 % CI: 5.447-85.636) respectively. P-ROSE demonstrated high accuracy for both diagnosis (80%-89.5 %) and adequacy of 83.1 %. Also, a high K coefficient for diagnosis and adequacy with a range (0.18-0.94) and (-0.02-0.72), respectively.

Conclusion: Our findings showed that P-ROSE had a high sensitivity, specificity, and accuracy, which means that following a brief training session, the pulmonologists can reliably assess prepared slides from TBNA of lymph nodes and lung masses, both in terms of sample adequacy and establishing an initial diagnosis. This can be especially valuable in healthcare settings where a Pathologist/cytopathologist is not consistently available.

Keywords: EBUS-TBNA; Mediastinal lymph nodes; Meta-analysis; P-ROSE; Pulmonologist; ROSE; TBNA.