DOP028. Development of red flags for early referral of adults with symptoms and signs suggestive of Crohn's disease: an IOIBD initiative
S. Danese1, G. Fiorino1, L. Peyrin-Biroulet2, P.L. Lakatos3, J.Y. Mary4, G. D'Haens5, L. Moja6,7, A. D'Hoore8, J. Panes9, W. Reinisch10, W.J. Sandborn11, S. Travis12, S. Vermeire13, J.F. Colombel14, 1Humanitas Research Hospital, IBD Center, Rozzano, Milan, Italy, 2Nancy University Hospital, Université de Lorraine, Gastroenterology and Hepatology, Vandoeuvre-les-Nancy, France, 3Semmelweis University, 1st Department of Medicine, Budapest, Hungary, 4Saint Louis Hospital, INSERM U717, Paris, France, 5Academic Medical Center, Department of Gastroenterology, Amsterdam, Netherlands, 6IRCCS Galeazzi Orthopedic Institute, Unit of Clinical Epidemiology, Milan, Italy, 7University of Milan, Department of Biomedical Sciences for Health, Milan, Italy, 8University Hospitals Leuven, Department of Abdominal Surgery, Leuven, Belgium, 9Hospital Clinic de Barcelona, IDIBAPS and CIBERehd, Gastroenterology Unit, Barcelona, Spain, 10AKH, Medizinische Universität, Klinische Abt. Gastroenterologie & Hepatologie, Wien, Italy, 11University of California, San Diego, Inflammatory Bowel Disese Center, Division of Gastroenterology, La Jolla, United States, 12John Radcliffe Hospital, Translational Gastroenterology Unit, Oxford, United Kingdom, 13Katholieke Universiteit and University Hospital Gasthuisberg, Division of Gastroenterology, Leuven, Belgium, 14Hemsley Inflammatory Bowel Disease Center, Icahn Medical School of Medicine at Mount Sinai, New-York, United States
Diagnostic delay >12 months is frequent in Crohn's disease (CD). This is associated with worse outcomes, including a complicated disease course and increased operation rates. We have developed a simple tool to help identify early CD and reduce diagnostic delay.
Twelve IBD specialists (11 gastroenterologists and 1 surgeon) were asked to generate a list of clinically relevant questions representing the most common symptoms or signs suspicious of CD, based on their clinical experience and on a systematic review of the literature. A pilot ‘Red Flag’ questionnaire was created by combining the most common questions and by online ranking. 21 final questions were administered to 20 healthy subjects and 71 patients with irritable bowel syndrome, the non-CD group, and 79 with early CD (diagnosis <18 months). Patients with CD were asked to report symptoms or signs in the 12 months before diagnosis. χ2 test was used to compare the frequency of symptoms and signs between CD and non-CD. Simple and multiple logistic regression analyses were used to identify and weigh symptoms in a Red Flag score separating significantly CD from non-CD. ROC curve assessed the threshold Red Flag score that discriminated CD from non CD. Logistic regression was performed to quantify through odds-ratios (OR) the predictive role for CD of each question and of the Red Flag score. Statistical significance was set as p < 0.05.
A total of 170 patients were enrolled. From the 21 pilot questions, 15 answers ('items') were more frequent in CD than non-CD. All 15 items were then included in the multivariate analysis, identifying 6 final questions (night time diarrhea, weight loss >5%, chronic abdominal pain, first-degree relatives with IBD, mild fever (<38°C), and no cramps after eating vegetables) which were significantly associated with a diagnosis of CD.
ROC curve analysis identified a minimum score of 6 as predictive for suspected CD (sensitivity 0.88, specificity 0.96, positive and negative likelihood ratios of 20.16 and 0.12, area under the curve 0.967). Logistic regression confirmed that a score >5 was likely to discriminate CD from non-CD (OR 169, 95% CI: 50 to 572.5, p < 0.0001).
|Item||Coefficient||Standard error||p value||Odds ratio||95% CI|
|First-degree relatives with IBD||3.2||1.14||0.004||27||21–129|
|Weight loss >5%||2.8||0.65||<0.0001||16||13–80|
|Chronic Abdominal pain||2.0||0.69||0.003||8||4–31|
|No cramps after eating vegetables||1.8||0.82||0.028||6||2–16|
These results are preliminary in the development of a Red Flag Index for CD. Validation of Red Flags among general practitioners that discriminate suspected CD from non-CD, will facilitate referral to IBD specialists, and could potentially reduce diagnostic delay in CD. This may ultimately lead to better outcomes for a chronic, disabling and progressive disease.