P130 Faecal calprotectin is a reliable tool to assess endoscopic activity in ulcerative colitis except for short-extent proctitis

Abstract Background Faecal calprotectin (Fcal) is now considered a reliablebiomarker of endoscopic inflammatory activity in patients with ulcerative colitis (UC). However, this biomarker remains imperfect with some false negative in some clinical situations. We aimed to search for endoscopic factors associated with faecal calprotectin level in patients with UC and to identify situations in which this tool might be less efficient. Methods From a database collecting consecutively all the patients with IBD with at least one Fcal dosage in our centre between June 2016 and June 2018, we included all patients who had an endoscopic evaluation within the month before or after dosing without intervention therapeutic during this interval. Fcal level was measured by immuno-chemiluminescence in all patients. Results One hundred and thirty-four patients were included (E1 in 12 patients (9.0%), E2 in 69 patients (51.5%) and E3 in 53 of them (39.5%)). There was a significant correlation between the UCEIS score and Fcal level (r = 0.55, p > 0.001). Fcal level was significantly associated with endoscopic Mayo score with higher values in severe cases (Mayo 0: 32 [15.5–75.3], Mayo 1: 119 [48.0–634.0], Mayo 2: 773 [150.8–1965.0] and Mayo 3: 1200 [572.0–3717.5] μg/g; p < 0.01 for all comparisons). Using an ROC curve, we have defined the best CF thresholds to define an endoscopic Mayo> 1. The performances are presented in the Table 1. In the same way, 100 μg/g was the best Fcal threshold to define a remission endoscopic (endoscopic Mayo score = 0): sensitivity = 79.3% [66.6%–86.3%], specificity = 88.9% [65.3%–98.6%], PPV = 97.9% [92.5%–99.7%], VPN = 40% [24.9%–56.7%]. In multivariate analysis, endoscopic factors associated with the values of CF are: the UCEIS sub-score – vascular pattern (p = 0.001), the UCEIS sub-score – ulcerations (p = 0.013) and the extent of the disease (p = 0.009). CF level in patients with endoscopic Mayo 0 or 1 is lower in the case of low proctitis (<5 cm) than in the case of more extensive RCH (45.5 [19.5–90.5] vs. 214.5 [71.5–730], p < 0.001). While the values of CF were significantly correlated with UCEIS score in patients with proctitis >5 cm (ρ = 0.3665, p < 0.001), they were not low rectal (<5 cm) (ρ = 0.1571, p = 0.71). Conclusion CF is a reliable tool for indirect and non-invasive assessment of activity endoscopic inflammatory disease in UC. Fcal is mainly influenced by severity and extent of the lesions. Its use should be careful in case of proctitis under 5 cm, a situation in which its performances are more questionable.