Transesophageal Ultrasonography for Lung Cancer Staging: Learning Curves of Pulmonologists

Lars Konge, Jouke Annema, Peter Vilmann, Paul Clementsen, Charlotte Ringsted

27 Citations (Scopus)

Abstract

INTRODUCTION:: Accurate mediastinal nodal staging is essential for patients with resectable non-small-cell lung cancer and is achieved by combined endobronchial ultrasound and transesophageal endoscopic ultrasound (EUS). Training requirements for EUS-guided fine-needle aspiration (FNA) for lung cancer staging are unknown. METHODS:: Pulmonologists from Denmark and The Netherlands were enrolled in a dedicated, supervised training program. They performed standardized EUS-FNA procedures for mediastinal nodal analysis and their performances were assessed by EUS experts using a validated EUS assessment tool. Data were collected prospectively and used to plot learning curves and relate these to procedures performed by experienced investigators. RESULTS:: Four participants performed 91 EUS-FNA procedures (range, 19-24). The performances of the participants improved significantly and became more consistent, but were still highly variable even in the latter part of the learning curves. Only two of the participants reached the mean score of experienced operators - after 17 and 23 procedures, respectively. CONCLUSIONS:: Pulmonologists with knowledge of lung cancer staging and experience in bronchoscopy quickly improved their performance of EUS-FNA. However, acquisition of skills varies between individuals, and certification should be based on assessment of performance of multiple cases. Twenty procedures were not enough to secure consistent and competent performance of all trainees.

Original languageEnglish
JournalJournal of Thoracic Oncology
Volume8
Issue number11
Pages (from-to)1402-1408
Number of pages7
ISSN1556-0864
DOIs
Publication statusPublished - Nov 2013

Fingerprint

Dive into the research topics of 'Transesophageal Ultrasonography for Lung Cancer Staging: Learning Curves of Pulmonologists'. Together they form a unique fingerprint.

Cite this