1040 Quality Improvement Initiative to Improve Receipt of Timely Surveillance Endoscopy After Variceal Bleeding

Abstract

INTRODUCTION: Among patients with cirrhosis who develop esophageal variceal bleeding, surveillance endoscopy (EGD) with variceal ligation within 4 weeks of index bleed decreases the risk of re-bleeding. However, many of these patients do not receive timely surveillance EGD. Therefore, we aimed to assess the rates of timely surveillance EGD after an episode of variceal hemorrhage (VH) before and after a quality improvement (QI) initiative to improve provision of surveillance EGD. METHODS: We identified patients with cirrhosis hospitalized at the University of North Carolina Medical Center from 7/1/2017-6/30/2018 with variceal bleeding. We excluded all patients with a transjugular intrahepatic portosystemic shunt placement or death during hospitalization, discharge to hospice, or patient preference for local follow-up. Following the initial EGD, we determined if surveillance EGD was ordered and scheduled, and if scheduled, whether or not patients underwent timely surveillance EGD, defined as <4 weeks from the initial EGD. Based on these results, we designed a QI intervention to reserve two open scheduling slots dedicated to surveillance EGD after VH. Post-intervention surveillance EGD rates were then calculated for hospitalized patients with cirrhosis who underwent EGD for VH from 12/1/2018-4/15/2019. RESULTS: We identified 33 patients with cirrhosis hospitalized from 7/2017-6/2018 with VH. Of these patients, 5 (15%) patients underwent surveillance EGD within 4 weeks from initial EGD. Of the 28 patients who did not undergo timely surveillance, the most common reason was the EGD being scheduled for a date >4 weeks after the index bleed (75%). Other reasons included inability to contact the patient (14%), cancellations/no-shows (7%), and failure to place an EGD order (4%). In the post-intervention period, 18 patients had a VH and met other inclusion criteria. Of these patients, 8 (44%) underwent timely surveillance EGD. The reasons for failure to undergo timely surveillance EGD were cancellations (50%), EGD scheduled >4 weeks from initial bleed (40%), and failure to place an EGD order (10%). CONCLUSION: A QI intervention to reserve post-VH scheduling slots was associated with a 29% absolute increase in timely surveillance EGDs. Steps to further increase procedure capacity and improve both patient and scheduler education on the importance of surveillance EGDs within 4 weeks of VH may further increase rates of timely surveillance EGDs.