Capsule Endoscopy or Angiography in Patients With Acute Overt Obscure Gastrointestinal Bleeding

A Prospective Randomized Study With Long-Term Follow-Up

Wai K Leung MD; FACG; Simon S M Ho MBBS; Bing-Yee Suen BN; Larry H Lai MBChB; Simon Yu MD; Enders K W Ng MD; Simon S M Ng MD; Philip W Y Chiu MD; Joseph J Y Sung MD; PhD; Francis K L Chan MD; James Y W Lau MD


Am J Gastroenterol. 2012;107(9):1370-1376. 

In This Article


Setting and Patients

Consecutive patients who presented with active overt OGIB to the Prince of Wales Hospital of Hong Kong between June 2005 and November 2007 were recruited. As a standard protocol, all patients presented with melena or hematochezia would receive upper gastrointestinal endoscopy within 24 h of admission. Patients who had nondiagnostic upper GI endoscopy were offered same-day colonoscopy. Polyethylene glycol (2 liters) was given for bowel cleansing before colonoscopy. All gastrointestinal endoscopies were performed by experienced endoscopists or trainees under the supervision of senior endoscopists. Overt OGIB were defined as patients who had nondiagnostic upper endoscopy and colonoscopy.

Patients who were <18 years, pregnant, in moribund conditions, or with terminal malignancy were excluded. Moreover, patients who had contraindications for CE including swallowing difficulties, suspected intestinal obstruction, presence of cardiac pacemaker, or implantable electromedical devices were excluded. Patients with contrast allergy and impaired renal function (serum creatinine >150 μmol/l) were also not eligible for this study. All patients provided written informed consent for participation in this study. They were randomized to receive either immediate mesenteric angiography or immediate wireless CE in a 1:1 ratio. Randomization was performed by computer-generated number in blocks of 4. To ensure concealed randomization, the randomization code was put in opaque envelope and kept by researchers not performing endoscopy. Both patients and investigators were unaware of the randomization sequence. The study protocol was approved by the Joint CUHK/NTEC Clinical Research Ethics Committee and was registered in (Identifier: NCT00153686).


Patients randomized to angiography received immediate three-vessel visceral angiography with selective cannulation of the celiac, superior, and inferior mesenteric artery using Seldinger technique, which was performed by experienced interventional radiologists (S.S.M.H. and S.Y.). Iodinated contrast media of concentration 200 mg/ml were used with sufficient coverage of region of interest. Active bleeding was defined as visible extravasation of contrast medium. Moreover, vascular anomalies such as angiodysplasia, arteriovenous malformation, or hypervascular masses and any suspicious lesions were noted. Further embolotherapy for hemostasis would be performed as determined appropriate by interventional radiologists.

Patients randomized to CE received immediate small-bowel CE (Pillcam SB, Given Imaging, Yoqneam, Israel). All capsules were swallowed by patients and there was no prokinetic agent given. Capsule endoscopy images were reviewed by two experienced endoscopists (W.K.L. and L.H.L.). Consensus was achieved if there was any discrepancy on individual endoscopist's findings. In this study, only CE findings that were reported to have a high probability of bleeding were considered as positive findings (e.g., area with fresh bleeding, ulcers, and erosions of at least 2 mm in size, tumors, and varices).[12] Other lesions with low probability of bleeding were regarded as negative (e.g., isolated red spots, single small erosions or mucosal breaks, and dilated venous vessels).


All patients were managed according to the results of their assigned investigation. Patients with positive findings on the assigned investigation would be offered further investigations to abort the bleeding diathesis or to clarify the nature of bleeding lesions. Those who developed rebleeding during the same hospitalization would be crossed over to the other investigation modality. Patients with negative findings on the initial assigned investigation but who developed rebleeding would undergo further investigation to localize the site of bleeding. Further bleeding was defined as continuous passage of melena or hematochezia with further drop in hemoglobin of >2 g/dl.

Patients were seen at 12-weekly interval for the first year for monitoring of rebleeding and hemoglobin levels. They were then monitored for further clinical bleeding, hospitalization for bleeding or anemia, and death for up to 5 years. All medical records were verified by the territory-wide hospital computer system (Clinical Management System, Hong Kong Hospital Authority), which captured all medical records including laboratory results, radiological, surgical, and endoscopy records of all public hospitals in Hong Kong.

Outcome Measures

The primary outcome of this study was the diagnostic yield of CE and mesenteric angiography in identifying the source of bleeding. Secondary outcome measures included long-term rebleeding rates, further hospital admissions for bleeding or anemia, further blood transfusion, and death.

Sample Size Estimation

Based on early studies published at the time of study design,[7,11] we estimated that the diagnostic yield of CE and angiography was 70% and 30%, respectively. Assuming a power (β) of 80% and a two-sided P value (α) of 0.05, 29 patients were required in each study arm.

Statistical Analysis

Analysis was based on intention to treat and all randomized patients were analyzed. Continuous variables were compared by Student's t-test and categorical data were analyzed by Fisher's exact test where appropriate. The proportions of patients with rebleeding in the two groups were compared by log-rank test. A two-sided P value of <0.05 is considered statistically significant.