Wireless capsule endoscopy (WCE) is increasingly used in the pediatric population. There are many indications for capsule endoscopy in children including, but not limited to, evaluation for and follow-up of inflammatory bowel disease, obscure gastrointestinal (GI) bleeding, suspected celiac disease, suspected small intestinal tumors, and unexplained iron deficiency anemia (1). The majority of adult patients are able to swallow the capsule endoscope; however, the pediatric population presents a number of challenges regarding capsule introduction. Many pediatric patients are extremely young to swallow the capsule. In addition, like adults with dysphagia or senile dementia, there is a large group of pediatric patients with either behavioral disorders or neurological conditions that preclude swallowing the capsule, thus requiring endoscopic placement.
The real-time (RT) viewer (Given Imaging, Yokneam, Israel) is a computer device that displays the images from the capsule endoscope as they are being recorded. The RT viewer is used with increasing frequency in the practice of capsule endoscopy to assist with increasing the rate of complete small bowel examinations by WCE, thus improving the diagnostic yield of the study (2). A capsule study is considered complete when it passes through the entire small bowel during the course of the study. In several studies, slow progress of the capsule, as measured by the RT viewer, allowed clinicians the opportunity to use specific interventions such as prokinetics, which provided for a significant increase in completion of the capsule study (3,4).
Endoscopic placement of the capsule endoscope can be challenging because of the size of the capsule delivery device and the difficulty of advancing the device and scope around the duodenal sweep. We describe here a novel use of a RT viewer to effectively and safely deploy the capsule endoscope in the pediatric population. We hypothesize that the use of the RT viewer will allow precise localization of the endoscopic capsule at the time of placement, and result in an increased rate of complete capsule endoscopy examinations.
A retrospective chart review was performed on all WCE studies completed at the Children's Memorial Hospital in Chicago, IL, from February 2010 to October 2011. The institutional review board of the Children's Memorial Hospital approved the study. The Given wireless video capsule endoscopy system with SB2 capsules and the Given Real-Time Imager (Given Imaging) were used to perform the WCE in all of the patients. The study was not supported by any external funding source, and specifically was not funded by the manufacturer of the device.
All procedures were performed under general endotracheal anesthesia with the patient in the left lateral decubitus position. All of the patients in the study had a diagnostic upper endoscopy to the extent of the second portion of the duodenum, either with or without biopsy, before WCE placement. The AdvanCE capsule delivery device (US Endoscopy, Mentor, OH) was used for deployment of the capsule endoscope in all cases. The device was fastened on to the gastroscope according to manufacturer instructions. The capsule endoscope was removed from packaging and placed in the capsule delivery device. Recognition of capsule was noted on the capsule device recorder. The RT viewer was attached to the capsule recorder, and the capsule image was noted on screen before insertion into the oral cavity. The RT viewer was positioned in close proximity to endoscopy monitor so that the endoscopist was able to view endoscope image and capsule image simultaneously (Fig. 1).
The endoscope and loaded capsule delivery device were inserted into oral cavity. The upper esophageal sphincter was visualized on RT viewer. Endoscope and delivery device were slowly advanced into stomach to the antrum. The loaded AdvanCE delivery device remained positioned tightly at the distal end of the endoscope throughout the procedure. The pylorus was visualized on RT viewer. The endoscope and device were then advanced through the pylorus and into the second portion of the duodenum. The duodenal lumen was visualized on the RT viewer. The AdvanCE delivery device was then engaged and the capsule was released. The RT viewer was observed during withdrawal of gastroscope for continual visualization of the duodenal mucosa while the endoscope and now empty AdvanCE device were withdrawn. Visual inspection of duodenal, gastric, and esophageal mucosa was performed during withdrawal to assess for any mucosal trauma incurred during placement of WCE. The patient underwent routine postanesthesia recovery and was then discharged with capsule recorder to be returned when study complete. All of the patients were asked to adhere to the standard diet schedule as recommended by the manufacturer. The recording device was disconnected after 8 hours or sooner if capsule noted to be in colon on RT viewer or had been passed in the stool. Clear instructions were given to patients to observe for capsule passage and patients were to call with any adverse events including abdominal pain, nausea, vomiting, or failure to pass the capsule.
Seventeen patients that had endoscopic placement of the capsule endoscope using the RT viewer were identified. Demographic information is detailed in Table 1. Two patients had received liver transplants for biliary atresia and each had a previous hepatoportoenterostomy. One of these had capsule done secondary to GI bleeding and 1 due to persistent anemia. One patient had short gut syndrome secondary to neonatal necrotizing enterocolitis, had a gastrojejunal (GJ) tube in place and was being evaluated for a GI bleed. Thirteen patients required endoscopic placement of capsule endoscope because of inability to swallow the capsule endoscope, whereas 4 were placed as part of a previously scheduled procedure to take advantage of sedation and airway protection. Although we have previously placed capsules endoscopically because of incomplete capsule studies and prolonged gastric transit time in patients with partial gastric outlet obstruction or delayed gastric emptying, none of the patients evaluated during the course of this study had either of these conditions.
The upper esophageal sphincter, pylorus, and duodenal lumen were successfully visualized in all of the 17 patients using the RT viewer, and the capsule was successfully deployed into the duodenal lumen in all of the patients. In each case the endoscopist was able to confirm capsule location within the duodenum during the procedure and delivery device withdrawal. In the patient with the GJ tube, the endoscopist noted proper location of the GJ tube on the RT viewer while gastroscope was withdrawn. One patient had retrograde transit of the capsule back into the stomach upon retching while waking up from anesthesia. There was no evidence of iatrogenic trauma or bleeding in any patient.
Nine patients had findings on their capsule study. Three patients had small bowel polyps noted. One patient had atrophic and erythematous mucosa consistent with celiac disease, 3 patients had findings consistent with Crohn disease (2 known Crohn with apthous ulcerations and pseudopolypoid lesions noted and 1 of the patients with anemia with denuded mucosa and ulcerations consistent with inflammatory bowel disease). One patient had a bleeding venous ectasia noted in mid-jejunum and 1 patient had small bowel varices noted. There were 5 incomplete studies. One of these was a 30-minute break in recording of the capsule owing to the patient accidentally misplacing the recording device. The capsule in this patient did reach the cecum within the recording time of the capsule endoscope. One patient was 1 of the post–liver-transplant patients and 1 patient had short gut with severely dilated and stagnant loops of bowel. There was prolonged delay of transit of the capsule within these loops; however, the final images were of the terminal ileum. The final 2 patients had prolonged intestinal transit, likely secondary to general anesthesia for the procedure, although the final images were of terminal ileal mucosa and a 12-hour capsule may have allowed for completion of the study. The completion rate was consistent with that seen in other pediatric studies of capsule endoscopy (5–7). All of the 13 patients whose capsules reached the cecum within the recording time of the capsule endoscope had a small bowel transit time of <6 hours, and 10 of 13 patients had a small bowel transit time of <4 hours. All of the patients passed the capsule without complications.
We performed a retrospective study examining the use of the RT viewer for endoscopic placement of the capsule endoscope in a pediatric population. We found that the RT viewer allowed for effective and safe deployment of the WCE in our population. In every case, we demonstrated the placement of the capsule within the duodenal lumen and successful deployment into the second portion of the duodenum. Additionally, we were able to confirm location of the capsule within the duodenum even as the scope was withdrawn.
There are several potential benefits to the RT viewer in pediatric capsule practice. First, with this technology, we have the ability to view capsule study in real time. This may be useful in evaluation of the patient with acute GI bleeding. A study of capsule endoscopy in the setting of acute GI bleeding demonstrated specific lesions in a majority of patients, and specific diagnostic and therapeutic procedures were able to be performed in many patients (8). The RT viewer may allow the clinician to monitor the capsule in real time and identify a possible lesion that may direct therapy before downloading and reading the capsule. In our practice, we use the RT viewer to track the progress of the capsule through the small bowel, regularly checking capsule location on the RT viewer at 4 and 6 hours. In all capsule studies performed on outpatients, we were able to disconnect and collect the recorder before completion of the 8-hour recording time, allowing the family to leave the medical center earlier while also assuring them that the capsule had successfully passed. Additionally, using the RT viewer, there is potential to intervene by administering a prokinetic medication or increasing patient mobility during the recording period if a capsule either travels retrograde into the stomach or is not progressing during the study.
Endoscopic placement of the capsule has been described and is widely performed. The earliest descriptions used a Roth net to advance the capsule beyond the pylorus (9,10). Barth et al (11) performed a study in pediatric patients looking at delivery of the capsule into the duodenum using a modified Roth net/bard adapter. A difficulty with the present study was that they had poor visualization and control of capsule requiring the assistance of anesthesiologist to advance through pharynx and upper esophageal sphincter. Although no serious adverse events were reported in these series, a significant amount of mucosal trauma was noted with these techniques. The utilization of the AdvanCE delivery device has improved the technique of endoscopic capsule placement (12), and Uko et al (13) demonstrated use of this device in pediatric patients. The use of this device was not associated with any significant mucosal trauma as described with other techniques.
The rate of incomplete capsule studies in the medical literature is between 20% and 30% (7). One study of factors associated with incomplete small bowel capsule studies demonstrated that studies performed for overt GI bleeding and patients with a prior history of abdominal surgery or bowel obstruction were significant risk factors for incomplete studies (6). A separate analysis of incomplete studies in capsule endoscopy demonstrates that slow gastric transit time is associated with a significant percentage of incomplete studies (14). Jensen et al (5) found that endoscopic placement was associated with incomplete capsule studies in a series of pediatric patients, potentially implicating a role of general anesthesia in causing slower small intestinal transit. Of our incomplete studies, 1 was owing to human error, misplacing the data recorder, and 2 patients had a history of major abdominal surgery including bowel resection and construction of a roux-en-Y anastomosis for hepatoportoenterostomy and liver transplant. Although our completion rate using the RT viewer did not vary from that noted in the medical literature, deployment of the capsule in the duodenal lumen allowed the capsule to spend nearly its entire recording time in the small bowel. Each of our incomplete studies demonstrated ileal mucosa on the final image. It is possible that a longer duration of capsule recording time, such as a capsule with a 12-hour recording time, may allow for completion of the study in these patients.
Attaining adequate visualization during the endoscopic placement of the capsule endoscope is a reasonable goal for any endoscopist performing the procedure. This would be of particular importance in patients with abnormal upper GI anatomy such as prior Nissen fundoplication, prior gastrostomy tube placement, Zenker diverticulum, or mechanical gastric outlet obstruction; however, even in patients with normal anatomy, our technique enabled us to direct the capsule directly to the pylorus without excessive searching or manipulation of the delivery device. The limitations of the delivery device involve the ability to visualize wherein the scope is moving during advancement. Additionally, the capsule adds 2.6 cm of length onto the end of the scope that cannot be visualized. Many endoscopists will push the delivery device away from the tip of the endoscope to obtain visualization and advance the capsule delivery device. Additionally, some endoscopists will deploy the capsule blindly, not being able to visualize the ejection of the capsule from the delivery device into the duodenal lumen. The capsule thus has potential to be deployed into the duodenal bulb and rebound into the stomach where it will either remain or need to be advanced using a different device such as the Roth net, increasing the potential for mucosal trauma. The capsule may also incompletely deploy and be withdrawn as the scope is withdrawn. The endoscopist is in effect introducing a foreign body, the capsule endoscope, into the GI tract. In foreign body removal, endoscopy textbooks recommend keeping the foreign body tight against the endoscope so that the endoscope and foreign body move as a unit (15). We suggest that it is reasonable to adhere to these principles of foreign body removal in the case of capsule introduction. The use of the RT viewer enables the endoscopist to move the delivery device and endoscope as a unit to have maximum control over both the device and the endoscope, to minimize mucosal trauma and to negotiate placement in patients with difficult anatomy. Our technique enables the endoscopist to confirm that the capsule will be deployed into the duodenal lumen and that the capsule remains there during withdrawal of the endoscope.
In conclusion, the use of the RT viewer is a useful and effective adjunct to endoscopic placement of the capsule endoscope in pediatric patients. It is highly effective in allowing the endoscopist to safely determine the site of deployment, and in selected patients, may help to increase the rate of successfully completed studies.
1. Shamir R, Hino B, Hartman C, et al. Wireless video capsule in pediatric patients with functional abdominal pain. J Pediatr Gastroenterol Nutr
2. Gao YJ, Ge ZZ, Chen HY, et al. Endoscopic capsule placement improves the completion rate of small-bowel capsule endoscopy and increases diagnostic yield. Gastrointest Endosc
3. Shiotani A, Honda K, Kawakami M, et al. Use of an external real-time image viewer coupled with prespecified actions enhanced the complete examinations for capsule endoscopy. J Gastroenterol Hepatol
4. Lai LH, Wong GL, Lau JY, et al. Initial experience of real-time capsule endoscopy in monitoring progress of the videocapsule through the upper GI tract. Gastrointest Endosc
5. Jensen MK, Tipnis NA, Bajorunaite R, et al. Capsule endoscopy performed across the pediatric age range: indications, incomplete studies, and utility in management of inflammatory bowel disease. Gastrointest Endosc
6. Lee MM, Jacques A, Lam E, et al. Factors associated with incomplete small bowel capsule endoscopy studies. World J Gastroenterol
7. Westerhof J, Weersma RK, Koornstra JJ. Risk factors for incomplete small-bowel capsule endoscopy. Gastrointest Endosc
8. Lecleire S, Iwanicki-Caron I, Di-Fiore A, et al. Yield and impact of emergency capsule enteroscopy in severe obscure-overt gastrointestinal bleeding. Endoscopy
9. Carey EJ, Heigh RI, Fleischer DE. Endoscopic capsule endoscope delivery for patients with dysphagia, anatomical abnormalities, or gastroparesis. Gastrointest Endosc
10. Hollerbach S, Kraus K, Willert J, et al. Endoscopically assisted video capsule endoscopy of the small bowel in patients with functional gastric outlet obstruction. Endoscopy
11. Barth BA, Donovan K, Fox VL. Endoscopic placement of the capsule endoscope in children. Gastrointest Endosc
12. Holden JP, Dureja P, Prau PR, et al. Endoscopic placement of the small-bowel video capsule by using a capsule endoscope delivery device. Gastrointest Endosc
13. Uko V, Atay O, Mahajan L, et al. Endoscopic deployment of the wireless capsule using a capsule delivery device in pediatric patients: a case series. Endoscopy
14. Hoog CM, Bark LA, Arkani J, et al. Capsule retentions and incomplete capsule endoscopy examinations: an analysis of 2300 examinations. Gastroenterol Res Pract
15. Gershman G, Ament M. Therapeutic upper gi endoscopy. In: Practical Pediatric Gastrointestinal Endoscopy
. Malden, MA: Blackwell Publishing; 2007:102–31.