Chapter 33. Wireless Capsule Endoscopy & Deep Small Bowel Enteroscopy

General Considerations

The first human ingestion of a video capsule endoscope occurred in 1999, and high-quality images from healthy human volunteers were described in 2000. A commercially available capsule was approved by the U.S. Food and Drug Administration (FDA) for use in the United States in August 2001, having been approved for use in Europe earlier that year. Between 2001 and 2007, more than a half million capsule endoscopy studies were performed.

Currently, three video capsules are available for use in the United States. Two are made by Given Imaging (Yoqneam, Israel). One capsule, the PillCam SB2, images the small bowel, while the other, the PillCam ESO2, is designed for imaging of the esophagus. Olympus Medical Systems Corporation (Tokyo, Japan) has also released a small bowel capsule, the EndoCapsule. Two other capsules have been developed and are being studied in clinical trials. The first is a colon capsule for colorectal cancer screening (Given Imaging), and the second is the MiRo capsule (Intro-Medic Co, Ltd, Seoul, South Korea), which uses electric field propagation to transmit the images.

The PillCam SB2 is composed of a complementary metal oxide silicon (CMOS) chip camera, a short focal length lens, six white light-emitting diode (LED) illumination sources, two silver oxide batteries, and a UHF band radio telemetry transmitter. Improvements in the design of CMOS image sensors, application-specific integrated circuits (ASICS) and white LED illumination, along with the development of nontoxic batteries made from silver oxide, allowed for the development of video capsules. The PillCam SB2 and the PillCam ESO2 capsules are 11 × 26 mm and weigh 3.7 g. The images acquired by the capsules have a field of view of 156 degrees, with eightfold magnification and a 1–30 mm depth of view. The capsules have a resolution of 0.1 mm that allows for the visualization of individual villi. The small bowel capsule transmits two 256 × 256 pixel color images per second for a total of 8 hours (~55,000 images), and the esophageal capsule transmits nine images per second from each of its two lenses, for a total of 18 images per second over the course of 20 minutes (~22,000 images). The EndoCapsule is similar to the PillCam SB2, but instead of a CMOS chip it has a charge-coupled device.

Indications

The indications for small bowel and esophageal capsule endoscopy are summarized in Table 33–1.

Obscure Gastrointestinal Bleeding

Obscure gastrointestinal bleeding refers to bleeding that persists or recurs, without a source identified after standard endoscopic evaluation with upper endoscopy and colonoscopy. Obscure gastrointestinal hemorrhage can be subdivided into obscure-overt and obscure-occult gastrointestinal bleeding. Patients with obscure-overt gastrointestinal bleeding present with visual evidence of bleeding, such as hematemesis, melena, or hematochezia. Obscure-occult bleeding is manifested by stool that is positive for occult blood, frequently with iron deficiency anemia. Obscure bleeding accounts for approximately 5% of patients with gastrointestinal bleeding, and the source of bleeding is frequently from the small bowel, between the ligament of Treitz and the ileocecal valve.

Common causes of small bowel bleeding include vascular ectasias, small bowel tumors, Crohn disease, and nonsteroidal anti-inflammatory drug (NSAID) enteropathy (Table 33–2; see also Plates 77 and 78). Vascular ectasias are the most common cause of obscure gastrointestinal bleeding, accounting for 30–40% overall, and are responsible for the majority of obscure gastrointestinal bleeds in older patients. Patients between the ages of 30 and 50 years are more likely to have a small bowel tumor as the source of their bleeding. Finally, young patients often are found to have an ulcerated Meckel diverticulum as the source, although it should be remembered that bleeding from a Meckel diverticulum may occur at any age.

Crohn Disease

In patients with Crohn disease limited to the small bowel, or in those with indeterminate colitis, arriving at a correct diagnosis can be difficult. Capsule endoscopy can help diagnose small bowel Crohn disease by providing mucosal detail that is not available radiologically (Plate 79). It also provides the opportunity to visualize areas of the bowel not accessible by standard endoscopy. Because capsule endoscopy has a resolution of 0.1 mm, it is able to detect small, superficial defects, such as aphthoid ulcers. It is important to note, however, that not all mucosal ulcerations occur as a result of Crohn disease (Table 33–3). Ten to 23% of normal volunteers who are not taking NSAIDs will have mucosal breaks and other lesions seen on capsule endoscopy, and 50–71% of NSAID users will have evidence of small bowel injury (red spots, erosions, and ulcerations). Therefore, it is important to use capsule endoscopy findings in combination with other clinical information to arrive at the correct diagnosis.

Small Bowel Tumors and Polyposis Syndromes

Benign and malignant small bowel tumors are found in approximately 6–8% of people and can be seen on capsule endoscopy (Plate 80). Capsule endoscopy may be able to detect tumors years before they would be detected by other imaging modalities, potentially increasing the chance to remove them while still localized.

Capsule endoscopy can also detect small bowel polyps in patients with hereditary polyposis syndromes. Peutz-Jeghers syndrome and familial adenomatous polyposis (FAP) are two of the hereditary polyposis syndromes that are associated with small bowel polyps and malignancies. Patients with Peutz-Jeghers syndrome form hamartomatous polyps within the gastrointestinal tract, along with mucocutaneous pigmentation. The polyps have a predilection for the small bowel and will develop in more than 90% of patients (Plate 81). Complications from the polyps include gastrointestinal bleeding, anemia, intestinal obstruction, intussusception, and development of cancer. Patients with Peutz-Jeghers syndrome have a 13% lifetime risk of developing small bowel cancer, so surveillance of the small intestine is recommended. Traditionally, patients have undergone radiographic imaging with a small bowel series or enteroclysis every 2 years, starting at age 10. This approach exposes a patient to significant amounts of ionizing radiation. In addition, radiologic evaluation with small bowel follow-through, computed tomography (CT) enteroclysis, or magnetic resonance enteroclysis lacks sensitivity for detecting small polyps. Push enteroscopy has also been employed, but is only able to visualize a portion of the upper small bowel. Capsule endoscopy on the other hand, is capable of detecting small polyps throughout the small bowel.

Celiac Disease

Celiac disease affects up to 0.3–1% of Caucasians. Celiac disease has traditionally been diagnosed by detection of antibodies (eg, antiendomysial and tissue transglutaminase antibodies), followed by endoscopy with biopsies of the small bowel to confirm the diagnosis. Capsule endoscopy has been proposed as a possible alternative to endoscopy for the diagnosis of celiac disease. Capsule endoscopy findings of celiac disease include scalloping, a mosaic mucosal pattern, loss of mucosal folds, visible vessels, and micronodularity (Plates 82 and 83). An advantage of capsule endoscopy is that it allows for visualization of the entire small bowel, and symptom severity in celiac disease is more closely related to the length of small bowel involved and not to the severity of the villous atrophy seen on biopsy.

Capsule endoscopy may have an even more important role in the evaluation of refractory celiac disease. Celiac disease can be complicated by small bowel adenocarcinoma, lymphoma (including enteropathy-associated T-cell lymphoma), and ulcerative jejunitis. In patients who fail to respond to a gluten-free diet or who have a recurrence of symptoms while on a gluten-free diet, further evaluation is needed to differentiate between refractory celiac disease (ie, celiac disease that does not respond to a gluten-free diet), ongoing gluten ingestion (intentional or unintentional), or a complication of celiac disease.

Other Applications

Capsule endoscopy has also been used to evaluate for evidence of small bowel graft-versus-host disease in patients following bone marrow transplantation, to look for evidence of rejection following small intestinal transplantation, and to evaluate for radiation enteritis. However, in the case of suspected radiation enteritis, the risk for capsule retention is increased. In addition, capsule endoscopy may detect small intestinal varices or portal hypertensive enteropathy in patients with portal hypertension.

The PillCam ESO2 was designed to visualize the esophagus and can be used to detect findings such as esophageal varices, erosive esophagitis, and Barrett esophagus.

Contraindications

Capsule endoscopy is contraindicated in some patients. Because cooperation is required, demented patients are not good candidates for capsule endoscopy. Capsule endoscopy is also contraindicated in patients with cardiac pacemakers or defibrillators (due to concerns that the capsule could interfere with the cardiac device) and in pregnant women (due to a lack of data on the effects of capsule endoscopy in pregnancy). Many centers, however, perform capsule endoscopy in patients with pacemakers or defibrillators, and studies have suggested that the capsules may not interfere with the devices. In addition, the cardiac devices have not been shown to disrupt the capsule study (with the exception of one instance of images being lost as a capsule passed the pulse generator of an abdominally implanted pacemaker). Finally, there have been no reports of a cardiac device malfunctioning due to a capsule study. Patients should be instructed not to undergo magnetic resonance imaging until passage of the capsule has been confirmed, which can be done with a plain abdominal radiograph if the capsule was not noted to pass naturally.

Patients with gastrointestinal tract obstructions, strictures, or fistulas (either suspected or demonstrated on imaging studies) should not undergo capsule endoscopy because of an increased risk of capsule retention or obstruction. A patency capsule (Agile Patency Capsule, Given Imaging) is available to establish small bowel patency in patients with suspected strictures or obstructions. The capsule is a dummy capsule with a transmitter. The patient swallows the capsule, and then after 30 hours a hand-held scanner or a plain abdominal film is used to determine if the capsule is still present in the small bowel. If it is not, then it is safe to proceed with the capsule endoscopy study. If the capsule is retained due to a stricture or obstruction it will begin to dissolve after 30 hours.

In patients with swallowing disorders, such as achalasia, esophageal strictures, esophageal diverticula, or gastroparesis, the capsule may need to be delivered using a special delivery device to ensure that it enters and traverses the small bowel during its recording time. Two devices are currently available, the AdvanCE device (US Endoscopy, Mentor, Ohio) and the PillCam Express (Given Imaging). Both devices are composed of a catheter that is passed through the channel of a gastroscope. A cap that holds the capsule is then screwed onto the end of the catheter. The scope is then advanced blindly into the esophagus (the cap obscures visualization). Once in the esophagus, the catheter can be advanced slightly to allow for limited visualization. The scope is then advanced to the small bowel, where the capsule is deployed. A standard upper endoscopy examination should be carried out before using the devices to detect any abnormalities that could complicate capsule delivery. Alternative delivery methods include placing the capsule in the stomach by inserting it through an over-tube and then advancing it to the small bowel using snares or nets.

Procedure

Small Bowel Examination

Patient Preparation

Studies can be performed on both inpatients and outpatients. For a small bowel study, a patient fasts for 8–12 hours before the examination. Some centers also have the patient consume only clear liquids for up to 24 hours prior to the examination. Data are conflicting, but some studies suggest that preparing the small bowel with 2–4 L of a polyethylene glycol solution may improve visualization (oral phospho soda should not be used, as it has been shown to increase gastric emptying time). A consensus panel at the 2005 International Conference on Capsule Endoscopy (ICCE), however, did not recommend a bowel preparation other than fasting prior to the study. In cases where there is concern for slowed gastric transit, such as in diabetic patients, a prokinetic agent such as metoclopramide or erythromycin is given approximately 30–60 minutes prior to the study. The ICCE consensus panel, however, did not recommend routine use of prokinetic agents. Some centers give patients simethicone prior to the study in an attempt to decrease air bubbles that can obscure the view of the mucosa, although this, too, was not recommended by the ICCE consensus panel.

Techniques

Prior to capsule ingestion, an eight-element sensor array is attached to the patient's abdomen, or a sensor belt is placed around the patient's abdomen. Removing the magnet that is packaged with the capsule activates the capsule. The patient then swallows the capsule in an upright position, or the capsule is placed endoscopically (see preceding text), and images are transmitted to a recording device worn about the patient's waist. During the study, patients are instructed to avoid activities that could lead to sensor detachment, such as exercise. After 2 hours, the patient is allowed to have clear liquids, and a light meal can be consumed after 4 hours. At the end of the study, the images are downloaded to a computer workstation equipped with proprietary software (either the RAPID Application, Given Imaging, Norcross, GA, or the Endo Capsule Software, Olympus, Tokyo, Japan). A physician can then read the study and generate a report, a process that takes on average 40–60 minutes. To aid in reading the study, the software has features such as a blood indicator that marks areas with suspected bleeding.

To allow real-time viewing of the images, hand-held devices have been developed. A possible disadvantage is that reading a study in real time takes significantly longer than reviewing downloaded images (the average small bowel transit time is ~4 hours). However, in certain situations, real-time imaging may have a role. For example, the capsule could be administered in the emergency department with immediate viewing to help guide further evaluation and management.

Esophageal Examination

Patient Preparation

For the esophageal examination, patients need to fast for 2 hours prior to ingesting the capsule.

Techniques

A three-element sensor array is attached to the patient's chest, and the patient consumes 100 mL of water with simethicone while standing. The patient then lies supine and the capsule is swallowed with a 10-mL sip of water. The patient remains supine for 2 minutes, and then progressively moves into an upright position (30 degrees for 2 minutes, 60 degrees for 1 minute, then upright for the remainder of the study). The images are then downloaded to the workstation and reviewed.

Due to difficulty with adherence to the ingestion protocol, an alternative simplified ingestion protocol has been developed for the esophageal examination. The patient lies in the right lateral decubitus position with his or her head on a pillow. The patient then swallows the capsule with a sip (approximately 15 mL) of water. The patient remains on the right side and takes an additional sip of water every 30 seconds for a total of 7 minutes. After 7 minutes, the patient sits upright and takes an additional sip of water. The patient may then get up and walk for the remainder of the study. The images are then downloaded to the computer workstation for review.

Outcomes

Obscure Gastrointestinal Bleeding

Obscure gastrointestinal bleeding is the most common indication for performing a capsule endoscopy. The overall yield of capsule endoscopy is 55–70%, and leads to alterations in management in approximately one third of patients (range, 25–71%). However, the yield is highly dependent on the indication for the study. In a 2004 study of 100 patients, if active bleeding was suspected, the yield was 92%, but the yield dropped to only 13% for patients with a previous overt gastrointestinal bleed (time between the bleed and capsule endoscopy ranged from 10 days to 1 year). For patients with occult gastrointestinal hemorrhage, the yield was intermediate at 44%. A second study of 47 patients from 2005, however, showed a yield of 100% for patients with ongoing bleeding, 67% for patients with prior overt bleeding (time between the bleed and capsule endoscopy ranged from 7 to 90 days), and 67% for occult gastrointestinal bleeding.

The difference in these two studies when it comes to patients with previous overt gastrointestinal bleeding may relate to the amount of time elapsed between the bleed and the capsule endoscopy. In the first study, 25 of the 31 patients had their capsule endoscopy studies at least 2 months after the bleeding episode, and the study found that the yield for capsule endoscopy dropped off in parallel with the length of time between the bleed and the capsule endoscopy. In patients who had an interval of 10–14 days, the yield was 67%, compared with 6% for those with an interval of 4–12 months. In the second study, the average time between the bleed and the capsule endoscopy was only 16 days, which may explain the much higher yield in this group. In the second study, when capsule endoscopy was compared with the gold standard of intraoperative enteroscopy, capsule endoscopy had a sensitivity of 95%, a specificity of 75%, a positive predictive value of 95%, and a negative predictive value of 86% for detecting a bleeding source.

Capsule endoscopy has also been compared with push enteroscopy, small bowel barium radiography (small bowel follow-through or enteroclysis), CT angiography, and mesenteric angiography in the evaluation of obscure gastrointestinal bleeding. A pooled analysis of seven prospective studies that compared capsule endoscopy with push enteroscopy demonstrated that capsule endoscopy had a yield of 71%, compared with 29% for push enteroscopy (although other studies have estimated the yield of push enteroscopy to be higher, at 40–65%). In a meta-analysis of 14 studies, the yield of capsule endoscopy in detecting clinically significant findings for obscure gastrointestinal bleeding was 56%, compared with 26% for push enteroscopy (P < .001). In that same meta-analysis, three studies that compared capsule endoscopy with small bowel barium radiography were analyzed. Capsule endoscopy had a 42% yield for detecting clinically significant lesions, compared with 6% for small bowel barium radiography (P < .001). A prospective study that compared capsule endoscopy with both CT angiography and mesenteric angiography in 25 patients who were able to complete all three studies found that capsule endoscopy was superior to CT angiography, detecting a bleeding source in 72% of patients compared with CT angiography, which detected a bleeding source in only 24% (P = .005). Capsule endoscopy also had a higher yield than mesenteric angiography, which had a yield of 56%, although the difference between the two studies did not reach statistical significance (P = .29).

Whether capsule endoscopy should be repeated in a patient with a negative examination has been examined. One small study found that 18 of 24 patients (75%) undergoing repeat capsule endoscopy for the evaluation of obscure-overt or obscure-occult gastrointestinal bleeding had new findings, and in 15 patients (62.5%), the findings resulted in a change in management. However, a second study with a median follow-up of 19 months found that in 18 patients with obscure-overt gastrointestinal bleeding who had negative capsule endoscopies, only one rebled (5.6%), compared with 15 of 31 patients (48.4%) who had positive capsule studies. Therefore, it is reasonable to wait for evidence of rebleeding before repeating a capsule endoscopy, since a large percentage of patients with a negative capsule endoscopy will not have further bleeding.

Capsule endoscopy has also been used in the evaluation of acute gastrointestinal bleeding in the emergency department. A study of 24 patients presenting with nonhematemesis gastrointestinal bleeding found that if capsule endoscopy was performed within 24 hours of presentation, the yield for identifying a bleeding source was 63% (10/16). In addition, capsule endoscopy demonstrated active bleeding in 54% (13/24) of the study patients.

Crohn Disease

Capsule endoscopy is superior to other diagnostic modalities in detecting small bowel disease when the suspicion for Crohn disease is high (Table 33–4). A meta-analysis demonstrated that capsule endoscopy is superior to small bowel radiography, colonoscopy with ileoscopy, CT enterography or enteroclysis, and push enteroscopy for the detection of nonstricturing small bowel Crohn disease (incremental yields of 40%, 15%, 38%, and 38%, respectively). Overall, capsule endoscopy had a yield of 46–72% for detecting small bowel Crohn disease. It had a yield of 33–70% in patients with suspected Crohn disease, and a yield of 68–86% in patients with established Crohn disease. It is possible, however, that the yield of capsule endoscopy for suspected Crohn disease is significantly lower in practice because patients being evaluated for possible Crohn disease do not always fulfill the selection criteria used in studies. In a study of patients being evaluated for abdominal pain who had undergone previous endoscopic or radiographic evaluations, a cause was detected by capsule endoscopy in only 6% of those with abdominal pain alone, and in 13% of those with abdominal pain and diarrhea. Capsule endoscopy can also help in the evaluation of indeterminate colitis. In patients with indeterminate colitis, two retrospective studies of capsule endoscopy have demonstrated small bowel ulcerations in 33–49% of patients, suggesting (but not proving) a diagnosis of Crohn disease.

In addition to detecting small bowel Crohn disease, capsule endoscopy can help in defining the extent of disease, diagnosing a Crohn flare, or detecting a postoperative recurrence in patients with established disease. In one study, patients who were suspected of having a Crohn flare underwent capsule endoscopy. In 20% there was no active disease, suggesting that the symptoms may have been due to other causes, such as a superimposed functional disorder.

Small Bowel Tumors and Polyposis Syndromes

Capsule endoscopy is capable of detecting tumors and polyps of all sizes throughout the small bowel; however, because many of the tumors are submucosal, it can be difficult at times to differentiate a tumor from the transient bulges that are frequently seen during capsule endoscopy. In addition, a tumor may only be seen tangentially on one frame of the study, making characterization of the mass difficult, especially when it comes to size. Because capsule endoscopy lacks biopsy capability, arriving at a definitive diagnosis is rarely possible using capsule endoscopy alone. However, a presumptive diagnosis can be made in some cases, such as in a patient with a small bowel mass and known metastatic melanoma (see Plate 80).

In a study of patients with polyposis syndromes, capsule endoscopy detected polyps in 10 of 11 patients (91%) with Peutz-Jeghers syndrome. That study also examined patients with FAP. It found that 24% of FAP patients (5 of 12 patients) with duodenal adenomas had distal jejunal or ileal polyps detected on capsule endoscopy. In patients without duodenal adenomas, however, more distal polyps occurred in only 12%. Of note, capsule endoscopy often failed to achieve adequate visualization of the ampulla of Vater, an area of frequent polyp and adenocarcinoma development in patients with FAP. Therefore, capsule endoscopy is not a substitute for standard surveillance using a side-viewing duodenoscope with ampullary biopsies.

Celiac Disease

In a small, blinded study of 20 patients (10 with celiac disease and 10 with controls), capsule endoscopy had a sensitivity of 70%, a specificity of 100%, a positive predictive value of 100%, and a negative predictive value of 77% for the diagnosis of celiac disease. Interobserver agreement for experienced capsule endoscopists was perfect (κ = 1.0). Patients with extensive small bowel involvement were more likely to have classic celiac disease symptoms, including diarrhea and weight loss, whereas those with only proximal involvement had mild, nonspecific symptoms.

Studies have also looked at using capsule endoscopy in patients with refractory celiac disease. In a study of 47 celiac disease patients with persistent abdominal pain, occult blood loss, or refractory iron deficiency anemia, capsule endoscopy identified lesions in 87% (41 of 47 patients). The capsule endoscopy studies detected findings consistent with celiac disease in a majority of patients (32 with villous atrophy, 29 with scalloping and fissuring, and 9 with a mosaic pattern). Additional findings included ulcerations (21 patients), nodularity (6 patients), an adenocarcinoma (1 patient), a polyp (1 patient), a stricture (1 patient), an intussusception (1 patient), and a submucosal mass (1 patient). This study suggests that in patients with ongoing symptoms, despite the report of adherence to a gluten-free diet, capsule endoscopy has a high yield for identifying abnormalities.

Esophageal Capsule Endoscopy

The PillCam ESO2 was designed to visualize the esophagus and can be used to detect findings such as esophageal varices, erosive esophagitis, and Barrett esophagus. In a study of 32 patients, there was 97% concordance between capsule endoscopy and upper endoscopy for the detection of varices, and 91% concordance for the diagnosis of portal hypertensive gastropathy.

Limitations

An advantage of capsule endoscopy is that it has the potential to image the entire length of the small bowel. However, in many instances there is poor visualization of areas of mucosa due to quick passage, inability to insufflate, tangential views, and debris. Additionally, in approximately 15% of cases, the capsule does not reach the colon prior to the battery running out (currently about 8 hours). A significant limitation is that capsule endoscopy lacks biopsy capability. This can be a problem because findings such as erythema, aphthous ulcerations, or frank ulcerations are seen in multiple disorders (see Table 33–3). In addition, nodules cannot be biopsied to determine if there is an underlying malignancy. Some of these limitations can now be addressed using deep small bowel enteroscopy if an abnormality is found on capsule endoscopy (see "Deep Small Bowel Enteroscopy," later in this chapter).

Complications

The most important complication related to capsule endoscopy is capsule retention. Overall, the risk of retention is 1–2%. Many of the disorders for which capsule endoscopy is being employed, such as Crohn disease or radiation enteritis, can increase the risk of retention. Patients with Crohn disease are at increased risk because of possible small bowel strictures (which may be missed on conventional imaging). In the setting of established Crohn disease, the risk or retention increases to 4–13%. In patients with established Crohn disease, small bowel radiographic imaging (small bowel follow-through or CT enteroclysis) should be performed prior to capsule endoscopy to decrease the risk of retention. The use of a patency capsule (Agile Patency System, Given Imaging Ltd, Yoqneam, Israel) can also decrease the risk of retention (see "Contraindications" discussed earlier).

Capsule retention is often associated with the identification of significant pathologic findings that often require further surgical or endoscopic intervention. In a study of 733 cases, capsule retention occurred in 1.9% (14 patients). Of these, all occurred at a site of pathology (Crohn disease [5], small bowel stenosis [5], small bowel neoplasm [3], and mesenteric ischemia [1]). Eleven patients underwent surgery for capsule removal, two had the capsule removed endoscopically, and one (with mesenteric ischemia) did not have the capsule removed. Deep small bowel enteroscopy is one option for retrieving capsules that are retained in the mid or distal small bowel and are thus out of reach of a standard enteroscopy (see the later section on "Deep Small Bowel Enteroscopy").

Uncommon complications of capsule endoscopy include aspiration of the capsule, impaction at the cricopharyngeus, or retention in a Zenker or Meckel diverticulum. A case of aspiration was noted in a patient who was part of the study of 733 capsule endoscopy examinations. In that case, the patient was able to expel the capsule by coughing. These complications reinforce the need to evaluate patients carefully for swallowing disorders prior to performing a capsule endoscopy study.

General Considerations

The small bowel is on average 430 cm (14 ft) long. Push enteroscopy with an over-tube, however, can at most be advanced to 160 cm beyond the ligament of Treitz. With the advent of wireless capsule endoscopy, it is now possible to visualize the entire length of the small bowel, but capsule endoscopy lacks biopsy and therapeutic capability. Lesions detected by capsule endoscopy that are beyond the reach of a push enteroscope can now be evaluated using deep small bowel enteroscopy. Total enteroscopy is possible with balloon- or spiral-assisted enteroscopy and has the advantage of allowing for biopsies and for therapeutic interventions. Yamamoto and colleagues first described double balloon enteroscopy in 2001, and it was FDA approved for use in the United States in the fall of 2004 (Double Balloon Technology, Fujinon, Wayne, NJ). Since then, two additional systems using single balloon enteroscopy (Single Balloon Enteroscope System, Olympus, Tokyo, Japan) and spiral enteroscopy (Spirus Endo-Ease Discover SB System, Spirus Medical, Stoughton, MA) have become available.

Indications

Deep small bowel enteroscopy is used most often for the evaluation of obscure gastrointestinal bleeding, small bowel radiographic abnormalities, abnormalities identified on capsule endoscopy, chronic diarrhea and malabsorption, and in polyposis syndromes to detect and remove polyps (Table 33–5). It has also been used to biopsy and dilate small bowel strictures, to screen for disease recurrence in patients with a history of small bowel malignancies, and to evaluate patients with refractory celiac disease. In patients who have undergone roux-en-Y gastric bypass, it can visualize the defunctionalized stomach and gain access to the bile and pancreatic ducts.

Procedure

Patient Preparation

Patients who are undergoing an antegrade (per os) study, fast for 8–12 hours prior to the study. For a retrograde (per anus) study, a standard colonoscopy preparation is employed.

Techniques

Double Balloon Enteroscopy

The double balloon enteroscopy system is composed of an enteroscope, an over-tube, and a balloon pump controller. The system uses two latex balloons, one on the end of the enteroscope and one on the end of the over-tube (Figure 33–1). Through a combination of antegrade and retrograde approaches, the entire small bowel can be examined in 42–86% of patients.

The examination can be carried out using conscious sedation or propofol, although some centers prefer general anesthesia due to the length of the study and the potential for patient discomfort. During an antegrade study, the scope and over-tube are advanced until both are within the duodenum. The balloon on the end of the over-tube is then inflated to anchor the small bowel. The scope is then advanced. When the scope can no longer be advanced, the balloon at the tip of the scope is inflated, again anchoring the small bowel. The balloon on the over-tube is then deflated and the over-tube is advanced until it reaches the end of the scope. At this point, the balloon on the end of the over-tube is again inflated. With both balloons inflated, the scope and the over-tube are gently withdrawn until resistance is met. In so doing, the small bowel is pleated onto the over-tube and loops are reduced. The balloon on the scope is then deflated and the scope is again advanced. This sequence is repeated until the lesion of interest is reached or until the scope can no longer be advanced. Fluoroscopic guidance may be used to aid with scope advancement and reductions. When a retrograde approach is employed, the scope and over-tube are advanced until they are both within the terminal ileum, and the same sequence is carried out.

Using the antegrade approach, an average of 240–270 cm of small bowel can be examined. With the retrograde approach, an average of 140–150 cm of small bowel can be visualized. The reported rates of complete small bowel visualization (often through a combination of antegrade and retrograde examinations) vary widely (4–86%), with higher rates being reported in Japan and lower rates in Europe and the United States.

The current therapeutic double balloon enteroscope (Fujinon EN-450T5; Fujinon Inc, Saitama, Japan) has 140-degree field of view and a working length of 200 cm. It has a forceps channel diameter of 2.8 mm, which will accommodate biopsy forceps, argon plasma coagulation probes, bipolar hemostasis probes, cytology brushes, Roth nets, snares, and injection needles. The over-tube is 135 cm with an outer diameter of 13.2 mm.

Single Balloon Enteroscopy

The single balloon enteroscopy system is similar to the double balloon system except that instead of having a balloon on the end of the enteroscope, the tip of the enteroscope can be angulated sharply to anchor the scope. Average depths of small bowel insertion are 133 to 270 cm for antegrade studies and 73 to 199 cm for retrograde studies.

Spiral Enteroscopy

Spiral enteroscopy uses an enteroscope designed for double or single balloon enteroscopy and an overtube with a soft, raised helical spiral. The enteroscope is advanced into the small bowel along with the overtube. As the overtube is rotated, the small bowel is pulled onto the overtube. The system was developed as an alternative to balloon-assisted enteroscopy in the hope that it would be a simpler and faster method. Currently it is only performed from an antegrade (per os) approach. Reported insertion depths are similar to those seen with antegrade studies performed using balloon-assisted enteroscopy.

Outcomes

The majority of data available on outcomes comes from research on double balloon enteroscopy, although early results reported for single balloon and spiral enteroscopy appear similar. The diagnostic yield for double balloon enteroscopy ranges from 43% to 80%, with a therapeutic yield of 18–55%. In a 2005 study of 137 patients by May and colleagues, double balloon enteroscopy had a diagnostic yield of 80%. The majority of the patients were being evaluated for obscure gastrointestinal bleeding (67%). Other indications included polyposis syndromes (10%), abdominal pain (8%), subileus or severe abdominal pain in the setting of Crohn disease (4%), chronic diarrhea or malabsorption (2%), non-Hodgkin lymphoma of the small bowel (2%), and intestinal obstruction by a foreign body (2%). A new diagnosis was made in 34% of patients and a diagnosis was confirmed in 30%. In 12% of the patients, the study determined the extent of a previously known diagnosis, and in 10% of the patients a previously made diagnosis was excluded. Findings included ulcerations, polyps, tumors, angiodysplasias, small bowel diverticula, foreign bodies (video capsules and dentures), hypertensive enteropathy, and bleeding sources outside of the small bowel. Endoscopic therapy was employed in 41.5% of the patients (argon plasma coagulation, polypectomy, foreign body extraction, dilation, and injection of dilute epinephrine). Medical treatments were started or changed in 17% (mainly in patients with Crohn disease of the small bowel). Surgical therapy was recommended as a result of the study in 17.5% of patients. Double balloon enteroscopy had no treatment implications in only 14.5% of the patients studied.

In a second study of 353 patients from 2007, May and associates found a similar diagnostic yield of 75% for small bowel lesions. Sixty percent of the patients were being evaluated for suspected small bowel bleeding, 10% had chronic abdominal pain, 9% had a polyposis syndrome, 8% had Crohn disease, and 13% underwent the study for other indications, including foreign body extraction. The findings influenced subsequent therapy in 67%. Endoscopic therapy was performed in 59%, and medical therapy was initiated or changed in 19%. Twenty-two percent of the patients required surgery. Not surprisingly, the majority of patients who received endoscopic therapy suffered from small bowel bleeding (74%).

Limitations

The primary limitation of deep small bowel enteroscopy is incomplete mucosal visualization. As noted above, the small bowel can be examined in 4–86% of patients using a combined antegrade and retrograde approach with double balloon enteroscopy, leaving a significant percentage of patients with incomplete small bowel visualization. Most commonly this occurs due to inability to advance the scope through the entire length of small bowel. Deep small bowel enteroscopy is also limited by the fact that it typically requires two operators (at least one of whom is a physician), and it is very time consuming, with an average procedure time of 73–115 minutes for balloon-assisted enteroscopy, though the time required for spiral-assisted enteroscopy may be less. In addition, because of potential patient discomfort, some centers use general anesthesia, which can make procedures logistically more difficult to arrange and expensive.

Complications

The most common adverse event is abdominal pain, the day of or the day after the procedure. This occurs in up to 20% of patients. Patients also may report a sore throat. Perforations have been reported, including multiple perforations following chemotherapy for lymphoma and following small bowel polypectomy. Post-procedure paralytic ileus and pancreatitis have also been reported. Bleeding has been seen following polypectomies.

The overall major complication rate in May's study of 353 patients, cited earlier, was 3.4%. Bleeding occurred in 1.1%, perforation in 1.7%, and enteritis in 0.6%. This, however, underestimates the risk associated with individual interventions. Of the 46 patients who underwent polypectomies, 5 (10.8%) suffered complications. Bleeding occurred in 2 (4.3%), and perforation occurred in 3 (6.5%). All of the complications in patients undergoing polypectomy occurred after the removal of polyps that were larger than 3 cm. Argon plasma coagulation had a lower complication rate of 1 in 108 (0.9%).

Thus, while the overall rate of complications is low, patients undergoing polypectomy should be advised that there is a significant complication rate associated with the removal of large polyps, as are often seen in Peutz-Jeghers syndrome. However, given that the alternative in these patients is intraoperative enteroscopy, which has a morbidity rate up to 30% and a mortality rate of 2%, deep small bowel enteroscopy is still an attractive option.