This article is more than 5 years old. Information may no longer be current.

Model accurately predicts cirrhosis-driven hepatic encephalopathy

A recently published study reports on the development of two risk-score models to predict the 1-year and 5-year probability for hepatic encephalopathy in patients with cirrhosis, based on statin use and total bilirubin and albumin.

Elliot B. Tapper

“We found that the risk of [hepatic encephalopathy] in patients with cirrhosis can be stratified by two readily available lab tests and a brief inventory of the medication list,” Elliot B. Tapper, MD, from the department of gastroenterology and hepatology, University of Michigan, and colleagues wrote. “The potential benefits of statins in preventing [hepatic encephalopathy] need to be studied in rigorously designed randomized controlled trials. This is particularly important for patients with cirrhosis for whom there is no effective treatment to eliminate or control the underlying cause.”

According to Tapper and colleagues, over 40% of patients with cirrhosis will develop hepatic encephalopathy and those who do have a 1-year overall mortality rate of 60% or higher.

The researchers enrolled 1,979 patients with cirrhosis from the Veterans Affairs Integrated Service Network who had been seen for any outpatient or inpatient visit between Jan. 1, 2005, and Dec. 31, 2010. Mean patient age was 58 years and most were men (98%) and white (74%).

During follow-up, 863 patients developed overt hepatic encephalopathy. Median time to development of hepatic encephalopathy was 340 days (range, 71-842 days) and median survival time was 747 days (range, 448-1,812 days).

Patients who did and did not develop hepatic encephalopathy had similar demographics, liver disease etiology and complications from cirrhosis. Those who did not develop hepatic encephalopathy had a significantly higher use of statins (19.2% vs. 13.3%; P = .0005) but lower use of non-selective beta-blockers (10.4% vs. 13%; HR = 1.27; 95% CI, 1.04-1.55).

After adjusted analysis, the researchers found that higher baseline bilirubin (aHR = 1.066; 95% CI, 1.048-1.088) and non-selective beta-blocker use (aHR = 1.268; 95% CI, 1.036-1.551), and lower albumin (aHR = 0.543; 95% CI, 0.493-0.597) and statin use (aHR = 0.74; 95% CI, 0.608-0.901) correlated with the development of hepatic encephalopathy. The researchers used these factors in the development of their risk-score models.

In the 5-year baseline risk-score model, patients had a low-risk (score of –10 or lower), intermediate-risk (score of –9 to 20) or high-risk score (score of 21 or higher). Similarly, the 1-year longitudinal model increased from a score of zero, to an intermediate score of 1 to 20, to a high score of 21 or higher. The median baseline risk score for those who developed hepatic encephalopathy were 8 (range, –23 to 62) and the longitudinal risk score was 9 (range, –17 to 59).

The area under the receiver operating characteristic curves for the baseline model was 0.68 (95% CI, 0.66-0.7) and for the longitudinal model it was 0.73 (95% CI, 0.71-0.75).

“This data provides clinicians caring for patients with cirrhosis a simple way to risk-stratify their patients at greater risk of developing hepatic encephalopathy, which frequently results in more hospital admissions, falls, and decreased survival,” Jessica L. Mellinger, MD, of the department of Internal Medicine-Hepatology at the University of Michigan, told Healio.com/Hepatology. “At a minimum, these high-risk patients should be offered hepatitis C therapy, if present, aggressive alcohol cessation treatment, and nutrition support.” – by Talitha Bennett

Disclosure: The authors report no relevant financial disclosures.