Effect Of Water Exchange Colonoscopy On Adenoma Detection Rate And Factors To Predict Pain During Unsedated Colonoscopy

Background:Colon adenomas are closely related to colorectal cancer,and it has been showed that lower adenoma detection rate is linked to increased risk of interval cancer and related deaths.Colonoscopy is one of the best tool for primary investigation of colorectal diseases,and the efficacy of colonoscopy for preventing CRC by virtue of the detection and removal of colorectal adenomas.However,there are around 30%adenomas missed during colonoscopies.Moreover,approximately 20%of colonoscopy is assumed to be painful procedures,which cause a lower chance of completing the examination and further increase the risk of missed adenomas.Water exchange(WE)colonoscopy reproducibly reduces insertion pain,enhances insertion rate and boston bowel preparation quality scale(BBPS).Whether WE could increase adenoma detection rate(ADR),the key quality indicator,is unknown.In addition,deep sedation with propofol has been increasingly used in colonoscopy in recent years.However,whether WE method can improve the ADR in propofol sedated patients remains currently inconclusive.Objective:1.To investigate the effect of WE on ADR;2.To investigate whethere WE can improves ADR among propofol-sedated patients;3.To investigate the possible risk factors associated with painful colonoscopy and develop a model of pre-stage risk factors to enable targeted offer an appropriate method of colonoscopy or with the aid of special techniques to help the completion of procedure.Methods:1.Effects of WE on ADR:In this prospective,randomized controlled trial at 6 centers in China,screening,surveillance and diagnostic patients were randomized to be examined by WE or traditional air insufflation(AI)method.The primary outcome was ADR.Secondary outcome included:Polyps detection rate(PDR),intubation rate,maximum pain score and quality of bowel preparation.Intention-to–treat(ITT)analysis was used to assess primary outcome and colonoscopic findings from all evaluable patients.To assess factors associated with ADR,logistic regression model was conducted.2.Effects of WE on ADR in patients receiving Propofol sedation:To optimize sample size,data of 2 previously published RCT were pooled to assess the impact of WE on ADR in propofol-sedated patients who received split-dose bowel preparation.Multiple logistic regression analysis was used to identify factors associated with higher ADR.3.Factors associated with pain during scope insertion:A cross-sectional study was performed.Patients'information were prospectively collected.The related factors of the pain during colonoscopy,such as age,gender,grade of education,previous surgery,anxiety levels.Multivariate stepwise regression analysis was used to identify high-risk factors associated with painful colonoscopy and a predicting model with the intubation discomfort score(IDS)was developed in training cohort,then validated in the validation cohort.Outcomes:1.From 4/2014 to 7/2015,3303 patients were randomized:WE(n=1653)and AI(n=1650).The baseline characteristics were comparable.Overall ADR was 18.3%(WE)and 13.4%(AI)(relative risk 1.45,95%confidential interval:1.20-1.75,p<0.001).Detection rate of advanced and sessile serrated adenoma was 6.2%(WE)and 4.5%(AI)(relative risk 1.16,95%confidential interval:1.02-1.32,p=0.046).ADR in screening patients aged>50 years old was 29.4%(WE)and 22.9%(AI)(relative risk 1.09,95%confidential interval:1.00-1.19,p=0.040).BBPS is higher in WE group,but maximum insertion pain score,number of times patient position change and need abdominal compression was lower in WE group.2.Propofol-sedated patients(n=510)were randomized to AI and WE.The baseline characteristics were comparable.WE had fewer patients with inadequate Boston Bowel Preparation Scale(BBPS)score of<6.Despite a significantly shorter inspection time,WE had significantly higher overall ADR than AI,especially in those with adequate BBPS of?6.Right colon ADR(17.5%vs.10.5%),flat ADR(32.3%vs.19.4%),combined advanced and sessile serrated ADR(13.1%vs.7.4%)of WE were significantly higher than those of AI.Multiple logistic regression analyses show that age,withdrawal time,indications(screening vs.diagnostic)and WE were significantly and independently associated with higher ADR.3.From 5/2017 to 10/2017,618 patients were included in training cohort.Lower BMI,constipation,anticipating pain was moderate or severe were identified as independent factors for painful colonoscopy and were used to develop the IDS.The results were externally confirmed in the validation cohort.Conclusion:1.WE colonoscopy can significantly improve ADR and this result is equally applicable in propofol-sedated procedure.2.Patients with BMI<18.5kg/m~2,constipation,and anticipating pain was moderate or severe were identified as independent factors for painful colonoscopy.For such patients,WE colonscopy is a better choice.