Endocrinology, Metabolism, and Geriatrics, Department of Internal Medicine, University of Modena and Reggio Emilia, Modena, Italy
Operating Unit Internal Medicine and Metabolism, Endocrinology, Metabolism, and Geriatrics, Department of Internal Medicine, University of Modena and Reggio Emilia, Nuovo Ospedale Civile Estense di Baggiovara, Modena 41100, Italy
Potential conflict of interest: Nothing to report.
A growing number of chronic liver disease patients, especially those with metabolic syndrome–associated nonalcoholic fatty liver disease or hepatitis C virus–associated dysmetabolic syndrome, will take statins to prevent cardiovascular disease. As a result, clinicians will weigh complex issues raised by the interaction of statins with liver metabolism in these disorders. In this article, we critically review data concerning statins and liver pathophysiology with an emphasis on nonalcoholic fatty liver disease and hepatitis C virus, while also touching on other chronic liver diseases. Basic research interests include statins' mechanism of action and their effects on cholesterol-related cell signaling pathways and angiogenesis. From the clinical standpoint, many chronic liver diseases increase cardiovascular risk and would undeniably benefit from sustained statin use. The false alarms and security accompanying aminotransferase monitoring, however, are disturbing in light of the scarcity of data on statins' long-term effects on liver histology. Although some actions of statins might eventually prove to be particularly useful in nonalcoholic steatohepatitis, hepatitis C virus, or hepatocellular carcinoma, others may prove harmful. The lack of definitive data makes a fully informed decision impossible. Research using histological endpoints is urgently needed to determine the indications and contraindications of this extraordinary class of agents in patients with chronic liver disease. (HEPATOLOGY 2008;48:662–669.)
Cholesterol metabolism is a major physiologic function of the liver, and inhibition of this central pathway by 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, that is, statins, has been accompanied by concerns of hepatotoxicity. Additionally, cardiovascular risk is likely increased in nonalcoholic fatty liver disease (NAFLD), hepatitis C virus (HCV) infection, and primary biliary cirrhosis (PBC).1–3 Therefore, selected populations of chronic liver disease patients may benefit from statins.
Continued concern for the rare complication of statin-induced acute liver failure, the uncertain impact of statins on fibrosis progression, and the clear limitations of aminotransferases in monitoring complicate statin therapy.4, 5 Chronic aminotransferase elevation and histological injury in patients using statins have never been connected convincingly, and this devalues aminotransferase monitoring in this setting. Moreover, reference ranges in hyperlipidemic and healthy patients differ.6 Recent studies have shown no significant difference in incidence of aminotransferase elevation in hyperlipidemic patients receiving moderately dosed statins7 or in patients with known liver disease receiving maximally dosed pravastatin8 (Table 1). No previous review of statins has focused on incorporating both basic and clinical research aspects regarding statins and liver physiology to evaluate the safety and potential therapeutic applications of statins in chronic liver disease. We aim to examine the effects of statins on the liver with a focus on HCV and NAFLD, the two most prevalent chronic liver conditions, while also addressing relevant aspects of this interaction in other liver diseases.
Table 1. Incidence of Aminotransferase Elevation with Statin Use for Cardiovascular Disease
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; NS, not significant.
Direct and Indirect Effects of Statins
Statins are undoubtedly the most important class of cholesterol-lowering agents.9 Through competitive inhibition of HMG-CoA reductase, the rate-limiting enzyme of cholesterol synthesis, statins up-regulate hepatic low-density lipoprotein (LDL) receptors and reduce proatherogenic circulating LDL cholesterol (Fig. 1). In addition to their cholesterol-lowering properties, statins exert cholesterol-independent, pleiotropic effects that are not mediated by HMG-CoA reductase inhibition, including improvement of endothelial dysfunction, increased nitric oxide bioavailability, antioxidant effects, and anti-inflammatory and immunomodulatory properties.10 Especially relevant to patients with NAFLD are studies demonstrating an enhanced cardiovascular response in patients with increased C-reactive protein levels,11 which may indicate more severe NAFLD.12 Accumulating evidence from rat models13, 14 and a single report for humans15 suggest that statins may decrease portal hypertension through nitric oxide pathways.
Statins' activity is primarily dependent on the lipophilicity of ortho-substituents and meta-substituents on the aryl/biphenyl moiety.16 Specific differences among statins have recently been reviewed.17 Simvastatin, lovastatin, fluvastatin, and atorvastatin are metabolized by cytochrome P450 systems, whereas pravastatin, rosuvastatin, and pitavastatin undergo minimal hepatic metabolism. A recent meta-analysis indicated that statins in larger doses and with lower degrees of lipophilicity are associated with a higher incidence of aminotransferase elevations18 (Table 2). Clinical trial data suggest that standard doses of pravastatin, simvastatin, and atorvastatin show no significant differences in reducing cardiovascular risk.19
Table 2. Lipophilicity of Statins and Effects on Incidence of Aminotransferase Elevations
95% Confidence Interval
Statins vary in their degree of lipophilicity, which appears to have an impact on their likelihood of being associated with aminotransferase elevations. Dale et al.'s meta-analysis18 demonstrated that less lipophilic statins result in an increased relative risk of aminotransferase elevation. An aminotransferase elevation episode was defined as an alanine aminotransferase or aspartate aminotransferase level greater than 3 times the upper limit of normal. Overall, additional work has suggested that the intensity of cholesterol lowering and anti–hepatitis C virus in vitro activity appear to be highest in statins with the lowest lipophilicity.
Highly lipophilic agents included cerivastatin, lovastatin, and simvastatin.
Mildly lipophilic agents included pravastatin, rosuvastatin, atorvastatin, and fluvastatin.
The mechanisms of some pleiotropic effects include statin-induced changes in intermediates of cholesterol metabolism, which lead to altered phosphorylation of membrane-associated proinflammatory proteins, reduced release of proinflammatory cytokines, and, consequently, suppression of innate immune function.20 A reduction in oxidized LDL has also been observed in statin-treated patients,21 and this is especially relevant to nonalcoholic steatohepatitis (NASH), in which oxidized LDL levels are increased.22 However, this property appears to be inconsistent: atorvastatin and pravastatin reduce lipid peroxidation for up to 3 months after initiation but have little additional effect with continued use.23
ALT, alanine aminotransferase; AST, aspartate aminotransferase; BA, bile acid; CoA, coenzyme A; CT, computed tomography; FFA, free fatty acid; FPP, farnesyl pyrophosphate; FXR, farnesoid X receptor; GGPP, geranylgeranyl pyrophosphate; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HMG-CoA, 3-hydroxy-3-methylglutaryl coenzyme A; LDL, low-density lipoprotein; MAP-K, mitogen-activated protein kinase; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; NF-κB, nuclear factor-kappa beta; ns or NS, not significant; PBC, primary biliary cirrhosis; PPAR, peroxisome proliferator-activated receptor; PXR, pregnane X receptor; RhoA, Ras homolog gene family member A; s, statistically significant; sHh, sonic hedgehog; SREBP, sterol regulatory element binding protein; TNF-α, tumor necrosis factor-alpha; US, ultrasound; VEGF, vascular endothelial growth factor; VLDL, very low density lipoprotein.
Statins' Effects on Cholesterol-Related Cell Signaling Pathways
Because bile acid (BA) metabolism is intimately related to cholesterol metabolism, inhibition of cholesterol synthesis by statins not surprisingly produces changes in this complex pathway. For example, inhibition of cholesterol/oxysterol synthesis by statins influences the activity of liver X receptor (LXR) and farnesoid X receptor (FXR), transcription factors that function as nuclear cholesterol sensors and key modulators of lipid metabolism, BA synthesis, and inflammatory signalling.24, 25 Statin treatment leads to activation of sterol regulatory element binding proteins (SREBPs), which increase the transcription of enzymes involved with lipid synthesis and increase the number of LDL receptors. This in turn decreases the concentration of hydroxyl sterols and increases LDL clearance. Statins' effects in vivo, however, may be different than in cell culture systems, and this may partially explain divergent findings of decreased steatosis with statin use in animals and humans.26, 27 In HepG2 cells, statins appear to inhibit biosynthesis of sterols and increase triglycerides in opposing fashion, both of which SREBP-2 mediates.28 These data should be interpreted cautiously; however, as HepG2 cells may possess defective mechanisms for mobilizing triglycerides for lipoprotein export. Pravastatin and colestimide, a BA binding resin, led to the activation of hepatic SREBP-2 in rats with a subsequent reduction in hepatic cholesterol and increased plasma high-density lipoprotein.29 Modulation of additional transcription factors with statins, including carbohydrate responsive element binding protein or forkhead box 01, may also have an impact on hepatic fatty acid synthesis and thus steatosis.30
Speculation concerning BA involvement in hepatic steatosis is derived from studies demonstrating reduced gene expression of components of the hepatic BA metabolism and transport pathways in an acute fatty liver mouse model, with effects related to the blockade of long-chain fatty acid beta-oxidation.31 In humans, the effects of statins on BA synthesis are inconclusive. Although early statin administration reduces BA synthesis,32 long-term administration produces mixed effects.33 Moreover, concurrent therapy with statins and ursodeoxycholic acid results in further reduced BA synthesis.34 These data suggest that the net result of statin treatment on BA synthesis might vary between statins with different combinations of statins and concurrent agents or with various underlying pathological conditions.
These complexities are especially relevant because BAs help to regulate body weight, energy homeostasis, insulin resistance, and lipid metabolism via obscure FXR-dependent35 and FXR-independent endocrine pathways.36 Animal studies point to these effects being critically dependent on the BA pool size and composition,37 which, in turn, may influence the hepatic stem cell niche, which is integral to the liver's response to injury.38 Despite the theoretical impact of statins on BA metabolism and related effects, the interaction between statins and BA synthesis in NAFLD is relatively undefined.
For patients with chronic liver disease such as NAFLD, there are further interactions that may have significance but lack adequate investigation (Fig. 2). For example, alteration of cholesterol metabolism likely has an impact on the sonic hedgehog (sHh) pathway, a sterol-dependent pathway that influences body composition, fibrosis, and repair pathways. Early work in rodents showed that direct inhibition of hedgehog proteins can decrease diet-induced weight gain39 and that low cholesterol levels induced by statins inhibit cholesterol-modified hedgehog activation of the patched-1 receptor. Moreover, this work identified sHh as an autocrine viability factor for myofibroblastic hepatic stellate cells and suggests a potential role for sHh signaling in the pathogenesis of cirrhosis.40
Statins and NAFLD
NAFLD and hyperlipidemia frequently coexist because of both conditions' association with metabolic syndrome.41 Although likely influenced by ethnic variation, hypercholesterolemia and hypertriglyceridemia are independent predictors of steatosis.42 In a population-based study, magnetic resonance spectroscopy detected steatosis in 60% of patients with mixed hyperlipidemia and in 83% of those with both mixed hyperlipidemia and abnormal alanine aminotransferase.43 NAFLD has been repeatedly linked with the entire spectrum of vascular disease, including endothelial dysfunction, intima-media thickening, and atherosclerotic plaque formation.1, 44, 45 Although the standardized mortality ratio in type 2 diabetics is higher for cirrhosis than for cardiovascular disease46 and the coexistence of diabetes in patients with NAFLD more than doubles the prevalence of cirrhosis on biopsy from 10% to 25%,47 the higher mortality of NASH patients is largely a result of cardiovascular disease.48 Thus, statins are likely to play an increasing role in the overall management of patients with NAFLD.
Despite their potential significance, the cellular effects of statins in NAFLD have not been elucidated. For instance, there are potential interactions with fundamental factors involved with fat metabolism such as the peroxisome proliferator-activated receptors (PPARs).49 In cell cultures, multiple statins have induced the activation of PPARγ in macrophages and monocytes via cyclooxygenase 2–mediated interactions with both mitogen-activated protein kinase and extracellular signal-regulated kinase inflammatory pathways.50 In rats, statins influence gene activation of PPARα, and this suggests a role in promoting fatty acid oxidation.51 Moreover, by reducing membrane cholesterol, statins enhance plasma membrane fluidity.52 Statins' net effect on hepatocyte membranes and associated receptor activation, however, remains unclear. Molecular studies examining statins' effects on insulin sensitivity pathways and epidemiological studies of the risk of developing diabetes with statin exposure have shown inconsistent results.53, 54
There are numerous other potential molecular targets by which statins may exert a more specific effect on disease activity in NAFLD. For instance, decreased serum levels of tumor necrosis factor-alpha, interleukin-6, and possibly C-reactive protein, associated with statin therapy, may be beneficial because elevated levels of these markers are associated with advanced histology in NASH.55, 56 Statins may also indirectly exert a positive effect on steatosis through reduced free fatty acid delivery to the liver57 or on insulin signaling through altered adiponectin metabolism.58 Several animal studies have further investigated statins' effects on apoptosis pathways. In cultured rat hepatocytes, pravastatin induced more hepatocyte apoptosis,59 possibly by causing a mitochondrial permeability transition, an effect favoring apoptosis.60 However, simvastatin and lovastatin led to reduced apoptosis by weakening the p53 tumor suppressor gene's response to DNA damage in rat hepatocytes.61 Statins' effects on BA metabolism may explain some of the variation in experimental results. BAs bind to numerous transcription factors, including FXR, which modulates lipid and glucose metabolism, and the pregnane X receptor, which is uniquely expressed on human stellate cells.35, 62 Variations in the expression of these transcription factors may explain divergent statin-induced stellate cell expression in rodents versus humans.63, 64 Dietary composition may also affect these pathways as both fatty acids and intermediates of BA synthesis bind to these receptors.65
Despite the potential positive and negative consequences of statin use in NAFLD, only six pilot studies66–71 have evaluated the potential therapeutic effects of statins in a limited number of patients, and histological endpoints were examined in only three analyses (Table 3). Short-term results have been promising, with several studies showing histological improvement in inflammation without changes in fibrosis. The absence of a clear warning signal of injury from these pilot studies provides guarded optimism.
Table 3. Pilot Studies of Statin Therapy for NAFLD or NASH
Two recent clinical studies contribute additional support to the safe use of statins in NAFLD patients. Lewis et al.'s8 controlled, prospective study of maximum-dose pravastatin therapy in a variety of clinically diagnosed chronic liver diseases, including 64% with NAFLD, showed equal effectiveness in lowering LDL cholesterol, with no difference in aminotransferase elevation incidence. Histological endpoints were not addressed. The retrospective study by Ekstedt et al.,72 although small and without data on dosages or target LDL levels, provides detailed histology-based data that provide reassurance as well as grounds for concern. Histological outcomes from 17 NAFLD patients who received statins for up to 16 years were compared to those from 51 NAFLD patients without statin exposure. After a comparable period of follow-up, no change in fibrosis score was seen in 11 of 17 (64%) statin-treated patients versus only 18 of 51 (37%) non–statin-treated patients. Concerning advanced fibrosis, however, 5 of 17 (29%) statin-treated patients had bridging fibrosis or cirrhosis at the end of observation versus only 6 of 51 (12%) non–statin-treated patients, despite similar baseline levels of the fibrosis stage between groups (Table 4). One could speculate from these results that there may be a subgroup with severe lipotoxicity and progressive fibrosis despite therapeutic measures such as statins or, alternatively, that there is a subgroup at risk for progressive fibrosis as a result of long-term statin therapy via an undetermined mechanism.
Table 4. Long-Term Histological Outcomes of Nonalcoholic Fatty Liver Disease Patients With and Without Statin Therapy
Statin (n = 17)
No Statin (n = 51)
Ekstedt et al.'s case-control study72 demonstrated that nonalcoholic fatty liver disease patients who received statins as treatment for cardiovascular disease or hyperlipidemia had lower rates of fibrosis progression than those without statin exposure. A greater proportion of patients in the group who received statins, however, developed advanced fibrosis over the long-term follow-up period. These findings raise two possibilities: there is a subset of nonalcoholic steatohepatitis patients in whom statins may accelerate fibrosis, or there is a subgroup in which the inflammation and fibrosis due to the underlying steatohepatitis will progress despite therapeutic intervention.
Mean % steatosis difference between baseline biopsy and final biopsy
Progressive ballooning (%)
Baseline fibrosis stage (stage 0/1/2/3/4)
Final fibrosis stage(stage 0/1/2/3/4)
Fibrosis progressed (%)
Fibrosis unchanged (%)
Fibrosis regressed (%)
Stage 3–4 fibrosis at baseline biopsy (%)
Stage 3–4 fibrosis at final biopsy (%)
Statins and HCV Infection
Compared to NAFLD, HCV infection is less frequently associated with metabolic syndrome. However, a newly described entity named HCV-associated dysmetabolic syndrome features the triad of insulin resistance, hypocholesterolemia, and steatosis, all three of which may have an impact on the natural history of HCV.73 For example, HCV infection predicts atherosclerosis and may be, in theory, an indication for statin therapy.2 Paradoxically, HCV-infected patients often have hypocholesterolemia, which is inversely proportional to the extent of steatosis74 and fully reversible, along with steatosis, after viral eradication.75 On the other hand, those HCV-positive individuals having higher serum LDL and total cholesterol pretreatment levels may actually have a better response to antiviral treatment.76 To understand these clinical findings, it is worth recalling that HCV RNA replication takes place on lipid rafts77 and that geranylgeranyl, a cholesterol intermediate, is an essential component of HCV replication in vitro.78 Inhibition of cholesterol synthesis by statins may therefore inhibit HCV replication. This question has been assessed in vitro in an HCV RNA replication system. Fluvastatin had the strongest anti-HCV activity, atorvastatin and simvastatin showed moderate reductions in HCV viral loads, and lovastatin exhibited the weakest anti-HCV activity.79 Conversely, statin therapy is associated with an up-regulation in LDL receptors,80 and because HCV may enter hepatocytes via interactions with LDL receptors,81 statins may actually enable HCV infection.82 This feature of HCV infection may counterbalance the antiviral effects observed in vitro. For example, a small pilot trial evaluated 10 HCV-infected hypercholesterolemic patients who received moderately dosed atorvastatin therapy with initiation of antiviral therapy. There was no statistically significant change in viral load at weeks 4 and 12 of therapy.83 As in NAFLD, statins have equivalent efficacy in lipid lowering and similar rates of aminotransferases elevation in HCV infection in comparison with patients without known liver disease.8
Statins in PBC and Other Liver Diseases
Statins are safe and effective in reducing serum cholesterol levels in PBC, but they modulate cholestasis only in those patients who respond to ursodeoxycholic acid administration.84–86 Current studies are inadequately powered to determine cardiovascular risk in PBC patients; however, 12% of PBC patients in one report died from cardiovascular causes, and this suggests that statins would likely benefit some patients.3 Publications regarding other liver diseases and statins mainly involve anecdotal evidence from case reports. Further study is necessary to better establish statins' safety profile and indications in PBC and other liver diseases.
Statins and Hepatocellular Carcinoma
Statins have been theorized to affect angiogenesis, tumor cell growth, apoptosis, and the metastatic potential of a variety of tumor types. Although controversial, statins' antiangiogenic effects vary by agent and dosage.87 Low-dosed statins tend in general to increase the proangiogenic effects of vascular endothelial growth factor (VEGF), whereas high doses inhibit VEGF-related angiogenesis.88, 89 Statins influence angiogenesis via effects on the production of protein Akt and B kinases, Ras, Ras homolog gene family member A (RhoA), and interleukin-18.90, 91 Statins also decrease geranylgeranyl pyrophosphate (PP) production, which prevents RhoA localization to the cell membrane, leading to decreased endothelial cell mobility and reduced angiogenesis.92 Conversely, statins may be proangiogenic through effects on the Akt–endothelial nitric oxide synthase–caveolin pathway, nitric oxide production, stimulation of VEGF production, and increased endothelial nitric oxide synthase activity by decreasing caveolin.93, 94 Several studies have evaluated statin use and hepatocellular carcinoma (HCC) in humans and rodent models. Although results are conflicting, current evidence favors a net anti-HCC effect of statins.87, 95 A recent abstract showed that diabetic patients with no known coexisting liver disease were 40% less likely to be diagnosed with HCC if they had received at least one statin prescription.96 Although limited in scope, this report is encouraging because of the emerging relationship between HCC and steatosis.
A growing number of chronic liver disease patients, especially those with metabolic syndrome–associated NAFLD or HCV dysmetabolic syndrome, will use statins to combat cardiovascular disease. As a result, clinicians will commonly face patients with progressive NASH or HCV in the setting of indications for long-term statin therapy. On the one hand, many patients with known liver disease, especially those with NAFLD or HCV who have increased cardiovascular risk, will undeniably benefit from sustained statin use. On the other hand, the inability to predict effects on fibrosis and the imprecision of aminotransferases monitoring are disturbing. Uncertainty regarding statins' interaction with potentially abnormal cell signaling pathways and the scarcity of data on statins' long-term effects on liver histology warrant further consideration. In the final analysis, research using histological endpoints is critical for determining the range of indications and contraindications of statins in patients with chronic liver disease.