Endoscopy 2020; 52(S 01): S220-S221
DOI: 10.1055/s-0040-1704689
ESGE Days 2020 ePoster Podium presentations
Saturday, April 25, 2020 11:30 – 12:00 Pancreatic cancer diagnosis ePoster Podium 3
© Georg Thieme Verlag KG Stuttgart · New York

INTEROBSERVER AGREEMENT BETWEEN TUTOR AND FELLOW AFTER 6 MONTHS OF TRAINING IN ENDOSCOPIC ULTRASOUND

E Dabizzi
1   Unit of Gastroenterology and Operative Digestive Endoscopy, Bellaria – Maggiore Hospital, AUSL Bologna, Bologna, Italy
,
S Landi
1   Unit of Gastroenterology and Operative Digestive Endoscopy, Bellaria – Maggiore Hospital, AUSL Bologna, Bologna, Italy
,
M Bassi
1   Unit of Gastroenterology and Operative Digestive Endoscopy, Bellaria – Maggiore Hospital, AUSL Bologna, Bologna, Italy
,
S Ghersi
1   Unit of Gastroenterology and Operative Digestive Endoscopy, Bellaria – Maggiore Hospital, AUSL Bologna, Bologna, Italy
,
M Riccò
2   AUSL IRCCS di Reggio Emilia, Reggio Emilia, Italy
,
V Cennamo
1   Unit of Gastroenterology and Operative Digestive Endoscopy, Bellaria – Maggiore Hospital, AUSL Bologna, Bologna, Italy
› Author Affiliations
Further Information

Publication History

Publication Date:
23 April 2020 (online)

Also available at
 

Aims Learning curve in Endoscopic Ultrasound (EUS) is extremely variable between operators and competence cannot be assessed just by the number of procedures performed. Interobserver agreement between trainees and tutors could be a reliable first step parameter to evaluate acquired skills during the training.

Aim of the study was to evaluate interobserver agreement between a fellow and his tutor assessing anatomical structures and final diagnosis during video sequences of biliopancreatic EUS.

Methods At the end of a 6-month training period, 60 consecutive videos were prospectively recorded by two operators. A third independent operator randomly selected 40 cases out of the 60 produced. All the videos were finally evaluated, according to a report form that contained quality indicators of EUS procedure as well as diagnostic hypothesis developed by the visualization of the exam. All procedures were conducted under conscious sedation, with a linear-array echo-endoscope (Pentax EG-3870UTK) and patients were examined in the left lateral position. During the EUS training, the fellow attended about 500 procedures and performed about 250 scans. Data were statistically analyzed assessing Fleiss’ kappa.

Results Of the selected videos, 15 were recorded by the trainee and 25 by the tutor. The statistical analysis showed an almost perfect agreement regarding the diagnostic hypothesis (κ = 0.881 p < 0.001), although the agreement on the quality of the video was only moderate (κ = 0.439 p = 0.002). The agreement on diagnosis remained high even by dividing the videos according to the operator who had recorded them (κ = 0.917 vs 0.858).

Conclusions At our knowledge, this is the first reported experience of interobeserver agreement assessment for evaluation of training competence. With a reported good diagnostic agreement between tutor and the trainee after 6 months of training, so it could be considered a reliable method to assess the level of competence in EUS.