PTH-054A Hemospray use in the management of upper gastrointestinal haemorrhage: a 2-year experience across 2 teaching hospitals in the north and south of england

Introduction Hemospray (Cook Medical) is a haemostatic powder licensed for the management of non-variceal upper gastrointestinal bleeding (UGIB). Our aim was to describe the effectiveness of hemospray monotherapy vs adjunctive and salvage therapy, haemostasis and re-bleed rates. Method A retrospective use of hemospray use in UGIB at 2 teaching hospitals in the north and south of England was evaluated from June 2014-June 2016. Data was extracted from the endoscopy reporting system. Results A total of 44 patients (19 females and 25 males), median age 77.5 years (max 94 years and min age 25 years) with UGIB were treated with hemospray across the 2 centres over the 2 year period. Bleeding was secondary to peptic ulcer disease in 29/44 (65.9%), varices in 3 (6.8%), post biopsy bleeds in 3 (6.8%), unidentified lesion in 3 (6.8%), post sphincterotomy bleed in 2 (4.5%), mallory weiss tear in 2 (4.5%), portal hypertensive gastropathy in 1 (2.3%) and diffuse large b-cell lymphoma in 1 (2.3%). Forrest classification was calculated where information was available (34/44). Forrest Ia-9 cases. Forrest Ib-15 cases. Forrest IIa-8 cases. Forrest IIb-2 cases.Hemospray was utilised as a monotherapy in 10/44 (22.7%) of patients. It was used as adjunctive treatment in 25/44 cases (56.8%) and as salvage treatment in 9/44 (20.5%) of cases where previous treatment had failed.Haemostasis was achieved in 41/44 patients (93.2%). 1/3 patients without haemostasis had hemospray monotherapy and 2/3 patients had hemospray as salvage treatment. 2/3 patients without hemostasis who received hemospray salvage therapy went for interventional radiology. 1/3 patient without hemostasis with hemospray monotherapy application due to profuse bleeding was unstable from admission and palliated after the gastroscopy. 7/44 (15.9%) patients had a re-bleed. 4/10 (40%) had been treated with hemospray as monotherapy, 2/25 (8%) as adjunctive treatment and 1/9 (11.1%) as salvage therapy. Of the 7 re-bleeds, 3 were Forrest IIa classification, 3 were Forrest Ib classification and the 7th Forrest IIb classification. Haemostasis was achieved in 5/7 (71.4%) of the rebleeds. 2/7 re-bleeds did not achieve hemostasis and went for surgery. 4/7 re-bleeds achieved hemostasis with a combination of adrenaline, diathermy and clips. 1/7 re-bleeds did not receive treatment as there were no stigmata of recent haemorrhage. No adverse events occurred. Conclusion Hemospray as monotherapy appears to be associated with high rates of re-bleeding. It appears to be more effective as an adjunctive treatment or salvage therapy. Further larger studies are needed to establish where it can most effectively be used, but this data does not support its use as a primary monotherapy. Disclosure of Interest None Declared