Thomas Jefferson UniversitySidney Kimmel Medical College

Main menu:

Ambulatory Swallowed Camera Capsule

Advances in Imaging of the Small Intestine

Gastrointestinal diseases are a very common problem in the world. Approximately 67,000,000 people in the United States are affected by gastrointestinal illness. GI diseases are responsible for more hospital admissions in the United States than any other disease group. This adds up to an annual cost of approximately one-hundred billion dollars. Over the last two decades, gastroenterologists and radiologists employed a multitude of new techniques to image the gastrointestinal tract. Certainly, examination of the esophagus, stomach, and proximal small bowel have been brought to an exceedingly high level of visualization with digital endoscopy. The same holds true for colonoscopy. However, examination of the small bowel has continued to be a difficult area to completely evaluate. For many years, the only examination that was available was an upper GI with a small bowel x-ray or enteroclysis. Although this does allow for imaging of the majority of the small bowel, it is clearly operator-dependent and even in the best of institutions, not all of the small bowel can be separated to ensure that no lesions were overlooked. In addition, as holds true for a barium enema, vascular ectasias cannot be diagnosed by barium studies. CT scan imaging of the small bowel again cannot completely visualize the entire length and is not adequate for imaging nor is MRI examination. Virtual colonoscopy is being examined for screening of that particular organ, but it is still considered experimental. There is no virtual small bowel endoscopy available at this time.

A number of years ago, two upper endoscopes were developed, one called a Push enteroscope and another called a Sonde enteroscope. The push enteroscopes are variable in length but generally are between 200 and 260 centimeters. They are passed as a traditional upper endoscope through the mouth and either using an overtube or with external compression can be passed to the proximal to midjejunum. This has been helpful for trying to identify obscure gastrointestinal bleeding, but again could only visualize the very proximal portion of the small bowel. Sonde enteroscopy is a much more labor-intensive and uncomfortable procedure for the patient and the doctor. The small thin Sonde scope is passed through the nose and then into the stomach. Another scope is then passed through the mouth and the Sonde scope is dragged into the small bowel using the larger scope that has been passed through the mouth. A balloon is then inflated and by peristalsis and with the patient advancing the scope much like a Cantor tube, it is hopefully advanced to the distal jejunum and/or ileum. The total passage time can be long varying from six to ten hours. At the end of that time, the gastroenterologist must slowly remove the Sonde scope. There are no therapeutic capabilities with this instrument and there is no ability to push back into an area once it has been removed except by using some means of external compression. It is quite labor-intensive and uncomfortable for the patients. Nasal bleeding is not uncommon.

Given the above limitations for the evaluation of the small bowel, a number of companies around the world began to develop swallowed camera capsule technology. In the 1950's, the Heidelberg capsule had been used to record gastric pressure and acidity during digestion. However, further development of true imaging of the small bowel was delayed until microelectronics including metal-oxide semiconductor video imaging chips and white light-emitting diodes were developed.

A small company from Israel named GIVEN Imaging, Ltd. with a base in Norcross, Georgia was the first company to obtain FDA approval for a swallowed camera capsule. The capsule measures 11 x 26 mm and contains a lens and color camera chip, two batteries, a radio frequency transmitter, and four LEDs light-emitting diodes. The camera utilizes a CMOS (complimentary metal-oxide silicone) chip. This chip requires low power similar to those found on the video endoscopes and digital cameras. Also, they operate at very low levels of illumination. As the capsule traverses the gastrointestinal tract, images are obtained per second and transmitted to a data recorder that is worn by the patient utilizing a Velcro belt.

On the Velcro belt, in addition to the data recording device, there is a battery that lasts approximately eight to ten hours. This is rechargeable. The video images are transmitted to aerials that are taped to the body utilizing high frequency-band radio telemetry at approximately 410 MHz. Trigonometric analysis of signal strength allows continuous monitoring of the capsule's position in the body. The camera takes and transmits about 2 images per second. (50,000 plus images in 8 hours.) By using a lens of short focal length, images are obtained as the optical window of the capsule sweeps past the gut wall without requiring air inflation of the gut lumen. The capsule is propelled by normal peristalsis. Once the images are acquired by the image data recorder, the patient comes to the hospital approximately eight hours later and removes the belt. The data is then downloaded onto a computer workstation. This utilizes RAPID (Recording and Processing of Images and Data) application software. Viewing of the gastrointestinal tract is then performed by the clinician which takes between forty-five minutes and an hour. One of the first successful trials was published in nature (2000;405:417). Ten subjects were studied. The M2A capsule successfully transmitted the images from the gastric pylorus to the cecum. Mean gastric transit time was eighty minutes (range 17 to 280 minutes). The average small bowel transit time was ninety minutes (range 45 to 140 minutes). The capsule reached the cecum in less than two hours on average with evacuation times ranging from ten to forty-eight hours. Another large clinical trial was completed by Blair Lewis et al on twenty-one patients. This was an IRB approved trial for patients referred for small bowel endoscopy. If they had a history of pregnancy, diabetes, pacemaker, or bowel obstruction, they were excluded. They underwent a capsule study after a twelve hour fast. They were all pretreated with simethicone. After swallowing, patients remained without food for four hours and the device was removed at eight hours. The capsule images were reviewed by two physicians, one that was blinded to the clinical background of the patient. There were twelve women and nine men. The average age was sixty-one. Swallowing of the capsule was also assessed by the patients and the investigator. No patient had any difficulty swallowing the capsule. Twenty of twenty-one patients judged swallowing as easy or very easy while one felt it was tolerable. Average transit times were similar to other studies. Twenty of the twenty-one capsules were retrieved and no microscopic damage was documented. Capsule integrity was maintained. Capsule images were judged as good to excellent by two physicians and no discrepancies of findings were noted. The average time for review during this study was fifty-six minutes. Capsule endoscopy was superior to push enteroscopy for the evaluation of obscure bleeding. The capsule made a diagnosis in fifty-five percent, that is eleven of twenty patients, and the findings included fresh blood, tumor, angiodysplasias, or an ileal ulcer. Push enteroscopy made a diagnosis in six of twenty or thirty percent of the patients. All findings were angiodysplasias. No additional diagnoses were made by push enteroscopy. The capsule identified lesions found distally in the small bowel not reachable by push enteroscopy. The study was statistically significant. Concerns initially about the capsule included image quality and tumbling of the capsule. Although tumbling may occur, it did not appear to happen often. Even if the capsule were to rotate, not much is missed as the capsule does tumble. Mucous coating the capsule limiting visibility also did not turn out to be a major problem. Lack of air insufflation to fully evaluate the small bowel was also a concern but as it is for sonde enteroscopy, this was not a huge problem. We believe that capsule camera technology is in its infancy. Future directions include the ability to motorize the capsule and perform biopsies or therapeutic capabilities utilizing miniature lasers.. These advances are presently under investigation The ability to examine the colon has also been investigated and posters were presented at digestive disease week in 2001.

Long-life batteries are in development of up to eighteen hours. In terms of reimbursement, at the present time, the insurance industry has not authorized payment for the capsule itself although in small pockets around the country that may have changed. We are presently performing a study at Jefferson for patients with unexplained abdominal pain who have already undergone small bowel imaging. We are comparing the results of small bowel imaging to camera capsule technology. Given the uncomfortable nature of an upper GI small bowel follow through, the x-ray exposure, etc., we believe that a swallowed camera capsule would be much more definitive and far better tolerated and may indeed have a higher yield for pathology than small bowel x-ray.

In conclusion, swallowed camera technology has arrived and has opened up a whole new world of endoscopic imaging. No longer will patients with obscure gastrointestinal bleeding need to undergo higher risk, painful and more costly procedures to visualize the small bowel. Although at the present time there is no therapeutic capability, certainly this can direct other therapies such as push enteroscopy with cautery and/or intraoperative enteroscopy and/or resection if necessary once pathology has been identified and the exact location noted.

Anthony Infantolino, MD, FACP
Clinical Assistant Professor of Medicine
Thomas Jefferson University Medical Center
Jefferson Medical College
Clinical Director of Endoscopic Ultrasound and
Photodynamic Therapy