In this interesting study, Estevez et al.  prospectively investigated the clinical impact of capsule endoscopy (CE) in detecting causes of obscure-overt (n=52) and obscure-occult (n=48) bleeding in 100 patients. Their finding that the need for transfusion before CE was the only independent variable predicting significant findings in patients with overt bleeding is new, and significantly contributes to our understanding of the role of CE in clinical practice.
CE (M2A Capsule; Given Imaging Ltd., Yogneam, Israel)  is a new technique that allows the direct visualization of the entire upper gastrointestinal tract also in regions that are usually not accessible with conventional endoscopy. Since its introduction into clinical practice [2,3], clinical experience in patients with obscure gastrointestinal bleeding has indicated that in contrast to conventional upper gastrointestinal work-up (oesophago-gastro-duodenoscopy, push-enteroscopy, small bowel follow-through, enteroclysis, abdominal computerized tomography, radionuclide scans, angiography, intraoperative endoscopy), CE is well tolerated by the patient . The overall diagnostic yield in iron deficiency anaemia or bleeding of obscure origin is reported to be between 50 and 70% , leading to a change of treatment modalities in approximately a third of cases.
The study by Estevez et al.  is one of two studies on a relatively large patient cohort that try to stratify the patients according to their grade of bleeding . Estevez et al.  were able to refine a subgroup of patients who could best benefit from CE. In their study, the overall diagnostic yield was 68%, increasing to 73.1% in patients with overt bleeding, to 79.1% in patients with a requirement for transfusion, and to 88.4% in patients with overt bleeding and a requirement for transfusion. These findings are comparable with a study by Pennazio et al. , which reported a diagnostic yield of 92.3% in patients with ongoing overt bleeding compared with 44.2% (obscure-occult bleeding) and 12.9% (past bleeding). It is noteworthy and clinically important that Estevez et al.  also identified a subgroup of patients in whom CE is clearly not indicated. These were patients with overt bleeding and no transfusion requirement (9%), in whom CE obtained no significant findings. This has to be confirmed by further studies.
Although the results are promising, some further comments on limitations appear appropriate. In only 79% of procedures did the capsule reach the caecum, and the recording time was rather short in some patients, indicating that the entire small intestine was accessible only in a subgroup of patients. In addition, the acquisition of two frames per second may be too low for some clinical questions and there is a necessity for further technical improvement.
More than half of the significant lesions (36/68) were located in regions accessible to conventional endoscopy (stomach, seven; duodenum, four; duodenum and jejunum, three; caecum, one; ileum, 21), and one may argue that coeliac disease (one), Meckel's diverticulum (one), jejunal diverticulosis (one), and ulcerated stenosis (two) may also have been detected by conventional diagnostic work-up. Conventional diagnostic work-up is a highly sensitive and reliable technique to identify the source of bleeding in approximately 95% in patients with chronic occult bleeding [6–8]. Of course, the success rate is crucially dependent on the skills of the gastroenterologist who performs the endoscopy, and that may need to be improved. In contrast to endoscopy, CE does not allow one to take biopsies or administer therapy. CE is still expensive and only few HMOs or insurance companies will cover the costs for this procedure. This is a particular problem in some countries such as Germany, where the ‘Medizinischer Dienst der Spitzenverbände der Krankenkassen’ provided a fundamental review of CE based on an extensive literature analysis in 2002 that was updated in 2004 . The conclusion was that the impact of CE within a diagnostic algorithm is still unresolved because of the lack of prospective studies, the low number of recruited patients, and the non-standardized procedures.
Those who advocate CE argue that in clinical practice repeated invasive examinations and rehospitalizations are needed to identify the bleeding source , thereby causing a significant financial burden to the health system. In particular, the indication for small bowel follow-through and enteroclysis is questionable because they provide a low level of sensitivity. A protocol in which a CE study is performed after a negative oesophago-gastro-duodenoscopy and colonoscopy to reduce the number of repeated examinations  could solve the dilemma but needs further cost–benefit evaluation.
One problem not solved so far is the clinical relevance of potentially bleeding lesions that were detected by CE in 76.5% of significant lesions in the study by Estevez et al. . Half of those were angiodysplasias (23/52). However, the presence of angiodysplasias and other ‘unusual’ findings per se does not imply a significant bleeding source, because their natural history, morphological criteria to characterize the bleeding risk, and their occurrence in asymptomatic populations are still unknown. This problem is reflected by the finding of Estevez et al.  that the percentage of resolved cases was not significantly different according to whether CE produced or failed to produce significant findings. In addition, most of the treatment modifications after CE did not reflect a specific impact of CE on changes in treatment modalities. For example, iron supplementation, the eradication of Helicobacter pylori, gluten-free diet, and the suspension of non-steroidal anti-inflammatory drugs are common treatment modalities in iron-deficient anaemia that should be initiated even without CE.
Estevez et al.  are to be congratulated for their excellent study and for identifying subgroups of patients with obscure bleeding who may or may not benefit from CE. The authors also merit credits for their contribution to our understanding of how to put this ‘Star Wars technology’  into a rational diagnostic algorithm. It seems to be clear that this decision will not be made by the believer or non-believer in CE alone, but also by health insurance companies.
Conflict of interest
1. Estevez E, Gonzalez-Conde B, Vazquez-Iglesias JL, de los Angeles Vazquez-Millan M, Pertega S, Alonso PA, Clofent J, et al.
Diagnostic yield and clinical outcomes after capsule endoscopy in 100 consecutive patients with obscure gastrointestinal bleeding. Eur J Gastroenterol Hepatol 2006; 18: in press.
2. Iddan G, Meron G, Glukhovsky A, Swain P. Wireless capsule endoscopy. Nature 2000; 405:417–420.
3. Meron GD. The development of the swallowable video capsule (M2A). Gastrointest Endosc 2000; 52:817–819.
4. Pennazio M, Santucci R, Rondonotti E, Abbiati C, Beccari G, Rossini FP, De Franchis R. Outcome of patients with obscure gastrointestinal bleeding after capsule endoscopy: report of 100 consecutive cases. Gastroenterology 2004; 126:643–653.
5. Fireman Z. The light from the beginning to the end of the tunnel. Gastroenterology 2004; 126:914–916.
6. Rockey DC, Cello JP. Evaluation of gastrointestinal tract in patients with iron deficiency anemia. N Engl J Med 1993; 329:1691–1695.
7. Ioannou GN, Spector J, Scott K, Rockey DC. Prospective evaluation of a clinical guideline for the diagnosis and management of iron deficient anemia. Ann J Med 2002; 113:281–287.
8. Rockey DC. Occult gastrointestinal bleeding. N Engl J Med 1999; 341:38–42.
9. Bauer S, Terlutter H, Hoffmann H-G. Capsule endoscopy – evaluation of methodology [in German]. Update October 2004. Unterarbeitsgruppe “Kapselendoskopie” der M4/SEG7. Essen, Germany: Medizinischer Dienst der Spitzenverbände der Krankenkassen (MDS) e.V.
10. Ell C, Remke S, May A, Helu L, Henrich R, Mayer G. The first prospective controlled trial comparing wireless capsule endoscopy with push enteroscopy in chronic gastrointestinal bleeding. Endoscopy 2002; 34:685–689.