Author information
1
Corporal Michael J. Crescenz VA Medical Center, Philadelphia PA; Division of Gastroenterology and Hepatology, Hospital of the University of Pennsylvania, Philadelphia, PA. Electronic address: dakaplan@pennmedicine.upenn.edu.
2
Corporal Michael J. Crescenz VA Medical Center, Philadelphia PA; Division of Gastroenterology and Hepatology, Hospital of the University of Pennsylvania, Philadelphia, PA.
3
VA Connecticut Healthcare System, West Haven, CT.
4
San Francisco VA Medical Center, San Francisco, CA.
5
Hunter Holmes McGuire VA Medical Center, Richmond, VA.
6
VA New York Harbor Health Care System, Brooklyn, NY.
7
Boston VA Healthcare System, Boston, MA.
8
James J. Peters VA Medical Center, Bronx, NY.
9
Minneapolis VA Medical Center, Minneapolis, MN.
Abstract
BACKGROUND & AIMS:
Concerns related to hepatotoxicity frequently lead to discontinuation or non-initiation of HMG-coA reductase therapy in patients with cirrhosis despite data supporting statin use. We aimed to investigate the independent effects of hyperlipidemia and statin exposure on mortality, hepatic decompensation, and hepatocellular carcinoma development in a large national cohort of patients with cirrhosis.
METHODS:
We performed a retrospective cohort study of patients with newly diagnosed cirrhosis from January 1, 2008 through June 30, 2016 in the Veterans Health administration. Subjects were divided into 2 cohorts: 21,921 patients with prior statin exposure (existing users) and 51,023 statin-naïve individuals, of whom 8,794 subsequently initiated statin therapy (new initiators) and 44,269 did not (non-initiators). Multivariable Cox proportional hazard models with inverse probability weighting were constructed to assess the effects of time-updating lipid profiles and cumulative exposure to statins on survival and hepatic decompensation. Statin-naïve new initiators were propensity matched with non-initiators to simulate a randomized controlled trial of statin use in cirrhosis.
RESULTS:
In statin-naïve subjects, every 10 mg/dl increase in baseline total cholesterol was associated with a 3.6% reduction in mortality. In existing users, each year of continued statin exposure was associated with a hazard ratio of 0.920 (95% CI, 0.0.897-0.943) for mortality. After risk-set matching, each year of statin exposure among new initiators was associated with a hazard ratio of 0.913 (95% CI, 0.890-0.937) for mortality.