P160 A novel PillCam Crohn’s capsule score (Eliakim score) for quantification of mucosal inflammation in Crohn’s disease

R. Eliakim, D. Yablecovitch, A. Lahat, B. Ungar, E. Shachar, D. Carter, L. Selinger, S. Neuman, S. Ben-Horin, U. Kopylov

Sheba Medical Center, Gastroenterology, Ramat Gan, Israel


Capsule endoscopy (CE) is an important modality for monitoring of Crohn’s disease (CD). We recently established that small bowel (SB) inflammation on CE quantified by a Lewis score >350 accurately predicts risk of relapse within 2 years in CD patients in clinical remission. Recently, a novel pan-enteric capsule (PillCam Crohn’s (PCCE), Medtronic, USA) was approved for use. However, no quantitative index for pan-enteric PCCE is currently available. The current study was undertaken as a sub-study of a prospective randomised controlled CURE-CD trial aiming to optimise treatment of CD patients in remission using a PCCE- based treat-to-target approach; the aim of this ancillary study was to compare the correlation and reliability of the novel PCCE inflamatory score (Eliakim score) with the Lewis score as performed by 2 independent experienced CE readers.


The study cohort includes CD patients in clinical remission (CDAI<150). The patients were prospectively enrolled and underwent patency capsule evaluation; if excreted within 30 h, PCCE was performed following bowel preparation. PCCE was repeated every 6 months; if no colonic disease was detected on first PCCE, subsequent examinations were performed without colonic preparation. Each PCCE was independently reviewed by 2 experienced readers (RE (reader 1);UK (reader 2)). All studies were scored using the Lewis score and Eliakim score (comprised of a sum of scores for most common and most severe lesions multiplied by percentage of involvement per bowel segment (3 for small bowel and 2 for colon) with an additional stricture score). Pearson’s and Spearman’ correlation, Cohen’s kappa and inter-rater reliability coefficient (IRC) between the scores and the readers were calculated as appropriate.


Forty PCCE exams were included. The median LS was 225 for both readers (interquartile range (IQR)—157–815 for reader 1, 33- 1125 for reader 2). Both readers identified significant SB inflammation (LS>350) in 17/40 (42.5%) of the patients with strong agreement between the readers (Spearman’s r = 0.87, p < 0.0001). The median PCCE score was 6 (4–14.75) and 4 (2–14.75) for reader 1 and 2, respectively. There was a high IRC between the two readers for LS (0.88, p < 0.0001 for absolute agreement) and PCCE score (0.91, p < 0.0001). For the small bowel, the correlation between LS and PCCE was moderate for reader 1 (Pearson’s r = 0.72, p < 0.0001 and strong for reader 2 Pearson’s r = 0.84, p < 0.0001).


There is a need for a quantitative pan-enteric score for the novel Pillcam Crohns capsule. The presently proposed score, while mandating further clinical validation, has strong inter-reader reliability and moderate-to-strong correlation with the validated small bowel capsule score (LS)