497 WHAT TREATMENT TO USE FOR SEVERE NON-VARICEAL UGI HEMORRHAGE: STANDARD ENDOSCOPIC HEMOSTASIS, DOPPLER GUIDED HEMOSTASIS, OR OVER-THE-SCOPE-CLIP (OTSC)?

Abstract

Non-variceal UGI hemorrhage (NVUGIH) is common but 26-28% of high risk patients with bleeding peptic ulcers (PUB’s) or Dieulafoy’s lesions (DL) rebled after standard, visually guided endoscopic hemostasis in 2 recently reported randomized controlled trials (RCT’s). See references in Table. These were separate comparisons of Doppler probe assisted hemostasis (DEP) or OTSC with STANDARD endoscopic hemostasis, but the 3 have never been compared in a head-to-head study before. Our purposes were to compare rates of recurrent bleeding and other 30 day clinical outcomes after endoscopic treatment with DEP, OTSC or STANDARD hemostasis of patients with severe NVUGIH. Individual patient data from 2 successive RCT’s were analyzed. Only patients with PUB’s & DL’s were included in 3 treatment groups: 69 DEP, 24 OTSC, & 101 STANDARD patients. We utilized SAS for data management & a biostatistician for the statistical analyses (JG). Patients were randomized if they had either major stigmata of hemorrhage – SRH (spurting arterial bleeding, non-bleeding visible vessel, or adherent clot) or lesser SRH (oozing bleeding or flat spot). In all treatments of spurting arterial bleeding or adherent clots, epinephrine injection was used prior to hemostasis with hemoclips (HC) &/or multipolar electrocoagulation (MPEC), or OTSC. DEP patients had visually guided hemostasis & if still positive for residual blood flow, more MPEC or HC’s was applied to obliterate underlying arterial blood flow, if feasible & safe. All patients received high dose proton pump infusions – PPI’s - X 72 hours & then BID PPI’s. Patients & their healthcare teams were blinded to endoscopic treatments & made all decisions after randomization about transfusions, further intervention for rebleeding, & level of care. Patients in the 3 treatment groups had similar baseline risk factors. However, there were significantly higher rates of 30 day rebleeding for STANDARD vs. DEP or OTSC (25.7% vs. 10.1% vs. 4.2%). The numbers needed to treat (NNT) for OTSC vs. STANDARD are 4.6 & for DEP vs. STANDARD are 6.4. Severe complications, number of RBC’s after randomization & number of bleeding episodes were also significantly higher in the STANDARD group – See Table & Kaplan-Meier Curves (p = 0.0141 for overall difference). For patients with severe UGIH from PUB’s or Dieulafoy’s lesions treated by the same investigators in 2 consecutive RCT’s: 1. STANDARD hemostasis resulted in significantly higher rates of further bleeding than DEP assisted hemostasis or OTSC. 2. Post-randomization rates of severe complications, units of RBC transfused, & number of bleeds were also significantly higher in the STANDARD treatment group. Either DEP assisted or OTSC are recommended as the new initial treatments of choice for severe NVUGIH from PUB’s or DL’s.Legend: Kaplan-Meier curves of time to first rebleed. Overall p = 0.0141 by log rank test.View Large Image Figure ViewerDownload Hi-res image Download (PPT)