P139. The learning curve of intestinal ultrasonography in assessing inflammatory bowel disease – preliminary results
M. Monteleone1, A. Friedman2, F. Furfaro1, A. Dell'Era1, C. Bezzio1, G. Maconi1, 1Gastrointestinal Unit – L. Sacco university Hospital, Clinical Sciences, Milan, Italy, 2The Alfred Hospital and Monash University, Department of Gastroenterology, Melbourne, Australia
Ultrasonography (US) is considered an inexpensive imaging technique that could expedite and improve patient care in inflammatory bowel diseases (IBD). It is still unknown how much training is required in order to become competent to perform intestinal US. The aim of our study is to assess the intestinal US learning curve.
This is a prospective cohort study of 4 trainees, 2 trained in abdominal US (>500 exams) but not in intestinal US, and 2 with little experience in US (<50 exams), undertaking an intestinal US fellowship. Participants initially underwent a theoretical course in the use of US in bowel disorders (8 hours), followed by a one-week training period (approximately 50 bowel examinations). Trainees then independently assessed >100 patients (100–150 patients) with firm (70%) or suspected IBD and recorded the results concerning several US parameters including increased bowel wall thickness (>4 mm), increased small bowel dilatation (>25 mm) and enlarged mesenteric lymph nodes (>8 mm) on a standardised data sheet, which was subsequently compared to that of a physician experienced in US (>30000 bowel examinations) blinded to the previous results. Statistical analysis was performed by K statistics, subdivided in 3 cluster of 33–50 consecutive patients each (k value: 0=poor to 1=excellent agreement).
After accounting for clustering of exams by operator, we showed that physicians previously trained in abdominal US, but not in bowel US, had a variable and increasing agreement with the experienced physician. From the first to the third clusters of patients, k values were: 0.82–0.94, 1 and 1 respectively for detection of increased wall thickness; 0.90–1, 1 and 1 respectively for abnormal dilatation; 0.69–0.81, 0.72–1 and 0.89–1 respectively for enlarged mesenteric lymph nodes. Physicians with no or little experience in US had a variable and increasing agreement with the experienced sonographer: 0.60–0.76, 0.86–0.94 and 1 respectively for bowel wall thickness; 0.72–0.90, 1 and 1 respectively for abnormal bowel dilatation; 0.37–0.46, 1, and 0.93–1 respectively for enlarged mesenteric lymph nodes.
A distinct learning curve was seen for trainees learning intestinal US. Physicians previously trained in abdominal US were able to achieve competency sooner than those with no previous experience. All physicians showed an increasing agreement with experienced sonographer during their training and achieved competency in intestinal US after 150–200 supervised examinations.