Article Text

Download PDFPDF

Original article
Premature dissolution of the Agile patency device: implications for capsule endoscopy
  1. Nicholas Wray1,
  2. Ailish Healy1,
  3. Vicky Thurston1,
  4. Melissa Fay Hale1,
  5. Reena Sidhu1,
  6. Tony Blakeborough2,
  7. Mark McAlindon1
  1. 1 Academic Department of Gastroenterology and Hepatology, Sheffield Teaching Hospitals NHS Trust, Sheffield, UK
  2. 2 Department of Radiology, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK
  1. Correspondence to Dr Mark McAlindon, Department of Gastroenterology, Sheffield Teaching Hospitals NHS Trust, Sheffield S102JF, UK; mark.mcalindon{at}sth.nhs.uk

Abstract

Background The main risk of capsule endoscopy is retention of the capsule behind a stricture. Passage of an intact Agile patency device (Medtronic, Dublin, Ireland) through the small bowel is widely used to ensure luminal patency, although capsule retention has occurred in patients who have had a reassuring patency study. The device is designed to remain intact for at least 30 hours postingestion, such that loss of signal from the radiofrequency identification tag contained within, or absence of the device on radiological imaging, implies unimpeded intestinal transit.

Aim To identify the rate of premature dissolution (<30 hours postingestion) of the Agile patency device.

Methods Outcomes of all consecutive patients having an Agile patency device were analysed.

Results Premature dissolution of the patency device occurred in 5 of 307 patients, an incidence of 1.3%. This was recognised by the detection of a persistent radiofrequency signal after radiological imaging had failed to identify the patency device, prompting a careful search for the radiofrequency tag on the CT scout film. The tag was difficult to detect because of an oblique lie making it appear smaller than its 13×3 mm size and confusion with intra-abdominal or other metallic fragments.

Conclusions In the absence of radiological evidence of an intact Agile patency device, premature dissolution should be suspected in patients registering a persistent radiofrequency signal and confirmed by identifying the radiofrequency identification tag. Failure to do so might result in false reassurance that capsule endoscopy could be performed without risk of retention.

  • small bowel enteroscopy
  • small bowel disease
  • diagnostic and therapeutic endoscopy
  • endoscopy
View Full Text

Statistics from Altmetric.com

Introduction

Capsule endoscopy is a first-line investigative modality of the small bowel. Retention of the capsule behind a stricture is the the most common complication.1 2 While this might be desirable in some settings, such as diagnosing a hitherto unidentified tumour which requires surgical resection, it might complicate management of Crohn’s disease, for example, in which medical management is preferred.

Radiological methods are not always reliable in excluding strictures.1–3 This led to the development of the patency capsule (Given Imaging, Yoqneam, Israel), a device the same size and shape as the 26×11 mm M2A capsule endoscope comprising a polyurethane membrane containing lactose and barium, a radiofrequency identification tag (3×13 mm) and sealed at one end by a wax plug. After ingestion, the contents degraded in small intestinal secretions, allowing the barium to escape, the device to crumple and the remaining empty membrane and the radiofrequency tag to pass through any stricture present.4 5

Studies of the first version of the patency device showed that dissolution occurred at between 40 and 110 hours postingestion and that absence of a radiofrequency signal detected by a handheld scanner passed around the abdomen at 30 hours reliably predicted a patent lumen.4–7 However, there were some reports of pain and transient obstruction during the procedure which was thought to be due to the wax plug impacting in the stricture, thereby reducing exposure to intestinal secretions and delaying breakdown.4–8 Thus, the devices were redesigned to incorporate wax plugs at both ends and allow dissolution to begin at about 30 hours, the Agile patency capsule (Medtronic, Dublin, Ireland; figure 1).9

Figure 1

Agile patency device showing the wax timer plugs at either end and compressed lactose content.

At 30 hours postingestion, the absence of the radiofrequency signal implies that the device has been excreted intact and studies show that this reliably predicts safe capsule endoscopy without retention.10 However, the radiofrequency signal is still present in about half the patients, necessitating further imaging to determine if the device is in the small bowel, raising the possibility of a stricture or if it has safely passed into the colon.11 12 It is easy to identify on plain X-rays and cross-sectional radiology as a white opacity of 26 mm by 11 mm (figure 2).

Figure 2

CT imaging of an intact patency device (containing the visible radiofrequency tag) shown in the descending colon (arrow).

False-positive (suggesting a stricture) and false-negative (suggesting luminal patency) studies have been reported. The Agile device may occasionally remain intact in the small bowel 30–36 hours postingestion in the absence of small bowel strictures.11–13 Of more concern are reports of occasional patients who suffered a capsule retention having had an earlier negative study.14 This could arise from incorrect handling or malfunction of the radiofrequency scanner. It may occur because of reliance of plain abdominal radiographs to determine location of the device, which has been shown to be unreliable,11–13 or a significant delay before performing capsule endoscopy during which stricturing develops. However, this study is the first description of premature dissolution of the Agile patency device occurring earlier than 30 hours postingestion in which we report the frequency with which it occurs and describe how it should be recognised.

Methods

Agile patency protocol

Patients swallowed the Agile patency device with a glass of water at 08:00 hours prior to assessment at 14:00 hours on the following day. No restrictions, dietary or otherwise, were imposed during this time. A radiofrequency scan was performed and, if a signal persisted, patients were referred for a limited CT scan that afternoon as previously described.11 In the absence of a radiofrequency signal, or following identification of the patency device in the colon on CT scan (figure 2), capsule endoscopy was performed. If no patency device was identified on the CT scout film and a radiofrequency signal persisted on return to the endoscopy unit, the scout film was reviewed to identify the radiofrequency tag.

Patients

In June 2017, a patient was referred to the radiology department as per the usual unit protocol,11 having registered a radiofrequency signal at 14:00 hours, 30 hours after swallowing an Agile patency device. The protocol was developed in collaboration with our gastrointestinal radiologists who report all films. Prior scout film performed at 16:00 hours to identify the level at which to target the limited CT scan failed to reveal any evidence of a patency device and the patient was returned to the endoscopy unit with the suggestion that the device had passed in the 2-hour interval between scans. The patient denied that this had occurred and the radiofrequency signal was still evident. Review of the scout film suggested that one of four radio-opaque markers thought to be sterilisation clips was, in fact, the radiofrequency tag (figure 3). Premature dissolution has not previously been reported.

Figure 3

CT scout film showing three sterilisation clips and the radiofrequency tag (arrow) lying just to the right of the patient’s midline.

Data presented in this study are from consecutive patients referred for an Agile patency study, which was performed in all those known to have Crohn’s disease, who had a history of abdominopelvic radiotherapy, took regular non-steroidal anti-inflammatory drugs for at least 6 months or had small bowel obstructive symptoms.

Results

Three hundred seven patients swallowed an Agile patency device between June 2017 and March 2018. Five patients who had a persistent radiofrequency signal after failure to identify the patency device on a scout film had radiofrequency tags identified on closer inspection, a false-negative rate of 1.3%. In each case, the oblique lie meant that the tag appeared smaller than the 3×13 mm size making it difficult to detect (figure 3). Furthermore, it could be difficult to distinguish from sterilisation clips (figures 3 and 4), metallic objects (zippers or body piercings; figures 4 and 5), cholecystectomy clips and component parts of an intrauterine device (figure 6). In one case, dilute barium seen in the region of the tag (figure 6) is more obvious on the CT image (figure 7).

Figure 4

CT scout film following a repeat study on the patient shown in figure 3 showing the intact patency device (containing visible radiofrequency tag) in the descending colon (arrow), three sterilisation clips and a central circular radio-opaque object.

Figure 5

A lateral view taken during the same examination shown in figure 4 demonstrating the intact patency device (arrow) and in which it is evident that the central ring-like opacity seen in figure 4 is an umbilical piercing on the abdominal wall.

Figure 6

CT scout film showing the radiofrequency tag in the patient’s left upper quadrant (arrow), a cholecystectomy clip in the right upper quadrant and an intrauterine coil device in the pelvis.

Figure 7

CT imaging showing the radiofrequency tag in the descending colon (arrow), which is separate from residual barium from the patency device (also present but less obvious in the scout film taken of the same patient shown in figure 6).

All five patients had successful repeat Agile patency studies and underwent capsule endoscopy. The images in figures 3-5 are from the same patient: in the repeat study there are three instead of four radio-opaque markers in the pelvis (figure 4), although the image is further confused by the presence of a ring-shaped object which was an umbilical piercing seen more clearly on the lateral film (figure 5) and the intact patency device is in the descending colon (figures 4 and 5). Three patients had evidence of active Crohn’s disease (two with strictures, one with an incomplete study) and two patients had normal studies. No capsule retentions occurred.

Conclusions

Premature dissolution of the Agile patency device occurs with an incidence of 1.3%. This has not previously been recognised and failure to do so would lead to false reassurance that capsule endoscopy could be performed safely without risk of retention. If there is a persistent radiofrequency signal 30 hours after swallowing the device and this cannot be identified on radiological imaging, a careful search for the tag should be conducted. Clinicians need to be aware that the tag may appear smaller than expected if lying obliquely and may be confused for surgical clips, intrauterine devices, clothing or other metallic objects attached to the skin.

Capsule endoscopy is a first-line small bowel imaging tool, but there is a risk of capsule retention in patients with stricturing disease. This occurs in 1.2% of patients with suspected small bowel bleeding, but has been described in as many as 5%–13% of patients known to have Crohn’s disease.1 Radiological imaging is not always reliable in excluding strictures, whereas confirmed passage of the patency device appears to predict safe passage of the capsule endoscope, even in high-risk patients.9 10 Our study of luminal patency in a series of 400 patients demonstrated a positive predictive value of 99.7% using the patency device and a limited CT scan protocol.11 However, these devices are not entirely without problems. They may occasionally precipitate obstructive symptoms if retained in a stricture prior to dissolution.8 15 This study demonstrates that premature dissolution may also occur and if not recognised, could result in subsequent capsule retention.

Some units which do not have access to a radiofrequency identification tag scanner rely on plain abdominal and/or CT radiology to determine whether or not the patency device has passed through the small bowel. Nakamura et al also used radiological assessments to assess the location of an Agile device, which did not contain a radiofrequency tag.16 In addition to increasing radiation exposure, these techniques risk missing cases of false-negative investigations occurring as a consequence of premature dissolution of the patency device.

It is possible that a patient might excrete the patency device between having a radiofrequency signal detected and the radiological assessment performed to locate it. However, our recommendation is to repeat the radiofrequency scan in any case in which the findings of the two tests are inconsistent: it is a simple, safe test which incurs no patient risk and little inconvenience (figure 8). In the event that the signal persists, the imaging should be reviewed by a radiologist familiar with the appearance of the radiofrequency tag to determine if premature dissolution might explain the discrepancy. A radiofrequency tag lying obliquely or end-on might be identified with greater confidence by imaging from different angles (eg, lateral views). However, this would increase radiation exposure and our experience has been that in the context of a persistent radiofrequency signal, a radiologist aware of the appearance of a tag can readily locate it even in the presence of surgical clips. Should this be the case, a repeat study should be considered.

Figure 8

Suggested protocol for ensuring a patent small bowel lumen using the Agile patency device. In the absence of a radiofrequency (RF) signal 30 hours postingestion, capsule endoscopy can be performed. Should the signal remain, a scout film will allow targeting of a limited CT scan through the device. If a colonic location is confirmed, capsule endoscopy can proceed but further imaging may be needed to confirm and define a stricture should the device remain in the small bowel. If the scout film fails to identify a patency device, the RF scan should be repeated and if negative, capsule endoscopy could proceed. If the RF signal remains, evidence of the RF tag should be sought, premature collapse of the device suspected and repeat study considered.

Why a patency device should collapse prematurely is unclear, but may relate to duration or intensity of exposure to lactase. The polyurethane membrane body contains compressed lactose which is dissolved over time by intestinal enzymes.4–10 13 Most lactase is expressed in the jejunum, but transit through the small bowel is non-uniform and very variable.8 17 Furthermore, there is a 10-fold variation in lactase activity between those with homozygous lactase persistence and non-persistence.18 It may be that those in whom the device collapses prematurely are high lactase producers, have slower jejunal transit or both.

Significance of this study

What is already known on this topic

  • The Agile patency device is widely used to predict small bowel luminal patency and avoid the risk of subsequent retention of capsule endoscopes.

  • However, apparently successful passage of patency devices through the small bowel have been followed by subsequent retention of capsule endoscopes behind strictures.

What this study adds

  • Premature dissolution of the patency device occurs in 1.3% of devices, which may explain unexpected capsule retention.

  • Clinicians can be alerted to the possibility of premature dissolution by recognising the radiofrequency tag on CT scout (or plain abdominal) films, if a radiofrequency signal persists in the absence of any obvious intact patency device.

How might it impact on clinical practice in the foreseeable future

  • Awareness of the possibility of premature dissolution of the patency device and recognition of the appearance of the radiofrequency tag on radiological imaging may reduce the risk of retention of capsule endoscopes.

References

View Abstract

Footnotes

  • Contributors AH, VT and NW collated the data. NW, AH, MH, RS, TB and MEM were involved in the design of the paper. Images were provided by TB. NW wrote the first draft of the paper. NW, AH, MH, RS and TB contributed to the second draft and MEM wrote the final draft.

  • Funding No funding was received for this study.

  • Competing interests MMcA has received financial support for research and conference attendance from Given Imaging Ltd, Intromedic Ltd and Ankon Ltd; research support from Jinshan Science and Technology Ltd and has acted as a consultant for Medtronic Ltd. NW, AH, VT, MFH, RS and TB have no conflicts of interests or financial ties to disclose.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement There is no additional unpublished data from the study.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.