Version July 2025
INTRODUCTION —
Wireless video endoscopy or video capsule endoscopy (VCE) is a noninvasive technology designed primarily to provide diagnostic imaging of the small intestine, an anatomic site that has proven peculiarly difficult to visualize. Limited views of the esophagus, stomach, and cecum may also be acquired by small bowel capsules. Images acquired are of excellent resolution and have an approximately 1:8 magnification, which is higher than that of conventional endoscopes. This magnification allows visualization of individual villi. VCE approaches the concept of physiological endoscopy since the capsule moves passively, does not inflate the bowel, and images most of the small intestinal mucosa. Double-ended wireless video capsules have also been developed for the examination of the colon.
A long-awaited artificial intelligence assistant that speeds up the reading of video capsules has also been FDA approved. In addition, a magnetically controllable capsule has become available, along with a tether for examination of the esophagus and stomach.
This topic will review the use of wireless video capsule endoscopy for evaluation of the esophagus, small bowel, and colon. The general approaches to patients with occult and/or suspected small bowel bleeding are discussed in detail elsewhere. (See "Evaluation of occult gastrointestinal bleeding" and "Evaluation of suspected small bowel bleeding (formerly obscure gastrointestinal bleeding)".)
AVAILABLE CAPSULES —
Several small intestinal video capsules are available and FDA approved in the United States (table 1):
●PillCam SB3 and the Crohn’s capsule (Medtronic, Minneapolis, MN)
●MiRoCam capsule (Intromedics, South Korea)
●CapsoCam (CapsoVision, Saratoga, CA)
●NaviCam SB system (AnX Robotica, Plano, TX)
The Olympus EndoCapsule EC-10 has been discontinued. A colonic capsule is also available in Europe, the United States, and Japan (PillCam Colon 2 [Medtronic]). (See 'Colon capsule endoscopy' below.)
All of the small intestinal capsule devices available have a battery life of 12 hours or greater. This extended battery life compared with earlier versions of the capsules reduces the chance of incomplete transit in the non-obstructed small intestine to nearly zero.
The US Food and Drug Administration (FDA) approved the original small bowel capsule, which was called the M2A capsule (Given Imaging, Yoqneam, Israel), in 2001 [1,2]. It has evolved into the PillCam SB3. The Crohn’s capsule, which has a camera at each end, can be used to examine the entire gastrointestinal tract. A two-camera colon capsule was developed and FDA-approved more than 10 years ago but has had limited use because it is typically not covered by insurance.
The MiRoCam capsule became available in many countries between 2007 and 2009 and was approved by the FDA in 2013. This device uses a novel mode of transmission called electric field propagation, which uses the human body as a conductive medium to transmit images [3]. This technology uses less energy than radio frequency, which increases the operation time of the capsule and allows for the acquisition of more image data. A trial that compared the MiRoCam capsule with the EndoCapsule (no longer available) in 50 patients found they were similar with regard to complete small bowel examination rates and diagnostic yield, though the findings of the two studies were concordant in only 68 percent (kappa = 0.50) [4]. This discordance is another demonstration that, while the capsule passes the length of the small bowel in most patients, imaging of the mucosa with current devices is still far from complete [5]. A hand-held magnetic controller for the MiRoCam capsule while it is in the stomach is available, but not yet FDA approved.
The CapsoCam has four cameras giving a 360° field of view and a variable frame rate. The battery life for the CapsoCam is up to 15 hours. It was FDA approved in 2016. A drawback to the device is that it generates so much data that it is not possible to use radiofrequency to transmit the data; the capsule has to be retrieved from the fecal stream using a catchment device in the toilet and a magnetic wand. The data are then downloaded and processed in a special device before reading. The initial trials suggest comparable diagnostic accuracy to the PillCam [6,7]. It does not require a workstation for capsule initiation and it can be read locally or at a central facility.
The NaviCam SB device is a small bowel capsule that includes an artificial intelligence assistant to speed reading time. AnX Robotics also makes a magnetically controlled capsule for gastric examination that is coupled with a tether for esophageal examination (NaviCam Stomach) [8]. (See 'Esophageal and gastric capsule endoscopy' below.)
SMALL BOWEL CAPSULE ENDOSCOPY
Indications — The primary indications for video capsule endoscopy (VCE) are for evaluation of suspected small bowel bleeding in adults (including adults with iron deficiency), for the diagnosis of suspected Crohn disease, and for the diagnosis of small bowel tumors. Its role in the assessment of mucosal healing in patients with small bowel Crohn disease is evolving. In addition, VCE is being used to detect small bowel injury associated with the use of nonsteroidal anti-inflammatory drugs (NSAIDs), to screen for polyps in patients with genetically determined syndromes such as Peutz-Jeghers syndrome and familial adenomatous polyposis, to screen for small bowel malignancies in patients with Lynch syndrome, and possibly in the assessment of celiac disease (picture 1) [9-15]. It can also be used to evaluate abdominal pain and chronic diarrhea of unclear etiology. There is growing experience in children over the age of 10 years for the above indications [16], and these indications are now approved for children as young as two years of age. Finally, VCE is being studied for the evaluation of both hematemesis and non-hematemesis acute bleeding. (See 'Efficacy' below.)
Contraindications — The procedure is be contraindicated in patients with the following conditions, but these contraindications may not be absolute:
●Dysphagia, an esophageal stricture, or swallowing disorder that could prevent the passage of the capsule (eg, Zenker's diverticulum). The capsule can be endoscopically placed in the duodenum to avoid this problem.
●Gastroparesis (the capsule can be endoscopically placed in the duodenum).
●Partial or intermittent small bowel obstruction unless the patient understands the risks of retention and the potential need for deep small bowel enteroscopy or surgery to remove the capsule.
●Patient is not a surgical candidate or refuses surgery.
●Presence of a defibrillator or pacemaker. While this is a recommendation in the package insert, it does not appear to be a significant clinical problem [17-20]. In a 2017 guideline from the Canadian Association of Gastroenterology, the presence of a pacemaker was no longer considered a contraindication for capsule endoscopy [21]. Left ventricular assist devices may interfere with image acquisition by the recorder. In this situation, the sensor should be placed as far away as possible from the device.
●Pregnancy.
●Dementia if the patient cannot cooperate with swallowing the capsule or may inadvertently damage the equipment.
Careful consideration must be given before performing VCE on any patient in whom there is the potential for capsule retention. Examples include patients with known Crohn disease (up to a 13 percent retention rate) [22], patients with intermittent small bowel obstruction secondary to adhesions, patients with radiation enteritis, patients with severe motility disorders, and patients with a Zenker's diverticulum. Patients with suspected Crohn disease have a much lower risk of retention (approximately 1 to 4 percent retention rate) [22,23].
In patients at high-risk for retention, decisions to proceed with VCE should be made in conjunction with a surgeon, who may be called upon to retrieve the capsule and treat the underlying problem. A normal small bowel barium study or CT scan does not exclude the possibility of retention, particularly in NSAID-associated diaphragm disease, which is nearly impossible to pick up radiologically because the diaphragms are the same width as normal small bowel folds. (See 'Capsule retention' below.)
Magnetic resonance enterography (MRE) may help identify patients with Crohn disease who should undergo additional evaluation (eg, a patency study) prior to capsule endoscopy. In a series of 57 patients with Crohn disease who underwent MRE prior to a patency study, MRE had a sensitivity of 92 to 100 percent and a specificity of 52 to 59 percent for predicting patency capsule retention [24]. This suggests that patients with a normal MRE are at low risk for capsule retention, and that patients with a positive MRE should undergo a patency study prior to capsule endoscopy. A limitation of this study is that there were only two radiologists reading the studies and their interobserver agreement was only fair (kappa 0.58). (See 'Patency capsule' below.)
Patency capsule — A capsule system to determine small bowel patency has been developed (Agile Patency Capsule, Medtronic, MN). The patency capsule is used in patients who are at high risk for having small bowel strictures, such as those with symptoms suggesting small bowel obstruction, imaging findings suggesting stricturing, a history of small bowel strictures, previous small bowel surgery, abdominal radiation for intestinal or gynecological malignancy, or heavy NSAID consumption.
The system is based upon a capsule (the same size as the PillCam SB3 video capsule) composed of lactose and barium enclosed in a thin plastic envelope that contains a radiofrequency identification tag that can be detected by a scanning device placed on the abdominal wall or by a plain abdominal film. The biodegradable plugs at each end start to dissolve after 30 hours, and fully dissolve 40 to 80 hours after ingestion, allowing the capsule to pass even in the presence of a stricture. The patency capsule does not have any image acquisition capability.
The patency capsule is used as follows:
●The patient is instructed to ingest a liquid diet starting at noon and to not eat or drink anything after 10 PM.
●The patient ingests the capsule the following morning (we suggest 7 or 8 AM). Following capsule ingestion, liquids may be taken after two hours, and food and medications may be taken after four hours.
●The patient is assessed at as close to 30 hours as possible, without exceeding 30 hours (if the capsule is ingested at 7 AM, the assessment should be done prior to 1 PM the next day). This can be done using the scanning device or a plain abdominal film. Our practice is to use a plain film. If the scanning device is used and indicates the capsule is in the abdomen, then follow-up imaging (also performed less than 30 hours after capsule ingestion) is used to localize the capsule to the small bowel or colon. The radiologist needs to understand the difference in the radiological appearance of the patency capsule compared with the video capsule, which is more radiologically opaque. The patency capsule is sometimes difficult to see if it overlays the spine.
●Patency is suggested if at or before 30 hours:
•The capsule is not present in the abdomen (determined either with the scanning device or radiographic evaluation)
•The capsule is in the abdomen, but imaging indicates that it is in the colon
•The patient is a reliable historian and reports that the capsule passed fully intact
The capsule is used to assure small bowel patency before VCE and can also be used as a diagnostic test for suspected small bowel strictures that might retain the video capsule and that cannot be identified by standard radiographic means [25-27]. Passage of the patency capsule into the colon by 30 hours suggests that there are no obstructions likely to impede passage of the video capsule. However, VCE retention following a patency study that suggested no significant strictures has been reported when capsule localization was determined using a plain abdominal film (presumably due to inaccurate interpretation of the patency capsule location) [28].
In a report of 106 patients with known strictures, the patency capsule suggested the gastrointestinal tract was sufficiently patent in 59 patients. All subsequently underwent VCE with no cases of capsule retention [29]. In a meta-analysis of patients with IBD, the retention rate was 2.7 percent in patients who had a patency capsule or CT enterography prior to the capsule study (see 'Capsule retention' below) [23].
Adverse events related to patency capsule ingestion were common with an earlier version of the patency capsule but not with the current version. The earlier version had a single dissolvable plug, leading to incomplete dissolution of the device, allowing it to become lodged within a stricture as it dissolved to the dimensions of the stricture. In a study of 2578 patients who received the PillCam patency capsule (which has dissolvable plugs at each end), there was suspected impaction of the patency capsule in 23 patients (0.89 percent) [30].
Procedure — The procedure can be performed in ambulatory or hospitalized patients.
Preparation — Typically, patients fast at least overnight (12 hours). Whether additional patient preparation or prokinetics should be given remains controversial because studies have reached different conclusions regarding the efficacy of using preparations. We use a 12-hour fast without preparation. However, a guideline from the Canadian Association of Gastroenterology does recommend the use of a bowel preparation, though no specific preparation was recommended [21]. In addition, there is still no objective measurement of what constitutes a well-prepared small bowel.
One option if a preparation is to be used is a full colonoscopy preparation (with a gallon of polyethylene glycol) [31,32]; however, our approach is to not use a purgative bowel preparation before capsule endoscopy because it has not been shown to provide clear benefit. The rationale behind using a purgative bowel preparation is to minimize the amount of dark bile, which can reduce visibility in the distal small bowel. Individual studies have reached variable conclusions regarding the efficacy of a purgative bowel preparation [31-42], as have meta-analyses [37,43,44]. In a 2018 meta-analysis of 12 randomized trials that included 1221 patients, the use of a purgative preparation did not improve diagnostic yield, small bowel mucosal visualization quality, or capsule endoscopy completion rate, though there was significant heterogeneity among the studies [43].
Other interventions that have been tried to improve small bowel visualization and/or capsule endoscopy completion rates include mannitol with or without simethicone [45], simethicone alone [39,46,47], and prokinetic agents such as metoclopramide [48,49]. As with purgative bowel preparations, these interventions have not consistently been shown to be beneficial.
Some studies have suggested that simethicone increases visibility. Two randomized trials suggested improved visibility with bowel preparation involving bowel cleansing and simethicone [46,47]. However, another trial found that bowel preparation with magnesium citrate was superior to simethicone for improving small bowel fluid transparency and hence diagnostic yield [39].
The use of prokinetic agents such as metoclopramide is also controversial and as battery lives for capsules increased from 8 to 12 or more hours, the need for agents to speed small bowel transit has decreased. One study of 150 patients suggested metoclopramide increased the likelihood of complete small bowel examination compared with patients who did not receive metoclopramide (97 versus 76 percent) [48]. However, a randomized trial of 95 patients assigned to either metoclopramide 15 minutes prior to the examination or no metoclopramide did not show a benefit. In that study, there were no differences between those who received metoclopramide and those who did not with regard to the rate of complete small bowel examinations (81 versus 77 percent), median gastric transit time (26 versus 28 minutes), mean small bowel transit time (221 versus 256 minutes), or mean number of findings (4.5 versus 4.7) [49].
Sensor array — For the PillCam SB3, the original eight-lead sensor array has been replaced with a three-part sensor array contained in a belt worn by the patient, thereby eliminating adhesives. In addition, the belt containing the sensors is washable, reducing the risk of transfer of nosocomial infection. Eight-lead sensor arrays may still be used for patients who are unable to wear the sensor belt or who have a large pannus, the movement of which can put the sensor array out of range of the capsule transmission. Real-time viewing is possible with the current recorder. This gives a frame-by-frame view of what the capsule is recording.
The MiroCam uses a sensor array attached to the abdomen. For these systems, the array is connected to a solid-state recorder, battery pack, and real-time viewer in an all-in-one device, worn on a belt.
CapsoCam does not require an external recording system as the data are stored in the capsule, necessitating its recovery from the fecal stream before processing.
The NaviCam SB capsule system uses a belt and shoulder strap to house the recorder and sensor array. It employs radiofrequency transmission. The recording, once downloaded, can be reviewed locally or remotely.
Capsule ingestion — All video capsules are swallowed with water. A device (AdvanCE Endoscope Delivery Device, Steris, OH) is available for the endoscopic introduction of a capsule into the small bowel in patients who cannot swallow the capsule, who have gastroparesis, or who have some other impediment that may prevent passage of the capsule into the small bowel in a reasonable time [50].
Following capsule ingestion, clear liquids may be taken after two hours, and food and medications may be taken after four hours.
The sensor arrays used with the PillCam SB, Mirocam, and NaviCam SB capsule may be removed after 12 hours to optimize small bowel completion rates, and the recorded images are downloaded and processed on workstations. The CapsoCam requires recovery from the fecal stream using a magnetic wand and strainer placed in the toilet. All of the capsules are single use and are excreted with bowel movements.
Image acquisition and review — All of the available capsules have variable frame rates. The rate depends on how fast the capsule is moving. For example, in the duodenal sweep where contractility is vigorous, the frame rate increases, whereas if the capsule becomes static, the frame rate decreases. The recorders acquire up to 90,000 images over approximately 12 hours. The field of view is approximately 150 to 170 degrees for single-camera devices. This limitation can lead to incomplete imaging of the small bowel mucosa.
The three systems that use a sensor array all incorporate a real-time viewer. The utility of this device has yet to be fully defined, but it can be used to see if the capsule is in the stomach or small intestine and directly detect the presence of blood.
Review of the video, selection of representative images, and generation of a report can take 30 to 90 minutes. Algorithms in the software allow for quick preliminary screening of the videos but do not replace careful review of the study. Clinically important abnormalities may be represented on only one or two frames out of 90,000; thus, significant concentration is required during the review of images. The video may be reviewed as slowly as one frame at a time up to 30 frames per second when viewed singly. Images may be viewed singly, in groups of two or four, or in a mosaic. The latter can be helpful in precisely defining anatomic landmarks or specific features or pathology. Anecdotally, the author uses a rate of 23 frames per second in the four-view mode. The introduction of artificial intelligence (AI) may decrease reading times. In one study of the NaviCam SB capsule, the AI assistant decreased reading time from a mean of 34 minutes to a mean of 4 minutes [51].
Efficacy — VCE has several possible advantages compared with other means of visualizing the small bowel. It is non-invasive and permits examination of the majority of the small bowel mucosa, which is not possible with push enteroscopy [52-61] and rarely by device-assisted enteroscopy. The main disadvantage of VCE is that it does not permit tissue sampling or therapeutic intervention. With newer capsules with longer recording times, the capsule does not reach the cecum within recording time in approximately five percent of cases [62]. The frequency of incomplete transit has been substantially reduced by the advent of 12-hour or longer battery life in most capsules. One study found that an incomplete exam was more likely in patients who had undergone small bowel surgery, were hospitalized, had moderate or poor bowel cleaning, or had a gastric transit time of longer than 45 minutes [63].
The overall detection rate (ie, the percentage of studies which yield a diagnosis) is approximately 50 to 60 percent [64,65]. In a retrospective study of 1214 patients who underwent VCE (most often for anemia or overt gastrointestinal bleeding), abnormalities were found in 52 percent (45 percent had angiectasias, 18 percent had erosions, and 9 percent had ulcers) [65]. VCE has been shown to have an impact on the management of patients with a variety of small bowel disorders. An illustrative study included 40 clinicians who were interviewed before and after VCE examinations in 98 patients [66]. Clinicians changed overall management plans in 67 percent of patients. Of these clinicians, 74 percent reported that they changed their plan directly as a result of the VCE findings.
In addition to decreasing reading time, artificial intelligence (AI) may also increase accuracy. In a study of 133 patients with suspected small bowel bleeding, AI-assisted reading had a diagnostic yield for potential bleeding sources of 74 percent compared with 62 percent for standard reading (p = 0.02) [51].
Clinical applications
Suspected small bowel bleeding/iron deficiency anemia — The most common indication for VCE is the evaluation of suspected small bowel bleeding (including iron deficiency anemia), previously referred to as obscure gastrointestinal bleeding [64]. However, even with VCE, it can be difficult to identify bleeding sources within the small bowel. (See "Evaluation of suspected small bowel bleeding (formerly obscure gastrointestinal bleeding)".)
VCE was able to identify causes of small bowel bleeding more often than push enteroscopy in most reports (picture 2) [52-54,56,67-70]. The overall yield of VCE for suspected small bowel bleeding has been reported to be in the range of 30 to 70 percent [52-54,57,61,64,67,71-78]. A large meta-analysis included 227 studies with 22,840 procedures, 66 percent of which were done for suspected small bowel bleeding. In that analysis, the detection rate for VCE in patients with suspected small bowel bleeding was 61 percent [64].
In a study of 911 patients with suspected small bowel bleeding published subsequent to the meta-analysis, 509 patients (56 percent) had a lesion identified on capsule endoscopy that was thought to be responsible for the bleeding [75]. The findings included:
●Small bowel angiodysplasia – 22 percent
●Small bowel ulcerations – 10 percent
●Small bowel tumors – 7 percent (benign and malignant)
●Small bowel varices – 3 percent
●Blood in the small bowel with no lesion identified – 8 percent
●Esophagogastric lesions (eg, esophagitis, gastritis) – 11 percent
●Colonic angiodysplasia – 2 percent
A meta-analysis of 14 observational studies compared capsule endoscopy with other tests for suspected small bowel bleeding. They estimated that the overall yield (ie, the yield of VCE for any small bowel findings) of VCE (63 percent) was significantly higher than for push enteroscopy (26 percent) and barium studies (8 percent) [76]. Comparison of the published reports is somewhat limited since the definition of a bleeding site was variable. In some cases, active bleeding was identified, while in others, bleeding was presumed to originate from an ulceration or mass.
The yield appears to be higher for patients who take antithrombotic agents who are undergoing small bowel VCE for suspected bleeding compared with patients who are not taking antithrombotics. In one retrospective study, a lesion thought to be responsible for bleeding was found in 1103 of 2535 of patients receiving an antithrombotic (44 percent) compared with 1113 of 5866 of patients who were not receiving an antithrombotic (18 percent, p <00001) [79].
The diagnostic yield of VCE (ie, the percentage of studies that provided a clear-cut explanation for the bleeding) is highest when it is performed as close as possible to the bleeding episode and in patients with overt (visible), rather than occult, bleeding [75,77,80,81]. One study included 100 consecutive patients with suspected small bowel bleeding [77]. Patients were categorized into three groups; the diagnostic yield of VCE was highest in the group with ongoing overt bleeding (92 percent), compared with those with previous overt bleeding (13 percent) or guaiac-positive stools and iron deficiency anemia (44 percent). The most common findings were angiodysplasia (29 percent) and Crohn disease (6 percent). Sixty-two patients underwent further examination that led to independent verification of the diagnosis in 56. Sensitivity, specificity, positive, and negative predictive values were 89, 95, 97, and 83 percent, respectively. The authors concluded that VCE is effective for evaluation of suspected small bowel bleeding and that the best candidates appear to be patients with ongoing overt bleeding or occult bleeding.
Other factors associated with an increased yield of VCE include older age, male sex, current hospitalization, increasing transfusion requirements, and the presence of connective tissue disease [75,82].
VCE appears to be more accurate for identifying small bowel pathology than barium small bowel radiography [55,72,83]. Two of the studies that have looked at this showed the following:
●One series included 22 patients suspected of having small bowel pathology who underwent both VCE and a standard upper gastrointestinal series with small bowel follow-through [55]. VCE was considered diagnostic in 45 percent compared with 20 percent for barium studies. VCE was significantly more sensitive for causes of suspected small bowel bleeding (31 versus 5 percent).
●In a randomized trial, 136 patients with suspected small bowel bleeding (occult or overt) were assigned to either VCE (n = 66) or dedicated small bowel barium radiography (n = 70) [72]. The diagnostic yield was higher for VCE compared with barium radiography (30 versus 7 percent). However, the improved diagnostic yield did not result in a significant difference in subsequent diagnostic or therapeutic interventions for bleeding (26 percent for VCE versus 21 percent for barium radiography). Additionally, patients who underwent VCE did not differ from patients who underwent barium radiography with regard to subsequent hospitalizations for bleeding (12 versus 6 percent) or the need for subsequent blood transfusions (8 versus 6 percent).
At least one randomized controlled trial (involving a total of 89 patients) suggested that performing VCE before push enteroscopy was a more effective strategy than beginning with push enteroscopy in patients with suspected small bowel bleeding [84]. The VCE first strategy reduced the percentage of patients needing the push enteroscopy study (25 versus 79 percent). Twelve months after evaluation, the strategy based on VCE first (followed by push enteroscopy as necessary) had a similar diagnostic yield, clinical outcome, and therapeutic impact compared with a strategy of push enteroscopy first (followed by VCE as needed).
A 2004 study was the first to note that early deployment of the VCE enhances the diagnostic yield of capsule endoscopy [77]. A retrospective study in 2013 expanded on this observation in an inpatient population and showed that deployment within 72 hours of admission enhanced the detection rate of active bleeding, increased the rate of therapeutic intervention, and reduced the length of stay by 40 percent. Capsule deployed after 72 hours had the same diagnostic yield as that of an outpatient population [80]. Thus, there is increasing support for early use of VCE, particularly in patients with overt bleeding.
Crohn disease — VCE can be useful in diagnosing Crohn disease in patients with symptoms suggestive of Crohn disease or in patients with indeterminate colitis to evaluate for possible Crohn disease [21]. It can also be used in patients with known Crohn disease to detect active disease (picture 3) and to evaluate responses to therapy. The overall detection rate by VCE in patients with known or suspected Crohn disease was 55 percent in a large meta-analysis [64].
VCE should not be used in patients with known or suspected strictures without careful consideration and pre-procedure evaluation [85-89]. A small bowel follow-through that does not reveal strictures does not necessarily exclude strictures. Capsule retention has been described in up to 13 percent of patients who underwent a capsule study for known Crohn disease, even after performing an initial small bowel study [86]. In most reports, retention was more likely in patients in whom the capsule study was being performed for known Crohn disease compared with those with suspected Crohn disease (5 to 13 versus 1 to 2 percent) [22,86]. This has led to the recommendation that patients with known small bowel Crohn disease have small bowel imaging or a patency capsule study prior to VCE. (See 'Patency capsule' above.)
Computed tomographic enterography (CTE) and magnetic resonance enterography (MRE) have good sensitivity for detecting Crohn disease-related strictures, ranging from 75 to 100 percent [90]. (See "Clinical manifestations, diagnosis, and prognosis of Crohn disease in adults", section on 'Small bowel imaging'.)
A patency capsule study should also be obtained in patients without Crohn disease who are at high risk for having strictures (eg, known strictures that have not been treated or symptoms of recurrent small bowel obstructions). Its use should also be considered in patients with a history of abdominal/pelvic radiation (often many years prior), heavy NSAID use, and previous small intestinal surgery. In patients at lower risk (eg, patients with a history of small bowel Crohn disease who are asymptomatic), evaluation with a computed tomographic (CT) or magnetic resonance imaging (MRI) enterography is an acceptable alternative to a patency capsule study.
Studies have been done comparing VCE with other modalities for small bowel Crohn disease. A meta-analysis of 12 studies found that capsule endoscopy had an overall yield of 50 to 70 percent for findings of Crohn disease [91]. The yield was higher than the yield for barium radiography (22 percent), ileo-colonoscopy (48 percent), push enteroscopy (8 percent), or CT enterography/CT enteroclysis (31 percent). In patients with known Crohn disease who were being evaluated for a suspected recurrence, the yield for VCE was 66 to 71 percent. The yield was lower for patients with suspected Crohn disease (33 to 68 percent).
A subsequent study compared VCE with CT enterography and MR enterography in patients without small bowel strictures [92]. Using ileoscopy or surgery as the gold standard, VCE had a sensitivity of 100 percent for detecting terminal ileal Crohn disease, which was significantly higher than that for CT enterography (76 percent) and showed a trend toward being higher than the sensitivity for MR enterography (81 percent). The specificities of the three studies were similar (91, 85, and 86 percent, respectively). Overall, the diagnostic yield of VCE for Crohn disease in any portion of the small bowel did not differ significantly from the other studies (30 versus 33 and 28 percent, respectively), but it did detect more cases of Crohn disease proximal to the ileum (18 versus six and two cases, respectively).
VCE findings may also help identify patients who are likely to have a disease relapse. In a study of 108 patients with Crohn disease, the presence of jejunal lesions on capsule endoscopy was associated with an increased risk of relapse (adjusted hazard ratio 1.99, 95% CI 1.10-3.61) [93]. An evolving use of capsule endoscopy is in the assessment of mucosal healing in Crohn disease [94]. Pan-enteric capsule endoscopy has been shown to identify proximal small bowel Crohn disease and can be a guide to upgrading management [95].
Small bowel tumors, polyps, and other pathology — A variety of small intestinal lesions have been detected with VCE, including small intestinal varices, tumors and polyps, and intestinal graft-versus-host disease [12,58,59,87,88,96-104]. VCE is comparable to EGD for the diagnosis of celiac disease when there are overt villous changes. In a study of 43 patients with suspected celiac disease, VCE had a sensitivity of 88 percent and a specificity of 91 percent for detecting celiac disease [105]. However, VCE was less sensitive in a study of patients with known celiac disease [106]. The study included 42 patients with refractory celiac disease, 84 patients without celiac disease, and 30 patients with uncomplicated celiac disease. Using histology as the gold standard, VCE was 56 percent sensitive and 85 percent specific for detecting villous atrophy. Importantly, VCE detected complications of celiac disease in two of the patients with refractory celiac disease, finding one case of ulcerative jejunitis and one adenocarcinoma.
VCE appears to be less sensitive for detecting small bowel tumors than CT enterography. In a study that included 17 patients with small bowel tumors who underwent both CT enterography and capsule endoscopy, CT enterography was more sensitive than capsule endoscopy for detecting small bowel tumors (94 versus 35 percent) [107]. Lesions in the duodenum and proximal jejunum are easily missed because of the rapid transit of the capsule through these areas. One problem encountered with VCE is that transient bulges into the small bowel lumen may appear to be submucosal masses [108-111]. Factors associated with true submucosal masses include a well-defined boundary with the surrounding mucosa, an acute angle of protrusion, a lesion that is taller than it is wide, visible lumen in the frame with the lesion, ulceration, attenuation of folds, and visualization of the lesion for more than 10 minutes of the capsule's recording time.
One study reviewed the literature for scoring systems for the clinical relevance of observed lesions [112]. The Saurin score is the most widely used, and classifies lesions as P0, P1, and P2:
●P0: no potential risk of bleeding (eg, phlebectasias, erythematous patches, diverticula without evidence of blood)
●P1: low/uncertain risk of bleeding (eg, red spots, small erosions)
●P2: high risk of bleeding (eg, large ulcerations, tumors, varices)
VCE may have a role in surveillance of patients with polyposis syndromes [97-101]. However, at least one report found that VCE underestimated the number of small bowel polyps and did not reliably detect large polyps in persons with familial adenomatous polyposis compared with push enteroscopy and lower endoscopy [113]. Given that VCE frequently fails to identify the ampulla of Vater [114,115], it should not be used for ampullary surveillance in patients with familial adenomatous polyposis.
Occasionally, a small number of small, benign-appearing polyps are detected by capsule endoscopy in a patient without a polyposis syndrome. There are no clear guidelines on the management of such polyps. We typically will repeat the capsule endoscopy in one year to ensure there has been no change in the polyps.
VCE has also been used for surveillance in patients with Lynch syndrome. In a study of 200 patients with Lynch syndrome, VCE detected one adenocarcinoma and one adenoma, both of which were within reach of a duodenoscope [15]. VCE also missed one duodenal cancer.
Chronic abdominal pain — The yield for small bowel capsule endoscopy in patients with chronic abdominal pain is lower than is seen in patients with suspected small bowel bleeding or suspected/known Crohn disease. In a systematic review that included 1520 patients from 21 studies, the per-patient yield for capsule endoscopy was 21 percent [116]. The most common findings were inflammatory lesions (78 percent) and tumors (9 percent).
Acute gastrointestinal bleeding — VCE is being studied for the evaluation of non-hematemesis gastrointestinal bleeding (GIB). In a randomized trial with 87 patients with non-hematemesis GIB, patients randomized to undergoing VCE shortly after admission had a higher rate of bleeding localizations compared with those who received routine care (64 versus 31 percent; OR 2.7) [117]. Much of the difference was due to an increase in the detection of vascular lesions among patients undergoing early VCE (19 versus 4.4 percent). However, differences were not seen between the groups concerning performance of therapeutic intervention, rebleeding, or mortality. A multicenter study of VCE performed during the COVID-19 pandemic included patients with upper and lower GI bleeding. Detection of active bleeding by early use of capsule was 60 percent compared with 30 percent by standard of care (EGD and colonoscopy first) using historical controls and propensity scoring. The study showed a 45 percent reduction of negative procedures [118]. Esophageal VCE has also been studied for the evaluation of acute upper GIB [119] (see 'Esophageal and gastric capsule endoscopy' below).
Risks — VCE is an extremely safe technology. No deaths have been attributed to the device, despite millions of ingestions. One of the main risks associated with VCE, although not inherently dangerous, is retention of the capsule. In addition, in some patients the battery runs out before the capsule passes through the ileocecal valve, making it unclear if the capsule has been retained until it is passed with a bowel movement. However, not all patients will note passage of the capsule in their stool. If there is clinical concern regarding capsule retention, it is generally recommended to obtain a plain abdominal radiograph to confirm passage two weeks after ingestion.
A rare risk associated with VCE is capsule aspiration (approximately 0.1 percent of cases). In a review of 63 cases of VCE aspiration, the overall rate was estimated to be 0.1 percent [120]. The aspiration was self-resolved in 31 patients, whereas 32 patients required intervention for retrieval. One death was reported.
Capsule retention — Capsule retention refers to the failure of the capsule to reach the colon during its recording time. It generally occurs in three forms:
●Long-term retention, which may be first suspected in patients in whom the capsule does not pass the ileocecal valve before the battery runs out, is formally defined as the capsule being retained for more than 14 days. In a meta-analysis that included 22,840 procedures, retention occurred in 1.4 percent of patients [64]. The capsules were removed surgically in 59 percent and endoscopically in 16 percent. However, with the increasing use of deep small bowel enteroscopy (eg, double balloon enteroscopy), the frequency with which capsules are being removed endoscopically is likely increasing.
In a second meta-analysis, retention rates were examined in patients undergoing capsule endoscopy for obscure gastrointestinal bleeding (5876 patients), suspected inflammatory bowel disease (IBD; 968 patients), established IBD (558 patients), or abdominal pain (111 patients) [23]. Capsule retention occurred in 2.1 percent of studies done for obscure bleeding, 3.6 percent of studies done for suspected IBD, 8.2 percent of studies done for established IBD, and 2.2 percent of studies done for abdominal pain. In studies done for established IBD, the retention rate was 2.7 percent if the patient underwent a patency capsule study or CT enterography prior to the capsule study. Three percent of patients with retentions developed obstructive symptoms. The most common reason for retention was small-bowel strictures (77 percent of retentions), though some studies did not report the etiology. Management (when reported) included surgery (60 percent of the patients with retained capsules), endoscopic removal (7 percent), and medical treatment with subsequent passage of the capsule (3 percent). In 5 percent of the patients, the capsule passed spontaneously without management.
In general, surgical intervention to remove the capsule should also address the underlying cause of the retention. Long-term retention is always associated with underlying pathology (eg, stricture, tumors). The most common cause of long-term retention is Crohn disease [64].
●Incomplete transit of the capsule during its recording time, often with transient retention at the ileocecal valve. This occurs in approximately five percent of procedures and is typically of no clinical consequence [62]. In some cases, the capsule may be retained within the stomach during its entire recording time. If the patient does not see the capsule pass in the stool, a plain abdominal radiograph should be obtained in two weeks to confirm passage. The prolonged battery life of most of the capsules should substantially reduce transient retention of the capsules and provide complete information on the majority of patients and the entire small intestinal lumen. (See 'Efficacy' above.)
●Transient retention at a stricture or mass (picture 4). The patient is rarely aware of this, but in this situation it is recommended that an abdominal plain film be taken two weeks after the procedure to confirm excretion of the capsule if it is not seen to pass in the stool.
A retained capsule rarely causes obstruction and more often tumbles around above the narrowed segment, where it may remain for as long as several months to years. It is not yet known how long the capsule can safely be left above a stricture. The development of pain usually heralds passage through a tight stricture.
In a significant number of patients with a retained capsule, surgery would have been indicated for the underlying condition, even in the absence of retention (eg, in a patient whose retention is due to a small bowel tumor). A collaborative approach may enable the team to use the capsule to both visualize and localize an obstructive lesion [22,77]. Palpation by the surgeon of the capsule retained above a stricture may be helpful in localizing the stricture, which may have no serosal signs, thus avoiding the need for intraoperative enteroscopy. However, there is a risk of incorrect localization if the capsule migrates proximal to the stricture during surgery. This can be avoided by having the surgeon gently "milk" the capsule distally until further progress is prevented by the stricture.
There are situations where a laparotomy and enterotomy to retrieve the capsule may not be in the patient's best interest (eg, a patient with Crohn disease who has a capsule retained at an asymptomatic stricture). In these cases, removal of the retained capsule by device-assisted enteroscopy may be an option [121,122].
Magnetic resonance imaging — Patients should not undergo magnetic resonance imaging (MRI) until passage of the capsule has been confirmed, due to concern that it could result in damage to the gastrointestinal tract.
ESOPHAGEAL AND GASTRIC CAPSULE ENDOSCOPY —
Currently, the role for esophageal and gastric capsule endoscopy is limited. An esophageal capsule (the PillCam ESO2) with imaging systems at both ends was developed for the detection of mucosal disease of the esophagus in patients complaining of heartburn [123] and for screening for esophageal varices [124]. It was also used to screen for Barrett's esophagus [123,125]. However, the PillCam ESO2 was subsequently replaced by the PillCam UGI, which was later discontinued.
AnX Robotics makes a magnetically controlled capsule for gastric examination that is coupled with a tether for esophageal examination (NaviCam Stomach) [8]. On release in the stomach, which is filled with water, the capsule can then be moved around the stomach to examine it either manually or by an artificial intelligence program [126], in which the capsule can be controlled by an external solid-state magnet. The NaviCam Stomach capsule includes a tether that allows it to be used for esophageal examination.
COLON CAPSULE ENDOSCOPY —
A colon capsule for colorectal cancer screening has been approved by the European Medicines Agency and by the US Food and Drug Administration, but its role is still uncertain [127,128]. In the United States, it is approved for use in patients who have had an incomplete colonoscopy. In a 2020 update to its 2012 guideline, the European Society for Gastrointestinal Endoscopy (ESGE) suggests that colon capsule endoscopy is a reasonable alternative to computed tomographic colonography (CTC) for follow-up of an incomplete colonoscopy [129]. It is also an alternative to CTC for patients with a positive fecal occult blood test or a fecal immunochemical test who are not able to undergo colonoscopy. As of the 2020 update, the ESGE no longer suggests colon capsule endoscopy for first-line colorectal cancer screening. Guidelines issued by the Canadian Association of Gastroenterology recommend against the routine substitution of colon capsule endoscopy for colonoscopy [21].
Like optical colonoscopy, a rigorous preparation is given prior to colon capsule endoscopy. One regimen that has been used consists of the patient taking a clear liquid diet following a light breakfast the morning prior to the procedure [130]. The evening prior to the examination, patients take 3 liters of polyethylene glycol (PEG). The morning of the procedure, the patient drinks another liter of PEG between 6:00 and 7:00 am, and the capsule is ingested at 8:00 am. Additional medications (phosphosoda and bisacodyl) are then given during the procedure to increase transit of the capsule. Of note, phosphosoda has been associated with renal dysfunction and is not currently recommended in the United States. (See "Bowel preparation before colonoscopy in adults", section on 'Sodium phosphate preparations'.)
Colonic capsule endoscopy does not allow for biopsy or polyp removal, so patients with lesions detected during the examination typically require subsequent colonoscopy for further evaluation and/or treatment.
Studies looking at the efficacy of colon capsule endoscopy compared with standard colonoscopy have reached variable results. Most studies report sensitivities for detecting polyps ≥6 mm between 70 and 88 percent (range 39 to 88 percent), and specificities between 80 and 90 percent (range 64 to 93 percent) [128,131-138].
The colon capsule has also been studied for evaluating disease activity in patients with Crohn disease. A study of 40 patients with colonic Crohn disease compared colon capsule endoscopy using the PillCam Colon 2 capsule with optical colonoscopy for assessing disease severity [139]. The colon capsule was 86 percent sensitive and 40 percent specific for detecting colonic ulcerations. Compared with optical colonoscopy, colon capsule endoscopy underestimated the severity of the disease.
SOCIETY GUIDELINE LINKS —
Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Gastrointestinal bleeding in adults".)
INFORMATION FOR PATIENTS —
UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Angiodysplasia of the GI tract (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Common applications – Video capsule endoscopy (VCE) is most commonly used for the diagnosis of small bowel disorders. Available capsules include the PillCam SB3, Crohn's capsule, Mirocam, CapsoCam and the NaviCam SB capsule for evaluation of the small bowel (table 1). A colon capsule, PillCam Colon 2, has also been developed and is available in Europe, the United States, and Japan. (See 'Available capsules' above.)
•Small bowel capsules – VCE may be used to evaluate suspected small bowel bleeding in adults, to evaluate patients with suspected Crohn disease, assess mucosal healing, and to detect small bowel tumors. In addition, VCE is being used to detect small bowel injury associated with the use of nonsteroidal anti-inflammatory drugs. (See 'Indications' above and 'Suspected small bowel bleeding/iron deficiency anemia' above and "Evaluation of suspected small bowel bleeding (formerly obscure gastrointestinal bleeding)" and 'Crohn disease' above.)
•Esophageal and gastric capsule – The role of VCE for the evaluation of the esophagus and stomach is limited. NaviCam MCCE is a magnetically controlled capsule for gastric examination that is coupled with a tether for esophageal examination.
•Colon capsule – The colon capsule is used to screen for colon cancer and polyps. In the United States, it is approved for use in patients who have had an incomplete colonoscopy. Guidelines issued by the European Society for Gastrointestinal Endoscopy (ESGE) state that it is a reasonable alternative to computed tomographic colonography (CTC) for follow-up of an incomplete colonoscopy. It is also an alternative to CTC for patients with a positive fecal occult blood test or a fecal immunochemical test who are not able to undergo colonoscopy. Guidelines issued by the Canadian Association of Gastroenterology and the ESGE recommend against the routine substitution of colon capsule endoscopy for colonoscopy.
●Additional applications – Additional applications for VCE continue to be described but experience with some is limited. These include evaluating patients with acute gastrointestinal bleeding (not just those with suspected small bowel bleeding), diagnosing celiac disease, evaluating patients with complicated celiac disease (picture 1), detecting rejection following small bowel transplantation, detecting graft–versus-host disease after bone marrow transplantation, and performing surveillance in patients with hereditary cancer syndromes including Lynch syndrome. The evaluation of patients with abdominal pain of unclear etiology has been reported as a low-yield indication. (See 'Acute gastrointestinal bleeding' above and 'Small bowel tumors, polyps, and other pathology' above and 'Chronic abdominal pain' above.)
