Hsieh YH, Koo M, Leung FW. A patient-blinded randomized, controlled trial comparing air insufflation, water immersion, and water exchange during minimally sedated colonoscopy. Am J Gastroenterol 2014;109:1390–1400. Even in the hands of an experienced endoscopist, insertion of the colonoscope through the colon can be technically challenging owing to factors such as body habitus, diverticular disease or previous pelvic surgery (Am J Gastroenterol 2000;95:2784–2787; Gastrointest Endosc 2001;54:558–562). Occasionally, the examination may prove too difficult to allow cecal intubation. Difficulty in scope insertion may also manifest in longer procedure times, increased patient discomfort, increased sedation medication requirement, and attendant maneuvers such as abdominal pressure or change. Water-assisted colonoscopy may facilitate the passage of the colonoscope to the cecum, especially in challenging cases (Gastrointest Endosc 2013;77:767–773). Unlike air, water may open the lumen without fully distending the lumen and creating tight angles. The water method may ease the passage of the colonoscope by straightening angulated sections, such as the sigmoid colon, and weighing the colon down and lubricating the interface between the scope and the mucosa (Gastrointest Endosc 2012;76:812–817). In both water immersion and water exchange, water is infused into the lumen while the air channel is shut off. Water is suctioned out of the lumen during insertion in the water exchange method and removed during the withdrawal phase in the immersion technique. Potential benefits of water-assisted colonoscopy include higher cecal intubation rates, lower sedation requirements, and lower patient pain scores, especially in unsedated examinations (Dig Endosc 2013;25:231–240). There are limitations in published studies examining water-assisted examinations. Many of the previous studies were performed in older men, thus limiting generalizability of the results (J Intervent Gastroenterol 2011;1:172–176). In addition, limited data suggest that water exchange may be associated with a less painful examination than water immersion (Gastrointest Endosc 2012;76:657–666), but there few studies comparing the 2 methods with conventional air. Hsieh et al randomized subjects to 1 of 3 insertion methods—air insufflation, water exchange, and water immersion—and compared the proportion of patients reporting no pain during the insertion phase of colonoscopy (Am J Gastroenterol 2014;109:1390–1400). Subjects were blinded to the insertion method. Pain experienced by the patient during insertion was the main outcome and was recorded every 2–3 minutes at the discretion of the nurse. The pain score ranged from 0 (no pain) to 10 (most severe pain). Other outcomes included insertion and withdrawal times, abdominal pressure, change in patient position, and cecal intubation rate. Abdominal pressure and change in position were used liberally to avoid excessive looping. At discharge, the subject reported the amount of abdominal pain and bloating as well as satisfaction and willingness to repeat the examination. The trial enrolled 270 individuals, and approximately one-third of these subjects were women (n = 99). The percentage of subjects experiencing no pain during insertion was higher in water exchange (61.1%) compared with the air insufflation (30%; P = .0001) or water immersion (43.3%; P < .0249). The proportion of subjects with painless insertion in the water immersion was greater than that for the air insufflation arm, but the difference was not significant (P = .0885). The median (minimum to maximum) maximum pain score was the lowest in the water exchange group (0; 0-10) compared with water immersion (1.8; 0-8.5) or air insufflation (3.5; 0-10; P < .001). Does the study provide convincing data to suggest that water exchange is associated with painless insertion more often than air insufflation or water immersion during unsedated colonoscopy? Overall, the methods employed by the investigators are quite sound. The study was designed with adequate power based on previous data generated by the same group. The trial also enrolled a large percentage of women, thus increasing the generalizability of the findings. In addition, because pain can be subjective, the authors took several important measures to decrease the chance of bias in pain reporting. Patient pain was recorded every 2–3 minutes at the nurse’s discretion, minimizing the possibility that the endoscopist could influence the results. In addition, the patients were blinded to the method of insertion technique. Another strength of the study is the collection of secondary outcome data, which help to substantiate the main findings. Because pain was presumed to be owing to looping, the authors collected data on maneuvers that can minimize this problem, such as abdominal pressure and change in patient position. There was no difference with regard to abdominal pressure, perhaps owing to the liberal use of this maneuver to avoid excessive looping. However, there was a higher rate of change in patient position in the air as compared with the water-assisted groups. Finally, the authors collected the length of the colonoscope inside the patient when the tip was at the cecum. Not surprisingly, the median length in the water exchange group (75 cm) was shorter than that observed for the water immersion (80 cm) or the air insufflation (85 cm) groups. These data support the premise that the water exchange method might be associated with less looping and thus less pain. One of the authors’ stated concerns was that pain experienced during sedation might reduce patient satisfaction as well as willingness of the patient to repeat the procedure. Despite the lower pain scores in the water exchange group, there were no differences among the 3 arms with respect to satisfaction or willingness to repeat the examination. However, 1 important outcome, cecal intubation rate, was higher in the water assisted groups as compared with the air insufflation group. Furthermore, water exchange salvaged most (14/19) of the cecal intubation failures. These data suggest that water exchange may be a useful method for optimizing cecal intubation rates in unsedated patients. One interesting finding was the high adenoma detection rate (ADR) observed in the water exchange group as compared with the other 2 groups. There was a statistically significant difference in ADR between water exchange subject and the other 2 methods for the cecum and the ascending colon. The main difference overall among the 3 groups was the number of adenomas detected during insertion. One reason for this difference may be the longer insertion time in the water exchange group. There was also a greater volume of water infused in the water exchange group, which likely explained the higher quality bowel preparation rating in this group. The longer insertion time and higher bowel preparation quality may explain the higher ADR. These issues may limit conclusions regarding the potential for increasing ADR using the water exchange technique. There are some limitations that should be acknowledged when examining the potential impact of this study. The enrolled subjects were willing to undergo a colonoscopy with minimal sedation, thus limiting generalizability in general endoscopy practice. The study also utilized a single endoscopist who performed all of the procedures. In addition, the nurse and endoscopist were not blinded to the method of insertion. Did these factors have an impact on the findings? The extra insertion time in the water exchange group may have been used for adequate suctioning of infused water. However, the time may have allowed the endoscopist to pass the colonoscope with less force, thus possibly influencing the pain experienced by the patient. The authors state that, because the pain scores in the air insufflation arm were lower than those observed previously, the likelihood that the endoscopist may have influenced the trial’s results seems low. Despite this reassuring observation, another trial employing more endoscopists is needed to substantiate the current study’s findings. The authors should be lauded for conducting a much-needed trial comparing 3 commonly used insertion techniques in unsedated colonoscopies. Their trial demonstrated lower insertional pain associated with water exchange. However, this lower amount of reported pain the in the water exchange group did not translate into higher patient satisfaction scores or proportion of subjects willing to undergo a repeat examination. The extra insertion time may be of concern for busy endoscopists owing to a possible decrease in endoscopy efficiency. However, an important outcome—cecal intubation rate—was higher when using the water exchange method. This potential benefit might useful for endoscopists when salvaging incomplete examinations. The choice of whether to use air insufflation or water-assisted colonoscopy ultimately depends on the endoscopist’s preference. I use water assistance to facilitate scope insertion in sedated and unsedated patients who have a low body mass index or diverticular disease. The current study used water exchange as a salvage technique. I also find this application of the water method useful in patients with redundant colons. When I encounter significant looping in the right colon, I withdraw the scope back to the hepatic flexure. After suctioning any air from the lumen, I reinsert the colonoscope using only water to distend the lumen and complete the examination. However, my observations are based only on anecdotal experience and more data are needed from randomized, controlled studies in sedated and unsedated patients. In addition, a comparison with CO2 may be more salient owing to data suggesting superiority over air insufflation (Endoscopy 2012;44:128–136). ReplyGastroenterologyVol. 148Issue 5PreviewWe appreciate the comprehensive summary and thoughtful commentary provided by Dr Anderson. He highlights several important issues that warrant additional comment. Earlier studies on colonoscopy using air insufflation found that longer insertion time was associated with more discomfort (Dis Colon Rectum 2005;48:1295–1300; Eur J Gastroenterol Hepatol 2007;19:695–699), implying that the duration of insertion time per se is not a reliable predictor of less discomfort. To minimize the insertion pain in the 3 current study groups, all possible measures—including gentle and slow insertion, loop reduction, patient position change, and abdominal compression—were used liberally. Full-Text PDF