The third patient requiring emergency surgery

presented with haematemesis to one of our local District General FGFR inhibitor Hospitals. Although endoscopy confirmed a bleeding gastric ulcer, the haemorrhage could not be controlled endoscopically. The patient proceeded to theatre for laparotomy and a 3 cm ulcer high on the greater curvature was found with a central bleeding vessel. This was under-run and biopsies taken which confirmed adenocarcinoma. The patient made a good recovery and was referred to our centre for definitive oncological management. A total gastrectomy was performed six weeks following his initial presentation, the final histology was T1N0 adenocarcinoma, Ro 61-8048 in vivo 0/39 nodes. The patient survived for two years following this procedure. Emergency procedures after 24 hours The remaining 39 emergency patients were managed without operative intervention over the first 24 hours. Fifteen patients presented with haematemesis. Nine received endoscopic intervention (injection, Argon-beam laser, heater probe) for bleeding PSI-7977 concentration control. Four

patients were not actively bleeding at the time of endoscopy, and no further procedure was performed at this time. One patient had a large bleeding polyp removed at endoscopy, and three patients required injection of adrenaline to bleeding ulcerated areas. In one of these patients an endoclip was applied and argon plasma coagulation (APC) successfully performed. In only one case was endoscopic therapy not successful in controlling bleeding and this patient proceeded to theatre as described above. Overall 29 patients had some form of operation after complete

staging, often on separate admission. Patients presenting with gastric outlet obstruction were managed conservatively via nasogastric decompression in the initial period whilst further investigations were undertaken to stage their disease and plan further intervention. In 2 cases expanding Rolziracetam metal stents were inserted endoscopically allowing oral intake and palliative oncological therapies. Subsequently 3 out of 42 emergency patients (7.1%) and 44 out of 249 elective patients (17.6%) had neoadjuvant chemotherapy after their initial assessment (p < 0.05). Survival Overall survival Twelve patients from the elective group and three patients from the emergency were lost to follow-up. One year survival for patients presenting as an emergency was 48.3% compared to 63.4% in elective patients (p = <0.02). By 3 years follow-up there were only two survivors from the emergency presentation group (14.3%), while 32.5% of the elective patients survived to 3 years (p = <0.006). The overall survival is shown on the Kaplan Meier plot on Figure 2. Figure 2 Kaplan-Meier curve showing comparison of survival between patients presenting as an emergency and electively.