CHICAGO – Water infusion during the insertion phase of colonoscopy improved overall and proximal-colon adenoma-detection rates by 11% over standard air insufflation in a prospective randomized trial involving 368 patients.
After controlling for age, body mass index, smoking, alcohol consumption, withdrawal time, and the quality of bowel preparation, researchers found 81% higher odds of detecting an adenoma using the water-method group than in the air-method patients (odds ratio 1.81), Dr. Felix Leung reported at the annual Digestive Disease Week.
A prospective randomized trial in 82 veterans reported last year by Dr. Leung as a proof of principle of water insufflation in unsedated colonoscopy demonstrated improved patient comfort and cecal intubation with that method (Gastrointest Endosc. 2010;72:693-700).
The use of water to aid colonoscope insertion avoids colon elongation, minimizes angulations, helps identify the lumen, and provides a clear view of the colon, explained Dr. Leung, who is with the Veterans Affairs Greater Los Angeles Healthcare System in North Hills, Calif.
Water had been used before in colonoscopy, but the method proved cumbersome because multiple syringes were required to deliver the liquid. In the current study, doctors delivered room temperature water via pedal pump and a tube connected to the base of the colonoscope, with its air feature turned off. Roughly 100 ccs of water were used if the bowel was fairly clean, and 1-1.5 liters if fairly dirty, Dr. Leung said.
In the other group, air insufflation was performed as usual. Doctors used air during the withdrawal of colonoscopes from the patients in both groups. All procedures were performed using high-definition adult colonoscopes (Olympus 180-H series).
The overall adenoma-detection rate was 46% among the patients that were treated with air during insertion and 57% among patients treated with water, a significant difference (P = .04), according to Dr. Leung and his coauthor Dr. Francisco C. Ramirez of the University of Arizona, Phoenix.
Patients in whom colonoscopy detected at least one proximal adenoma were more likely to be in the water than the air group (46% vs. 35%, P = .03). This advantage held for adenomas less than 10 mm in size (42% vs. 31%, P = .04) but not for those 10 mm or larger (9% vs. 10%).
The mean time to the cecum was longer during colonoscopy initiated with water than with air, but only by 1.6 minutes (6.9 minutes vs. 5.3 minutes, P < .0001), Dr. Leung said.
Colonoscopy was accomplished in the water group with lower mean doses of fentanyl (68.8 mcg vs. 76.5 mcg, P = .0006) and midazolam (2.8 mg vs. 3.1 mg, P = .0007), and fewer patients in the water group received additional sedation (17.5% vs. 27%, P = .03).
The use of external pressure was also lower with water vs. air (12% vs. 28%, P < .0001).
Limitations of the study include a mostly male population (96%) and that a single endoscopist with a relatively high baseline adenoma-detection rate (46%) performed the procedures.
Good and excellent bowel preparation was reported in 79% of the 191 air-method patients and in 84% of the 177 water-method patients. Cecal intubation rates were similar at 100% in the air group and 92% in the water group. One adverse event was reported in each group.
Dr. Leung said the next randomized trial should compare the use of water for insertion and carbon dioxide for withdrawal – with polypectomy allowed at physician discretion – vs. a traditional air technique. He speculated that the water–carbon dioxide method would reduce pain during and after colonoscopy, salvage the procedure in patients with suboptimal bowel prep, and perhaps increase adenoma detection.
"This will impact post-colonoscopy pain-related [emergency room] visits and hospitalization, and because of the reduction in pain, perhaps colleagues will consider minimizing sedation and therefore decrease the overall patient burden," he said.
Also presented during the session were preliminary results from 313 patients in a randomized trial of unsedated colonoscopy comparing carbon dioxide insufflation during both colonoscope insertion and withdrawal , warm water infusion (98.6º F) during insertion and air insufflation during withdrawal, and standard air insufflation during both phases.
The proportion of patients requesting sedation or analgesia, the study’s primary end point, was 15.5% in the carbon dioxide group, 13.2% in the water-air group, and 25.6% in the all-air group, lead author Dr. Arnaldo Amato said. These differences were statistically significant for carbon dioxide vs. all air (P = .04) and for water-air vs. all-air (P = .03).
Median pain scores assessed by a visual analogue scale were 30 for the carbon dioxide group, 28 for the water-air group, and 46 for the all-air group. The differences were significant between the carbon dioxide and all-air groups and between the water-air and all-air groups (both at P less than .001).