P1066 - What Is the Endoscopy Yield in Abnormal Hepatobiliary Imaging and Factors Predicting the Yield? A Retrospective Single Center Study

Mitchell J. Eyerman, MD¹, Anthony R. Khoury, DO², Sanjeevani Kumar Tomar, MD², Brian M. Quigley, PhD³, Sultan Mahmood, MD⁴, Muhammad Tahir, MD^4
1University Hospitals Case Medical Center, Case Western Reserve University, Cleveland, OH; ²University at Buffalo, Buffalo, NY; ³Jacobs School of Medicine and Biomedical Sciences, Buffalo, NY; ⁴University at Buffalo Gastroenterology, Buffalo, NY

Introduction: The decision whether to proceed directly to endoscopy for patients with abnormal imaging of hepatobiliary symptoms can often be challenging. Endoscopists must consider a wide range of factors including patient history, symptoms and imaging findings. We performed this retrospective review to evaluate the yield of endoscopy in the presence of abnormal imaging of hepatobiliary system, and determining the patient and imaging related risk factors predicting the yield.

Methods: We included 108 patients (55 male, 53 female) who underwent endoscopic procedure (EGD, ERCP & EUS) for abnormal hepatobiliary imaging. Percent yield for endoscopy was compared with a dependent samples t-test; independent samples t-tests and ANOVAs examined group differences in yield between those with/without symptoms and between imaging modalities. Percent malignant findings were examined with follow-up independent t-tests examining differences in malignancy findings among individual pairs of imaging modalities and pairs of imaging outcome categories. Effect sizes (Cohen's d) and 95% confidence intervals were calculated for all comparisons.

Results: Endoscopic yield was not significantly higher for patients who were ever symptomatic (83.5%) than for patients who were always asymptomatic (68.8%) patients (Image 1a). Endoscopic yield was significantly higher for patients who were symptomatic (86.1%) at the time of endoscopy than asymptomatic (65.4%) at time of endoscopy (Image 1b). Number of days between imaging and date of endoscopy did not impact endoscopic yield. There were also significant differences in % of malignancy found based on the type of imaging findings. A finding of mass was significantly more likely to detect a malignancy than a finding of inflammatory changes or dilated bile duct (Table 1).

Discussion: Our analysis demonstrated that the endoscopy should not be delayed for the symptomatic patients and the abnormal imaging in the presence of these findings should be taken seriously. Future studies and artificial intelligence may explore other factors that may impact yield in abnormal imaging.




Disclosures:


Mitchell J. Eyerman, MD¹, Anthony R. Khoury, DO², Sanjeevani Kumar Tomar, MD², Brian M. Quigley, PhD³, Sultan Mahmood, MD⁴, Muhammad Tahir, MD⁴. P1066 - What Is the Endoscopy Yield in Abnormal Hepatobiliary Imaging and Factors Predicting the Yield? A Retrospective Single Center Study, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.