Water Exchange Is the Least Painful Colonoscope Insertion Technique and Increases Completion of Unsedated Colonoscopy.

Background & Aims: Unsedated colonoscopy is acceptable for diagnostic, surveillance, and screening indications worldwide. However, insertion of the colonoscope can be painful; it is not clear which technique is least painful and thereby increases the likelihood of colonoscopy completion. We performed a head-to-head comparison of air insufflation (AI), carbon dioxide (CO2) insufflation, water immersion (WI), and water exchange (WE) to determine which combination of insertion techniques produces the least amount of pain.

Methods: In a patient-blinded prospective trial, 624 subjects were assigned randomly to groups that underwent colonoscopy with AI-AI, CO2-CO2, WI-AI, WE-AI, WI-CO2, or WE-CO2 insertion and withdrawal techniques, including on-demand sedation, at the St. Barbara Hospital (Iglesias, Italy) or the Vìtkovice Hospital (Ostrava, Czech Republic), from October 2013 through June 2014. The primary outcome was real-time maximum insertion pain (0 = none, 10 = worst), recorded by an unblinded nurse assistant. At discharge, a blinded observer recorded the recalled maximum insertion pain and patients' and investigators' guesses about method or gas used.

Results: Patients and investigators correctly guessed the method used for fewer than 44% of procedures, confirming adequate blinding. The correlation between real-time and recalled maximum insertion pain (r = 0.9; P < .0005) confirmed internal validation of the primary outcome. The WE group had the lowest scores: mean pain values were 5.2 for AI-AI (95% confidence interval [CI], 4.6-5.8), 4.9 for CO2-CO2 (95% CI, 4.3-5.4), 4.3 for WI-CO2 (95% CI, 3.8-4.9), 4.0 for WI-AI (95% CI, 3.5-4.5), 3.1 for WE-CO2 (95% CI, 2.7-3.4), and 3.1 for WE-AI (95% CI, 2.7-3.6) (P < .0005). The highest proportions of patients completing unsedated colonoscopy were in the WE groups. WE groups also had significantly better colon cleanliness, particularly in the transverse and right colon (P < .0005). One limitation of the study was that colonoscopists and assistants were not blinded to water-aided insertion methods.

Conclusions: In a prospective study of colonoscopy insertion methods, CO2 insufflation did not reduce real-time maximum insertion pain. Compared with AI or CO2, WI and WE reduced insertion pain. The least painful technique was WE, which significantly increased completion of unsedated colonoscopy and bowel cleanliness without prolonging insertion time. ClinicalTrials.gov number: NCT01954862.