Current issues of ACP Journal Club are published in Annals of Internal Medicine


Capsule endoscopy was superior to push enteroscopy for identifying sources of obscure gastrointestinal bleeding


ACP J Club. 2007 Sep-Oct;147:39. doi:10.7326/ACPJC-2007-147-2-039

Clinical Impact Ratings

Gastroenterology: 5 stars

Source Citation

de Leusse A, Vahedi K, Edery J, et al. Capsule endoscopy or push enteroscopy for first-line exploration of obscure gastrointestinal bleeding? Gastroenterology. 2007;132:855-62. [PubMed ID: 17324401]



In patients referred for obscure gastrointestinal bleeding, does capsule endoscopy (CE) have a higher diagnostic yield than push enteroscopy (PE) when used first?


Design: Randomized controlled trial.

Allocation: Concealed.*

Blinding: Unblinded.*

Follow-up period: Median 12 months.

Setting: 2 clinical centers in Paris, France.

Patients: 89 patients who had overt bleeding within the previous 6 months or chronic (> 3 mo) iron-deficiency anemia (hemoglobin < 10 g/dL [6.2 mmol/L]) without obvious vaginal or rectal bleeding and had previous negative findings (including upper gastrointestinal endoscopy, colonoscopy, and small-bowel barium series or computed tomography enteroclysis).

Intervention: CE, and if negative, PE (n = 47); or PE, and if negative, CE (n = 42). 40 patients actually received CE first, and 38 patients received PE first (mean age 54 y, 62% men).

Outcomes: Diagnostic yield.

Patient follow-up: 80% completed 12-month follow-up.

Main results

A definitive source of bleeding was identified in more patients in the CE group than in the PE group (Table). Fewer lesions were missed by CE than by PE (Table). CE missed no lesions in the small bowel, whereas all missed lesions with PE were located in the small bowel. Patients who started with CE were less likely to require the second test than were patients who started with PE. Groups did not differ for rate of clinical remission at 1 year.


In patients referred for obscure gastrointestinal bleeding, first-line use of capsule endoscopy identified a definite bleeding source in more patients and missed fewer lesions than did first-line use of push enteroscopy.

*See Glossary.

Source of funding: Comité d’Evaluation et de Diffusion des Innovations Technologiques (CEDIT), Hôpitaux de Paris.

For correspondence: Dr. R. Jian, Hôpital Européen Georges Pompidou, Paris, France. E-mail:

Table. Capsule endoscopy vs push enteroscopy for first-line exploration of obscure gastrointestinal bleeding†

Outcomes at 12 months Type of lesion Capsule endoscopy Push enteroscopy RBI (95% CI) NNT (CI)
Definite bleeding source identified All 50% 24% 11% (14 to 310) 4 (3 to 21)
Small-bowel 43% 11% 304% (61 to 977) 4 (2 to 8)
Gastric 7.5% 11% 29% (−169 to 81) Not significant
Colonic 0% 2.6% 100% (−258 to 100) Not significant
Missed lesions All 7.5% 26% 72% (13 to 91) 6 (3 to 43)
Clinical remission at 1 year 70% 58% 20% (−14 to 73) Not significant

†Abbreviations defined in Glossary. RRR, NNT, and CI calculated from data in article.


The study by de Leusse and colleagues is among the latest studies showing that CE is more accurate than other methods (PE, small-bowel radiography, and computed tomographic or standard angiography) at finding the source of bleeding in patients with obscure gastrointestinal bleeding. Although I have no problem accepting this conclusion, I believe that several issues should be considered before providers accept the primary role of CE in such patients. First, CE is time-consuming, expensive, and in some studies has been associated with “retained capsules” in a substantial number of patients. Second, all but 1 study (1) so far have failed to use a gold-standard comparator. Third, in many of the studies to date, lesions found by CE should have been found by careful standard upper and lower endoscopy. Fourth, most studies have not properly examined the clinical utility of finding a source of blood loss. In patients with occult obscure bleeding, no one has compared use of CE with use of small-bowel enteroclysis to rule out a small-bowel tumor and then simply placing the patient on iron supplementation.

In my view, only patients who then do not maintain an adequate blood count should undergo CE. Finally, CE is only a diagnostic test and cannot provide therapy. Double-balloon enteroscopy (2) provides a means of both diagnosing a bleeding lesion and treating it, if appropriate. CE is clearly a valuable technique to find obscure bleeding lesions; we just need to determine when it should be used.

Walter L. Peterson, MD
University of Colorado Health Sciences Center
Denver, Colorado, USA


1. Hartmann D, Schmidt H, Bolz G, et al. A prospective two-center study comparing wireless capsule endoscopy with intraoperative enteroscopy in patients with obscure GI bleeding. Gastrointest Endosc. 2005;61:826-32. [PubMed ID: 15933683]

2. May A, Nachbar L, Ell C. Double-balloon enteroscopy (push-and-pull enteroscopy) of the small bowel: feasibility and diagnostic and therapeutic yield in patients with suspected small bowel disease. Gastrointest Endosc. 2005;62:62-70. [PubMed ID: 15990821]