The Study Of Factors Affecting The Endoscopic Detection Rate Of Early Cancer And Precancerous Lesions In The Gastrointestinal Tract

BackgroundGastrointestinal malignancies account for 26%of the global cancer incidence and 35%of all cancer-related deaths.Gastrointestinal endoscopy(GIE)is an important tool for early diagnosis,but its detection rate of early cancers needs to be further improved.The early cancer(EC)detection rate is affected by a variety of factors,which are important to be explored and studied for further improvement of endoscopic quality.In terms of cancer risk,the risk of patients’ own disease,as opposed to endoscopy and endoscopist-related factors,is influencing the detection of EC and precancerous lesions(PL)by GIE.Therefore,it is effective to anchor patients at high risk of cancer to perform accurate screening.Not all patients undergoing gastrointestinal screening are at high risk for malignancy.An efficient way should be established for all patients participating in gastrointestinal endoscopy to be automatically assessed for cancer risk,which could then be systematically alerted to the endoscopists.Such a system can combine the patient’s risk factors without additional costs,resulting in efficient and accurate screening of early cancers.In terms of endoscopy type,multiple endoscopy types or examinations including imageenhanced endoscopy influence the detection rate.Sedated endoscopy is an anesthesia-based type of endoscopy that is widely used.However,the difference in the detection rate of EC and PL between unsedated endoscopy and sedated endoscopy has not been fully explored.Cohort studies are needed to investigate the detection rate and its causes,which can provide a reference and basis for future clinical decisions regarding the choice of endoscopy type.This study focuses on the improvement of EC and PL detection rate by GIE.We build a pre-endoscopy risk prediction system that may improve the quality of examination and proved its ability.The unclear detection rate of early cancer by sedated endoscopy was also studied to explore the feasible means for further improvement of GIE quality.Part 1 A pre-endoscopy risk prediction system and its evaluationObjective:Build a pre-endoscopy risk prediction system through data analysis and validated it in the population.Methods:Predictive models were established through modeling and internal validation.Patients from June 2019 to April 2021 at Qilu Hospital of Shandong University were prospectively included for external validation.Results:The modeling cohort included 18,354 gastroscopies and 10,062 colonoscopies.The area under the receiver operating characteristic curves(AUC)for the internal validation of the esophageal,gastric and colorectal risk prediction models were 0.796,0.847 and 0.737,respectively;Hosmer-Lemeshow test>0.05.The prospective cohort included 12,874 gastroscopic and 8,656 enteroscopic cases.The AUCs of externally validation of esophageal,gastric,and colorectal risk prediction models were 0.755,0.811,and 0.718,respectively;Hosmer-Lemeshow test P>0.05.Highgrade intraepithelial neoplasia(HGIN)in the esophagus,HGIN in the stomach,colorectal adenoma,advanced adenoma(AA),and non-advanced adenomas were detected at significantly higher rates(P<0.05).A higher proportion of early cancer and precancerous lesions in the gastric body was detected in the risk-indicating group(P<0.05).Proportion of 0-Ⅱa colorectal EC and adenomas increased in the risk prediction group(P<0.05).There were no other significant changes in lesion sites and Paris classification.The risk-indicating group had significantly longer gastroscopy time(P<0.001),colonoscopy withdrawal time(P<0.001)and significantly more biopsies taken(P<0.001).Conclusions:1.The pre-endoscopic EC risk prediction model suggested good diagnostic effectiveness,significantly increasing the detection rate of EC and PL in the esophagus and stomach,and EC and adenoma in the colorectum.2.The above enhancement may be due to longer examination time and increased number of biopsies.3.The model suggested a significant increase in EC and PL of the gastric body and colorectal EC and adenomas of 0-IIa type,with no significant change in other sites and Paris classification.Part 2 Study on sedated endoscopy and detection rate of EC and PL in the upper gastrointestinal tractObjective:To investigate the effect of sedated endoscopy on the detection rate of esophageal and gastric EC and PL,and to analyze the rationale.Methods:Gastroscopies at four endoscopy centers from January 2012 to July 2019 were included to create a propensity index score-matched(PSM)cohort and to compare endoscopic factors.Results:Gastroscopies of 432,202 cases were included.The detection rate of esophageal EC,esophageal HGIN,gastric EC,and gastric HGIN was higher in the sedated group(P<0.05).The percentage of EC and HGIN detected at gastric body in the sedated endoscopy group was higher(P<0.05);the percentage of esophageal type 0-Ⅱb lesions detected was elevated(P<0.05).In addition,there were no other significant differences in lesion sites and Paris classification.The examination time was significantly longer(P<0.001)and the number of biopsies was significantly higher in the sedated group(P<0.001).Iodine staining,electron staining,magnification,and other chemical staining were increased in sedated group(P<0.001).Conclusions:1.Sedated endoscopy increases the detection rate of EC and HGIN in the esophagus and stomach,with no significant effect on the detection of low grade intraepithelial neoplasia.2.Sedated endoscopy increases application of staining,image enhancement and magnification technique,prolongs the examination time and increases the number of biopsies thus increasing the detection rate.3.Sedated endoscopy increases the detection rate of esophageal EC and HGIN in type 0Ⅱb and EC and HGIN in the body of the stomach.Part 3 Study on sedated endoscopy and detection rate of colorectal adenomaObjective:To investigate the effect of sedated colonoscopy on the detection rate of colorectal adenoma,and to analyze the rationale.Methods:Colonoscopies at four endoscopy centers from January 2012 to July 2019 were included to create a propensity index score-matched(PSM)cohort and to compare endoscopic factors.Results:Colonoscopies of 216,400 cases were included.The detection rate of adenoma,including AA and nonadvanced adenoma,and polyp was increased in the sedated endoscopy group(P<0.05,P<0.05),especially flat adenomas(P<0.05)and adenomas of 0-5mm(P<0.05).There was no significant difference in lesion site.The number of biopsies was significantly higher in the sedated group(P<0.001).There was a significant increase in electronic(P<0.05)and chemical staining(P<0.001)in the sedated group.Conclusions:1.Sedated colonoscopy increased the detection rate of colorectal adenomas and polyps,especially flat adenomas and adenomas of 0-5mm.2.Sedated colonoscopy increases the frequency of staining and image enhancement techniques and increases the number of biopsies.