P39 An analysis of rebleeding rates for variceal haemorrhage at a regional centre: what is the applicability and potential cost for early TIPS?

Abstract

Introduction: A recent randomised controlled trial by Garcia- Pagan et al demonstrated that the early use of TIPS in patients with Child-Pugh class B and C cirrhosis presenting with acute variceal haemorrhage was associated with a reduction in rebleeding and mortality. It remains unclear whether an additional economic benefit exists with their approach compared to the current standard of care using pharmacological and endoscopic therapy, and rescue TIPS. We looked at clinical data from our unit to ascertain how many patients would benefit from early TIPS and the economic implications of introducing this into practice. Methods: Consecutive patients admitted in 2009 with oesophageal variceal haemorrhage to a tertiary care liver unit at Nottingham University Hospitals (NUH) NHS Trust were identified retrospectively using a dedicated endoscopy database, cross-checked with the emergency medicine database. Standard management protocols including endoscopic therapy within 24 hours, glypressin and prophylactic antibiotics were used. Data was collected on aetiology, rebleeding related admissions and mortality at 12 months. Rebleeding costs were analysed for all patients meeting inclusion criteria for the original study and included subsequent inpatient care costs and endoscopic/radiological intervention (figures were supplied by the NUH finance and procurement department and based on known national tariffs). The actual cost of rebleeding in our Child Pugh score 7-13 patients was compared to the theoretical cost of introducing early TIPS in this group. Results: 51 cirrhotic patients were admitted to our unit with oesophageal variceal bleeding. 20% of this cohort had Childs A, 40% Childs B and 40% Childs C. The rebleeding rate was 15% at 28 days and 34% at one year follow up. Survival rates were 82% at 28 days and 40% at one year. 35 patients (70% of the cohort) had a Child Pugh score of 7-13. In this subgroup there was a 31% rebleeding rate requiring hospital admission over 12 months and 8% required a TIPS procedure within 12 months. The actual cost of rebleeding episodes for the selected subgroup was £138,446, (£3,955 per patient). Theoretical costs of early TIPS in this group were calculated as £117,670, (£3,362 per patient). Assuming a rebleeding rate of 3% with early TIPS, this strategy has a potential cost reduction of 7%. Conclusions: The proportion of variceal bleed patients benefitting from early TIPS could approach 70% in regional centres. This has implications for the provision and organisation of interventional radiology services. Our retrospective analysis suggests marginal cost benefit and prospective studies are needed to confirm this.