P185 Capsule endoscopic findings for the diagnosis of Crohn’s disease: a case-control study
M. Esaki*1, T. Matsumoto2, S. Yamamoto3, T. Kinjo4, K. Kinjo5, K. Togashi6, K. Watanabe7, S. Nouda8, S. Ashizuka9, M. Nakamura10, Y. Suzuki11
1Kyushu University, Medicine and Clinical Science, Fukuoka, Japan, 2Iwate Medical University, Internal Medicine, Morioka, Japan, 3University of Miyazaki, Gastroenterology and Hepatology, Miyazaki, Japan, 4University of Ryukyus, Endoscopy, Okinawa, Japan, 5Fukuoka University Chikushi Hospital, Gastroenterology, Chikushino, Japan, 6Aizu Medical Centre, Fukushima Medical University, Coloproctology, Fukushima, Japan, 7Osaka City University, Gastroenterology, Osaka, Japan, 8Osaka Medical College, Second Department of Internal Medicine, Takatsuki, Japan, 9University of Miyazaki, Circulatory and Body Fluid Regulation, Miyazaki, Japan, 10Nagoya University Graduate School of Medicine, Gastroenterology and Hepatology, Nagoya, Japan, 11Toho University, Sakura Medical Centre, Internal Medicine, Chiba, Japan
Diagnosis of Crohn’s disease (CD) at the early stage without intestinal complication is expected to improve its clinical course. We aimed to find small bowel capsule endoscopy (SBCE) findings, which can lead to early diagnosis of CD.
This study was based on a collection of nationwide SBCE data from institutions majoring in inflammatory bowel diseases. We accumulated clinical and SBCE data of 85 subjects with or without final diagnosis of CD. The inclusion criteria were negative history of chronic NSAIDs use, being suspected of having inflammatory bowel diseases, and positive small bowel mucosal injury under total enteroscopy by SBCE. SBCE data were reviewed by an observer who was blinded to their final diagnosis. Small bowel mucosal injuries were classified into erosion, ulcer, and cobblestone appearance. Configuration of erosion or ulcer was subclassified into oval, irregular, linear, or circular. Alignment of diminutive lesions was assessed as being circular or linear. Patients were classified into CD and non-CD groups according to the final diagnosis and SBCE findings were compared between the groups.
In the study, 49 patients were classified into CD group, and 36 patients were into non-CD group. Subjects in CD group were younger, and they had perianal lesions more frequently than those in non-CD group did (p < 0.001 and p = 0.004, respectively). Comparisons of ulcers between the groups revealed that cobblestone appearance was more frequent in the CD group than in the non-CD group (35% vs 3%, p = 0.0003). In addition, total number of ulcers was greater in the CD group than in non-CD group (8 [2–24] vs 2 [0–7], p = 0.016), and linear (76% vs 14%, p < 0.0001) or irregular (84% vs 58%, p = 0.009) ulcers were more frequent in the former group than in the latter. The number of erosions was also greater in the CD group than in non-CD group (36 [17–96] vs 7 [4–26], p < 0.0001), and linear (88% vs 31%, p < 0.0001) or irregular (86% vs 64%, p < 0.05) erosions were more frequent in the former group than in the latter. Further, circular (73% vs 6%, p < 0.0001) or longitudinal (59% vs 6%, p < 0.0001) alignment of diminutive lesions was noted more frequently in the CD group than in non-CD group. In the CD group, erosions were more frequent, whereas ulcers were less frequent in the upper third segment of the small bowel than in the lower third (p < 0.01, p < 0.05, respectively). Sensitivity, specificity, positive predictive value, and negative predictive value of linear erosions plus regular alignment of diminutive lesions under SBCE for the diagnosis of CD were 71%, 92%, 92%, and 70%, respectively.
Linear erosions plus regular alignment of diminutive lesions in the small bowel under SBCE may be diagnostic of early stage of CD.