Short- and long-term outcomes for patients with variceal haemorrhage in a tertiary hospital

To determine short- and long-term outcomes among a cohort of patients with variceal haemorrhage at a tertiary referral centre, and to determine the predictive value of the model for end-stage liver disease (MELD) score for mortality in these patients.

Prospective database hospital audit that captured patients who presented with or were transferred with variceal haemorrhage between 2004 and 2008, and a retrospective review of long-term outcomes. Patients who presented to or were transferred to John Hunter Hospital, a tertiary referral hospital, with confirmed variceal bleeding were included. The main outcome measures were in-hospital, 6 weeks and end-of-audit mortality. We also recorded cause, location and degree of planning surrounding the deaths in this patient group. We analysed the MELD score for patients with complete survival data.

We recorded 93 episodes of variceal haemorrhage from 78 unique patients during the initial study period. The in-hospital mortality, 6 weeks mortality and end-of-audit mortality were 2.6, 9.0 and 59, respectively, and median survival time was 3.2 years (95% confidence interval 0.0, 6.1). The most frequent cause of death was related to complications of end-stage liver disease at 74%, followed by variceal bleeding (19%) and unknown (6%). A Cox proportional hazard model showed that the risk of mortality is increased by 1.06 (1.01-1.11) for each unit increase in MELD score.

Short-term outcomes for patients with variceal bleeding continue to improve, but long-term prognosis remains guarded and should prompt further emphasis on advanced care planning to optimise patient care.