P76 Faecal sulcatone: a new marker of disease activity in ulcerative colitis (UC)?

Le résumé fourni par la source

Introduction UC disease assessment is based on symptoms, faecal and endoscopic analysis. Symptoms are subjective, and faecal calprotectin is elevated when blood is lost (e.g., due to mucosal bleeding in UC but also other conditions such as piles). We investigated whether faecal volatile organic compounds (VOCs) could be used to determine disease activity. Methods 38 newly presenting children with UC and controls with non-IBD intestinal disorders were recruited. Disease severity was assessed by PUCDAI. Stool samples were analysed using calprotectin assays and VOCs by GC mass spec using our optimised protocol (Reade et al, J Anal Bioanal Tech 2014;5:184) Results Median age was 13 years (range 5–16) and 13 (range 3–16) with 66% and 63% males in UC and controls respectively. UC was mild in 29%, moderate in 45% and severe in 26% of cases. More VOCs were identified in controls than cases (Wilcoxon rank-sum test p=0.02) and less VOCs in severe than mild or moderate UC (Kruskal-Wallis, p=0.002). 27 VOCs had significant variances in UC (p<0.05) according to disease activity. The abundance of 23 VOCs was decreased in UC, including 6-methylhept-5-en-2-one (sulcatone), which decreased in a stepwise manner as severity increased (figure 1). Phenol also showed an inverse correlation with disease severity and was included in the model. There was less sulcatone in UC patients than controls (p<0.005). Sulcatone was inversely correlated with PUCDAI (r=-0.58). This correlation was greater than for calprotectin in the same samples (r=0.27). A multiple linear regression model based on sulcatone and phenol was strongly associated with PUCDAI (p=0.0001) and could classify disease severity with 66% accuracy. Increased stool frequency was not significantly associated with sulcatone abundance and so is unlikely to be responsible for the correlation with disease activity seen. Conclusion VOCs such as sulcatone may have the potential to quantitatively assess disease severity in UC. Although any such model would need further work to improve its level of accuracy, we believe that the 66% accuracy shown here is demonstrative of its possible use. Not only did the model correlate with disease severity better than calprotectin, the most used biomarker for IBDs at present, but its accuracy was assumed by comparing it to the PUCDAI scores, which may be biased. In addition, the use of GC-MS was untargeted and could be further improved to quantify sulcatone and phenol with more accurately in future. Such changes could lead to the development of an accurate and quantitative method of assessing disease activity in paediatric UC.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.