Mo1579 Duration of Endoscopic Examination Significantly Impacts Detection Rates of Neoplastic Lesions During Diagnostic Upper Endoscopy

Abstract

Mo1579 Duration of Endoscopic Examination Significantly Impacts Detection Rates of Neoplastic Lesions During Diagnostic Upper Endoscopy Jun Liang Teh, Mikael Hartman, Linus Lau, Jin-Rong Tan, Annie Wong, Jun-Jie Ng, Nakul Saxena, Asim Shabbir, Jimmy B. So Surgery, National University Health System, Singapore, Singapore; National University of Singapore, Singapore, Singapore Background: Upper endoscopy is the gold standard for detection of pathological lesions in the upper gastrointestinal tract. However, there is poor standardization of upper endoscopic examination worldwide. We hypothesize that the quality of upper endoscopy affects the detection of neoplastic lesions. Objectives: To audit the quality control of upper endoscopy in our institution and identify factors that predict the detection of pathology during upper endoscopy. Methods: In this interim report of 492 patients of a projected 1000 total study population, we audited all upper endoscopies performed in a university hospital over period of one month. Study variables included patients’ demographics, indications and cancer history, endoscopists’ experience and specialty, duration of endoscopy and number of images captured during endoscopy. Histological diagnoses of helicobacter pylori, intestinal metaplasia, dysplasia and cancer as well as endoscopic diagnoses of positive urease test, gastric ulcers, polyps and cancers were considered as positive findings. Logistic regression analysis was used to determine the odds ratios of detecting a positive histological or endoscopic finding. Results: In this preliminary analysis of 492 consecutive upper endoscopies, the duration of endoscopy varied from less than a minute to 60 minutes (mean; 29.2 mins). The mean number of images captured was 9 images per endoscopy with a range of 2 to 20 images. Multivariate logistic regression analysis showed that patients aged 60 years (odds ratio 5.0; 95% CI, 1.4 to 17.2; P 0.01) and males (odds ratio 3.3; 95% CI, 1.6 to 6.9; P 0.001) were more likely to be diagnosed with a positive histological finding. Patients undergoing an upper endoscopy examination for 15 20 minutes were also more likely to be diagnosed with a positive histologic finding (odds ratio 9.0; 95% Cl 1.4 to 58.3; P 0.02). The multivariate logistic regression analysis also showed an increasing trend in OR (for detecting both histological and endoscopic end points) with increasing number of pictures taken but this finding was not statistically significant. Conclusions: There was a wide variation in the practice of upper endoscopic examination in our institution. Our preliminary study suggests that longer examination time and more pictures taken may improve the detection of lesions. There is a need for standardization of examination to improve the quality of upper endoscopy. In order to strengthen our findings, a larger number of patients will be needed for analysis.