Outcomes after hepatic encephalopathy in population-based cohorts of patients with cirrhosis

Abstract

Background: Hepatic encephalopathy is a devastating complication of cirrhosis.

Aim: To describe the outcomes after developing hepatic encephalopathy among contemporary, aging patients.

Methods: We examined data for a 20% random sample of United States Medicare enrolees with cirrhosis and Part D prescription coverage from 2008 to 2014. Among 49 164 persons with hepatic encephalopathy, we evaluated the associations with transplant-free survival using Cox proportional hazard models with time-varying covariates (hazard ratios, HR) and incidence rate ratios (IRR) for healthcare utilisation measured in hospital-days and 30-day readmissions per person-year. We validated our findings in an external cohort of 2184 privately insured patients with complete laboratory values.

Results: Hepatic encephalopathy was associated with median survivals of 0.95 and 2.5 years for those ≥65 or <65 years old and 1.1 versus 3.9 years for those with and without ascites. Non-alcoholic fatty-liver disease posed the highest adjusted risk of death among aetiologies, HR 1.07 95% CI (1.02, 1.12). Both gastroenterology consultation and rifaximin utilisation were associated with lower mortality, respective adjusted-HR 0.73 95% CI (0.67, 0.80) and 0.40 95% CI (0.39, 0.42). Thirty-day readmissions were fewer for patients seen by gastroenterologists (0.71 95% CI [0.57-0.88]) and taking rifaximin (0.18 95% CI [0.08-0.40]). Lactulose alone was associated with fewer hospital-days, IRR 0.31 95% CI (0.30-0.32), than rifaximin alone, 0.49 95% CI (0.45-0.53), but the optimal therapy combination was lactulose/rifaximin, IRR 0.28 95% CI (0.27-0.30). These findings were validated in the privately insured cohort adjusting for model for endstage liver disease-sodium score and serum albumin.

Conclusions: Hepatic encephalopathy remains morbid and associated with poor outcomes among contemporary patients. Gastroenterology consultation and combination lactulose-rifaximin are both associated with improved outcomes. These data inform the development of care coordination efforts for subjects with cirrhosis.

Figure 1.. The Effect of Gastroenterology Consultation and Rifaximin Utilization On Mortality

The effect on mortality of gastroenterology consultation and rifaximin use are modeled for the subset of patients who were first diagnosed with hepatic encephalopathy as inpatients, in persons with 1 year of ‘washout’ after enrolling in hepatic encephalopathy without recorded hepatic encephalopathy, as well as after a landmark analysis to extinguish the risk of immortal-time bias and an analysis accounting for the competing-risk of liver transplantation (the effect estimate in this case is a sub-hazard distribution ratio). HE = hepatic encephalopathy

Figure 2:. The Effect of Gastroenterology Consultation and Rifaximin Utilization On Healthcare Utilization After Hepatic Encephalopathy

The effect on hospitalization associated with both gastroenterology consultation and rifaximin use. Gastroenterology consultation is associated with fewer 30-day readmissions but a higher number of hospital-days in the cohort after an hepatic encephalopathy-discharge as well the cohort with 1 year of ‘washout’. Rifaximin, by contrast is associated with fewer hospitalizations in each analysis. HE = hepatic encephalopathy