High rate of re-bleeding after application of Hemospray for upper and lower gastrointestinal bleeds.

ABSTRACT

BACKGROUND & AIMS: Hemospray (TC-325, Cook Medical) has recently been approved for use in GI bleeding. Specific clinical indications and predictors of success or failure have not been well delineated. METHODS: We conducted a retrospective cohort study of Hemospray use at a tertiary center. We assessed demographics and characteristics of Hemospray use. We analyzed outcomes of hemostasis, rebleeding, need for embolization or surgery, and death. RESULTS: 86 applications of Hemospray were identified. The most common etiology of upper GI bleeds were ulcers (67.1%) whilst the etiology of lower GI bleeds varied. Hemospray was applied as monotherapy in 28 procedures (32.6%). Immediate hemostasis rate was 88.4%, but there was a high rate of re-bleeding (33.7%). Most re-bleeds occurred within 7 days (86.2%). Syncope was an independent predictive factor re-bleeding at 7 days for EGD (OR = 12.16, 95% CI = 1.51-97.75, P = 0.019). Bleeding refractory to endoscopic treatment with hemospray required radiological embolization in 9 instances, and surgery in 9 instances. Hemospray therapy was protective against need for embolization (p < 0.05). 2 patients underwent liver transplantation and there was a total of 5 deaths. Hepatic disease was an independent predictor of death (OR = 47.15, 95% CI = 2.42-916.89, P = 0.011). CONCLUSION: Hemospray is effective in achieving immediate hemostasis but is plagued by high rates of rebleeding. Syncope is a predictor of rebleeding, and hepatic disease is a predictor of death in patients undergoing Hemospray therapy. Despite high rates of embolization and surgery, Hemospray may reduce need for embolization. Hemospray use during endoscopy should prompt physicians to consider early re-look endoscopy and more aggressive therapy.