Declining Mortality of Cirrhotic Variceal Bleeding Requiring Admission to Intensive Care: A Binational Cohort Study

Abstract

Objectives: We aimed to describe changes over time in admissions and outcomes, including length of stay, discharge destinations, and mortality of cirrhotic patients admitted to the ICU for variceal bleeding, and to compare it to the outcomes of those with other causes of ICU admissions.

Design: Retrospective analysis of data captured prospectively in the Australian and New Zealand Intensive Care Society Centre for Outcome and Resource Evaluation Adult Patient Database.

Settings: One hundred eighty-three ICUs in Australia and New Zealand.

Patients: Consecutive admissions to these ICUs for upper gastrointestinal bleeding related to varices in patients with cirrhosis between January 1, 2005, and December 31, 2016.

Interventions: None.

Measurements and main results: ICU admissions for variceal bleeding in cirrhotic patients accounted for 4,003 (0.6%) of all 720,425 nonelective ICU admissions. The proportion of ICU admissions for variceal bleeding fell significantly from 0.8% (83/42,567) in 2005 to 0.4% (53/80,388) in 2016 (p < 0.001). Hospital mortality rate was significantly higher within admissions for variceal bleeding compared with nonelective ICU admissions (20.0% vs 15.7%; p < 0.0001), but decreased significantly over time, from 24.6% in 2005 to 15.8% in 2016 (annual decline odds ratio, 0.93; 95% CI, 0.90-0.96). There was no difference in the reduction in mortality from variceal bleeding over time between liver transplant and nontransplant centers (p = 0.26).

Conclusions: Admission rate to ICU and mortality of cirrhotic patients with variceal bleeding has declined significantly over time compared with other causes of ICU admissions with the outcomes comparable between liver transplant and nontransplant centers.