DOI: 10.1148/rg.253045134
RadioGraphics 2005;25:697-711
© RSNA, 2005
Imaging of Small Bowel Disease: Comparison of Capsule Endoscopy, Standard Endoscopy, Barium Examination, and CT1
Amy K. Hara, MD,
Jonathan A. Leighton, MD,
Virender K. Sharma, MD,
Russell I. Heigh, MD and
David E. Fleischer, MD
1 From the Departments of Radiology (A.K.H.) and Gastroenterology (J.A.L., V.K.S., R.I.H., D.E.F.), Mayo Clinic, 13400 E Shea Blvd, Scottsdale, AZ 85259. Recipient of a Cum Laude award for an education exhibit at the 2003 RSNA Scientific Assembly. Received June 16, 2004; revision requested July 16 and received August 18; accepted August 20. All authors have no financial relationships to disclose.
Address correspondence to A.K.H. (e-mail: hara.amy{at}mayo.edu).
 |
Abstract
|
|---|
Capsule endoscopy is a revolutionary new diagnostic tool for the detection of small bowel disease. As the name implies, capsule endoscopy makes use of a swallowable video capsule; as such, it is the only technique that allows noninvasive endoscopic examination of the entire small bowel without sedation. Obscure gastrointestinal bleeding is the most common indication for capsule endoscopy, which commonly depicts arteriovenous malformations, small bowel tumors, and ulcers missed at standard endoscopy and imaging examinations. However, capsule endoscopy is not optimal for the localization of small bowel lesions. In addition, lesions can be missed due to poor bowel preparation, rapid or delayed small bowel transit, or orientation of the camera away from a lesion. Computed tomography and barium examinations are useful for detecting these missed lesions and for localizing lesions detected at capsule endoscopy. Other limitations of capsule endoscopy are the inability to treat lesions and its limited use in patients with small bowel strictures or obstruction. Nevertheless, this new technique is easy to perform, is well tolerated by patients, and, for the first time, allows noninvasive endoscopic evaluation of the entire small bowel.
© RSNA, 2005
Abbreviations: EGD = esophagogastroduodenoscopic, FDA = Food and Drug Administration, NSAID = nonsteroidal anti-inflammatory drug
|
Introduction
|
|---|
The small intestine is the most difficult part of the gastrointestinal tract to evaluate due to its length and complex loops. Disadvantages of conventional endoscopic techniques such as push enteroscopy and colonoscopy with ileoscopy include limited endoscopic examination of the small bowel and sedation requirements. A complete endoscopic evaluation was previously possible only with intraoperative endoscopy.
Capsule endoscopy is a revolutionary new diagnostic tool for the detection of small bowel disease that makes use of a swallowable video capsule. Unlike conventional endoscopy, capsule endoscopy allows examination of the entire small bowel and does not require sedation. First introduced at the 2000 Digestive Disease Week Conference in San Diego, California, the Given Imaging M2A capsule (Yoqneam, Israel) subsequently received approval from the U.S. Food and Drug Administration (FDA) in mid-2001 for use in the United States. The primary clinical indication for capsule endoscopy is obscure gastrointestinal bleeding, but this technique is also indicated for evaluation of early Crohn disease. Over 10,000 examinations have been performed worldwide—more than 400 at our institution—with an overall complication rate of only 0.75% (1). In this article, we discuss capsule endoscopy in terms of technique, advantages and disadvantages, indications (eg, obscure gastrointestinal bleeding, Crohn disease) and contraindications, types of lesions detected (arteriovenous malformations, small bowel tumors, ulcers), false-negative and false-positive findings, and complications.
|
Capsule Endoscopic Technique
|
|---|
The endoscopy capsule measures 26 mm (length) by 11 mm (diameter) and weighs 4 g (Fig 1). It contains a video camera, light source (four white light–emitting diodes), radio transmitter, and batteries.
Prior to swallowing the capsule, patients fast for 10 hours. Eight sensors are attached to the patient’s abdomen (similar to electrocardiographic sensors). This sensor array is attached by means of a coaxial cable to a portable recorder. The recorder is battery operated and is composed of a receiver, a processor, and a hard disk to store the images. The patient wears both the recorder and the battery on a belt around the waist (Fig 2).
As soon as the capsule is swallowed, it begins acquiring images at a rate of two images per second. These color video images (256 x 256 pixels) are transmitted by means of ultra-high-frequency radio telemetry to the sensor array, which subsequently sends the data to the recorder. The capsule can acquire about 50,000 images. The sensors also allow a rough approximation of the capsule location along a trajectory that is based on the length of time the capsule has been traveling through the gastrointestinal tract.
The capsule batteries have an 8-hour life, during which time the patient is allowed to resume normal activities, including the ingestion of medications after 2 hours and of food after 4 hours. After 8 hours, the patient returns the recorder, and the data are downloaded to a computer workstation (RAPID workstation, Given Imaging), where the images can be evaluated at a rate of 1–40 images/sec by a physician. The capsule is excreted naturally, usually after 8–72 hours, although occasionally excretion can take as along as 2 weeks. The capsule is then discarded.
|
Advantages and Disadvantages of Capsule Endoscopy
|
|---|
The main advantage of capsule endoscopy is the ability to noninvasively evaluate the entire small bowel mucosa. Conventional endoscopic techniques allow examination of only the most proximal and distal portions of the small bowel, leaving the majority of the small bowel unexamined. Complete small bowel evaluation was previously possible only with indirect methods such as barium examination, computed tomography (CT), and magnetic resonance (MR) imaging. Suboptimal bowel distention and overlapping bowel loops make radiologic evaluation challenging. It is not surprising, then, that multiple studies have concluded that capsule endoscopy is superior in the detection of significant mucosal disease such as angiodysplasias, ulcers, and polyps (2,3).
The main disadvantage of capsule endoscopy is the inability to definitively localize or treat small bowel lesions. The method of roughly approximating capsule location described earlier is obviously prone to inaccuracy due to differences in small bowel transit time or variant anatomy. Anatomic landmarks such as the stomach and cecum are also used for lesion localization; however, this technique is useful mainly for only the proximal and distal small bowel. Unlike standard endoscopy, capsule endoscopy does not allow treatment or biopsy sampling of abnormalities, cleansing of poorly prepared areas, or a change in camera orientation to further evaluate an abnormality. In addition, capsule endoscopy yields lower-quality images compared with standard endoscopy.
|
Indications for Capsule Endoscopy
|
|---|
Obscure Gastrointestinal Bleeding
The most common indication for capsule endoscopy is obscure gastrointestinal bleeding, defined as bleeding of unknown origin that persists or recurs after negative endoscopy of the upper and lower gastrointestinal tract. Obscure gastrointestinal bleeding can be further categorized as overt (hematemesis, hematochezia, melena) or occult (iron-deficiency anemia, blood in stools not grossly apparent). Typically, a small bowel follow-through study or enteroclysis is performed to exclude a small bowel cause during work-up for obscure gastrointestinal bleeding. However, these examinations are typically low-yield studies in this population, with positive findings in less than 10% of cases (4). Up to 30%–50% of patients continue to suffer from unexplained gastrointestinal blood loss after undergoing complete work-up with these standard endoscopic and radiologic studies (5).
Several studies of patients with obscure gastrointestinal bleeding have demonstrated superior results with capsule endoscopy compared with radiologic studies. In a study by Ell et al (6) of 32 patients with obscure gastrointestinal bleeding, small bowel barium studies were negative in all patients, whereas a definitive source of bleeding was diagnosed with capsule endoscopy in 21 patients (66%). A retrospective comparison of capsule endoscopy, barium examinations, and CT in 52 patients (43 with obscure gastrointestinal bleeding) also reported a low diagnostic yield for barium examinations (3%) and standard CT (21%) compared with capsule endoscopy (55%–63%) (2). Capsule endoscopy established the diagnosis more often than even a dedicated CT examination of the small bowel (CT enteroclysis) (n = 4 vs n = 1) in eight patients with obscure gastrointestinal bleeding (7).
Push enteroscopy, which provides endoscopic visualization approximately 120 cm past the ligament of Treitz, also has a lower diagnostic yield for obscure gastrointestinal bleeding than does capsule endoscopy. In a study of 50 patients, push enteroscopy yielded the diagnosis in 32% of cases compared with 68% for capsule endoscopy (8). The study by Ell et al (6) showed similar results (28% for push enteroscopy vs 66% for capsule endoscopy).
Crohn Disease
Another common indication for capsule endoscopy is Crohn disease. Although the diagnosis of advanced Crohn disease with strictures, fistulas, and abscess can be made with barium examinations and CT, the diagnosis of early Crohn disease can be challenging. In several small studies, capsule endoscopy has outperformed barium studies and CT in the diagnosis of early Crohn disease. For example, in one study of 20 patients, Crohn disease was diagnosed with capsule endoscopy in six patients with normal findings at small bowel follow-through examination or CT enteroclysis and was ruled out in three patients with suspicious radiologic findings (9). In addition, extended involvement by Crohn disease was discovered at capsule endoscopy in three patients. The overall diagnostic yield was 70% for capsule endoscopy and 35% for radiologic studies, a statistically significant difference (9). In another study of 17 patients, Crohn disease was diagnosed in 70%, all of whom had a negative small bowel follow-through study (10).
Additional Potential Indications
Capsule endoscopy may also be used in the evaluation of patients with hereditary polyposis syndromes, small bowel damage induced by nonsteroidal anti-inflammatory drugs (NSAIDs), chronic diarrhea, or chronic abdominal pain.
|
Contraindications for Capsule Endoscopy
|
|---|
A known small bowel stricture or obstruction is a contraindication for capsule endoscopy, since capsules that are not excreted naturally will require surgical removal. Patients with strictures due to NSAIDs or Crohn disease may improve with medical treatment and not require surgery, so that a known stricture would obviate capsule endoscopy. On the other hand, patients with obstructive symptoms of unknown cause may benefit from capsule endoscopy even if the capsule is not excreted. In these patients, the capsule will be retained at the point of obstruction, providing endoscopic images of the cause or localizing the abnormal area prior to surgery.
Because of a lack of data, capsule endoscopy is not recommended in pregnant patients or in patients requiring MR imaging before the capsule is expelled. A pacemaker is also a relative contraindication, since there is a theoretic risk that the capsule could interfere with pacemaker function. However, this risk has not been found to occur clinically (11). Patients with dysphagia may require the capsule to be placed endoscopically in the small intestine. Prior intestinal surgery is a relative contraindication, and a barium study prior to capsule endoscopy is often recommended in these patients.
|
Lesions Detected at Capsule Endoscopy
|
|---|
Arteriovenous Malformations
Arteriovenous malformations, also referred to as angiodysplastic lesions, telangiectasias, or angioectasias, are the most common abnormality accounting for obscure gastrointestinal bleeding, seen in 21%–53% of patients who undergo capsule endoscopy (Fig 3) (2,3,6,8,12,13). They occur more frequently with increasing age and can be identified at endoscopy as spiderlike lesions.
View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Bleeding angioectasia in a patient with obscure gastrointestinal bleeding. (a) Capsule endoscopic image shows angioectasia (circled) in an area of gross blood. (b) Angiogram shows angioectasia in the jejunum. (c) Intraoperative endoscopic image helps confirm the presence of bleeding angioectasia in the jejunum.
|
|
View larger version (157K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Bleeding angioectasia in a patient with obscure gastrointestinal bleeding. (a) Capsule endoscopic image shows angioectasia (circled) in an area of gross blood. (b) Angiogram shows angioectasia in the jejunum. (c) Intraoperative endoscopic image helps confirm the presence of bleeding angioectasia in the jejunum.
|
|
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. Bleeding angioectasia in a patient with obscure gastrointestinal bleeding. (a) Capsule endoscopic image shows angioectasia (circled) in an area of gross blood. (b) Angiogram shows angioectasia in the jejunum. (c) Intraoperative endoscopic image helps confirm the presence of bleeding angioectasia in the jejunum.
|
|
Actively bleeding arteriovenous malformations can be detected radiologically with angiography and technetium-99m scintigraphy. Unfortunately, these studies can be difficult to perform during an active episode of gastrointestinal bleeding and therefore can yield false-negative findings in a patient with intermittent bleeding. CT angiography has not been used to detect small bowel arteriovenous malformations, but they may be detectable with multidetector scanners and fast scanning techniques. In fact, in one study, 70% of angiodysplasias in the colon were correctly identified with CT angiography (14).
It is controversial whether all arteriovenous malformations that are identified are clinically significant. In one study of 26 patients with ongoing overt bleeding, arteriovenous malformations were identified at capsule endoscopy in 13 patients, 11 of whom were treated with cauterization or hormone therapy (13). Treatment resulted in resolution of the bleeding in all 11 patients; the two patients who were not treated continued to bleed. Conversely, the same study found that in patients who did not have ongoing overt bleeding but rather had previous overt bleeding, guaiacpositive stools, or iron-deficiency anemia, the treatment or nontreatment of arteriovenous malformations did not affect the outcome. These results suggest that arteriovenous malformations may be significant only in patients with active bleeding at endoscopy or with ongoing overt bleeding.
Small Bowel Tumors
Small bowel neoplasms are a less common but nonetheless important cause of obscure gastrointestinal bleeding (Figs 4–11), accounting for only 1.4% of gastrointestinal cancers (15) but responsible for 75% of the symptomatic small bowel lesions that require surgery (16,17). In 872 patients undergoing capsule endoscopy in multiple studies, small bowel tumors were found in 3.8% (18).
View larger version (175K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Gastrointestinal stromal tumor. Coronal CT scan shows a gastrointestinal stromal tumor (circled), a finding that was confirmed surgically. Capsule endoscopy was nondiagnostic due to retained food.
|
|
View larger version (88K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Familial polyposis in a patient with negative findings at small bowel follow-through examination and CT. Capsule endoscopic (a) and esophagogastroduodenoscopic (EGD) (b) images show multiple small polyps in the proximal small bowel (arrows in a).
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Familial polyposis in a patient with negative findings at small bowel follow-through examination and CT. Capsule endoscopic (a) and esophagogastroduodenoscopic (EGD) (b) images show multiple small polyps in the proximal small bowel (arrows in a).
|
|
View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Lymphangioma. (a) Capsule endoscopic image shows multiple punctate white lesions (circled) in the proximal small bowel. (b) Intraoperative endoscopic image shows markedly thickened small bowel folds. (c) CT scan shows circumferential low-attenuation wall thickening in a jejunal segment (circled) causing narrowing of the lumen and enlargement of the small bowel loop. (d) Intraoperative photograph shows marked distention of a jejunal loop, a finding that corresponds to the abnormality seen at CT.
|
|
View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Lymphangioma. (a) Capsule endoscopic image shows multiple punctate white lesions (circled) in the proximal small bowel. (b) Intraoperative endoscopic image shows markedly thickened small bowel folds. (c) CT scan shows circumferential low-attenuation wall thickening in a jejunal segment (circled) causing narrowing of the lumen and enlargement of the small bowel loop. (d) Intraoperative photograph shows marked distention of a jejunal loop, a finding that corresponds to the abnormality seen at CT.
|
|
View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6c. Lymphangioma. (a) Capsule endoscopic image shows multiple punctate white lesions (circled) in the proximal small bowel. (b) Intraoperative endoscopic image shows markedly thickened small bowel folds. (c) CT scan shows circumferential low-attenuation wall thickening in a jejunal segment (circled) causing narrowing of the lumen and enlargement of the small bowel loop. (d) Intraoperative photograph shows marked distention of a jejunal loop, a finding that corresponds to the abnormality seen at CT.
|
|
View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6d. Lymphangioma. (a) Capsule endoscopic image shows multiple punctate white lesions (circled) in the proximal small bowel. (b) Intraoperative endoscopic image shows markedly thickened small bowel folds. (c) CT scan shows circumferential low-attenuation wall thickening in a jejunal segment (circled) causing narrowing of the lumen and enlargement of the small bowel loop. (d) Intraoperative photograph shows marked distention of a jejunal loop, a finding that corresponds to the abnormality seen at CT.
|
|
View larger version (76K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Bleeding cavernous hemangioma in a patient with obscure gastrointestinal bleeding. (a) Capsule endoscopic image shows hemorrhage in the proximal small bowel. (b) EGD image shows a suspected submucosal mass in the jejunum. (c) Axial CT scan shows a small soft-tissue mass in the jejunum (circled). (d) Coronal CT scan shows the small soft-tissue mass in the jejunum (circled) with punctate calcifications. (e) Intraoperative photograph shows a small bleeding cavernous hemangioma in the jejunal wall.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Bleeding cavernous hemangioma in a patient with obscure gastrointestinal bleeding. (a) Capsule endoscopic image shows hemorrhage in the proximal small bowel. (b) EGD image shows a suspected submucosal mass in the jejunum. (c) Axial CT scan shows a small soft-tissue mass in the jejunum (circled). (d) Coronal CT scan shows the small soft-tissue mass in the jejunum (circled) with punctate calcifications. (e) Intraoperative photograph shows a small bleeding cavernous hemangioma in the jejunal wall.
|
|
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. Bleeding cavernous hemangioma in a patient with obscure gastrointestinal bleeding. (a) Capsule endoscopic image shows hemorrhage in the proximal small bowel. (b) EGD image shows a suspected submucosal mass in the jejunum. (c) Axial CT scan shows a small soft-tissue mass in the jejunum (circled). (d) Coronal CT scan shows the small soft-tissue mass in the jejunum (circled) with punctate calcifications. (e) Intraoperative photograph shows a small bleeding cavernous hemangioma in the jejunal wall.
|
|
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7d. Bleeding cavernous hemangioma in a patient with obscure gastrointestinal bleeding. (a) Capsule endoscopic image shows hemorrhage in the proximal small bowel. (b) EGD image shows a suspected submucosal mass in the jejunum. (c) Axial CT scan shows a small soft-tissue mass in the jejunum (circled). (d) Coronal CT scan shows the small soft-tissue mass in the jejunum (circled) with punctate calcifications. (e) Intraoperative photograph shows a small bleeding cavernous hemangioma in the jejunal wall.
|
|
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7e. Bleeding cavernous hemangioma in a patient with obscure gastrointestinal bleeding. (a) Capsule endoscopic image shows hemorrhage in the proximal small bowel. (b) EGD image shows a suspected submucosal mass in the jejunum. (c) Axial CT scan shows a small soft-tissue mass in the jejunum (circled). (d) Coronal CT scan shows the small soft-tissue mass in the jejunum (circled) with punctate calcifications. (e) Intraoperative photograph shows a small bleeding cavernous hemangioma in the jejunal wall.
|
|
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Adenocarcinoma. (a) Capsule endoscopic image demonstrates blood in the proximal small bowel. (b) EGD image shows a bleeding lobulated mass. (c) Image from a small bowel follow-through study demonstrates an irregular jejunal bowel loop (circled) representing adenocarcinoma. The examination was initially thought to be negative. (d) CT scan shows diffuse, irregular thickening of the jejunal loop (circled), a finding that corresponds to the tumor.
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Adenocarcinoma. (a) Capsule endoscopic image demonstrates blood in the proximal small bowel. (b) EGD image shows a bleeding lobulated mass. (c) Image from a small bowel follow-through study demonstrates an irregular jejunal bowel loop (circled) representing adenocarcinoma. The examination was initially thought to be negative. (d) CT scan shows diffuse, irregular thickening of the jejunal loop (circled), a finding that corresponds to the tumor.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8c. Adenocarcinoma. (a) Capsule endoscopic image demonstrates blood in the proximal small bowel. (b) EGD image shows a bleeding lobulated mass. (c) Image from a small bowel follow-through study demonstrates an irregular jejunal bowel loop (circled) representing adenocarcinoma. The examination was initially thought to be negative. (d) CT scan shows diffuse, irregular thickening of the jejunal loop (circled), a finding that corresponds to the tumor.
|
|
View larger version (153K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8d. Adenocarcinoma. (a) Capsule endoscopic image demonstrates blood in the proximal small bowel. (b) EGD image shows a bleeding lobulated mass. (c) Image from a small bowel follow-through study demonstrates an irregular jejunal bowel loop (circled) representing adenocarcinoma. The examination was initially thought to be negative. (d) CT scan shows diffuse, irregular thickening of the jejunal loop (circled), a finding that corresponds to the tumor.
|
|
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9. Carcinoid tumor. Capsule endoscopic image shows a carcinoid tumor measuring 1.7 cm in diameter that was not detected at barium examination or CT. (Reprinted, with permission, from reference 2.)
|
|
View larger version (82K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10a. Lymphoma. (a) Capsule endoscopic image shows small nodules in the jejunum. (b) EGD image demonstrates an irregular lobulated mass. (c) Image from a small bowel follow-through study demonstrates an abnormal jejunal small bowel loop (circled). The examination was initially thought to be negative. (d) CT scan shows eccentric irregular thickening of the jejunal loop (circled), a finding that represents lymphoma.
|
|
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10b. Lymphoma. (a) Capsule endoscopic image shows small nodules in the jejunum. (b) EGD image demonstrates an irregular lobulated mass. (c) Image from a small bowel follow-through study demonstrates an abnormal jejunal small bowel loop (circled). The examination was initially thought to be negative. (d) CT scan shows eccentric irregular thickening of the jejunal loop (circled), a finding that represents lymphoma.
|
|
View larger version (172K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10c. Lymphoma. (a) Capsule endoscopic image shows small nodules in the jejunum. (b) EGD image demonstrates an irregular lobulated mass. (c) Image from a small bowel follow-through study demonstrates an abnormal jejunal small bowel loop (circled). The examination was initially thought to be negative. (d) CT scan shows eccentric irregular thickening of the jejunal loop (circled), a finding that represents lymphoma.
|
|
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10d. Lymphoma. (a) Capsule endoscopic image shows small nodules in the jejunum. (b) EGD image demonstrates an irregular lobulated mass. (c) Image from a small bowel follow-through study demonstrates an abnormal jejunal small bowel loop (circled). The examination was initially thought to be negative. (d) CT scan shows eccentric irregular thickening of the jejunal loop (circled), a finding that represents lymphoma.
|
|
View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11a. Ovarian cancer invading the terminal ileum. (a) Capsule endoscopic image shows a small bowel stricture of unknown origin (arrows) causing obstruction of the capsule. (b) CT scan shows the capsule (circled) in the distal ileum. (c) CT scan shows a large, heterogeneous mass (arrow) invading the terminal ileum and causing a small bowel stricture. (d) CT scan shows the invasive mass arising from the right ovary.
|
|
View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11b. Ovarian cancer invading the terminal ileum. (a) Capsule endoscopic image shows a small bowel stricture of unknown origin (arrows) causing obstruction of the capsule. (b) CT scan shows the capsule (circled) in the distal ileum. (c) CT scan shows a large, heterogeneous mass (arrow) invading the terminal ileum and causing a small bowel stricture. (d) CT scan shows the invasive mass arising from the right ovary.
|
|
View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11c. Ovarian cancer invading the terminal ileum. (a) Capsule endoscopic image shows a small bowel stricture of unknown origin (arrows) causing obstruction of the capsule. (b) CT scan shows the capsule (circled) in the distal ileum. (c) CT scan shows a large, heterogeneous mass (arrow) invading the terminal ileum and causing a small bowel stricture. (d) CT scan shows the invasive mass arising from the right ovary.
|
|
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11d. Ovarian cancer invading the terminal ileum. (a) Capsule endoscopic image shows a small bowel stricture of unknown origin (arrows) causing obstruction of the capsule. (b) CT scan shows the capsule (circled) in the distal ileum. (c) CT scan shows a large, heterogeneous mass (arrow) invading the terminal ileum and causing a small bowel stricture. (d) CT scan shows the invasive mass arising from the right ovary.
|
|
The detection of small bowel tumors is challenging due to limited evaluation of the small bowel with standard endoscopy and the suboptimal performance of small bowel follow-through studies. Although abnormalities may be seen at small bowel follow-through studies in up to 83% of patients with small bowel tumors, direct evidence of tumor is found in only 30%–44% (19–21). Enteroclysis is the examination of choice for detecting small bowel neoplasms, outperforming both small bowel follow-through examinations and CT in several studies (22,23), but is a more difficult examination for both the radiologist and the patient, requiring nasojejunal intubation and oral administration of large volumes of contrast material. The thinner collimation now available with multidetector CT scanners and the three-dimensional imaging capabilities of CT may improve detection of small bowel tumors (Figs 4, 6–8, 10, 11) (24), but this modality is still unlikely to outperform a technique like capsule endoscopy that provides direct mucosal evaluation.
To our knowledge, no study has yet been conducted that specifically compares capsule endoscopy with radiologic imaging in the diagnosis of small bowel neoplasms. However, several studies have demonstrated that capsule endoscopy helps detect more tumors than does radiologic imaging. In one study, two of four surgically confirmed small bowel tumors were detected at capsule endoscopy; the two remaining tumors were missed due to poor bowel preparation. In the same study, no tumors were identified at barium examination, and a tumor was detected with CT in only one of three patients with tumors who underwent CT (2). In the study by Ell et al (6), two malignancies were detected at capsule endoscopy but none at small bowel follow-through examination.
Ulcers
Small bowel ulcers are another common abnormality detected at capsule endoscopy (Fig 12). Although the majority of small bowel ulcers detected at capsule endoscopy are due to Crohn disease (Figs 13, 14) or NSAIDs, other causes include infection, ischemia, trauma, or vasculitis.
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13a. Early Crohn disease. (a) Image from a small bowel follow-through study shows mild nodularity in the terminal ileum (arrows). (b) CT scan shows mild diffuse wall thickening and mucosal enhancement of the terminal ileum (arrow). (c) Capsule endoscopic image shows a small apthous ulcer in the terminal ileum (circled).
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13b. Early Crohn disease. (a) Image from a small bowel follow-through study shows mild nodularity in the terminal ileum (arrows). (b) CT scan shows mild diffuse wall thickening and mucosal enhancement of the terminal ileum (arrow). (c) Capsule endoscopic image shows a small apthous ulcer in the terminal ileum (circled).
|
|
View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13c. Early Crohn disease. (a) Image from a small bowel follow-through study shows mild nodularity in the terminal ileum (arrows). (b) CT scan shows mild diffuse wall thickening and mucosal enhancement of the terminal ileum (arrow). (c) Capsule endoscopic image shows a small apthous ulcer in the terminal ileum (circled).
|
|
View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14a. Early Crohn disease in another patient. (a) Image from a small bowel follow-through study shows mild nodularity in the terminal ileum (arrow). (b) CT scan shows mild diffuse wall thickening and markedly increased mucosal enhancement of the terminal ileum (arrow). (c) Capsule endoscopic image shows a small apthous ulcer in the terminal ileum (circled).
|
|
View larger version (118K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14b. Early Crohn disease in another patient. (a) Image from a small bowel follow-through study shows mild nodularity in the terminal ileum (arrow). (b) CT scan shows mild diffuse wall thickening and markedly increased mucosal enhancement of the terminal ileum (arrow). (c) Capsule endoscopic image shows a small apthous ulcer in the terminal ileum (circled).
|
|
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14c. Early Crohn disease in another patient. (a) Image from a small bowel follow-through study shows mild nodularity in the terminal ileum (arrow). (b) CT scan shows mild diffuse wall thickening and markedly increased mucosal enhancement of the terminal ileum (arrow). (c) Capsule endoscopic image shows a small apthous ulcer in the terminal ileum (circled).
|
|
Ulcers due to Crohn Disease.—
As mentioned previously, multiple small studies have demonstrated that Crohn disease is detected more often at capsule endoscopy than at barium examination. Capsule endoscopy allows direct visualization of subtle mucosal inflammatory changes or erosions present in early Crohn disease that are not commonly seen at barium studies. Barium examinations are more successful in demonstrating later findings of Crohn disease, including deep ulcers with a cobblestone appearance, wall thickening, strictures, and fistulas. Capsule endoscopy is also superior to standard endoscopy in helping identify changes throughout the bowel, not just in the proximal or distal portion.
It is likely that some small bowel ulcers are missed at radiologic imaging due to inherent limitations of the technique and not solely to the radiologist’s technique or interpretive skills. In one study, even when a world-renowned specialist in enteroclysis was informed that three patients had small bowel ulcers identified at capsule endoscopy, the lesions could not be demonstrated at subsequent enteroclysis (25). The apparent superiority of capsule endoscopy over radiologic imaging may also be somewhat misleading due to selection bias in multiple studies (26). This selection bias is due to the fact that, in general, only patients with negative radiologic findings undergo capsule endoscopy. Therefore, patients with positive radiologic findings are often excluded from analysis.
Although capsule endoscopy often demonstrates findings of Crohn disease in patients with negative barium examinations and CT studies, radiologic imaging continues to play an important role. Barium examinations are often needed to exclude a small bowel obstruction or stricture prior to capsule endoscopy. In addition, barium studies and CT are often performed in patients with known strictures or fistulas who cannot undergo capsule endoscopy due to risk of capsule retention.
MR enteroclysis is another technique used in the evaluation of Crohn disease (27–29). In one study of 25 patients undergoing MR enteroclysis and conventional enteroclysis, findings of Crohn disease were at least equally well visualized at MR enteroclysis as at conventional enteroclysis in 24 patients (29). However, CT enteroclysis may be even better than MR enteroclysis in this setting. Another study comparing MR enteroclysis with CT enteroclysis found the latter to be more sensitive in detecting small bowel disease, with better interobserver agreement (30). MR enteroclysis has inherent advantages in younger patients in that, unlike CT or barium examination, it does not involve repeated radiation exposure. However, MR enteroclysis is likely not routinely used for the evaluation of Crohn disease owing to cost and availability issues. To our knowledge, no study has yet been conducted comparing MR enteroclysis with capsule endoscopy.
NSAID-induced Ulcers.—
NSAID use is another cause of small bowel ulcers and strictures (Fig 15). Clinical manifestations of NSAID-induced small bowel disease include obscure gastrointestinal bleeding, iron-deficiency anemia, diarrhea, protein-losing enteropathy, abdominal pain, obstruction, or perforation. This disease is induced by chronic use of NSAIDs (>6 months), causing small bowel ulceration, hemorrhage, and, eventually, strictures ("diaphragm disease"). Although ulcerations and erosions are nonspecific endoscopic findings, the intestinal diaphragms causing the stricture are thought to be pathognomonic for NSAID-induced small bowel injury (31). These lesions are very difficult to detect radiologically because they resemble normal plicae circulares (32) and usually do not manifest as a complete obstruction. NSAID-induced strictures are a common cause of nonnatural excretion of the capsule, accounting for five of seven cases in one study (1). These strictures rarely cause an acute obstruction, although there may be transient pain as the bowel attempts to pass the capsule through the narrowed area.
View larger version (81K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15a. NSAID-induced stricture resulting in a retained capsule. (a) Capsule endoscopic image shows a small bowel stricture. The capsule became impacted at this point. (b) CT scan shows the capsule (circled) impacted in the ileum. (c) Photograph shows the retained capsule lodged in the stricture.
|
|
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15b. NSAID-induced stricture resulting in a retained capsule. (a) Capsule endoscopic image shows a small bowel stricture. The capsule became impacted at this point. (b) CT scan shows the capsule (circled) impacted in the ileum. (c) Photograph shows the retained capsule lodged in the stricture.
|
|
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15c. NSAID-induced stricture resulting in a retained capsule. (a) Capsule endoscopic image shows a small bowel stricture. The capsule became impacted at this point. (b) CT scan shows the capsule (circled) impacted in the ileum. (c) Photograph shows the retained capsule lodged in the stricture.
|
|
|
False-Negative and False-Positive Findings at Capsule Endoscopy
|
|---|
Lesions can be missed at capsule endoscopy due to rapid small bowel transit, orientation of the camera away from a lesion, poor bowel preparation (Fig 4), delayed small bowel transit, small bowel stricture causing a downstream lesion to be missed, and perceptual error. Fistulas are also commonly not detected at capsule endoscopy. In the 2004 study of 52 patients by Hara et al (2), two surgically confirmed small bowel tumors, one ulcer, and one coloenteric fistula were not detected at capsule endoscopy. In another study of 56 patients who underwent follow-up (push enteroscopy, intraoperative endoscopy, surgical resection) for positive capsule endoscopic findings, capsule endoscopy proved to have demonstrated four false-negative findings (three duodenal arteriovenous malformations, one small bowel adenocarcinoma) and one false-positive finding (bleeding polyp). Although the adenocarcinoma was not identified at capsule endoscopy, the capsule became impacted at this point, and the tumor was diagnosed at surgery. The sensitivity, specificity, positive predictive value, and negative predictive value of capsule endoscopy in this study were 89%, 95%, 97%, and 83%, respectively (13).
The false-positive rate for capsule endoscopy is difficult to confirm, since the only applicable standard of reference is intraoperative endoscopy, and false-negative results may occur at laparoscopy due to physiologic changes. Two studies involving 25 patients who underwent intraoperative endoscopy for positive capsule endoscopic findings demonstrated false-positive findings in four patients (16%) and true-positive findings in 21 patients (84%). False-positive findings included a venous lake and a submucosal lesion (33). Because most studies do not compare capsule endoscopy with the standard of reference (intraoperative endoscopy), most capsule endoscopic studies report diagnostic yields that do not allow differentiation between true-positive and false-positive findings. In fact, in a study of 356 asymptomatic healthy patients, 13.8% had abnormal capsule endoscopic findings such as small bowel mucosal breaks (34). These results suggest that current capsule endoscopy is markedly superior to standard diagnostic techniques, although the actual clinical relevance of many of the capsule endoscopic findings remains to be demonstrated.
|
Complications of Capsule Endoscopy
|
|---|
Complications of capsule endoscopy are uncommon. The overall nonnatural excretion rate reported by Given Imaging is 0.75% and is based on over 10,000 capsule endoscopic examinations (9). Retained capsules requiring surgical removal are most commonly seen in patients with Crohn disease or diaphragm disease (Fig 15 ) associated with NSAID use (35,36). Other potential reasons for capsule retention are postoperative changes with a blind-ending loop, diverticula, or tumor obstruction.
|
Conclusions
|
|---|
Capsule endoscopy is a revolutionary technique that allows, for the first time, a complete, noninvasive endoscopic examination of the small bowel. Most studies have demonstrated superior detection of small bowel disease with capsule endoscopy compared with standard endoscopic or radiologic techniques. Capsule endoscopy appears to be most useful in the difficult evaluation of obscure gastrointestinal bleeding when barium examinations and standard endoscopy are negative. Limitations of capsule endoscopy include difficulty in localizing lesions and restricted use in patients with strictures or obstruction.
|
Acknowledgments
|
|---|
The authors thank Bonnie L. Schimek for her assistance in manuscript preparation.
|
Footnotes
|
|---|
See the commentary by Maglinte following this article.
|
References
|
|---|
Top
Abstract
Introduction
