Statins Linked to Slower Liver Disease Progression

Statin use was linked to lower rates of progression to severe liver disease in patients with non-cirrhotic chronic liver disease (CLD) in a prospective national population-based study.

Among 3,862 CLD patients who took statins and were followed for approximately 14 years, risk of developing severe liver disease was 40% lower (HR 0.60, 95% CI 0.48-0.74) compared with matched patients who didn't take statins, researchers led by Rajani Sharma, MD, of Columbia University in New York City, reported in Clinical Gastroenterology and Hepatology.

Statins were also associated with lower rates of progression to cirrhosis (HR 0.62, 95% CI 0.49-0.78), hepatocellular carcinoma (HR 0.44, 95% CI 0.27-0.71), and liver-related mortality (HR 0.55, 95% CI 0.36-0.82), the study found.

"To our knowledge, this is the largest study to date, comprised of ~3800 individuals exposed to statins with liver histopathology data for each individual, allowing for more accurate CLD diagnoses, staging of liver disease, and exclusions of cirrhotics at CLD diagnosis," Sharma's group wrote.

The findings held and were statistically significant across alcohol-related liver disease and non-alcoholic fatty liver disease and in both pre-fibrosis and fibrosis stages of liver disease, although not in viral hepatitis and autoimmune hepatitis.

It is possible that potential anti-inflammatory, vascular, and tissue healing benefits of statins could play a role in preventing severe liver disease, but mechanistic studies would be needed to demonstrate that, the researchers noted.

Sharma's team analyzed data on CLD patients from the ESPRESSO study (Epidemiology Strengthened by Histopathology Reports in Sweden). That population-based study examines the etiology and prognosis of gastrointestinal disease in which histopathology plays a prominent role. Sharma's group obtained data on statin prescriptions from the Swedish Prescribed Drug Register.

The researchers matched statin and non-statin users 1:1 with direct matching, using factors such as age, gender, year of CLD diagnosis, type of CLD, and liver histology findings at diagnosis. They also used propensity score matching, which accounted for number of inpatient and outpatient healthcare visits, country of birth, level of education, presence of ischemic heart disease, and many other health-related factors.

The team used Cox regression to estimate hazard ratios for the primary outcome of incident severe liver disease, which was a composite of cirrhosis, hepatocellular carcinoma, and liver transplantation or liver-related mortality. Each of those outcomes was also analyzed separately.

Furthermore, the researchers conducted a time-varying sensitivity analysis. With statin treatment as a time-dependent exposure, they found the risk of incident severe liver disease remained lower in statin-users compared to statin non-users (HR 0.87, 95% CI 0.75-1.00). However, they noted the confidence interval in the time-varying analyses did not overlap with that in the main analyses (0.48-0.74), likely because "careful propensity score matching to address confounding was not possible in this sensitivity analysis and the sample size was smaller," they said.

Previous studies of statins and risk of severe liver disease in patients with CLD have yielded mixed results, Sharma's group noted. The current study overcame limitations of past research in that it included non-cirrhotic patients, had stricter matching criteria, and included liver histopathology data, they explained.

Nevertheless, the study also had important limitations, the authors noted. They did not have access to laboratory data, medication indications, or clinical notes, and therefore could not rule out some unmeasured confounding. They also could not rule out a healthy-user effect in that statin users are more likely to exercise, lose weight, and stop drinking. Finally, because the study required all patients to have a liver biopsy, it may have excluded individuals too sick to undergo a biopsy.

"In conclusion, this large prospective, nationwide population-based study with liver histopathology data demonstrates an inverse association between statin use in non-cirrhotic individuals with CLD and the development of severe liver disease," Sharma and colleagues said. "While, this study provides the most robust estimates available thus far, prospective randomized controlled trials are necessary in order to recommend statin use in clinical practice."

Comment