Fecal tagging improves detection in virtual colonoscopy

Aug 28, 2002

The best virtual colonoscopy results begin with a clean colon. But VC's intensive cleansing regimen has always been troublesome, inasmuch as the discomfort, lost sleep, and hunger that patients must endure have been shown to discourage compliance in several studies.

Many radiologists believe that virtual colonoscopy's very future as a high-volume screening tool for colorectal cancer depends on the development of alternatives to cathartic cleansing, including fecal tagging and so-called electronic cleansing, a technique that relies on software-based analysis of CT data to separate polyps from fecal residue.

In a study published in the July issue of Radiology, researchers may have found a promising solution to the rigors of colonic cleansing, using a technique that combines barium fecal tagging (FT) with an easier prep. The Belgian researchers reported that virtual colonoscopy combined with the FT protocol yielded slightly higher sensitivity for colorectal polyps than VC with the standard bowel prep. Perhaps just as important, the more comfortable FT protocol left patients more satisfied with the screening experience overall (Radiology, July 2002, Vol. 224:2, pp. 393).

Cathartic bowel cleansing is not only difficult for patients, it still leaves fecal residue, which the radiologist must then differentiate from polyp in virtual colonoscopy, wrote study authors Dr. Philippe Lefere, Dr. Stefaan Gryspeerdt, Dr. Jef Dewyspelaere and colleagues from the departments of radiology and gastroenterology at Stedelijk Ziekenhuis in Roeselare, Belgium.

"Our efforts have been focused on finding a technique that first reduces the need for colonic cleansing," they wrote. "Second, the technique was required to label residual fecal residue with barium, allowing visual differentiation from polyp without the use of dedicated electronic cleansing software.... The purpose of our study was to compare reduced colonic cleansing with fecal tagging (FT) and colonic cleansing with regard to patient acceptance, sensitivity, and specificity."

In the study, 100 patients referred for colorectal cancer screening were divided into two groups: 50 who underwent the FT prep, and 50 who underwent standard prep, followed in all patients by both virtual and conventional colonoscopy. Virtual colonoscopy results were compared to conventional colonoscopy.

The 50 non-FT patients underwent standard colonoscopic cleansing 1 day before the examination. The protocol consisted of a low-residue diet, 59 grams of polyethylene glycol dissolved in 1 liter of water per 30 kg of body weight, and 2 bisacodyl tablets to reduce the amount of fluid ingested by the patient.

The 50 FT patients began their preparation 2 days before the procedure, when they were asked to avoid high-fiber and high-residue foods (including flavored cheese, condiments, whole-grain bread, crackers, cereal, popcorn, raw fruit, jam, and garlic). On the second day the FT group received a low-residue nutritional kit, along with orally administered magnesium citrate and barium solutions (E-Z-EM, Westbury, NY), and 4 bisacodyl tablets. On the day of the exam, the patients fasted and used a bisacodyl suppository (Dulcolax bisacodyl tablets and suppositories, Novartis, Basel, Switzerland).

After the prep and just before CT scanning, both groups of patients were interviewed in an effort to determine their acceptance of the preparation in terms of global discomfort, sleep, and diet disturbance, and side effects such as nausea, vomiting, cramps, dizziness, or palpitations. They were also sent a follow-up questionnaire two weeks later in order to assess their memory of the entire experience.

Following manual room-air insufflation of the colon, patients underwent CT virtual colonoscopy screening in both prone and supine positions (Tomoscan AV-EU; Philips Medical Systems, Best, the Netherlands). The protocol included 5-mm slice thickness, a 7-mm table feed, pitch 1.4, and 3-mm reconstructions, with a 350-mm field of view and a 512 x 512 matrix.

The FT group had additional colonic cleansing, and all patients underwent conventional colonoscopy by an experienced endoscopist within 2-3 hours after VC. Two radiologists experienced in VC interpretation evaluated the CT images for the presence of polyps and fecal residue. The latter was deemed either labeled or not labeled by barium. All 100 patients completed the study, and 81 returned the postprocedural questionnaires.

Per-lesion results compared to conventional colonoscopy showed that FT left more fecal residue than non-FT, but improved the differentiation of residue from polyps, resulting in comparable results for both groups.

In the non-FT group, virtual colonoscopy had an overall sensitivity of 78% (45/58 lesions), compared with overall sensitivity of 77% (34/44 lesions) in the FT group. For lesions 6-9 mm, sensitivity was 89% (16/18) in the non-FT group, compared to 92% (12/13 lesions) in the FT group. For lesions 5 mm and smaller, sensitivity was 56% (14/25 lesions) in the non-FT group, compared to 57% (12/13 lesions) in the FT group.

In per-patient results, overall specificity was 88% among FT patients, and 77% among non-FT patients. Per-patient sensitivity was 88% (14 of 16 patients) overall in the FT group, and 85% in the non-FT group, according to the authors.

Patient views

"When interviewed the morning after the procedure, patients in the non-FT group reported a significantly higher level of sleep disturbance and global discomfort, although in both groups nearly all patients reported at least some level of sleep disturbance caused by bowel movements," the authors wrote. Moreover, side effects such as nausea, vomiting, and cramps were significantly more common in the non-FT group. The FT group reported positive results from the dietary kit.

A questionnaire mailed two weeks after the exams put the FT group even further ahead of their non-FT counterparts, who reported that they had suffered greater overall discomfort and more sleep loss. The FT patients were also more willing to consider repeating the procedure than the non-FT group, though this result did not reach statistical significance (P > 0.5), according to the authors.

"Our technique thus reduced discomfort by reducing colonic cleansing and eliminated the need for electronic cleansing," Lefere and colleagues wrote. They noted, however, that electronic cleansing of the unprepared colon is expected to further increase patient compliance.

"In conclusion, dietary FT appears feasible, allowing reduced colonic cleansing, which results in improved patient compliance, and improving specificity by reducing false-positive results caused by fecal residue."

By Eric Barnes
AuntMinnie.com staff writer
August 29, 2002

Future looks bright for virtual colonoscopy, June 3, 2002

Prepless virtual colonoscopy shows early promise, June 3, 2002