Diagnosis of small-bowel varices by capsule endoscopy Shou-jiang Tang, MD, Simon Zanati, MD, Elena Dubcenco, MD, Maria Cirocco, MSc, Dimitrios Christodoulou, MD, Gabor Kandel, MD, Gregory B. Haber, MD, Paul Kortan, MD, Norman E. Marcon, MD Background: Capsule endoscopy is being used increas- ingly to investigate GI bleeding of obscure origin and disorders of the small bowel. Methods: Four cases of small-bowel varices of various etiologies diagnosed by capsule endoscopy are de- scribed: a bleeding small-bowel varix because of hepatic portal hypertension, oozing small-bowel anastomotic or adhesion-related varices, small-bowel varices secondary to mesenteric vein thrombosis, and ‘‘idiopathic intestinal varices.’’ Observations: Over a 12-month period, small-bowel varices were found in 4 of 46 patients (8.7%) who underwent capsule endoscopy for GI bleeding. Fresh blood adjacent to the varices was documented in 3 patients. The small-bowel varices had serpiginous or nodular shapes, with or without a bluish coloration. The variceal mucosa appeared mosaic-like, shining, or normal compared with surrounding mucosa. Conclusions: Capsule endoscopy is invaluable for the diagnosis of small-bowel varices. It is highly sensitive for detection of fresh blood in the small bowel. Clinical suspicion, capsule endoscopy image recognition, and alertness during capsule endoscopy interpretation are keys to diagnosis. Capsule endoscopy (CE) has revolutionized exam- ination of the small bowel (SB). 1 Currently, the most common indication for CE is GI bleeding of obscure etiology. 2 For this indication, the yield of CE is reportedly about 67%. 2 The most common findings include angioectasia, fresh blood, ulceration, tumor, and varices. ‘‘Varix’’ is defined simply as a tortuous enlargement of a vein. 3 The mucosal color of SB varices may differ minimally from that of the surrounding mucosa. 4,5 Before the advent of CE, an endoscopic diagnosis of SB varices distal to the duodenum was extremely infrequent. 4,5 In most instances, the endoscopic diagnosis of SB varices is impossible beyond the limit of insertion of an enteroscope, whereas CE examines the entire SB. During a 12-month period, 4 cases of SB varices caused by different etiologies were diagnosed by CE. PATIENTS AND METHODS All CE procedures performed over a 12-month period (July 2002 to June 2003) were analyzed for cause of bleeding. GI bleeding of obscure etiology was defined as persistent or recurrent GI bleeding in the absence of explanatory findings at upper endoscopy and colonoscopy. Capsule endoscopy Patients were instructed to withhold orally adminis- tered iron supplements for 5 days before CE. To improve SB emptying and visualization, they took one dose of an over- the-counter laxative (e.g., 30 mL of milk of magnesia) followed by several glasses of water during the evening of the day before CE. After an overnight fast (12 hours), the patient ingested the capsule endoscope (M2A; Given Imaging, Ltd., Yoqneam, Israel) after attachment of the sensor array. The capsule examination time was set for 7 hours. The CE digital image stream was assessed and interpreted by an endoscopy fellow (S.T.) and was reviewed by one of 4 staff endoscopists (N.M., P.K., G.H., G.K.). Images acquired within the esophagus, the stomach, and the SB were viewed and assessed at a frame rate of 15 per second; those acquired in the colon were read at a frame rate of 25 seconds. The entire recording was assessed in each patient. According to our diagnostic criteria and classification system, SB varices without stigmata of bleeding are considered a suspected source of bleeding. 6 If bleeding or stigmata (e.g., platelet plug associated with the varix) is observed, the varices are considered the source of bleeding. Capsule regional transit abnormality (RTA) is defined as significantly delayed transit (remains within a SB seg- ment for at least 15 minutes), with or without evident mucosal abnormality. 6 Capsule RTA usually is accompa- nied by mucosal collapse, often with the capsule pressing against the mucosa, and usually signifies an underlying abnormality. OBSERVATIONS During a 12-month period, 50 patients underwent CE. For 47 patients (28 men, 19 women; mean age 62 years, range 33-86 years), the indication for CE was GI bleeding of obscure origin. The remaining 3 patients underwent CE to investigate SB lesions noted on CT or barium contrast radiography. Before CE, these patients had undergone a total of 283 GI procedures: EGD, 109; push enteroscopy, 24; barium contrast radiography of the SB, 33; and colonoscopy, C ASE S TUDIES Received September 2, 2003. For revision December 10, 2003. Accepted January 27, 2004. Current affiliations: Center for Therapeutic Endoscopy and Endoscopic, Oncology, St Michael’s Hospital, University of Toronto, Ontario, Canada. Reprint requests: Norman E. Marcon, MD, Center for Therapeutic Endoscopy and Endoscopic Oncology, Victoria Wing 16-062, St. Michael’s Hospital, Toronto, Ontario, Canada M5B 1W8. Copyright Ó 2004 by the American Society for Gastrointestinal Endoscopy 0016-5107/$30.00 PII: S0016-5107(04)01458-0 VOLUME 60, NO. 1, 2004 GASTROINTESTINAL ENDOSCOPY 129