risk of cardiovascular disease in patients

8/7/2019 Risk of Cardiovascular Disease in Patients

http://slidepdf.com/reader/full/risk-of-cardiovascular-disease-in-patients 1/10

T he n e w e n g l a n d j o u r n a l o f medicine

n engl j med 363;14 nejm.org september 30, 2010 1341

review article

Current Concepts

Risk of Cardiovascular Disease in Patientswith Nonalcoholic Fatty Liver Disease

Giovanni Targher, M.D., Christopher P. Day, M.D., Ph.D.,and Enzo Bonora, M.D., Ph.D.

From the Section of Endocrinology andMetabolism, Department of Medicine,University of Verona, Verona, Italy (G.T.,E.B.); and the Institute of Cellular Medi-cine, Newcastle University, Newcastleupon Tyne, United Kingdom (C.P.D.). Ad-dress reprint requests to Dr. Targher at theSection of Endocrinology, Ospedale CivileMaggiore, Piazzale Aristide Stefani 1,37126 Verona, Italy, or at [email protected]; or to Dr. Day at the Faculty Of-fice, Medical School, Framlington Pl.,Newcastle upon Tyne, NE2 4HH, UnitedKingdom, or at [email protected].

Drs. Targher and Day contributed equallyto this article.

N Engl J Med 2010;363:1341-50.Copyright © 2010 Massachusetts Medical Society.

Nonalcoholic fatty liver disease encompasses a spectrum of

pathologic conditions, ranging from simple steatosis to nonalcoholic steato-hepatitis and cirrhosis. The disease has reached epidemic proportions and

is the most common cause of chronic liver disease in Western countries.1-4 Ap-proximately 20 to 30% of adults in the general population in Western countrieshave nonalcoholic fatty liver disease, and its prevalence increases to 70 to 90%

among persons who are obese or have diabetes; such patients are also at increasedrisk for the development of advanced fibrosis and cirrhosis.1-4

Recognition of the importance of nonalcoholic fatty liver disease and its strongassociation with the metabolic syndrome1-4 has stimulated interest in its putativerole in the development and progression of cardiovascular disease.5 Accumulatingevidence suggests that cardiovascular disease dictates the outcome (or outcomes)in patients with nonalcoholic fatty liver disease more frequently and to a greaterextent than does the progression of liver disease.2,4,5

This review focuses on the rapidly expanding body of clinical evidence that supports a strong association between nonalcoholic fatty liver disease and the risk of cardiovascular disease. Because of the link between the two disorders, morecareful surveillance of these patients will be needed.

Increased Prevalence of Cardiovascular Disease

Markers of Subclinical Cardiovascular Risk

Patients with nonalcoholic fatty liver disease, both adults and children, typically meet the diagnostic criteria for the metabolic syndrome (i.e., abdominal obesity,hypertension, atherogenic dyslipidemia, and dysglycemia) and therefore have mul-tiple risk factors for cardiovascular disease.1-5 As compared with control subjectswho do not have steatosis, patients with nonalcoholic fatty liver disease have im-paired flow-mediated vasodilatation6 and increased carotid-artery intimal medialthickness7-12 — two reliable markers of subclinical atherosclerosis — that are in-

dependent of obesity and other established risk factors. Although some recent stud-ies have shown no significant association between nonalcoholic fatty liver diseaseand either carotid-artery intimal medial thickness or carotid-artery calcium (asquantified on computed tomography 13,14), a systematic review and meta-analysis of seven cross-sectional studies (involving a total of 3497 subjects) confirmed that nonalcoholic fatty liver disease diagnosed on ultrasonography is strongly associat-ed with increased carotid-artery intimal medial thickness and an increased preva-lence of carotid atherosclerotic plaques.15

In a study from 2006, we found that carotid-artery intimal medial thicknesswas greatest in patients with nonalcoholic steatohepatitis, intermediate in thosewith simple steatosis, and lowest in healthy controls matched for age, sex, and

The New England Journal of Medicine

Downloaded from www.nejm.org by Maduka Sanjeewa on October 7, 2010. For personal use only. No other uses without permission.

Copyright © 2010 Massachusetts Medical Society. All rights reserved.



Th e n e w e n g l a n d j o u r n a l o f medicine

n engl j med 363;14 nejm.org september 30, 20101342

body-mass index (Fig. 1A).10

In addition, thehistologic severity of nonalcoholic steatohepati-tis was associated with the degree of carotid-artery intimal medial thickness, independently of classic cardiovascular risk factors, insulin re-sistance, and metabolic syndrome components(Fig. 1B). Larger studies will be needed to con-firm the reproducibility of these results. Youngpatients with nonalcoholic fatty liver diseasewho are not obese and who do not have diabetesor hypertension have echocardiographic features

of early left ventricular dysfunction16 and im-paired left ventricular energy metabolism, asmeasured on cardiac phosphorus-31 magneticresonance spectroscopy.17

Clinically Manifested Cardiovascular Disease

Given the strong association between nonalco-

holic fatty liver disease and markers of subclini-cal cardiovascular disease, it is certainly not sur-prising that patients with ultrasonographically diagnosed nonalcoholic fatty liver disease have ahigher prevalence of clinically manifested car-diovascular disease than do control subjectswithout steatosis.18-20 In a large study involvingapproximately 3000 unselected patients withtype 2 diabetes, the prevalence of coronary, cere-brovascular, and peripheral vascular disease wasremarkably higher among patients with nonalco-holic fatty liver disease than among those with-

out this disease, independently of traditional risk factors, duration of diabetes, extent of glycemiccontrol, use of lipid-lowering, hypoglycemic, an-tihypertensive, or antiplatelet medications, andcomponents of the metabolic syndrome.18 Thefindings were similar in a study of adults withtype 1 diabetes.19 In a community-based cohort of 2088 male workers, the presence of ultrasono-graphically diagnosed nonalcoholic fatty liver dis-ease was independently associated with an in-creased prevalence of ischemic heart disease.20 In patients consecutively referred for elective cor-onary angiography, nonalcoholic fatty liver dis-ease was associated with more severe coronary artery disease independently of established risk factors.21 Moreover, nonalcoholic fatty liver dis-ease, as assessed by magnetic resonance spec-troscopy, was associated with reduced myocardialperfusion in patients with type 2 diabetes whowere known to have coronary artery disease, inde-pendently of traditional risk factors, visceral fat mass, and insulin sensitivity (as assessed with theuse of the euglycemic hyperinsulinemic clamp).22

Finally, an autopsy study involving 742 childrenshowed that the prevalence of coronary heart dis-ease was increased by a factor of 2 among thosewith nonalcoholic fatty liver disease.23

Increased Incidence of

Cardiovascular Disease

The main prospective and retrospective studiesassessing the relationship between nonalcoholicfatty liver disease (as detected by means of biop-

IntimalMedialThickness

(mm)

2.0

1.2

1.6

0.8

0.4

0.0HealthyControls(N=160)

SimpleSteatosis(N=16)

NASH(N=69)

B

A

P<0.001

IntimalM

edialThickness

(mm)

2.0

1.2

1.6

0.8

0.4

0.0Stage 0(N=23)

Stage 1(N=25)

Stage 2(N=13)

Stage 3(N=8)

P<0.001

FibrosisStageinNASH

Figure 1. Carotid-Artery Intimal Medial Thicknessin Patients with Nonalcoholic Fatty Liver Disease.

Panel A shows the mean (±SD) intimal medial thick-ness of the common carotid artery, as measured on

ultrasonography, in healthy controls, patients withsimple steatosis, and patients with nonalcoholic steato-

hepatitis (NASH). Panel B shows the mean intimal me-dial thickness in patients with NASH according to the

histologic stage of hepatic f ibrosis (from 0, indicatingno fibrosis, to 3, indicating advanced fibrosis; patients

with cirrhosis [i.e., those with stage 4 fibrosis] were notincluded in the study). All data have been adjusted for

age, sex, status with respect to smoking history, body-

mass index, waist circumference, hypertension, dia-betes status, levels of low-density and high-density

lipoprotein cholesterol and triglycerides, and insulinresistance (as estimated on the basis of a homeostasis

model assessment). Data are from Targher et al.10








markers of nonalcoholic fatty liver disease (andshould therefore be interpreted cautiously)1-5 have shown that this disease is associated withan increased risk of cardiovascular disease inde-pendently of alcohol consumption and severalestablished cardiovascular risk factors. In a sys-tematic review and meta-analysis of 11 prospec-

tive studies, Fraser et al. confirmed that an ele-vated serum γ-glutamyltransferase level was anindependent, long-term predictor of incident car-diovascular events in both men and women.35 The meta-analysis of the only 2 prospective stud-ies that used an elevated serum alanine amino-transferase level as a surrogate marker of non-alcoholic fatty liver disease failed to show any independent association with cardiovascular dis-ease outcomes.35 Some of the studies,37,38 but not all of them,32,36,39 confirmed that an increasedserum alanine aminotransferase level is less pre-

dictive of incident cardiovascular disease than isan increased serum γ-glutamyltransferase level,which is thought to be a marker not only of non-alcoholic fatty liver disease but also of oxidativestress.3-5

Liver Ultrasonography

Hamaguchi et al. reported that nonalcoholic fat-ty liver disease diagnosed on ultrasonography ina community-based cohort of healthy adults wasassociated with an increased risk of nonfatal car-diovascular events independently of cardiometa-bolic risk factors (Table 1),40 and nonalcoholicfatty liver disease has also been found to be anindependent predictor of incident cardiovascularevents in patients with type 2 diabetes.41,42 Morerecently, in a population-based study of 4160middle-aged subjects, Haring et al. found that ultrasonography of the liver was useful in pa-tients with increased levels of γ-glutamyltrans-ferase not only for the diagnosis of nonalcoholicfatty liver disease but also for better cardiovascu-lar-risk stratif ication.43

To date, the evidence from published pro-spective studies suggests that patients with non-alcoholic fatty liver disease have multiple risk factors for cardiovascular disease; that cardio-vascular disease is much more common thanliver disease as a cause of death in such patients,especially those with nonalcoholic steatohepati-tis; and that nonalcoholic fatty liver disease islinked to an increased risk of cardiovascularevents both in patients without diabetes and inthose with type 2 diabetes. However, further

study is needed to determine whether nonalco-holic fatty liver disease poses an independent risk above and beyond known cardiovascularrisk factors. Although the evidence for this issuggestive, too few studies have been carried out and they have not been methodologically rigor-ous. Additional large-scale prospective studiesof a more extensive panel of known risk factorsare needed to draw firm conclusions about any independent hepatic contribution to the increasedcardiovascular risk observed among patients withnonalcoholic fatty liver disease.

Putative Mechanisms Linking

Nonalcoholic Fatty Liver Disease

to Car diovascular Disease

From a pathophysiological perspective, there aretwo key questions that should be addressed.First, is nonalcoholic fatty liver disease associat-ed with cardiovascular disease as a consequenceof the shared risk factors, or does nonalcoholicfatty liver disease contribute to cardiovasculardisease independently of these factors? Second,

is the risk of cardiovascular risk also increased inpatients with simple steatosis, or is the necroin-flammatory milieu of nonalcoholic steatohepati-tis a necessary proatherogenic stimulus?

The close correlations among nonalcoholicfatty liver disease, abdominal obesity, and insulinresistance make it extremely difficult to distin-guish the precise causal relationships underlyingthe increased risk of cardiovascular disease amongpatients with nonalcoholic fatty liver disease.

As shown in Figure 2, the biologic mecha-

Figure 2 (facing page). Possible Mechanisms Leadingto Cardiovascular Disease in Patients with Nonalcoholic

Fatty Liver Disease.

The putative underlying mechanisms that link non-

alcoholic fatty liver disease and cardiovascular diseasemight originate from the expanded and inflamed vis-

ceral adipose tissue, with the liver functioning as both

the target of the resulting systemic abnormalities andthe source of several proatherogenic factors. Non-

alcoholic fatty liver disease — especially its necro-inflammatory form, nonalcoholic steatohepatitis —

might play a part in the pathogenesis of cardiovasculardisease through the systemic release of several inflam-

matory, hemostatic, and oxidative-stress mediators orthrough the contribution of nonalcoholic fatty liver dis-

ease to insulin resistance and atherogenic dyslipidemia.HDL denotes high-density lipoprotein, LDL low-density

lipoprotein, NAFLD nonalcoholic fatty liver disease,and NASH nonalcoholic steatohepatitis.






current concepts


Expanded and inflamedvisceral fat mass

Atherothrombosis

NAFLD or NASH

Increased inflammatory cytokines

Increased insulin resistance

Increased free fatty acids

Increased inflammatory cytokines

Increased insulin resistance

Chronic inflammation(e.g., increases in C-reactive protein, interleukin-6, tumor necrosis factor α,

and other acute-phase proteins)

Hypercoagulation and hypofibrinolysis(e.g., increases in fibrinogen, factor VII, plasminogen activator inhibitor 1,

and other coagulation factors)

Atherogenic dyslipidemia(e.g., increased triglycerides, decreased HDL cholesterol, increased small, dense LDL cholesterol,

postprandial lipemia)

Dysglycemia and (hepatic) insulin resistance








nisms potentially responsible for acceleratedatherogenesis in nonalcoholic fatty liver diseaseprobably have their origin in the expanded vis-ceral adipose tissue, with the liver being boththe target of the resulting systemic abnormali-ties and a source of proatherogenic moleculesthat amplify the arterial damage.

Visceral Obesity, Inflammation, and Insulin

Resistance

Expanded and inflamed visceral adipose tissuereleases a wide array of molecules potentially in-volved in the development of insulin resistanceand atherosclerosis, including free fatt y acids, in-terleukin-6, tumor necrosis factor α (TNF-α),monocyte chemotactic protein 1 (also known asCC chemokine ligand 2), and other proinflam-matory cytokines.45-49 These cytokines may de-rive from adipocytes themselves, infiltrating

macrophages, or both.45-49 As reviewed in detailelsewhere,47,49 the resulting adipose-tissue in-flammation is one of the earliest steps in thechain of events leading to insulin resistance, es-pecially in obese and overweight persons. Activa-tion of proinflammatory pathways is mediatedby cytokine receptors and pattern-recognition re-ceptors, including toll-like receptors and recep-tors for advanced glycation end products, whichare gatekeepers of the innate immune sys-tem.47,49,50 These pathways converge on two mainintracellular transcription factor–signaling path-ways — namely, the nuclear factor κB (NF-κB)pathway, which is activated by the inhibitor of NF-κB kinase beta, and the c-Jun N-terminal ki-nase (JNK) pathway.46-49 Experimental findingsin mice indicate that the activation of JNK 1 inadipose tissue can translate into insulin resis-tance within the liver.51

Some evidence suggests that in lean persons,insulin resistance may be dissociated from adi-pose-tissue inf lammation in the earliest phases,and this dissociation seems to be due mainly to

cellular lipid accumulation in skeletal muscleand inhibition of the insulin-signaling cascade.52 Insulin resistance in skeletal muscle is in turnassociated with hyperinsulinemia in peripheraland portal veins, which promotes hepatic insu-lin resistance and hepatic steatosis, at least inpart by inducing hepatic lipogenesis mediated by sterol regulatory element–binding protein 1c,and by inhibiting fatty acid oxidation.52,53

Inflammation, Coagulation, and Disordered

Lipid Metabolism

Hepatic steatosis results from increased hepaticuptake of free fatty acids derived mainly from thehydrolysis of adipose-tissue triglycerides (in-creased because of insulin resistance) but alsofrom dietary chylomicrons and hepatic lipogen-

esis.1-4,46-49 Insulin resistance is a pathogenicfactor in the development and progression of nonalcoholic fatty liver disease1-4,46-49 and alsoplays a major role in the development of the met-abolic syndrome and cardiovascular disease.54

In the presence of increased free fatty acidflux and chronic, low-grade inflammation, theliver is again both the target of and a contributorto systemic inflammatory changes. Activation of the NF-κB pathway in the liver of patients withnonalcoholic steatohepatitis leads to increasedtranscription of several proinflammatory genes

that amplify the systemic, low-grade inflamma-tion.48,49 Hepatic steatosis is associated withincreased production of interleukin-6 and otherproinflammatory cytokines by hepatocytes andnonparenchymal cells, including Kupffer cellsand hepatic stellate cells.46-49 Increased intrahe-patic cytokine expression results from local acti-vation of the NF-κB pathway, as mediated by hepatocellular damage and fat-derived factors,and is likely to play a key role in the progressionof nonalcoholic fatty liver disease46-49 and car-diovascular disease.4,5

As recently reviewed in detail,55 some studieshave shown that a number of the genes involvedin fatty acid metabolism, lipolysis, monocyte andmacrophage recruitment, coagulation, and inflam-mation are overexpressed in patients with non-alcoholic fatty liver disease. Moreover, in a num-ber of case–control studies, circulating levels of several inflammatory markers (e.g., C-reactiveprotein, interleukin-6, monocyte chemotactic pro-tein 1, and TNF-α), procoagulant factors (e.g.,plasminogen activator inhibitor 1 [PAI-1], fibrino-

gen, and factor VII), and oxidative stress mark-ers (e.g., oxidized low-density lipoprotein cho-lesterol, thiobarbituric acid–reacting substances,and nitrotyrosine) are highest in patients withnonalcoholic steatohepatitis, intermediate in thosewith simple steatosis, and lowest in control sub-jects without steatosis, and the differences areindependent of obesity and other potentially con-founding factors.55 Notably, some studies also






current concepts


showed a strong, graded relationship betweenintrahepatic messenger RNA expression of C-reactive protein, interleukin-6, or PAI-1 and theseverity of histologic changes in patients withnonalcoholic fatty liver disease.56-58 Recently, wealso reported that men with nonalcoholic steato-hepatitis had higher plasma high-sensitivity C-reactive protein, fibrinogen, and PAI-1 activity

levels and lower adiponectin levels than did over-weight men without steatosis but with similarlevels of visceral adiposity, suggesting that non-alcoholic steatohepatitis can contribute to a moreatherogenic risk profile over and above the con-tribution of visceral adiposity.59 This hypothesiswas supported by the strong, graded relation-ships of these plasma inflammatory and proco-agulant markers with the histologic severity of

nonalcoholic steatohepatitis, independently of age, visceral adiposity, and other metabolic ab-normalities (Fig. 3).59

The atherogenic role of hepatic necroinflam-mation, which is a feature of nonalcoholic steato-hepatitis, is supported by the observation that cardiovascular risk is greater among patientswith nonalcoholic steatohepatitis than among

those with simple steatosis24,26-29

and by the ob-servation that the risk of incident cardiovascularevents is strongly associated with elevated serumliver enzyme levels — a marker of hepatic necro-inflammation. We have also found that patientswith nonalcoholic steatohepatitis and those withchronic viral hepatitis both have markedly great-er carotid-artery intimal medial thickness thando healthy control subjects, which is consistent

hs-CRP

(mg/liter)

6

5

4

3

1

2

0

A

Stage0(N=13)

Stage1(N=16)

Stage2(N=10)

Stage3(N=6)

P<0.001

PAI-1Activity(AU/ml)

50

40

30

10

20

0

B

Stage0(N=13)

Stage1(N=16)

Stage2(N=10)

Stage3(N=6)

P<0.001

Adiponectin(µg

/liter)

14

12

10

8

6

2

4

0

C

Stage0(N=13)

Stage1(N=16)

Stage2(N=10)

Stage3(N=6)

P<0.001

Fibrinogen(g/

liter)

6

5

4

3

1

2

0

D

Stage0(N=13)

Stage1(N=16)

Stage2(N=10)

Stage3(N=6)

P<0.001

Figure 3. Inflammatory and Procoagulant Biomarkers in Patients with Nonalcoholic Steatohepatitis.

Mean (±SD) plasma high-sensitivity C-reactive protein (hs-CRP), plasminogen-activator inhibitor 1 (PAI-1) activity,adiponectin, and fibrinogen in relation to the histologic stage of fibrosis are shown for 45 overweight male patients

with histologically defined nonalcoholic steatohepatitis (NASH). AU denotes arbitrary units. P values for trends weredetermined by means of one-way analysis of variance. Data are from Targher et al. 59








with the hypothesis that liver inflammation playsa role in the pathogenesis of cardiovascular dis-ease.60

Ample evidence indicates that nonalcoholicfatty liver disease, especially in its necroinflam-matory form (nonalcoholic steatohepatitis), canexacerbate both hepatic and systemic insulin

resistance and promote the development of atherogenic dyslipidemia,3-5,10,47,48 thus favoringprogression of cardiovascular disease. Finally,nonalcoholic fatty liver disease may also contrib-ute to cardiovascular risk through abnormal lipo-protein metabolism, especially during the post-prandial phase.61,62

Further research is required to define the ma-jor sources of some proinflammatory and proco-agulant mediators (i.e., to determine the relativecontributions of visceral adipose tissue and theliver itself), as well as to uncover other specific

mechanisms by which nonalcoholic fatty liverdisease and nonalcoholic steatohepatitis may con-tribute to the development and progression of cardiovascular disease.

Conclusions

Nonalcoholic fatty liver disease has emerged as agrowing public health problem worldwide. In-creases in morbidity and mortality from cardio-vascular disease are probably among the most important clinical features associated with non-alcoholic fatty liver disease. To date, there is agrowing body of evidence suggesting that car-diovascular disease is the leading cause of deathin patients with advanced nonalcoholic fatty liverdisease and that nonalcoholic fatty liver diseaseis associated with an increased risk of incident cardiovascular disease that is independent of therisk conferred by traditional risk factors andcomponents of the metabolic syndrome. Althoughadditional research is required to draw a defini-tive conclusion, these observations raise the

possibility that nonalcoholic fatty liver disease— especially its necroinflammatory variant,nonalcoholic steatohepatitis — not only is amarker of cardiovascular disease but also may beinvolved in its pathogenesis. This process may occur through the systemic release of proathero-genic mediators from the steatotic and inflamedliver or through the contribution of nonalcoholicfatty liver disease itself to insulin resistance andatherogenic dyslipidemia, which are important risk factors for cardiovascular disease.

The treatment strategies for nonalcoholicfatty liver disease and cardiovascular disease aresimilar, aimed primarily at reducing insulin re-sistance and modifying the associated cardio-metabolic risk factors.1-5,63 Pharmacotherapy fornonalcoholic fatty liver disease should probably be reserved for patients with nonalcoholic ste-

atohepatitis who are at highest risk for diseaseprogression. The lack of data from large, ran-domized, controlled trials with both histologicfollow-up and cardiovascular end points makesit difficult to offer definitive recommendationsregarding the treatment of nonalcoholic steato-hepatitis. Current recommendations are limitedto weight reduction by means of diet and exer-cise and to the treatment of individual compo-nents of the metabolic syndrome with the use of therapies that may have beneficial hepatic effects,including bariatric surgery for obesity, insulin

sensitizers (metformin and thiazolidinediones)for type 2 diabetes, and drugs directed at therenin–angiotensin–aldosterone system to controlhypertension.4,63-66 Pioglitazone is probably thethiazolidinedione of choice, since most of theevidence supporting a beneficial effect of thisclass of drugs on nonalcoholic steatohepatitiscomes from studies of pioglitazone. Unlike rosi-glitazone, pioglitazone has not been associatedwith an increased risk of cardiovascular events.67,68 There is no convincing evidence that lipid-lower-ing agents, including statins, are beneficial forpatients with nonalcoholic steatohepatitis; how-ever, they can be safely prescribed for conven-tional indications, such as diabetes and highcardiovascular risk, since there is no evidencethat patients with preexisting nonalcoholic fatty liver disease are at increased risk for statin-induced idiosyncratic hepatotoxicity or that statinsare associated with an increased frequency of he-patic steatosis or serum alanine aminotransferaseabnormalities in these patients.69 Preliminary evi-dence also supports a role for antioxidants, anti-

cytokine agents, and hepatoprotectants, includ-ing bile acids4,70; however, there are insufficient data to either support or refute the use of theseagents as standard therapy for patients withnonalcoholic fatty liver disease.

It is not known whether ameliorating nonal-coholic fatty liver disease will ultimately prevent or slow the development and progression of car-diovascular disease. Moreover, the prognosticvalue of nonalcoholic fatty liver disease in car-diovascular risk stratification remains debatable.






current concepts


Nevertheless, the strong association betweennonalcoholic fatty liver disease and cardiovascu-lar risk deserves particular attention in view of itspotential implications for screening and surveil-lance strategies in clinical practice. The current body of evidence argues for careful monitoringand evaluation of the risk of cardiovascular dis-

ease in all patients with nonalcoholic fatty liverdisease. Such patients, especially those with non-alcoholic steatohepatitis, are candidates not only for early treatment of their liver disease but also

for early and aggressive treatment aimed at theirassociated cardiovascular risk factors, becausemany patients with more severe forms of nonal-coholic fatty liver disease will have major cardio-vascular events and will ultimately die fromcardiovascular disease before advanced liver dis-ease develops.

No potential conflict of interest relevant to this article wasreported.

Disclosure forms provided by the authors are available withthe full text of this art icle at NEJM.org.

References

1. Angulo P. Nonalcoholic fatty liver dis-ease. N Engl J Med 2002;346:1221-31.2. Marchesini G, Moscatiello S, Di Dom-izio S, Forlani G. Obesity-associated liverdisease. J Clin Endocrinol Metab 2008;93:Suppl 1:S74-S80.3. Kotronen A, Yki-Järvinen H. Fatty liver:a novel component of the metabolic syn-drome. Arterioscler Thromb Vasc Biol2008;28:27-38.4. de Alwis NMW, Day CP. Non-alcoholicfatty liver disease: the mist gradually clears. J Hepatol 2008;48:Suppl 1:S104-S112.5. Targher G, Marra F, Marchesini G.Increased risk of cardiovascular diseasein nonalcoholic fatty liver disease: causaleffect or epiphenomenon? Diabetologia2008;51:1947-53.6. Villanova N, Moscatiello S, Ramilli S,et al. Endothelial dysfunction and cardio-vascular risk profile in nonalcoholic fatty liver disease. Hepatology 2005;42:473-80.

7. Targher G, Bertolini L, Padovani R,Zenari L, Zoppini G, Falezza G. Relationof nonalcoholic hepatic steatosis to early carotid atherosclerosis in healthy men:role of visceral fat accumulation. DiabetesCare 2004;27:2498-500.8. Brea A, Mosquera D, Martin E, AriztiA, Cordero JL, Ros E. Nonalcoholic fatty liver disease is associated with carotidatherosclerosis: a case-control study.Arterioscler Thromb Vasc Biol 2005;25:1045-50.9. Volzke H, Robinson DM, Kleine V, et al. Hepatic steatosis is associated with anincreased risk of carotid atherosclerosis.World J Gastroenterol 2005;11:1848-53.

10. Targher G, Bertolini L, Padovani R, et al. Relations between carotid artery wallthickness and liver histology in subjectswith nonalcoholic fatty liver disease. Dia-betes Care 2006;29:1325-30.11. Fracanzani AL, Burdick L, Raselli S,et al. Carotid artery intima-media thick-ness in nonalcoholic fatty liver disease.Am J Med 2008;121:72-8.12. Kim HC, Kim D, Huh KB. Associationbetween nonalcoholic fatty liver diseaseand carotid intima-media thickness ac-cording to the presence of metabolic syn-drome. Atherosclerosis 2009;204:521-5.

13. McKimmie RL, Daniel KR, Carr JJ, et al. Hepatic steatosis and subclinical car-diovascular disease in a cohort enrichedfor type 2 diabetes: the Diabetes Heart Study. Am J Gastroenterol 2008;103:3029-35.14. Petit JM, Guiu B, Terriat B, et al. Non-alcoholic fatty liver is not associated withcarotid intima-media thickness in type 2diabetic patients. J Clin Endocrinol Metab2009;94:4103-6.15. Sookoian S, Pirola CJ. Non-alcoholicfatty liver disease is strongly associatedwith carotid atherosclerosis: a systematicreview. J Hepatol 2008;49:600-7.16. Goland S, Shimoni S, Zornitzki T, et al.Cardiac abnormalities as a new manifes-tation of nonalcoholic fatty liver disease:echocardiographic and tissue Dopplerimaging assessment. J Clin Gastroenterol2006;40:949-55.17. Perseghin G, Lattuada G, De CobelliF, et al. Increased mediastinal fat and im-

paired left ventricular energy metabolismin young men with newly found fatt y liver.Hepatology 2008;47:51-8.18. Targher G, Bertolini L, Padovani R, et al. Prevalence of nonalcoholic fatty liverdisease and its association with cardio-vascular disease among type 2 diabeticpatients. Diabetes Care 2007;30:1212-8.19. Targher G, Bertolini L, Padovani R, et al. Prevalence of non-alcoholic fatty liverdisease and its association with cardio-vascular disease in patients with type 1diabetes. J Hepatol 2010;53:713-8.20. Lin YC, Lo HM, Chen JD. Sonographicfatty liver, overweight and ischemic heart disease. World J Gastroenterol 2005;11:

4838-42.21. Mirbagheri SA, Rashidi A, Abdi S,Saedi D, Abouzari M. Liver: an alarm forthe heart? Liver Int 2007;27:891-4.22. Lautamäki R, Borra R, Iozzo P, et al.Liver steatosis coexists with myocardialinsulin resistance and coronary dysfunc-tion in patients with type 2 diabetes. AmJ Physiol Endocrinol Metab 2006;291:E282-E290.23. Schwimmer JB, Deutsch R, Behling C,Lavine JE. Fatty liver as a determinant of atherosclerosis. Hepatology 2005;42:Suppl:610A. abstract.

24. Matteoni CA, Younossi ZM, GramlichT, Boparai N, Liu LC, McCullough AJ.Nonalcoholic fatty liver disease: a spec-trum of clinical and pathological severity.Gastroenterology 1999;116:1413-9.25. Dam-Larsen S, Franzmann M, Ander-sen IB, et al. Long-term prognosis of fatty liver: risk of chronic liver disease anddeath. Gut 2004;53:750-5.26. Adams LA, Lymp JF, St Sauver J, et al.The natural history of nonalcoholic fatty liver disease: a population-based cohort study. Gastroenterology 2005;129:113-21.27. Rafiq N, Bai C, Fang Y, et al. Long-term follow-up of patients with nonalco-holic fatty liver. Clin Gastroenterol Hepa-tol 2009;7:234-8.28. Ekstedt M, Franzén LE, MathiesenUL, et al. Long-term follow-up of patientswith NAFLD and elevated liver enzymes.Hepatology 2006;44:865-73.29. Söderberg C, Stål P, Askling J, et al.Decreased survival of subjects with elevat-

ed liver function tests during a 28-yearfollow-up. Hepatology 2010;51:595-602.30. Ruttmann E, Brant LJ, Concin H,Diem G, Rapp K, Ulmer H. Gamma-glu-tamyltransferase as a risk factor for car-diovascular disease mortality: an epide-miological investigation in a cohort of 163,944 Austrian adults. Circulation 2005;112:2130-7.31. Lee DH, Silventoinen K, Hu G, et al.Serum gamma-glutamyltransferase pre-dicts non-fatal myocardial infarction andfatal coronary heart disease among 28,838middle-aged men and women. Eur Heart J2006;27:2170-6.32. Schindhelm RK, Dekker JM, Nijpels

G, et al. Alanine aminotransferase pre-dicts coronary heart disease events: a 10-year follow-up of the Hoorn Study. Athero-sclerosis 2007;191:391-6.33. Lee DS, Evans JC, Robins SJ, et al.Gamma-glutamyltransferase and meta-bolic syndrome, cardiovascular disease,and mortality risk: the Framingham Heart Study. Arterioscler Thromb Vasc Biol2007;27:127-33.34. Wannamethee SG, Lennon L, ShaperAG. The value of gamma-glutamyltrans-ferase in cardiovascular risk prediction inmen without diagnosed cardiovascular




risk of cardiovascular disease in patients

Documents