P326 PREFAB-study: PRediction tool for Early identification of patients at risk of Crohn's disease in perianal Fistulas and ABscesses: interim analysis of a prospective pilot study at a non-academic, IBD-expert centre in the Netherlands
Munster, L.(1)*;de Groof, E.J.(1);van Dieren, S.(2);Mundt, M.W.(3);D’Haens, G.R.A.M.(4);Bemelman, W.A.(5);Buskens, C.J.(5);van der Bilt, J.D.W.(1);
(1)Flevoziekenhuis/Amsterdam UMC- location AMC, Department of Surgery, Almere/Amsterdam, The Netherlands;(2)Amsterdam UMC- location AMC, Department of Clinical Epidemiology, Amsterdam, The Netherlands;(3)Flevoziekenhuis, Department of Gastroenterology, Almere, The Netherlands;(4)Amsterdam UMC- location AMC, Department of Gastroenterology, Amsterdam, The Netherlands;(5)Amsterdam UMC- location AMC, Department of Surgery, Amsterdam, The Netherlands;
Background
Perianal abscesses (PAA) and perianal fistulas (PAF) are life impairing conditions associated with Crohn’s disease (CD). A retrospective study at Flevohospital showed a reduced median delay in diagnosis of CD from 33 months (IQR 23-64) in the period 2007-2016 to 3 months (IQR 0-5) in the period 2017-2021, with <20% of patients diagnosed <1 year after the first perianal surgical procedure in 2007-2016 versus 79% in 2017-2021. This study also showed CD in up to 10% of patients with PAA/PAF, which was reason to continue data collection in a larger prospective pilot study to identify risk factors for CD in patients presenting with PAA/PAF.
Methods
All consecutive patients ≥16 years presenting with PAA/PAF were included. Patients were prospectively screened for risk factors/red flags from a ‘perianal Red Flag Index Questionnaire’ (pRFI) (identified from previous literature searches/expert opinion) and fecal calprotectin (FC-)samples were taken in all patients. Colonoscopy was performed in case of ≥5 positive answers in the pRFI and/or FC-values ≥150mcg/g.
Results
Sixty-nine patients were included (72,5% male) with median age of 41 years (IQR 30,5-54,6). Thirty-five patients (50,7%) presented with PAA, whereas 34 patients presented with PAF (49,3%). Thirty-one patients (44,9%) had recurrent PAA/PAF. Additional colonoscopy was performed in 11 patients, in whom 5 patients (7,2%) eventually were diagnosed with CD with a median delay of 44 months (IQR 7,5-87,5, still 5 patients in diagnostic work-up). Median FC-values were 552mcg/g (IQR 198,5-4585) in patients with confirmed CD and 31,5mcg/g (IQR 10-102,75) in patients without CD (p=0,002). Risk factors associated with CD were a younger age, presence of PAF (compared to PAA only), higher number of previous perianal interventions, multiple internal fistula openings, fissures and proctitis (p<0,05 in all).
Conclusion
This pilot study reveals several risk factors associated with CD in patients presenting with PAA/PAF and forms the basis for development of a clinical decision tool that incorporates both the pRFI questionnaire and selected FC-measurement to early identify patients at risk of CD when presenting with perianal disease. This clinical decision tool will be subject of a larger prospective multicentre study to reduce diagnostic/treatment delay, thereby improving outcomes in patients with Crohn’s fistulas.