P278 Real-time histology at endoscopy: virtual electronic chromoendoscopy and probe confocal laser endomicroscopy can assess fine details of inflammatory changes in ulcerative colitis patients
M. IACUCCI*1, S. Gui X2, A. Oluseyi1, R. Panaccione1, M. Fort Gasia1, S. Ghosh1
1University of Calgary, Gastroenterology, Calgary, Canada, 2University of Calgary, Pathology, Calgary, Canada
Objective assessment of inflammation is considered increasingly important in the assessment of disease activity and therapeutic response in ulcerative colitis (UC). Although endoscopic assessment is generally the most widely accepted endoscopic method, histological assessment is being standardised to reflect disease activity, even when endoscopic features are subtle. We have reported that such subtle changes can be detected by iSCAN virtual electronic chromoendoscopy (VCE) scoring. Probe confocal laser endomicroscopy (pCLE) provides real-time images of the cellular and vascular elements in the mucosa and submucosa of the intestine and therefore might provide histology-like images, but be less prone to sampling errors inherent in some biopsies. We aimed to determine whether pCLE reflected VCE and histological abnormalities in UC patients.
In the study, 81 patients (73 UC and 8 controls) (male: 48, median age 50 years; range 20–79 yr) were assessed with high-definition–iSCAN virtual chromoendoscopy (VCE) (Pentax, Japan) and pCLE (Cellvizio, Paris) after IV fluorescein. The endomicroscopy findings were graded as follows: A) Crypt architecture, (1) normal, (2) irregular, (3) drop-out, (4) cryptitis, (5)necrosis with crypt abscess; B) Leakage of fluorescein, (1) normal, (2) low density in the lumen of the crypts, (3) amongst the cells, (4) high density in the lumen of the crypts; C) Vessel architecture, (1) normal, (2) branchless, (3) tortuous, (4) dilated; D) Blood flow, (1) normal, (2) low back-and-forth flow, (3) high back-and-forth flow, (4) stagnant. Harpaz grade (1–4) was used to reflect the histological grade of inflammation.
The Mayo endoscopy sub-score was significantly correlated with pCLE score (rs = 0.76, 95% CI 0.65–0.84; p < 0.001). The overall mucosal and vascular pattern iSCAN endoscopic score1 was significantly correlated with pCLE score (rs = 0.83, 95% CI 0.75–0.90; p < 0.0001). pCLE features of leakage of fluorescein (rs = 0.84, 95% CI 0.76–0.89; p < 0.00001), vascular architecture (rs = 0.80, 95% CI 0.69–0.86; p < 0.0001), and blood flow (rs = 0.76, 95% CI 0.64–0.83; p < 0.00001) reflected the endoscopic iSCAN vascular pattern. The Harpaz histology score was significantly correlated with pCLE (rs = 0.57, 95% CI 0.40–0.70; p < 0.0001). The sensitivity of pCLE to detect histological inflammation was 92.6%.
pCLE can assess grade of inflammation in UC patients, both at VCE and histology levels. The leakage of fluorescein, vessels architecture, and blood flow determined by pCLE reflect well the vascular changes seen by VCE. Technological advances in endoscopy are rapidly approximating histology.