PTU-019 Outcomes from mesenteric angiography and embolisation in non-variceal upper gi bleeding; a single centre experience

Introduction Upper GI bleeding remains an important cause of morbidity and mortality. Mortality rates in non-variceal upper GI bleeding have remained relatively static over recent decades despite an ever-increasing range of therapeutic options. Interventional radiology (IR) has become an increasingly available tool for management of upper GI bleeding when endoscopic haemostasis has failed. However, literature is lacking surrounding the technical success and long term outcomes of mesenteric embolisation in patients with non-variceal upper GI haemorrhage. We therefore wished to assess the overall technical efficacy and outcomes of interventional radiology in patients presenting with upper GI haemorrhage who had undergone initial endoscopy at the Royal Infirmary of Edinburgh. Methods We retrospectively analysed the interventional radiology database for all patients who had undergone embolisation procedures. We then focussed on patients who had presented with non-variceal haemorrhage and assessed their outcomes using computer-based records. Patients were followed up for a minimum of 1 year. Results Data were available from 2007 onward. We assessed patient’s mortality outcomes at 30 days and 1 year. In total, 24 patients had undergone mesenteric embolisation for non-variceal upper GI haemorrhage (15 female, 9 males). Median age was 72 (range 52–96). Over half of patients (14 of 24, 58.3%) had an ASA grade of III or IV (figure 1). 19 of 24 had information available to calculate Glasgow-Blatchford score, with a median score of 15 (figure 2). Mean length of hospital stay in survivors was 31.75 days (5–148). 23 lesions were located in the duodenum, 1 in the oesophagus. There were 2 Dieulafoy lesions, 1 tumour vessel, 1 unclear bleeding point and 20 ulcers with varying stigmata. 5 patients had already undergone surgical management of their bleeding lesion prior to IR. IR was technically successful in 22 of 24 patients. 4 out of 24 patients rebled following embolisation. No patients developed an acute kidney injury following angiography. 6 out of 24 (25%) of patients died within 30 days of their IR procedure (figure 3). 8 out of 24 (33%) died within 1 year. 3 of these were due to bleeding, 3 due to sepsis and 2 due to malignancies (figure 4). Conclusions Mesenteric embolisation in patients with significant non-variceal upper GI bleeding has high technical success rates with low rebleeding rates, in a patient population that often is elderly with significant comorbidity. Approximately one third of patients who undergo interventional radiology procedures for non-variceal upper GI haemorrhage will be dead at 1 year; the majority from non-bleeding related causes.