P202 Performance characteristics of serum FGF19 measurement compared with the Se-HCAT retention test in the diagnosis of bile acid diarrhoea in Crohn’s Disease

Ruiz-Cerulla, A.(1);Blat Serra, R.(1);Sánchez-Pastor, E.(1);Notta, P.C.(2);Rodríguez-Alonso, L.(1);Aràjol Gonzalez, C.(1);Serra Nilsson, K.(1);Antón Güell, S.(1);Serrano Santacruz, I.(3);Luque Gómez , A.(3);Aran, J.M.(3);Rodríguez-Moranta, F.(1);Guardiola Capon, J.(1);

(1)Bellvitge University Hospital, Gastroenterology, L'Hospitalet de Llobregat, Spain;(2)Bellvitge University Hospital, Nuclear Medicine, L'Hospitalet de Llobregat, Spain;(3)IDIBELL, Immune-Inflammatory Processes and Gene Therapeutics Group, L'Hospitalet de LLobregat, Spain;

Background

Bile acid diarrhoea (BAD) is common in patients with Crohn’s disease (CD), but its recognition is challenging since symptoms are similar to those of active inflammatory disease. The current clinical gold standard for diagnosing BAD is the Se-HCAT test, however, it is inconvenient to the patient and has limited availability. FGF19 is a hormone produced in the enterocytes of the ileum in response to absorbed BAs. Serum FGF19 levels are a direct marker of BA absorption, and they are reduced in BAD. The aim of this study was to evaluate the performance of FGF19 as a diagnostic tool for BAD in Crohn’s disease.

Methods

Fasting serum FGF19 levels and Se-HCAT retention were measured before bowel preparation in 63 consecutive Crohn’s disease patients referred for an ileocolonoscopy. Clinical, endoscopic and biologic data were prospectively recorded. BAD was defined as an abdominal retention <10% and severe BAD as <5% at day seven.

Results

BAD was present in 60% of non-resected CD (NR-CD) patients and in 93% of ileal-resected (IR-CD) patients. FGF19 levels were lower in IR-CD patients (median 23 pg/ml; IQR, 3-44) than in the NR-CD patients (61 pg/ml; IQR, 18-121) (p = 0.02). FGF19 levels were inversely related with ileal resection length in IR-CD patients (rs = -0.52, P = 0.01). FGF19 and Se-HCAT values were positively related (rs = 0.57, P < 0.0001), whereas FGF19 was inversely related with the number of bowel movements (rs = -0.31, P=0.01) and Bristol scale (rs = -0.27, P= 0.04). No significant relation was found between FGF19 and clinical (CDAI) nor endoscopic (SES-CD) scores. Area under the ROC curve for FGF19 to detect SeHCAT at <10% and <5% were 0.75 (95% CI 0.62–0.88) and 0.81 (95% CI 0.71–0.92) respectively.




Conclusion

Serum FGF19 can be used as a simple blood test to the diagnostic of BAD in CD. FGF19 measurement could be an adjunct in guiding treatments for diarrhoea in CD.