Accuracy of CT scan parameters for detecting endoscopic inflammation in small bowel incrohn' disease

970

Multicenter Capsule Endoscopy Study of Small Bowel Crohn's DiseasePatients in Clinical Remission: Long-Term Follow-up and Correlation WithFaecal Biomarkers and Clinical OutcomeVipul Aggarwal, Andrew S. Day, Susan J. Connor, Steven T. Leach, Gregor J. Brown,Rajvinder Singh, Antony Friedman, Michael C. Grimm, Philip I. Craig

Background: Conventional parameters, including Crohn's Disease Activity Index [CDAI] andcolonoscopy are limited in assessing small bowel (SB) Crohn's disease (CD) remission. Recentstudies suggest that CD patients in clinical but not endoscopic remission have worse long-term outcomes. Capsule endoscopy (CE) is the most sensitive test to diagnose SB CDhowever, no studies have reported its role in assessing remission. Aims: This study extendsour initial baseline findings (Gastroenterology 2010, 138, 5, S-114) and reports on extended(12 month) follow-up to determine if CE findings and/or faecal biomarkers offer prognosticinformation about future relapses. Methods: Adult SB CD patients in clinical remission(CDAI <150) were prospectively enrolled. CE studies (Olympus EndoCapsule) were reportedusing a Capsule Endoscopy Scoring Index (CESI) which numerically assesses mucosalinflammation: CESI <135 normal, 135-790 mild inflammation & >790 moderate-severeinflammation. CDAI and CRP were measured at baseline and at 3, 6, 9 and 12 months or,clinical relapse. Faecal calprotectin and S100A12 levels were obtained at baseline and atstudy endpoint. Baseline CESI findings were correlated with CDAI, CRP, calprotectin andS100A12. Referring physicians were blinded to CE and faecal biomarker findings and therapywas only changed with a clinical relapse. Results: 39 pts (19 M) median age 41 (range 20-67) yrs were studied. At enrolment, median CDAI was 73 (14-141); CRP 5.2 (0.5-31)(NR<5); calprotectin 114 (2-6390) (NR<100) mg/kg and; S100A12 4 (0-2000) (NR<10)mg/kg. As assessed by CESI, 62% of pts in clinical remission had inflammatory changes.All pts without inflammation by CESI had normal calprotectin levels and all with mod-severe inflammation had an elevated calprotectin. CESI and baseline faecal biomarkers weresignificantly correlated, calprotectin R=0.7505 (p<0.0001) and S100A12, R=0.4136 (p<0.01).There was no correlation between CESI and CDAI or CRP. 28 pts reached study end point,4 of whom had a clinical flare. Of these 4, all had mucosal inflammation at CE and 3 hadelevated baseline calprotectin levels. Of interest, 24 of the 28 pts remained in clinicalremission, 15 (63%) of whom had CE evidence of mucosal inflammation. Conclusions: Insmall bowel CD pts assessed in clinical remission: (1) 62% had mucosal inflammation onCE which significantly correlated with faecal calprotectin and S100A12 levels; (2) During12 months follow-up only a minority of these pts developed clinical relapse; (3) Thesefindings may have important clinical & therapeutic implications.Sensitivity and Specificity of baseline calprotectin levels in predicting mucosal inflammation

Comparison of data between patients with or without flare

S-169 AGA Abstracts

971

Impact of Magnetic Resonance Enterography in the Management of SmallBowel Crohn's DiseaseDanny Cheriyan, Eoin Slattery, Shaunagh McDermott, Aoife Kilcoyne, Denise Keegan,Hugh Mulcahy, Diarmuid P. O'Donoghue, Glen A. Doherty, Dermot E. Malone, Seamus J.Murphy

Background: Crohn's disease (CD) is an autoimmune, chronic inflammatory condition ofthe gastrointestinal tract that typically affects young adults. The disease has a wide spectrum ofclinical impact on patients, ranging from relatively asymptomatic disease to severe pathologyrequiring powerful immunomodulators and biological therapy. Radiological imaging is auseful means of assessing extent and severity of small bowel disease. With its lack of ionizingradiation and high tissue penetration, magnetic resonance imaging (MRI) has become apopular modality for assessing CD. Magnetic resonance enterography (MRE) is a relativelynew imaging modality that involves small bowel distention with orally administered fluid.While there is sufficient data to support the use of MRE as an effective diagnostic tool, thereis little to demonstrate its impact on patient management. Aims: The aim of this study isto determine if MRE influences the medical and surgical management of patients with smallbowel CD. Methods: From a prospectively maintained database of 2,655 IBD patients, weidentified patients with histologically confirmed CD who underwent MRE between 2007and 2010 in St. Vincent's University Hospital, Ireland. Demographic characteristics, diseaseactivity indices (Harvey Bradshaw Index- HBI), C-reactive protein (CRP), smoking status,previous surgical history and prior medical therapy were analysed. The results of MRE andsubsequent changes in patient management were evaluated. Results: 30 females and 27males with histologically confirmed CD were studied. The median age of the patients was36 years (range 16-68), and median duration of disease before MRE was 9.5 years (range1-35). 40 patients (70%) had ileo-colonic disease, and 17 patients (30%) had small boweldisease. 39 patients (68%) had previous resection for CD. 25% were smokers at the timeof examination. The median HBI was 8 (range 2-15), and median CRP 10 (range 1-350).68% of patients had a history of steroid use, 37% thiopurine use, and 19% had been onbiological therapy. 7 patients (12%) had a normal MRE, 6 (11%) had fibrotic disease, 31(54%) had active inflammatory disease, and 12 (21%) had both active and fibrotic disease.42 out of 57 patients (74%) had a change in management as a direct result of the MRE. Ofthese 42 patients, 22 (53%) patients had changes in medical therapy, and 20 (47%) patientsunderwent surgery. An abnormal MRE result yielded significantly more changes in patientmanagement compared to a normal study (p=0.0008). Conclusion: The majority of patientswith small bowel CD who undergo MRE have a change in management as a result of theexamination. MRE is a safe imaging modality, has a high clinical impact on patient manage-ment, and should therefore become the small bowel evaluation of choice in CD patients.

972

Small Bowel Ultrasound: Diagnostic Yield in Established Small Bowel Crohn'sDiseaseDaniel S. Pearl, Antony Higginson, Amanda Quine

Introduction: Crohn's disease is a chronic, relapsing intestinal inflammatory disorder ofunknown aetiology which frequently involves the small intestine. Accurate localizationof disease is important to direct targeted therapy. Video capsule endoscopy (VCE) hasrevolutionized clinical assessment of small intestinal Crohn's disease. Small bowel ultrasound(SB USS) is a rapid, inexpensive, interactive and non-invasive alternative method for assessingsmall bowel Crohn's disease, which is in routine use only at selected UK institutions. Thisis the first study to evaluate diagnostic yield of SB USS in VCE determined Crohn's disease.Methods: A retrospective assessment of patients who had undergone VCE in 2008-2010was carried out. Patients investigated for suspected small bowel Crohn's disease, or whohad findings of small bowel Crohn's on VCE were included, if they had also had a SB USSwithin 12 months. VCE findings were graded as mild (apthous ulcers only), moderate(apthous ulcers with mucosal distortion) or severe (apthous ulcers with mucosal distortionand strictures/stenosis). SB USS was graded positive or negative for small bowel Crohn'sdisease. Both assessments were single operator. Either investigation could predate the other.Results were expressed as sensitivity, specificity, positive and negative predictive value (PPVand NPV) of SB USS compared with VCE for detection of small bowel Crohn's. Sub-analysisof SB USS findings for VCE-defined severity of small bowel Crohn's disease was carried out.Results: 196 VCE procedures were reviewed, of which 22 fulfilled the inclusion criteria. 10patients had SB Crohn's on VCE; this was detected in 4 patients by SB USS (sensitivity40%). 12 patients had no evidence of SB Crohn's on VCE; none of these had SB USS findingsof Crohn's disease (specificity 100%). Of 18 patients with no evidence of SB Crohn's on SBUSS, VCE findings of Crohn's disease were apparent in 6 patients (negative predictive value67%); however, all patients with positive findings of Crohn's disease on SB USS had evidenceof SB Crohn's on VCE (positive predictive value 100%). Sub-analysis for severity of inflamma-tion on VCE was carried out. Of 4 patients with positive findings at SB USS, 3 were severeand 1 moderate on VCE. 1 patient with severe Crohn's on VCE was missed by SB USS;however, the patient's body habitus was unfavourable. Conclusion: SB USS has excellentpositive predictive value (100%) and specificity (100%) for detection of SB Crohn's disease,with only moderate negative predictive value (67%). In addition, all detected cases weremoderate or severe, which may complicate VCE. It therefore seems a safe, quick, relativelycheap initial investigation in expert hands, which may obviate more costly, invasive investi-gations. A prospective evaluation of these diagnostic modalities should be carried out.SMALL BOWEL USS FOR DIAGNOSIS OF SMALL INTESTINAL CROHN'S DISEASE

AG

AA

bst

ract

s