Narrative Review CKD and Nonalcoholic Fatty Liver Disease Giovanni Targher, MD,1 Michel B. Chonchol, MD,2 and Christopher D. Byrne, MB BCh3,4 The possible link between nonalcoholic fatty liver disease and chronic kidney disease (CKD) recently has attracted considerable scientific interest. Accumulating clinical evidence indicates that the presence and severity of nonalcoholic fatty liver disease is associated significantly with CKD (defined as decreased estimated glomerular filtration rate and/or proteinuria) and that nonalcoholic fatty liver disease predicts the development and progression of CKD, independently of traditional cardiorenal risk factors. Experimental evidence also suggests that nonalcoholic fatty liver disease itself may exacerbate systemic and hepatic insulin resistance, cause atherogenic dyslipidemia, and release a variety of proinflammatory, procoagulant, pro-oxidant, and profibrogenic mediators that play important roles in the development and progression of CKD. However, despite the growing evidence linking nonalcoholic fatty liver disease with CKD, it has not been definitively established whether a causal association exists. The clinical implication for these findings is that patients with nonalcoholic fatty liver disease may benefit from more intensive surveillance or early treatment interventions to decrease the risk of CKD. In this review, we discuss the evidence linking nonalcoholic fatty liver disease with CKD and the putative mechanisms by which nonalcoholic fatty liver disease contributes to kidney damage. We also briefly discuss current treatment options for this increasingly prevalent disease that is likely to have an important future impact on the global burden of disease. Am J Kidney Dis. 64(4):638-652. ª 2014 by the National Kidney Foundation, Inc. INDEX WORDS: Nonalcoholic fatty liver disease; chronic kidney disease; metabolic syndrome; review; risk factors.

N

onalcoholic fatty liver disease has reached epidemic proportions and is the most common cause of chronic liver disease in developed countries.1-3 During the last decade, it has been shown that the global health burden of nonalcoholic fatty liver disease is confined not only to potentially progressive liver disease. Accumulating evidence indicates that nonalcoholic fatty liver disease may not only increase risk of developing type 2 diabetes mellitus and cardiovascular disease,3-5 but also may increase the risk of chronic kidney disease (CKD).6,7

DIAGNOSIS AND EPIDEMIOLOGY OF NONALCOHOLIC FATTY LIVER DISEASE Nonalcoholic fatty liver disease refers to a wide spectrum of liver damage, ranging from simple From the 1Section of Endocrinology, Diabetes and Metabolism, Department of Medicine, University and Azienda Ospedaliera Universitaria Integrata of Verona, Verona, Italy; 2Division of Renal Diseases and Hypertension, University of Colorado Denver, Aurora, CO; 3Nutrition and Metabolism, Faculty of Medicine, University of Southampton; and 4Southampton National Institute for Health Research Biomedical Research Centre, Southampton, United Kingdom. Received March 18, 2014. Accepted in revised form May 30, 2014. Originally published online July 30, 2014. Address correspondence to Giovanni Targher, MD, Division of Endocrinology, Diabetes and Metabolism, Department of Medicine, University and Azienda Ospedaliera Universitaria Integrata, Piazzale Stefani, 1, 37126 Verona, Italy. E-mail: [email protected]  2014 by the National Kidney Foundation, Inc. 0272-6386/$36.00 http://dx.doi.org/10.1053/j.ajkd.2014.05.019 638

steatosis to nonalcoholic steatohepatitis, advanced fibrosis, and cirrhosis. Nonalcoholic fatty liver disease is characterized histologically by lipid droplets and is defined by the accumulation of liver fat (. 5% per liver weight). The characteristic histology of nonalcoholic fatty liver disease resembles that of alcohol-induced liver injury, but occurs in people who do not drink or consume only a small amount of alcohol (ie, a threshold of , 20 g/d for women and , 30 g/d for men typically is adopted) and do not have other secondary causes of chronic liver disease, such as viral hepatitis, autoimmune hepatitis, hemochromatosis, or use of steatogenic medications.1-3 At present, liver biopsy remains the gold-standard investigation for diagnosing and assessing the severity of disease progression because histologic examination of liver tissue is the best technique for assessing the severity of key components of liver disease pathology in nonalcoholic fatty liver disease, for example, hepatic inflammation and fibrosis.1,3 However, there are problems with this diagnostic test and interpretation of the results, and liver biopsy is an imperfect gold standard. It is widely accepted that nonalcoholic fatty liver disease is a patchy liver disease that can produce sampling variability depending on the site of the biopsy and alter the diagnosis and fibrosis staging of the disease.1 Additionally, the rate of severe morbidity related to liver biopsy in large series ranges from 0.1%-0.2%, whereas the fatality rate ranges from 0%-0.2%. Am J Kidney Dis. 2014;64(4):638-652

Nonalcoholic Fatty Liver Disease and CKD

Other noninvasive tests are available that are able to assess liver fat only within the spectrum of nonalcoholic fatty liver disease. Ultrasonography remains the recommended first-line imaging modality for diagnosing hepatic lipid accumulation (steatosis) in clinical practice, which is noted when a hyperechogenic or bright appearance is seen on imaging. Ultrasonography provides a subjective and qualitative assessment of hepatic fat content and generally is believed to be of only limited sensitivity (60%-90%) if , 30% of hepatocytes are steatotic.1,3,4 A recent meta-analysis has shown that the overall sensitivity and specificity of ultrasonography for the detection of moderate to severe fatty liver compared to histology were 84.8% and 93.6%, respectively.8 By contrast, magnetic resonance imaging and proton magnetic resonance spectroscopy are more sensitive techniques than ultrasonography for detecting liver fat. Magnetic resonance imaging techniques (eg, assessing decreased liver signal intensity when out of phase compared with in phase) and magnetic resonance spectroscopy techniques (eg, using single-voxel magnetic resonance spectroscopy to measure the area under the lipid spectrum relative to the water spectrum) are the best available diagnostic techniques for assessing quantitatively the severity of liver fat accumulation. However, both techniques are resource intensive and cannot reliably discriminate simple steatosis from steatosis with inflammation and fibrosis (ie, nonalcoholic steatohepatitis).3,4 Current estimates are that approximately 30%40% of people with nonalcoholic fatty liver disease can develop nonalcoholic steatohepatitis.1-3 Nonalcoholic steatohepatitis often is progressive, with development of advanced fibrosis in 30%-40% of patients, cirrhosis in 15%-20%, and liver failure in 2%-4%.1-3,9 Nonalcoholic fatty liver disease as an indication for liver transplantation is on the rise,10 and it is projected to be the principal cause for liver transplantation within the next decade.1,3,11 In at least one large United Network for Organ Sharing database, nonalcoholic fatty liver disease was responsible for a higher rate of simultaneous liver-kidney transplantation than other indications.12 Patients with nonalcoholic fatty liver disease who receive a liver transplant alone may have worse kidney function after transplantation compared with other indications.13 Notably, growing evidence also indicates that cardiovascular disease dictates the outcome(s) in patients with nonalcoholic fatty liver disease more frequently and to a greater extent than does progression of liver disease,3-5,14 and that other serious extrahepatic complications, including CKD, may occur more frequently during the life course of patients with nonalcoholic fatty liver disease.6,7,14 Am J Kidney Dis. 2014;64(4):638-652

EPIDEMIOLOGIC EVIDENCE LINKING NONALCOHOLIC FATTY LIVER DISEASE TO CKD Increased Prevalence of CKD Patients with nonalcoholic fatty liver disease, both adults and adolescents, have multiple traditional and nontraditional risk factors for CKD (Fig 1).1-7,15-18 As summarized in Table 1, several large crosssectional population and hospital-based studies involving both adults without diabetes and patients with diabetes consistently have shown that the prevalence of CKD (defined as either decreased estimated glomerular filtration rate [eGFR] and/or abnormal albuminuria and/or overt proteinuria) is increased in people with nonalcoholic fatty liver disease.19-31 These studies have used either ultrasonography or biopsy to diagnose nonalcoholic fatty liver disease and have excluded patients with end-stage renal disease and those with secondary causes of chronic liver disease (alcohol abuse, viral hepatitis, and use of hepatotoxic drugs in all studies and also hemochromatosis and autoimmune hepatitis in some studies) or cirrhosis of any cause. In these studies, the prevalence of CKD in patients with nonalcoholic fatty liver disease ranged from approximately 20%-55% compared to 5%-35% in patients with non–nonalcoholic fatty liver disease. Importantly, most of these studies reported that nonalcoholic fatty liver disease was associated independently with the presence of CKD even after adjusting for traditional cardiorenal risk factors. Case-control studies that used liver biopsy (the gold standard) to diagnose nonalcoholic fatty liver disease have shown that patients with histologically defined nonalcoholic steatohepatitis have

Figure 1. Patients with nonalcoholic fatty liver disease exhibit the typical features of metabolic syndrome and have a myriad of other emerging risk factors and risk markers for chronic kidney disease. Abbreviations: CRP, C-reactive protein; IL-6, interleukin 6; PAI-1, plasminogen activator inhibitor 1; TGF, transforming growth factor; TNF, tumor necrosis factor. 639

640

Table 1. Principal Cross-sectional Studies of the Association Between Nonalcoholic Fatty Liver Disease and the Prevalence of CKD

Study Characteristics

Definition of Prevalent CKD or of Abnormal Kidney Function

Targher et al19 (2008)

Outpatient-based cohort of 2,103 pts with type 2 diabetes without CVD, cirrhosis, or viral hepatitis; overall prevalence of HTN 63%

Prevalent CKD defined as eGFR , 60 mL/min/1.73 m2 and/or overt proteinuria (macroalbuminuria)

Age, sex, BMI, waist circumference, HTN, alcohol consumption, diabetes duration, hemoglobin A1c, LDL cholesterol, triglycerides, smoking, medication use (hypoglycemic, antihypertensive, antiplatelet or lipid-lowering drugs)

NAFLD pts had significantly lower eGFRs (91 6 32 vs 96 6 31 mL/min/1.73 m2) and higher prevalence of microalbuminuria (20% vs 15%), macroalbuminuria (6% vs 3%), and CKD (15% vs 9%) than pts without NAFLD; NAFLD independently associated with increased risk of prevalent CKD (OR, 1.87; 95% CI, 1.3-4.1)

Targher et al20 (2010)

Outpatient sample of 202 adults with type 1 diabetes without secondary causes of chronic liver disease; overall HTN prevalence 35%

Prevalent CKD defined as eGFR , 60 mL/min/1.73 m2 and/or ACR $ 30 mg/g

Age, sex, BMI, systolic BP, alcohol consumption, diabetes duration, hemoglobin A1c, triglycerides, and medication use

NAFLD pts had significantly lower eGFRs (98 6 25 vs 107 6 28 mL/min/1.73 m2) and higher prevalence of abnormal albuminuria (36% vs 4.4%) and CKD (37.8% vs 10%) than pts without NAFLD; NAFLD independently associated with increased risk of prevalent CKD (OR, 3.29; 95% CI, 1.2-9.1)

Hwang et al21 (2010)

Health examination survey of 1,361 pts with impaired glucose tolerance or newly diagnosed diabetes (on oral glucose tolerance test) without CVD, malignancy, cirrhosis, or viral hepatitis; overall HTN prevalence 20%

Abnormal albuminuria defined as urinary ACR $ 30 mg/g; no eGFR measurement (ie, those with eGFRs , 60 mL/min/ 1.73 m2 were excluded from the study)

Age, sex, BMI, waist circumference, smoking, hemoglobin A1c, triglycerides, LDL cholesterol, liver enzymes, insulin resistance, metabolic syndrome

Prevalence of abnormal albuminuria significantly higher in those with than without NAFLD irrespective of glucose tolerance status (19% vs 6.3% in those with prediabetes and 33% vs 5% in those with newly diagnosed diabetes); NAFLD independently associated with increased risk of prevalent abnormal albuminuria (OR, 3.66; 95% CI, 1.3-10.2 in those with prediabetes; OR, 5.47; 95% CI, 1.01-29.6 in those with newly diagnosed diabetes)

Targher et al22 (2012)

Outpatient sample of 343 adults with type 1 diabetes without secondary causes of chronic liver disease; overall HTN prevalence 42%

Prevalent CKD defined as eGFR , 60 mL/min/1.73 m2 and/or ACR $ 30 mg/g

Age, sex, BMI, smoking, physical activity, diabetes duration, hemoglobin A1c, systolic BP, lipids, and use of antihypertensive and lipid-lowering drugs

NAFLD pts had significantly lower eGFRs (83 6 27 vs 93 6 29 mL/min/1.73 m2) and higher prevalence of abnormal albuminuria (48% vs 20%) and CKD (54.4% vs 24.2%) than pts without NAFLD; NAFLD independently associated with increased risk of prevalent CKD (OR, 1.93; 95% CI, 1.1-3.6); consistent results found for each component of kidney damage (abnormal albuminuria or decreased eGFR)

Study

Risk Factors Adjusted for in Analysis

Main Findings

Diagnosed by Ultrasonography

Targher, Chonchol, and Byrne

Am J Kidney Dis. 2014;64(4):638-652

(Continued)

Study

Study Characteristics

Definition of Prevalent CKD or of Abnormal Kidney Function

Risk Factors Adjusted for in Analysis

Main Findings

Li et al23 (2012)

Community-based sample of 1,412 adults without HTN, diabetes, or secondary causes of chronic liver diseases

Prevalent mild CKD defined as eGFR , 90 mL/min/1.73 m2 and/or overt proteinuria

Age, sex, alcohol intake, smoking, sleep quality, physical activity, BMI, systolic BP, diastolic BP, fasting glucose, serum total cholesterol, triglycerides, and aminotransferases

NAFLD independently associated with increased risk of mild CKD (OR, 2.31; 95% CI, 1.04-5.17)

Sirota et al24 (2012)

NHANES (National Health and Nutrition Examination Survey) 1988-1994; n 5 11,469 men and women; overall HTN prevalence 24%

Prevalent CKD defined as eGFR , 60 mL/min/1.73 m2 and/or ACR $ 30 mg/g

Age, sex, race, HTN, diabetes, waist circumference, HDL cholesterol, triglycerides, insulin resistance

Prevalence of CKD significantly higher in those with than without NAFLD (42.2% vs 34.5%); NAFLD significantly associated with increased risk of prevalent CKD after adjustment for age, race, and sex (OR, 1.21; 95% CI, 1.04-1.40) but not after further adjustment for insulin resistance and other metabolic syndrome features (OR, 1.04; 95% CI, 0.88-1.23)

Ahn et al25 (2013)

Health examination survey of 1,706 Korean adult men and women without viral hepatitis or excessive alcohol consumption; overall HTN prevalence 38%

Prevalent CKD defined as eGFR , 60 mL/min/1.73 m2 and/or overt proteinuria (dipstick $ 11)

Age, sex, smoking, abdominal obesity, HTN, diabetes, HDL cholesterol, triglycerides, GGT, and serum aminotransferases

NAFLD pts had comparable eGFRs (65 6 8 vs 66 6 8 mL/min/1.73 m2) but significantly higher CKD prevalence (31.7% vs 21.6%) than pts without NAFLD; NAFLD independently associated with increased risk of prevalent CKD (OR, 1.68; 95% CI, 1.3-2.2); consistent results found for those with and without HTN or diabetes

Mikolasevic et al26 (2013)

62 pts with moderate to severe CKD (53% CKD stage 4) without viral hepatitis or excessive alcohol consumption; overall HTN prevalence 100%

All patients had CKD stage 3 or 4 (mean eGFR: 32.8 6 14 mL/ min/1.73 m2)

Targher et al27 (2010)

Hospital-based sample of 80 pts with NASH and 80 age-, sex-, and BMI-matched controls without hepatic steatosis; overall HTN prevalence 27%

Prevalent CKD defined as eGFR , 60 mL/min/1.73 m2 and/or ACR $ 30 mg/g

Diagnosed by Elastography With Controlled Attenuation Parameter No adjustments made

Severity of hepatic steatosis, but not liver stiffness, significantly associated with decreasing eGFR (eGFR for steatosis scores 1, 2, and 3: 42 6 6 vs 37 6 5 vs 25 6 6 mL/min/1.73 m2, respectively)

Diagnosed by Biopsy Age, sex, BMI, waist circumference, smoking, systolic BP, triglycerides, insulin resistance

(Continued)

NAFLD pts had significantly lower eGFRs (75 6 12 vs 87 6 6 mL/min/1.73 m2) and higher prevalence of abnormal albuminuria (14% vs 2.5%) and CKD (25% vs 3.7%) than pts without NAFLD; NASH independently associated with increased risk of prevalent CKD (OR, 6.14; 95% CI, 1.6-12.8); eGFR and albuminuria decreased progressively in relation to severity of NASH histology

Nonalcoholic Fatty Liver Disease and CKD

Am J Kidney Dis. 2014;64(4):638-652

Table 1 (Cont’d). Principal Cross-sectional Studies of the Association Between Nonalcoholic Fatty Liver Disease and the Prevalence of CKD

641

642

Table 1 (Cont’d). Principal Cross-sectional Studies of the Association Between Nonalcoholic Fatty Liver Disease and the Prevalence of CKD

Study

Study Characteristics

Definition of Prevalent CKD or of Abnormal Kidney Function

Risk Factors Adjusted for in Analysis

Main Findings

Yilmaz et al28 (2010)

Hospital-based sample of 87 NAFLD pts; predominantly normotensive (mean BP, 126/ 77 mm Hg)

24-h urinary albumin excretion rate (16% of pts had microalbuminuria, 84% had normoalbuminuria); no eGFR measurement

Age, sex, BMI, systolic BP, triglycerides, HDL cholesterol, aminotransferases, insulin resistance

Increasing albumin excretion rate independently associated with severity of hepatic fibrosis

Yasui et al29 (2011)

Hospital-based sample of 174 NAFLD pts; overall HTN prevalence 30%

Prevalent CKD defined as eGFR , 60 mL/min/1.73 m2 and/or overt proteinuria (urinary dipstick $ 11)

Age, sex, BMI, HTN

NASH pts had comparable median eGFRs (82 vs 83 mL/min/1.73 m2) but significantly greater frequency of proteinuria (13% vs 6%) and CKD (21% vs 6%) than those with simple steatosis; association mediated primarily by HTN

Park et al30 (2011)

Hospital-based sample of pts with NASH (n 5 71) or other chronic liver diseases (n 5 472) consecutively referred for liver transplantation evaluation; overall HTN prevalence 35%

Serum creatinine level; no eGFR measurement

No adjustments were made

NASH pts had significantly higher serum creatinine (mean creatinine, 1.26 vs 0.98 mg/dL) than those with other chronic liver diseases

Machado et al31 (2012)

148 consecutive morbid obese pts with NAFLD underwent bariatric surgery (37 had NASH); overall HTN prevalence 67%

Prevalent CKD defined as eGFR , 60 mL/min/1.73 m2

Age, sex, HTN, diabetes, dyslipidemia

NASH pts had significantly lower eGFRs (97 6 22 vs 105.7 6 16 mL/min/1.73 m2) and higher proportion of CKD (8% vs 1%) than those with simple steatosis independent of potential confounders (OR, 9.7; 95% CI, 1.0-96.4); eGFR decreased progressively in relation to histologic severity of NASH

Targher, Chonchol, and Byrne

Am J Kidney Dis. 2014;64(4):638-652

Note: Insulin resistance was estimated by a homeostasis model assessment (HOMA); GFR was estimated by using the MDRD (Modification of Diet in Renal Disease) Study equation (except for the study by Li et al, in which GFR was estimated using both the MDRD Study and the CKD-EPI equations, and for the study by Machado et al, in which GFR was estimated by using the CKD-EPI study equation). Abbreviations: ACR, albumin-creatinine ratio; ALT, alanine aminotransferase; BMI, body mass index; BP, blood pressure; CI, confidence interval; CKD, chronic kidney disease; CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; GGT, gamma-glutamyltransferase; HDL, high-density lipoprotein; HTN, hypertension; LDL, low-density lipoprotein; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; OR, odds ratio; pts, patients.

Nonalcoholic Fatty Liver Disease and CKD

lower eGFRs and a greater prevalence of CKD or abnormal albuminuria than matched controls (Table 1).27-31 Additionally, these studies show a graded positive relationship between the histologic severity of nonalcoholic fatty liver disease and kidney disease independent of several risk factors for CKD, including metabolic syndrome features. However, as reported in Table 1, it is important to note that all these published studies have used the MDRD (Modification of Diet in Renal Disease) Study equation or other creatinine-based GFR estimating equations (which do not perform well in patients with obesity or cirrhosis), and that the adjustment for potential confounders has not always been complete. Further studies in larger cohorts of patients with biopsy-confirmed nonalcoholic fatty liver disease are needed to confirm these findings and better elucidate whether nonalcoholic fatty liver disease severity adversely affects the risk for the development and progression of CKD. Increased Incidence of CKD The relationship between nonalcoholic fatty liver disease and increased prevalence of CKD appears to be robust and has been replicated consistently across populations of different ethnicities; however, the specific contribution of nonalcoholic fatty liver disease per se to the development and progression of CKD is much more controversial. The question of whether nonalcoholic fatty liver disease is only a risk marker or a mediator (ie, pathogenic factor) of CKD has not yet been answered. It also is plausible to argue that nonalcoholic fatty liver disease does not cause CKD and that both nonalcoholic fatty liver disease and CKD are caused by obesity and other shared risk factors. Moreover, uncertainty exists about the prognostic role of nonalcoholic fatty liver disease in risk stratification for CKD. However, with those caveats, accumulating evidence suggests that CKD is a serious threat to patients with nonalcoholic fatty liver disease. As summarized in Table 2, 4 retrospective and prospective studies with reasonably long follow-up have assessed the relationship between nonalcoholic fatty liver disease, as diagnosed by ultrasonography, and the risk of developing incident CKD (defined as either decreased eGFR and/or abnormal albuminuria and/or overt proteinuria).32-35 In this review, we do not discuss the large population-based cohort studies that used serum liver enzyme levels (ie, surrogate markers of nonalcoholic fatty liver disease) to diagnose nonalcoholic fatty liver disease and that confirmed that mildly elevated serum liver enzyme levels predicted the development of CKD.36,37 It is known that serum liver enzyme levels are not sensitive for the diagnosis of nonalcoholic fatty liver disease and that the full histopathologic Am J Kidney Dis. 2014;64(4):638-652

spectrum of nonalcoholic fatty liver disease may be present in patients with normal liver enzyme levels.1,3 In all studies reported in Table 2, patients with overt cirrhosis or secondary causes of chronic liver diseases were excluded. In addition, in the 2 studies published by our group, no participant had ultrasonographic findings suggestive of cirrhosis.32,35 As specified in Table 2, in the Valpolicella Heart Diabetes Study, which included 1,760 individuals with type 2 diabetes with preserved kidney function (mean baseline eGFR, 92 6 10 mL/min/1.73 m2 in those who developed incident CKD during follow-up and 96 6 7 mL/min/1.73 m2 in those who did not) and no macroalbuminuria at baseline, we found that nonalcoholic fatty liver disease was associated with a 6.5-year increased incidence of CKD (adjusted hazard ratio [HR], 1.49; 95% confidence interval [CI], 1.1-2.2) independent of sex, age, body mass index, waist circumference, systolic blood pressure, smoking, diabetes duration, hemoglobin A1c level, triglyceride level, highdensity lipoprotein cholesterol level, low-density lipoprotein cholesterol level, baseline eGFR, and use of lipid-lowering, hypoglycemic, antihypertensive, or antiplatelet drugs. Notably, similar results were found for each component of the kidney disease outcome: overt proteinuria and eGFR , 60 mL/min/1.73 m2, separately (adjusted HRs of 1.45 [95% CI, 1.05-2.6] and 1.57 [95% CI, 1.2-2.5], respectively).32 Similarly, nonalcoholic fatty liver disease was associated with an increased incidence of CKD (adjusted HR, 1.85; 95% CI, 1.03-3.27) independent of age, sex, duration of diabetes, hypertension, hemoglobin A1c level, presence of microalbuminuria, and baseline eGFR in an outpatient cohort of 261 adults with type 1 diabetes who were followed up for a mean of 5.2 years.35 Of note, patients had a mean baseline eGFR of 92 6 23 mL/min/ 1.73 m2; w90% had normoalbuminuria and 10% had microalbuminuria. Approximately 4.5% of participants progressed every year to eGFR , 60 mL/min/1.73 m2 or macroalbuminuria. The annual eGFR decline for the entire cohort was 2.68 6 3.5 mL/min/1.73 m2 per year. Interestingly, patients with nonalcoholic fatty liver disease had a significantly greater annual decline in eGFR than those without nonalcoholic fatty liver disease at baseline (3.28 6 3.8 vs 2.10 6 3.0 mL/min/1.73 m2 per year). Similarly, the frequency of a decline in kidney function (arbitrarily defined as $25% loss of baseline eGFR) was significantly greater for those with nonalcoholic fatty liver disease than those without the disease (26% vs 11%). Notably, the addition of nonalcoholic fatty liver disease to traditional risk factors for CKD significantly improved the discriminatory capability of the regression models for predicting incident CKD in this cohort of patients.35 As also shown in Table 2, Chang et al,33 following up an occupational cohort of 8,329 nondiabetic 643

644

Table 2. Principal Retrospective and Prospective Studies of the Association Between NAFLD (as diagnosed by ultrasonography) and Incidence of CKD

Study

Study Characteristics

Followup

Definition of Incident CKD

Risk Factors Adjusted for in Analysis

Main Findings

Valpolicella Heart Diabetes Study; n 5 1,760 Italian type 2 diabetic outpatients with normal or near-normal kidney function (ie, eGFR $ 60 [mean baseline eGFR, 94 6 10] mL/min/ 1.73 m2 and no macroalbuminuria), who did not have cardiovascular disease, cirrhosis, or viral hepatitis at baseline; at baseline, 66% of pts had hypertension and 34% had microalbuminuria

6.5 y

Incident CKD defined as eGFR , 60 mL/min/ 1.73 m2 and/or overt proteinuria (macroalbuminuria); 547 pts developed incident CKD (mean eGFR, 55 6 12 mL/min/1.73 m2) during follow-up (428 developed decreased eGFR alone, 112 developed proteinuria, irrespective of eGFR, and 7 developed ESRD; no pt developed nephrotic syndrome)

Sex, age, BMI, waist circumference, blood pressure, smoking, diabetes duration, hemoglobin A1c, triglycerides, HDL cholesterol, LDL cholesterol, baseline eGFR, and use of lipidlowering, hypoglycemic, antihypertensive, antiplatelet drugs

Overall cumulative incidence of CKD significantly higher in pts with than without NAFLD (48% vs 29%); NAFLD independently associated with increased risk of incident CKD (HR, 1.49; 95% CI, 1.1-2.2)

Chang et al33 (2008)

Community-based cohort of 8,329 nondiabetic and nonhypertensive Korean men with normal kidney function (ie, eGFR $ 60 [median baseline eGFR, 79] mL/min/1.73 m2) and no overt proteinuria at baseline; at baseline, no pts had hypertension

3.2 y

Incident CKD defined as eGFR , 60 mL/min/ 1.73 m2 and/or overt proteinuria (urinary dipstick $ 1 1); 324 pts developed incident CKD during follow-up (no data available for no. of pts developing reduced eGFR or overt proteinuria or ESRD)

Age, BMI, alcohol consumption, blood pressure, smoking, fasting glucose, baseline eGFR, triglycerides, HDL cholesterol, LDL cholesterol, insulin resistance, C-reactive protein, and incident cases of hypertension and diabetes

NAFLD independently associated with increased risk of incident CKD (HR, 1.60; 95% CI, 1.3-2.0); consistent results observed in all subgroups analyzed

Arase et al34 (2011)

Retrospective cohort of 5,561 Japanese patients with NAFLD and no CKD at baseline (ie, eGFR $ 60 [mean eGFR, 75 6 12] mL/min/1.73 m2 and no overt proteinuria); at baseline, 13% of pts had hypertension

5.5 y

Incident CKD defined as eGFR , 60 mL/min/ 1.73 m2 and/or overt proteinuria (urinary dipstick $ 1 1); 263 pts developed incident CKD during follow-up (no data available for no. of pts developing reduced eGFR or overt proteinuria or ESRD)

Age, sex, hypertension, diabetes, total cholesterol, triglycerides, HDL cholesterol, aminotransferases, GGT, hemoglobin, white blood cells, platelets, baseline eGFR

Among pts with NAFLD, elevated serum GGT independently associated with increased risk of incident CKD (HR, 1.35; 95% CI, 1.02-1.8)

(Continued)

Targher, Chonchol, and Byrne

Am J Kidney Dis. 2014;64(4):638-652

Targher et al32 (2008)

Am J Kidney Dis. 2014;64(4):638-652

Note: Insulin resistance was estimated by a homeostasis model assessment (HOMA). GFR was estimated by using the MDRD (Modification of Diet in Renal Disease) Study equation (except for Targher et al,33 in which GFR was estimated using both the MDRD Study and CKD Epidemiology Collaboration equations). Abbreviations: BMI, body mass index; CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; GGT, gammaglutamyltransferase; HDL, high-density lipoprotein; HR, hazard ratio; LDL, low-density lipoprotein; NAFLD, nonalcoholic fatty liver disease; pts, patients.

Overall cumulative incidence of CKD significantly higher in pts with than without NAFLD (35% vs 11%); NAFLD independently associated with increased risk of incident CKD (HR, 1.85; 95% CI, 1.03-3.3); addition of NAFLD to traditional cardiorenal risk factors improved discriminatory capability of regression models for predicting incident CKD Age, sex, diabetes duration, hemoglobin A1c, hypertension, baseline eGFR, microalbuminuria Incident CKD defined as eGFR , 60 mL/min/ 1.73 m2 and/or overt proteinuria (macroalbuminuria); 61 pts developed incident CKD (mean eGFR, 59 6 13 mL/ min/1.73 m2) during follow-up; 28 developed decreased eGFR with abnormal albuminuria, 21 developed reduced eGFR alone, and 12 developed macroalbuminuria alone; no pt developed ESRD; no pt developed nephrotic syndrome 5.2 y Outpatient cohort of 261 Italian type 1 diabetic adults with normal or nearnormal kidney function (ie, eGFR $ 60 [mean baseline eGFR, 92 6 23] mL/min/ 1.73 m2 and not macroalbuminuria), who did not have cardiovascular disease, cirrhosis, or viral hepatitis at baseline; at baseline, 44% of pts had hypertension and 10% had microalbuminuria Targher et al35 (2014)

Main Findings Risk Factors Adjusted for in Analysis Definition of Incident CKD Followup Study Characteristics Study

Table 2 (Cont’d). Principal Retrospective and Prospective Studies of the Association Between NAFLD (as diagnosed by ultrasonography) and Incidence of CKD

Nonalcoholic Fatty Liver Disease and CKD

nonhypertensive men with normal kidney function and no overt proteinuria at baseline, showed that nonalcoholic fatty liver disease was associated with an increased incidence of CKD (adjusted HR, 1.60; 95% CI, 1.3-2.0), even after adjusting for several potential confounders, including insulin resistance and C-reactive protein level. Interestingly, the association between nonalcoholic fatty liver disease and incident CKD was consistent in all subgroups evaluated and remained significant after further adjustment for incident cases of hypertension or diabetes that occurred during the study follow-up period.33 Finally, Arase et al34 reported that mildly elevated serum gamma-glutamyltransferase levels were associated independently with increased incidence of CKD in a large retrospective cohort of patients with ultrasounddiagnosed nonalcoholic fatty liver disease who did not have CKD at baseline. It is important to remark that in all these studies, nonalcoholic fatty liver disease was diagnosed by ultrasonography, and future prospective studies in large cohorts of patients with histologically defined nonalcoholic fatty liver disease are needed to examine whether the severity of nonalcoholic fatty liver disease histology is associated with greater risk of CKD development and progression. It also is important to note that there is marked heterogeneity in results across the different published studies that may be explained at least in part by differences in age, ethnicity, and characteristics of the patient cohorts (diabetic in 2 and predominantly nondiabetic in 2). In addition, the mean duration of follow-up of these studies also was variable, ranging from approximately 3-6.5 years, and the investigators used varying degrees of baseline adjustments for potential confounders, as specified in Table 2. In particular, only a few studies adjusted results for important risk factors such as abdominal obesity or insulin resistance, which play important roles in the pathogenesis of nonalcoholic fatty liver disease and CKD. Accurate assessment of abdominal visceral fat and insulin resistance would be particularly important in order to elucidate whether the relationship between CKD and nonalcoholic fatty liver disease is affected by these 2 risk factors. It also is important to highlight that in the 4 published prospective studies, the investigators have used eGFR (by using either the MDRD Study equation, CKD-EPI [CKD Epidemiology Collaboration] equation, or both) instead of direct GFR measurement to define CKD. The use of direct GFR measurements should be encouraged because creatinine-based equations are not accurate in estimating GFR, especially for patients with obesity or cirrhosis.38,39 None of these published studies has specifically assessed whether a change in nonalcoholic fatty liver disease status (either development of 645

Targher, Chonchol, and Byrne

new fatty liver, progression to cirrhosis, or resolution of existing fatty liver) during the follow-up period modified the risk of incident CKD. Moreover, no detailed information is available in these studies about specific renal pathology/morphology associated with nonalcoholic fatty liver disease. Finally, only 2 studies have specified the number of patients who developed the individual outcome measures of the study (eGFR reduction or macroalbuminuria) and have reported the number of those who developed nephrotic syndrome or end-stage renal disease.32,35 Notwithstanding these limitations, the 4 prospective studies that have used ultrasonography to diagnose nonalcoholic fatty liver disease have reported an independent association between nonalcoholic fatty liver disease and increased incidence of CKD, with HRs for CKD that ranged from approximately 1.3-1.9.32-35 However, further longer follow-up studies in larger cohorts of patients with histologically confirmed nonalcoholic fatty liver disease are needed to confirm these findings and determine whether improvement in nonalcoholic fatty liver disease (or future treatments for nonalcoholic fatty liver disease) ultimately will prevent or delay the development and progression of CKD. Moreover, because CKD has many potential causes, it also will be of great interest to characterize the renal injury manifestations associated with nonalcoholic fatty liver disease and define, in the future, whether nonalcoholic fatty liver disease may selectively contribute to the pathogenesis of different types of kidney disease.

PUTATIVE MECHANISMS LINKING NONALCOHOLIC FATTY LIVER DISEASE TO CKD Although the pathogenic mechanisms linking nonalcoholic fatty liver disease and CKD are not fully understood, understanding the potential pathways between these 2 pathologic conditions is important because this might lead to new therapeutic strategies to slow kidney function decline. The tight inter-relationship of each disease state with abdominal obesity and insulin resistance makes it challenging to identify the main causal factor(s) that underlines the increased risk of kidney disease in patients with nonalcoholic fatty liver disease.5-7,14 To date, there is still uncertainty about whether nonalcoholic fatty liver disease is a marker or a mediator (ie, a susceptibility factor) of CKD. However, as discussed next, although more research is needed, we consider that mounting epidemiologic and experimental evidence supports the assertion that nonalcoholic fatty liver disease is a mediator rather than simply a marker of kidney disease. From the data available in the literature, the association between nonalcoholic fatty liver disease and CKD seems to 646

have strength, consistency, specificity, temporality, and biological plausibility, satisfying many of the established criteria for a causal relationship. The liver is a master regulator of glucose and lipid metabolism and the primary source of a number of circulating coagulation and inflammatory factors, many of which are involved in the development of vascular and kidney disease. As shown in the schematic Fig 2, the putative underlying mechanisms that link nonalcoholic fatty liver disease and CKD probably have their origin in the expanded and inflamed visceral adipose tissue, with the liver being both the target of the resulting systemic abnormalities and an important source of several pathogenic mediators that further amplify kidney and vascular damage. It is likely that there is pathogenic cross-talk between the liver and the expanded/inflamed adipose tissue, which releases increased free fatty acids and secretes multiple proinflammatory adipocytokines.4,5,40-42 The resulting inflammatory process in adipose tissue is one of the earliest events that leads to systemic insulin resistance.4,5,40-42 These proinflammatory pathways converge on 2 main intracellular transcription factor signaling pathways, the nuclear factor-kB (NF-kB) pathway and C-Jun-N-terminal kinase (JNK) pathway.40-42 Experimental data in animals suggest that JNK-1 activation in adipose tissue causes insulin resistance in the liver.43 In addition, insulin resistance is a key factor causing an increase in free fatty acid uptake from the hydrolysis of adipose tissue triglycerides,4,40-42 and insulin resistance is likely to be an important pathogenic factor in the development of nonalcoholic fatty liver disease and associated complications.3-5,42 It also has been suggested that insulin resistance is a susceptibility factor of CKD that may play an important role in the development of kidney disease, making insulin resistance a possible mechanistic link between nonalcoholic fatty liver disease and CKD.15-18 Activation of the NF-kB pathway in the livers of patients with nonalcoholic steatohepatitis leads to increased transcription of several proinflammatory genes that amplify systemic chronic inflammation.44,45 Hence, the increased intrahepatic cytokine expression that is mediated by hepatocellular damage and fatderived factors likely plays a role in kidney disease progression. Nonalcoholic fatty liver disease, especially its necroinflammatory form (nonalcoholic steatohepatitis), may exacerbate systemic and hepatic insulin resistance, cause atherogenic dyslipidemia, and release a myriad of proinflammatory, procoagulant, pro-oxidant, and profibrogenic factors (Fig 2).3-6,46-48 Moreover, the release of key components of the reninangiotensin-aldosterone system (ie, angiotensinogen) Am J Kidney Dis. 2014;64(4):638-652

Nonalcoholic Fatty Liver Disease and CKD

Figure 2. Schematic representation of the putative mechanisms underlying the contribution of nonalcoholic fatty liver disease (NAFLD) to the increased risk of chronic kidney disease (CKD). The complex and intertwined interactions among NAFLD, abdominal obesity, and insulin resistance make it extremely difficult to dissect out the specific role of the liver and the underlying mechanisms responsible for the association between NAFLD and the risk of developing CKD. NAFLD may contribute to the development and progression of CKD by atherogenic dyslipidemia, systemic/hepatic insulin resistance, dysglycemia (increased hepatic glucose production), and the systemic release of numerous potentially pathogenic mediators (ie, proinflammatory biomarkers, procoagulant and profibrogenic factors). Abbreviations: CRP, C-reactive protein; CTGF, connective tissue growth factor; FGF-21, fibroblast growth factor 21; HDL-C, high-density lipoprotein cholesterol. IGF, insulin-like growth factor; IL-6, interleukin 6; LDL-C, lowdensity lipoprotein cholesterol; NASH, nonalcoholic steatohepatitis; PAI-1, plasminogen activator inhibitor 1; TGF, transforming growth factor; TNF, tumor necrosis factor; VWF, von Willebrand factor.

that may contribute to the pathophysiology of hypertension also is increased in nonalcoholic steatohepatitis.3-6,46 The experimental findings that nonalcoholic steatohepatitis is associated with abnormal intrahepatic messenger RNA expression of these potential mediators of kidney and vascular injury further support the conclusion that the increased circulating levels of the aforementioned biomarkers result from upregulation of their own synthesis in the steatotic and inflamed liver.5,44,48 Some experimental studies also have shown that a number of the genes involved in fatty acid metabolism, lipolysis, monocyte and macrophage recruitment, coagulation, and inflammation are overexpressed in livers of patients with nonalcoholic steatohepatitis.49-51 In this dangerous scenario (as also shown in Fig 2), some evidence from studies of animals and humans also suggests that the coexistence of obesity-related increases in fat accumulation in the renal sinus and renal parenchyma additionally may exert local adverse effects that result in structural and functional changes in the kidney and vasculature.52-54 From a pathophysiologic point of view, it is important to emphasize that chronic inflammation, lipotoxicity, oxidative stress, and hypercoagulation increasingly are recognized for their role in the pathogenesis of CKD in Am J Kidney Dis. 2014;64(4):638-652

animal models.55-57 It is well established that in the context of CKD, generation and metabolism of various pro- and anti-inflammatory cytokines are disturbed. Although the exact mechanisms by which chronic inflammation, lipotoxicity, and oxidative stress may cause kidney damage are not well understood, experimental data in animals suggest that cytokine imbalance contributes to CKD pathogenesis through a number of local adverse effects, such as activation of proinflammatory pathways, upregulation of adhesion molecules, induction of endothelial dysfunction, and decreased adiponectin expression.55-57 Patients with nonalcoholic fatty liver disease have lower plasma adiponectin levels than control individuals without steatosis, and plasma adiponectin levels are related inversely to the severity of nonalcoholic fatty liver disease histology independent of other important confounders.46,47,58 An intriguing and attractive hypothesis supports a central role for adiponectin (ie, a fatsecreted protein with antiatherogenic/anti-inflammatory effects) and fetuin-A (a liver-secreted protein that regulates plasma adiponectin levels) in the mechanisms linking obesity, CKD, and nonalcoholic fatty liver disease. Some recent studies suggest that mechanisms leading to CKD and nonalcoholic fatty liver disease may be interlinked through cross-talk between fat, kidney, and liver through at least 2 plasma proteins, namely 647

Targher, Chonchol, and Byrne Table 3. Summary of Potential Benefits of Different Treatments or Intervention Trials in Adult Patients With NAFLD/NASH Intervention/ Treatment

Benefits

Risk Factor Modification and Comments

Lifestyle intervention (eg, weight loss)

Y serum liver enzymes, Y hepatic fat (MRS and US), [ insulin sensitivity, improved or unchanged NAFLD histologic staging; NAFLD improvement (hepatic steatosis and inflammation) with w10% weight loss; rapid weight loss could lead to [ hepatic fat content

Y blood pressure, plasma glucose, and triglycerides; [ HDL-C; dietary restriction of calorie intake is of proven benefit in decreasing hepatic fat content; some studies suggest a benefit of specific dietary changes on hepatic fat content (eg, reducing intakes of saturated fatty acids and total carbohydrates)

Physical exercise

Y serum liver enzymes, Y hepatic fat (MRS and US), [ insulin sensitivity, Y inflammation; regular exercise showed improvements in NAFLD independent of body weight or visceral fat changes; resistance training beneficial in decreasing liver fat independent of changes in body weight

Y blood pressure, plasma triglyceride, and glucose; [ HDL-C; improved VO2max with more intense physical activity; preliminary studies suggest a marked benefit of resistance training; however, type of activity needs to be tailored to individual patient

Metformin

Y serum liver enzymes, unchanged or Y hepatic fat (US), [ insulin sensitivity; no change in NAFLD histologic staging (overall)

Thiazolidinediones

Y serum liver enzymes, Y hepatic fat content (MRS), [ insulin sensitivity; improved or unchanged NAFLD histologic staging and possible decrease in hepatic inflammation

Y plasma glucose; however, gastrointestinal symptoms (diarrhea and abdominal cramping) common with higher doses (ie, . 2 g/d); risk of lactic acidosis (rare); contraindicated if eGFR , 35-40 mL/min/1.73 m2 [ risk of nonfatal myocardial infarction (rosiglitazone only) and heart failure; pioglitazone: Y risk of major adverse CVD events (excluding congestive heart failure) in type 2 diabetic patients; however, both cause weight gain (by increased peripheral fat accumulation) and edema and increase risk of bone fractures and bladder cancer (slightly)

Glucagon-like peptide (GLP)-1 analogues

Y serum liver enzymes and hepatic fat content (CT and US), possible improvement in insulin sensitivity, insulin secretion; possible improvement in hepatic steatosis and inflammation

GLP-1 analogues decrease plasma glucose and may decrease appetite and facilitate weight loss; may also have direct action to decrease hepatic fat content by improving insulin signaling; most evidence of benefit to date is from animal studies; however, gastrointestinal symptoms (nausea, vomiting, diarrhea) common, cases of acute pancreatitis observed; long-term safety unknown

Statins

Y serum liver enzymes, unchanged or Y hepatic fat (US); no change in NAFLD histologic staging

Y risk of adverse CVD events and death in primary and secondary prevention regardless of plasma lipid levels; safe in NAFLD, no need for liver enzyme monitoring but small increase (w10%) in risk of diabetes

Fibrates

Y serum liver enzymes; no change in NAFLD histologic staging

Y triglycerides, [ HDL-C, Y in CVD events seen in only subgroup with atherogenic dyslipidemia

Vitamin E

Y serum liver enzymes, insulin sensitivity improved or unchanged; improved or unchanged NAFLD histology

Omega-3 fatty acids

Y serum liver enzymes, Y hepatic fat (MRS and US), [ insulin sensitivity or unchanged

Overall, no conclusive benefit but recent evidence suggests that vitamin E at 800 IU/d dosage improves liver histology in nondiabetic adults with biopsy-proven NASH; vitamin E not recommended to treat NASH in diabetic patients, NAFLD without liver biopsy, NASH cirrhosis, or cryptogenic cirrhosis; dose and duration uncertain and likely to be important; long-term safety unproven Y triglycerides, Y mortality post–myocardial infarction, possible Y in atrial fibrillation burden; possible carotid artery plaque stabilization; may [ ventricular arrhythmias in angina patients; dose and duration of omega-3 fatty acids uncertain; uncertain whether there is benefit on hepatic inflammation

(Continued)

648

Am J Kidney Dis. 2014;64(4):638-652

Nonalcoholic Fatty Liver Disease and CKD Table 3 (Cont’d). Summary of Potential Benefits of Different Treatments or Intervention Trials in Adult Patients With NAFLD/NASH Intervention/ Treatment

Benefits

Risk Factor Modification and Comments

Angiotensin receptor blockers

Y serum liver enzymes, [ insulin sensitivity; improved NAFLD histologic staging (telmisartan and losartan only), may Y liver fibrosis

Y blood pressure; may improve impaired glucose tolerance; most evidence of benefit to date from animal studies or small pilot studies in humans; no randomized clinical trials have specifically examined the effect of different antihypertensive agents on the liver in hypertensive patients with NAFLD

Bariatric surgery

Y serum liver enzymes, [ insulin sensitivity; improved NAFLD histologic staging (hepatic steatosis and inflammation); may improve hepatic fibrosis

Benefit derives mainly from weight loss; effects of bypass surgery probably greater than gastric band; potential initial worsening of NAFLD with very rapid weight loss

Note: Data summarized in the table are based on recent reviews and guidelines.1,4,60-62,65,67 Abbreviations: CT, computed tomography; CVD, cardiovascular disease; HDL-C, high-density lipoprotein cholesterol; MRS, magnetic resonance spectroscopy; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; US, ultrasonography; VO2max, maximal oxygen consumption. Source: Modified from Targher G et al60 with permission of Thieme.

adiponectin and fetuin-A.59 In the liver and kidney, low adiponectin levels reduce activation of the energy sensor 50 -adenosine monophosphate activated protein kinase, which is pivotal in directing hepatocytes and podocytes to compensatory and potentially deleterious pathways that lead to inflammatory and profibrogenic cascades culminating in end-organ damage (ie, end-stage liver and kidney diseases).59 However, further research is required to uncover other specific mechanisms by which nonalcoholic fatty liver disease may contribute to CKD pathogenesis and to characterize the specific renal pathology/ morphology associated with nonalcoholic fatty liver disease and how the specific mechanisms of nonalcoholic fatty liver disease may contribute to the pathologic findings associated with CKD.

NONALCOHOLIC FATTY LIVER DISEASE TREATMENT Detailed discussion of the interventions that have been tested for nonalcoholic fatty liver disease/ nonalcoholic steatohepatitis is beyond the scope of this brief review and have been discussed elsewhere.1,4,60-62 Consequently, we have summarized the evidence supporting the use of these interventions in Table 3. Data summarized in the table are based on recent reviews and guidelines.1,4,60-62 Currently, there is no standard treatment for liver disease per se in nonalcoholic fatty liver disease. However, nonalcoholic fatty liver disease and CKD share numerous cardiometabolic risk factors, and prevention and treatment strategies for nonalcoholic fatty liver disease and CKD should be similar and aimed primarily at reducing insulin resistance and modifying the associated cardiometabolic risk factors. Current recommendations for nonalcoholic fatty liver Am J Kidney Dis. 2014;64(4):638-652

disease/nonalcoholic steatohepatitis therapy are limited to weight reduction through diet and physical exercise (a 5%-10% weight loss reduces hepatic steatosis, whereas up to a 10% weight loss is needed to improve hepatic necroinflammation) and the treatment of individual components of metabolic syndrome, possibly with the use of therapies that may have potential beneficial liver effects, including bariatric surgery for severe obesity, insulin-sensitizing agents for type 2 diabetes (especially pioglitazone), and drugs directed at the renin-angiotensin-aldosterone system to control hypertension (especially some angiotensin receptor blockers, such as telmisartan).1,4,5,60-62 However, as reported in Table 3, no large randomized clinical trials have examined the effect(s) of different hypoglycemic or antihypertensive agents on liver histology in patients with nonalcoholic fatty liver disease with diabetes or hypertension. To date, there is no convincing evidence that lipidlowering agents, including statins, are beneficial for patients with nonalcoholic fatty liver disease/nonalcoholic steatohepatitis. However, statins can be prescribed safely for conventional indications because there is no evidence that patients with pre-existing nonalcoholic fatty liver disease/nonalcoholic steatohepatitis are at increased risk for statin-induced idiosyncratic hepatotoxicity or that statins are associated with an increased frequency of hepatic steatosis or abnormal serum aminotransferase levels in these patients.1,4,5,60-62 Post hoc analyses of randomized intervention trials recently have suggested that the cardiovascular benefit of statins is greater in patients with established cardiovascular disease and mild to moderately abnormal aminotransferase levels (that are potentially attributable to nonalcoholic fatty liver disease) than in those with normal serum 649

Targher, Chonchol, and Byrne

aminotransferase levels.63,64 Thus, despite previous hepatotoxicity concerns and a reluctance to prescribe statins in patients with nonalcoholic fatty liver disease, statins are now considered safe in this patient group. Use of these agents has a significant positive benefit to risk ratio and some research has even suggested a potential benefit to reverse hepatic steatosis.1,4,60,62 In patients who fail to implement the lifestyle changes or those with progressive nonalcoholic steatohepatitis, pharmacologic treatments specifically directed at the liver for improving the histologic features of nonalcoholic steatohepatitis might be necessary.1,4,5,60-62 The major problem in this field is the scarcity of definitive clinical trials. To date, there are few high-quality, randomized, blinded, adequately powered, controlled trials of sufficient duration and with adequate histologic outcomes. Thiazolidinediones have the best evidence-based data for efficacy in nonalcoholic steatohepatitis, but long-term adverse cardiovascular (eg, congestive heart failure) and noncardiovascular side effects (eg, weight gain and increased risks of bladder cancer and bone fractures) attributed to this class of drugs are a serious issue and are likely to prevent approval of thiazolidinediones as a treatment for nonalcoholic fatty liver disease.1,4,60-62,65 Preliminary evidence suggests some benefit of vitamin E (principally in nondiabetic adults with biopsy-proven nonalcoholic steatohepatitis), high-dose omega-3 fatty acids, antioxidants, and other hepatoprotectants, but to date, there are insufficient data to advocate the use of any of these agents in nonalcoholic fatty liver disease/ nonalcoholic steatohepatitis (Table 3).1,4,60-62,66,67 Future therapeutic research in nonalcoholic steatohepatitis should focus on exploring innovative molecules that have better efficacy and a better safety profile.

CONCLUSIONS The prevalence of nonalcoholic fatty liver disease is increasing globally, mainly due to the increasing burden of obesity and metabolic syndrome. Despite the growing evidence of an independent association between nonalcoholic fatty liver disease and CKD, it remains to be definitively established whether a causal relationship exists and the precise pathogenesis of nonalcoholic fatty liver disease–mediated CKD is unclear. To date, there is uncertainty about whether nonalcoholic fatty liver disease poses an independent risk for CKD above and beyond known risk factors for CKD. Although there is a suggestion that nonalcoholic fatty liver disease adds to the risk of CKD, existing studies are too few and methodologically not sufficiently rigorous. Additional large-scale prospective studies are needed to draw firm conclusions about 650

any independent hepatic contribution to the increased risk of CKD observed in patients with nonalcoholic fatty liver disease. Further studies also are needed to characterize the specific renal pathology/morphology and better elucidate the progression rate of kidney disease associated with nonalcoholic fatty liver disease. In the meantime, until larger long-term prospective studies fully establish the nature of the association between nonalcoholic fatty liver disease and CKD, we recommend for patients with nonalcoholic fatty liver disease: (1) careful surveillance for the development of kidney disease and (2) careful monitoring of patients with established kidney disease for accelerated atherosclerosis and deterioration in kidney function.

ACKNOWLEDGEMENTS Support: Dr Targher is supported in part by grants from the School of Medicine of the Verona University. Dr Byrne is supported in part by the Southampton National Institute for Health Research Biomedical Research Centre. Financial Disclosure: The authors declare that they have no relevant financial interests.

REFERENCES 1. Chalasani N, Younossi Z, Lavine JE, et al. The diagnosis and management of non-alcoholic fatty liver disease: practice guideline by the American Association for the Study of Liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association. Hepatology. 2012;55: 2005-2023. 2. Loomba R, Sanyal AJ. The global NAFLD epidemic. Nat Rev Gastroenterol Hepatol. 2013;10:686-690. 3. Anstee QM, Targher G, Day CP. Progression of NAFLD to diabetes mellitus, cardiovascular disease or cirrhosis. Nat Rev Gastroenterol Hepatol. 2013;10:330-344. 4. Targher G, Byrne CD. Nonalcoholic fatty liver disease: a novel cardiometabolic risk factor for type 2 diabetes and its complications. J Clin Endocrinol Metab. 2013;98:483-495. 5. Targher G, Day CP, Bonora E. Risk of cardiovascular disease in patients with nonalcoholic fatty liver disease. N Engl J Med. 2010;363:1341-1350. 6. Targher G, Chonchol M, Zoppini G, Abaterusso C, Bonora E. Risk of chronic kidney disease in patients with nonalcoholic fatty liver disease: is there a link? J Hepatol. 2011;54: 1020-1029. 7. Armstrong MJ, Adams LA, Canbay A, Syn WK. Extrahepatic complications of nonalcoholic fatty liver disease. Hepatology. 2014;59:1174-1197. 8. Hernaez R, Lazo M, Bonekamp S, et al. Diagnostic accuracy and reliability of ultrasonography for the detection of fatty liver: a meta-analysis. Hepatology. 2011;54:1082-1090. 9. Michelotti GA, Machado MV, Diehl AM. NAFLD, NASH and liver cancer. Nat Rev Gastroenterol Hepatol. 2013;10:656-665. 10. Charlton MR, Burns JM, Pedersen RA, Watt KD, Heimbach JK, Dierkhising RA. Frequency and outcomes of liver transplantation for nonalcoholic steatohepatitis in the United States. Gastroenterology. 2011;141:1249-1253. 11. Nascimbeni F, Pais R, Bellentani S, et al. From NAFLD in clinical practice to answers from guidelines. J Hepatol. 2013;59: 859-871. Am J Kidney Dis. 2014;64(4):638-652

Nonalcoholic Fatty Liver Disease and CKD 12. Singal AK, Guturu P, Hmoud B, Kuo YF, Salameh H, Wiesner RH. Evolving frequency and outcomes of liver transplantation based on etiology of liver disease. Transplantation. 2013;95:755-760. 13. Houlihan DD1, Armstrong MJ, Davidov Y, et al. Renal function in patients undergoing transplantation for nonalcoholic steatohepatitis cirrhosis: time to reconsider immunosuppression regimens? Liver Transpl. 2011;17:1292-1298. 14. Bonora E, Targher G. Increased risk of cardiovascular disease and chronic kidney disease in NAFLD. Nat Rev Gastroenterol Hepatol. 2012;9:372-381. 15. Kurella M, Lo JC, Chertow GM. Metabolic syndrome and the risk for chronic kidney disease among nondiabetic adults. J Am Soc Nephrol. 2005;16:2134-2140. 16. Weiner DE, Tighiouart H, Elsayed EF, et al. The relationship between nontraditional risk factors and outcomes in individuals with stage 3 to 4 CKD. Am J Kidney Dis. 2008;51:212-223. 17. Kendrick J, Chonchol MB. Non-traditional risk factors for cardiovascular disease in patients with chronic kidney disease. Nat Clin Pract Nephrol. 2008;4:672-681. 18. Kronenberg F. Emerging risk factors and markers of chronic kidney disease progression. Nat Rev Nephrol. 2009;5:677-689. 19. Targher G, Bertolini L, Rodella S, Zoppini G, Lippi G, Day C. Non-alcoholic fatty liver disease is independently associated with an increased prevalence of chronic kidney disease and proliferative/laser-treated retinopathy in type 2 diabetic patients. Diabetologia. 2008;51:444-450. 20. Targher G, Bertolini L, Chonchol M, Rodella S, Zoppini G, Lippi G. Nonalcoholic fatty liver disease is independently associated with an increased prevalence of chronic kidney disease and retinopathy in type 1 diabetic patients. Diabetologia. 2010;53: 1341-1348. 21. Hwang ST, Cho YK, Yun JW, et al. Impact of NAFLD on microalbuminuria in patients with prediabetes and diabetes. Intern Med J. 2010;40:437-442. 22. Targher G, Pichiri I, Zoppini G, Trombetta M, Bonora E. Increased prevalence of chronic kidney disease in patients with type 1 diabetes and non-alcoholic fatty liver. Diabet Med. 2012;29:220-226. 23. Li G, Shi W, Hug H, Chen Y, Liu L, Yin D. Nonalcoholic fatty liver disease associated with impairment of kidney function in nondiabetes population. Biochem Med (Zagreb). 2012;22:92-99. 24. Sirota JC, McFann K, Targher G, Chonchol M, Jalal DI. Association between nonalcoholic liver disease and chronic kidney disease: an ultrasound analysis from NHANES 1988-1994. Am J Nephrol. 2012;36:466-471. 25. Ahn AL, Choi JK, Kim MN, et al. Non-alcoholic fatty liver disease and chronic kidney disease in Koreans aged 50 years or older. Korean J Fam Med. 2013;34:199-205. 26. Mikolasevic I, Racki S, Bubic I, Jelic I, Stimac D, Orlic L. Chronic kidney disease and nonalcoholic fatty liver disease proven by transient elastography. Kidney Blood Press Res. 2013;37:305-310. 27. Targher G, Bertolini L, Rodella S, Lippi G, Zoppini G, Chonchol M. Relationship between kidney function and liver histology in subjects with nonalcoholic steatohepatitis. Clin J Am Soc Nephrol. 2010;5:2166-2171. 28. Yilmaz Y, Alahdab YO, Yonal O, et al. Microalbuminuria in nondiabetic patients with nonalcoholic fatty liver disease: association with liver fibrosis. Metabolism. 2010;59:1327-1330. 29. Yasui K, Sumida Y, Mori Y, et al. Nonalcoholic steatohepatitis and increased risk of chronic kidney disease. Metabolism. 2011;60:735-739. 30. Park CW, Tsai NT, Wong LL. Implications of worse renal dysfunction and medical comorbidities in patients with NASH

Am J Kidney Dis. 2014;64(4):638-652

undergoing liver transplant evaluation: impact on MELD and more. Clin Transplant. 2011;25:E606-E611. 31. Machado MV, Gonçalves S, Carepa F, Coutinho J, Costa A, Cortez-Pinto H. Impaired renal function in morbid obese patients with nonalcoholic fatty liver disease. Liver Int. 2012;32:241-248. 32. Targher G, Chonchol M, Bertolini L, et al. Increased risk of CKD among type 2 diabetics with nonalcoholic fatty liver disease. J Am Soc Nephrol. 2008;19:1564-1570. 33. Chang Y, Ryu S, Sung E, et al. Nonalcoholic fatty liver disease predicts chronic kidney disease in nonhypertensive and nondiabetic Korean men. Metabolism. 2008;57:569-576. 34. Arase Y, Suzuki F, Kobayashi M, et al. The development of chronic kidney disease in Japanese patients with non-alcoholic fatty liver disease. Intern Med. 2011;50:1081-1087. 35. Targher G, Mantovani A, Pichiri I, et al. Nonalcoholic fatty liver disease is independently associated with an increased incidence of chronic kidney disease in patients with type 1 diabetes. Diabetes Care. 2014;37:1729-1736. 36. Lee DH, Jacobs DR, Gross M, Steffes M. Serum gammaglutamyltransferase was differently associated with microalbuminuria by status of hypertension and diabetes: the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Clin Chem. 2005;51:1185-1191. 37. Ryu S, Chang Y, Kim DI, Kim WS, Suh BS. Gammaglutamyltransferase as a predictor of chronic kidney disease in nonhypertensive and nondiabetic Korean men. Clin Chem. 2007;53:71-77. 38. Stevens LA, Coresh J, Greene T, Levey AS. Assessing kidney function—measured and estimated glomerular filtration rate. N Engl J Med. 2006;8:2473-2483. 39. Stevens LA, Coresh J, Feldman HI, et al. Evaluation of the Modification of Diet in Renal Disease Study equation in a large diverse population. J Am Soc Nephrol. 2007;18:2749-2757. 40. Badman MK, Flier JS. The adipocyte as an active participant in the energy balance and metabolism. Gastroenterology. 2007;132:2103-2115. 41. Shoelson SE, Herrero L, Naaz A. Obesity, inflammation and insulin resistance. Gastroenterology. 2007;132:2169-2180. 42. Byrne CD. Dorothy Hodgkin Lecture 2012: non-alcoholic fatty liver disease, insulin resistance and ectopic fat: a new problem in diabetes management. Diabet Med. 2012;29:1098-1107. 43. Sabio G, Das M, Mora A, et al. A stress signaling pathway in adipose tissue regulates hepatic insulin resistance. Science. 2008;322:1539-1543. 44. de Alwis NMW, Day CP. Nonalcoholic fatty liver disease: the mist gradually clears. J Hepatol. 2008;48(suppl 1):S104-S112. 45. Tilg H, Moschen AR. Evolution of inflammation in nonalcoholic fatty liver disease: the multiple parallel hits hypothesis. Hepatology. 2010;52:1836-1846. 46. Targher G, Chonchol M, Miele L, Zoppini G, Pichiri I, Muggeo M. Nonalcoholic fatty liver disease as a contributor to hypercoagulation and thrombophilia in the metabolic syndrome. Semin Thromb Hemost. 2009;35:277-287. 47. Targher G, Bertolini L, Rodella S, et al. NASH predicts plasma inflammatory biomarkers independently of visceral fat in men. Obesity (Silver Spring). 2008;16:1394-1399. 48. Stefan N, Häring HU. The role of hepatokines in metabolism. Nat Rev Endocrinol. 2013;9:144-152. 49. Westerbacka J, Kolak M, Kiviluoto T, et al. Genes involved in fatty acid partitioning and binding, lipolysis, monocyte/macrophage recruitment, and inflammation are overexpressed in the human fatty liver of insulin-resistant subjects. Diabetes. 2007;56:2759-2765.

651

Targher, Chonchol, and Byrne 50. Greco D, Kotronen A, Westerbacka J, et al. Gene expression in human NAFLD. Am J Physiol Gastrointest Liver Physiol. 2008;294:G1281-G1287. 51. Sookoian S, Gianotti TF, Rosselli MS, Burgueno AL, Castano GO, Pirola CJ. Liver transcriptional profile of atherosclerosisrelated genes in human nonalcoholic fatty liver disease. Atherosclerosis. 2011;218:378-385. 52. Deji N, Kume S, Araki S, et al. Structural and functional changes in the kidneys of high-fat diet-induced obese mice. Am J Physiol Renal Physiol. 2009;296:F118-F126. 53. Bobulescu IA, Dubree M, Zhang J, McLeroy P, Moe OW. Effect of renal lipid accumulation on proximal tubule Na/H exchange and ammonium secretion. Am J Physiol Renal Physiol. 2008;294:F1315-F1322. 54. Foster MC, Hwang SJ, Porter SA, Massaro JM, Hoffmann U, Fox CS. Fatty kidney, hypertension, and chronic kidney disease: the Framingham Heart Study. Hypertension. 2011;58:784-790. 55. Guebre-Egziabher F, Alix PM, Koppe L, et al. Ectopic lipid accumulation: a potential cause for metabolic disturbances and a contributor to the alteration of kidney function. Biochimie. 2013;95:1971-1979. 56. Massy ZA, Stenvinkel P, Drueke TB. The role of oxidative stress in chronic kidney disease. Semin Dial. 2009;22:405-408. 57. Vlassara H, Torreggiani M, Post JB, Zheng F, Uribarri J, Striker GE. Role of oxidants/inflammation in declining renal function in chronic kidney disease and normal aging. Kidney Int Suppl. 2009;114:S3-S11. 58. Moschen AR, Wieser V, Tilg H. Adiponectin: key player in the adipose tissue-liver crosstalk. Curr Med Chem. 2012;19: 5467-5473.

652

59. Ix JH, Sharma K. Mechanisms linking obesity, chronic kidney disease, and fatty liver disease: the roles of fetuin-A, adiponectin, and AMPK. J Am Soc Nephrol. 2010;21:406-412. 60. Targher G, Byrne CD. Diagnosis and management of nonalcoholic fatty liver disease and its hemostatic/thrombotic and vascular complications. Semin Thromb Hemost. 2013;39:214-228. 61. Day CP. Clinical spectrum and therapy of non-alcoholic steatohepatitis. Dig Dis. 2012;30(suppl 1):69-73. 62. Ratziu V. Pharmacological agents for NASH. Nat Rev Gastroenterol Hepatol. 2013;10:676-685. 63. Athyros VG, Tziomalos K, Gossios TD, et al; GREACE Study Collaborative Group. Safety and efficacy of long-term statin treatment for cardiovascular events in patients with coronary heart disease and abnormal liver tests in the Greek Atorvastatin and Coronary Heart Disease Evaluation (GREACE) Study: a post-hoc analysis. Lancet. 2010;376:1916-1922. 64. Tikkanen MJ, Fayyad R, Faergeman O, et al; on behalf of the IDEAL Investigators. Effect of intensive lipid lowering with atorvastatin on cardiovascular outcomes in coronary heart disease patients with mild-to-moderate baseline elevations in alanine aminotransferase levels. Int J Cardiol. 2013;168:3846-3852. 65. Ratziu V, Caldwell S, Neuschwander-Tetri BA. Therapeutic trials in nonalcoholic steatohepatitis: insulin sensitizers and related methodological issues. Hepatology. 2010;52:2206-2215. 66. Sanyal AJ, Chalasani N, Kowdley KV, et al; for the NASH. Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. N Engl J Med. 2010;362:1675-1685. 67. Scorletti E, Byrne CD. Omega-3 fatty acids, hepatic lipid metabolism, and nonalcoholic fatty liver disease. Annu Rev Nutr. 2013;33:231-248.

Am J Kidney Dis. 2014;64(4):638-652