REVIEW

Diabetes and liver disease: An ominous

association
Simona Moscatiello, Rita Manini, Giulio Marchesini*
Unit of Metabolic Diseases, Alma Mater Studiorum University,
Policlinico S. Orsola, Via Massarenti 9, I-40138 Bologna, Italy
Received 28 July 2006; accepted 24 August 2006
KEYWORDS
Liver disease;
Cirrhosis;
Non-alcoholic
fatty liver;
Type 2 diabetes
mellitus;
Liver transplantation;
Viral hepatitis
Abstract Diabetes mellitus and advanced liver disease are associated with each
other more frequently than expected by chance, and such an association carries
a signicant risk of morbidity and mortality. A metabolic pathway leading to ad-
vanced liver disease via fatty liver and steatohepatitis has been demonstrated, fur-
ther supporting the possibility that cirrhosis may be a late complication of diabetes.
In addition, an interaction between hepatitis C virus (HCV) and insulin resistance
increases the overall prevalence of associated diseases, through largely unidenti-
ed mechanisms. Extensive prospective monitoring of non-alcoholic fatty liver dis-
ease cases, analysis of insulin signaling in HCV-infected patients using molecular
biology techniques, and intervention studies, will help to clarify the mechanisms
of action of the possible clinical strategies, the predictive value of biochemical, his-
tological, and clinical markers, and the effectiveness of treatments available.
2006 Elsevier B.V. All rights reserved.
Introduction
The association between liver disease and diabetes
mellitus (DM) is well known, the overall prevalence
being signicantly higher than that expected by
a chance association of two very common diseases.
The key role of the liver in blood glucose
controldthe hepatic metabolism of insulindas
well as the inuence of liver disease on peripheral
glucose metabolism and whole-body insulin sensi-
tivity, contribute to DM in the presence of ad-
vanced liver disease. More recently, new insights
into this association came from the recognition
that: (a) DM itself may be a cause of liver disease,
via non-alcoholic fatty liver disease (NAFLD), non-
alcoholic steatohepatitis (NASH), cirrhosis, and
ultimately hepatocellular carcinoma; (b) hepatitis
C virus may have direct diabetogenic effects; and
(c) post-transplantation DM is a major cause of
morbidity and mortality in subjects following liver
transplantation. We shall revise the most recent
evidence for an association between DM and liver
* Corresponding author. Tel.: 39 051 6364889; fax: 39 051
6364502.
E-mail address: giulio.marchesini@unibo.it (G. Marchesini).
0939-4753/$ - see front matter 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.numecd.2006.08.004
Nutrition, Metabolism & Cardiovascular Diseases (2007) 17, 63e70
www.elsevier.com/locate/nmcd
disease, with particular emphasis on the thera-
peutic aspects.
Diabetes in advanced liver disease
An insulin-resistant state may be demonstrated in
approximately 80% of patients with cirrhosis, and
20e63% of them will develop DM (hepatogenous
diabetes). The prevalence is largely dependent on
the diagnostic criteria, etiology, and the time
from the diagnosis of cirrhosis. Once cirrhosis is
established, hyperglycemia develops in up to 20%
of cases within 5 years, but the clinical character-
istics and the course of diabetes are different
from those observed in the absence of liver
disease.
Patients acquiring diabetes as a result of
cirrhosis differ from typical type 2 DM patients
by having a lower prevalence of family history
of diabetes and a lower risk of macro- and
microangiopathic complications [1]. In a point-
prevalence study, the prevalence of micro- and
peripheral macroangiopathy and coronary heart
disease in cirrhosis with DM was comparable to
that of controls, and was signicantly lower
than that observed in randomly selected patients
with type 2 DM [2]. In a retrospective/prospec-
tive study, the prognostic signicance of DM
was analyzed in a group of almost 400 patients
with cirrhosis (25% with DM) [3]. The larger mor-
tality rate in DM patients was not due to the
classical diabetes-related complications, but to
an increased risk of hepatocellular failure. The
prognostic signicance of DM was particularly rel-
evant when patients who died of gastrointestinal
bleeding were excluded. The long-term prognosis
is thus determined to a greater extent by the
primary hepatic disease and its complications,
rather than by DM-related complications. The
low prevalence of micro- and macrovascular dis-
ease may be related to a shorter duration of DM,
which may be related in turn to a reduced life
expectancy, as well as to liver disease-induced
abnormalities (low cholesterol, low platelet
count, etc.) protecting the cardiovascular system
from atherosclerosis.
The ultimate stage of cirrhosis may be the
development of hepatocellular carcinoma (HCC).
DM is one of the most common complications
observed in HCC, but very few data exist on the
impact of DM on the survival of HCC patients. In
HCC patients undergoing surgical or non-surgical
therapy, DM is associated with a poorer long-term
prognosis. This is not the result of DM enhancing
the progression of HCC or its recurrence, but is
due to the induction of a rapid decline in residual
liver function; such an effect is inuenced by
treatment strategies and by tumor and/or cirrho-
sis-related factors [4,5].
Therapeutic implications
Any therapeutic intervention in patients with DM
and cirrhosis must be evaluated in terms of risks
and benets. The relatively low impact of DM on
prognosis is a reason for adopting a non-aggressive
approach to metabolic control. It can rarely be
achieved solely by diet, particularly as dietary
restrictions should be avoided in hypercatabolic
patients with cirrhosis. Oral hypoglycemic agents
are contraindicated in the presence of advanced
disease due to the possible risk of lactic acidosis
(biguanides) or the potential toxic effects in the
presence of reduced hepatic or renal metabolism
(sulfonylureas) [6]. Thiazolidinediones are rarely
used after the unfortunate experience with trogli-
tazone [7]. Thus, insulin remains the elective
treatment; however, insulin-resistance, a variable
hepatic insulin metabolism, and difculties in die-
tary management, make glucose control extremely
difcult to achieve.
Acarbosemay beareasonabletherapeutic option
to improve postprandial glucose control in cirrhotic
patients with insulin-treated DM [8]. In a recent
study [9], acarbose was also tested for the treat-
ment of cirrhotic patients with DM and low-grade
hepatic encephalopathy in a placebo-controlled
trial. Acarbose improved glucose controldmainly
postprandial glucosedand reduced glycosylated
hemoglobin without deleterious effects on liver
function. This therapeutic option might thus be
particularly appealing for improving hepatic
encephalopathy.
Diabetes as a cause of liver disease
DM per se may generate liver disease of metabolic
origin (the so-called NAFLD) (Table 1), in associa-
tion with obesity, dyslipidemia and hypertension.
NAFLD is now considered as the hepatic manifesta-
tion of the metabolic syndrome, and is present in
approximately 80% of type 2 DM. The epidemiol-
ogy, etiology and pathogenesis of NAFLD are
beyond the scope of this review, and we invite
the readers to refer to an excellent recent review
article [10].
In NAFLD, DM constitutes a risk factor for NASH
and advanced, progressive liver disease [11,12]. In
103 patients who underwent serial liver biopsies,
diabetes (P 0.007) and a low initial brosis stage
64 S. Moscatiello et al.
(P < 0.001) were associated with a higher rate of
brosis progression, as was higher body mass index
(BMI; P 0.008), when patients with cirrhosis were
excluded [12]. Diabetic patients with elevated
BMI are thus particularly at risk for higher rates
of brosis progression; indeed, obesity per se is
an additional risk factor for advanced disease
[13]. Similarly, in subjects with a previous diagno-
sis of fatty liver, DM increases the risk of signicant
liver disease and HCC in the long term [14].
Very few data are so far available on the natural
history of NAFLD in the community. In Olmsted
County, Minnesota, the survival of subjects with
a diagnosis of NAFLD was lower than that for the
general population (standardized mortality ratio,
1.34; 95% CI 1.003e1.76; P 0.03), and the higher
mortality was associated with age, impaired fasting
glucose (hazard ratio, 2.6; 95% CI 1.3e5.2), and cir-
rhosis [15]. Liver disease was the third leading cause
of death (as compared with the thirteenth cause of
death in the general Minnesota population), al-
though the absolute risk remained low. The nal
prognosis of NAFLD may be dictated by the severity
of necroinammation and brosis. In a Danish se-
ries, the mortality for liver-related complications
was reported to be increased in NAFLD [16], but
not in subjects with pure fatty liver [17].
Liver disease constitutes a signicant comorbid-
ity in DM. The prevalence of ultrasound-assessed
fatty liver is very high and, considering the low
sensitivity of ultrasounds for fatty liver detection,
we can assume that nearly 100% of cases have
excess fat deposition. Aminotransferases are ele-
vated in several cases, but the histology of severe
liver disease may also be present in subjects
with normal liver enzymes [18]. In a recent point-
prevalence study in over 9000 type 2 DM patients in
eight Italian centers, the overall prevalence of
raised alanine and aspartate aminotransferases
(ALT and AST) was 16% and 8.8%, respectively
(G. Marchesini, unpublished). Although a few cases
may have viral hepatitis (see below), there is evi-
dence that in virus-related liver disease DM may
be an additional damage-amplifying factor.
The long-term effects of liver disease in DM
have been tested in an Italian series. In patients
enrolled in the Verona Study [19], the age-stan-
dardized cirrhosis-related death was increased by
a factor of 2.5 compared with the general Verona
populationda gure exceeding the standardized
mortality ratio for cardiovascular disease. The
original 5-year results have recently been con-
rmed in a 10-year analysis [20], where a signi-
cant risk for HCC was also reported in association
with obesity.
A comprehensive analysis of the association of
DM with HCC has recently been published [21]. The
risk appears to be mediated by obesity [22] and
hepatitis viruses (see below), but a specic role
of insulin resistance or glucose/insulin dysregula-
tion cannot be excluded.
Therapeutic implications
Several issues must be considered in the treatment
of diabetes-associated NAFLD (Table 2), and the
nal therapeutic options are not necessarily
mutually exclusive (Table 3):
Table 1 Principal clinical, laboratory, histologic and diagnostic characteristics of non-alcoholic fatty liver disease
1. Clinical data (a) Usually asymptomatic, sometimes mild
right-upper quadrant discomfort
(b) Association with diabetes, obesity, dyslipidemia
(hypertriglyceridemia, low HDL-C, hypobetalipoproteinemia),
metabolic syndrome
2. Laboratory data (a) Elevated alanine and aspartate aminotransferases
(ALT, AST), elevated gamma-glutamyltransferase
(usually (2e5 times upper normal limits)
(b) ALT to AST ratio >1
(c) Hyperinsulinemia and insulin resistance
(d) Dyslipidemia (see above)
3. Liver histology
a
(a) Steatosis (fatty inltration >5% hepatocytes)
(b) Necroinammation (lobular or portal inammation,
Mallory bodies, ballooning)
(c) Fibrosis (perisinusoidal, perivenular, bridging, cirrhosis)
4. Diagnostic criteria (a) Imaging indicative of fatty inltration (ultrasonography,
magnetic resonance, magnetic resonance spectroscopy)
(b) Alcohol intake <140 g/week (females) or 140e210 g/week (males)
(c) Exclusion of liver disease of viral, autoimmune, genetic origin
a
The presence of brosis and necroinammation characterizes non-alcoholic steatohepatitis (NASH).
Diabetes and the liver 65
Diet, exercise, and lifestyle changes
The amount of fat in the diet may regulate hepatic
triglyceride deposition [23]. A few studies have
testedtheeffectiveness of intensivebehavior treat-
ment in NAFLD. Huang et al. [24] reported histologic
improvement in NASH patients who successfully
completed a 1-year intense dietary intervention
study. In a large epidemiologic study of a Japanese
cohort, the adherence to lifestyle advice paralleled
the changes in aminotransferase levels, but no data
were reported on histology [25,26]. There is evi-
dence that a healthy lifestyle may also reduce the
burden of liver disease in DM patients, but no spe-
cic studies are available. Lifestyle changes used
in the prevention of DM are reasonable targets for
NAFLD treatment [27,28].
Insulin-sensitizers
Improved insulin sensitivity is expected to remove
liver fat, with possible benecial effects on necro-
inammation and brosis. Both metformin and
glitazones proved effective in reducing amino-
transferase levels in all but one study [29e33], and
in improving histology [29e32]; however, only met-
formin has shown positive results in a randomized,
controlled trial [34], although in this case the lack
of comparative histologic data in the control group
is a serious drawback. Aminotransferase levels re-
turn to pre-treatment levels after the treatment is
terminated [30,31], suggesting that it be continued
indenitely to avoid disease progression. Insulin-
sensitizers are the rst-line agents in type 2 DM
and most DM patients probably receive these, irre-
spective of liver disease. We need to know whether
such benecial effects also extend to DM patients.
Antioxidant and cytoprotective agents
Vitamins (E and C), betaine, taurine, n-acetyl-
cysteine, sibilin, ursodeoxycholic acid and other
drugs have been tested in many pilot, uncontrolled
studies. Vitamin E is probably the treatment of
choice in obese children with NAFLD, since no side
effects are expected. However, there is no clear
evidence of the effectiveness of vitamin E in pilot
studies, and in controlled studies its advantage
over placebo is doubtful [35].
Table 2 Problems in the treatment of diabetes-
associated non-alcoholic fatty liver disease
(a) Unhealthy lifestyles should be corrected by both
dietary counseling and physical activity
(b) Diabetes treatment, according to specic
guidelines, and the achievement of a good
metabolic control are mandatory
(c) Only continuous treatments are expected to be
effective
(d) The effectiveness of specic treatment of liver
disease should be measured on the basis of hard
outcome measures (histology)
(e) Repeated liver biopsies are not feasible in the
majority of cases
Table 3 Therapeutic options in non-alcoholic fatty liver disease/non-alcoholic steatohepatitis (for a detailed
review, see [10])
1. Lifestyle changes (a) Weight reduction
(b) Reduce total fat intake to <30% energy
(c) Replace saturated with unsaturated fats
(d) Increase ber intake to >15 g/day
(e) Increase physical activity to >150 min/week
2. Insulin-sensitizing agents (a) Metformin
(b) Thiazolidinediones
(i) Pioglitazone
(ii) Rosiglitazone
3. Antioxidants and cytoprotective agents (a) Vitamin E and C
(b) Betaine
(c) Taurine
(d) N-Acetylcysteine
(e) Sibilin
(f) Ursodeoxycholic acid
(g) Fibrates and statins
(h) Orlistat
4. Other treatments and future areas of research (a) Anti TNF-a
(b) Pentoxifylline
(c) Angiotensin II receptor antagonists
66 S. Moscatiello et al.
Ursodeoxycholic acid (UDCA) treatment was the
sole pharmacologic agent tested in a randomized,
placebo-controlled study using histology as the
outcome measure [36]. In 166 patients with biopsy-
proven NASH randomized to receive 13e15 mg/kg
per day of UDCA or placebo for 2 years, serum liver
biochemistries were stable or improved in both
the experimental and the control arm. Changes in
the degree of steatosis, necroinammation or bro-
sis at biopsy (107 patients) in the experimental arm
were not signicantly different from changes in
placebo. In conclusion, UDCA does not constitute
an effective treatment for NASH, but its potential
use in pure fatty liver to prevent disease progression
to the brotic stage remains to be determined.
Other treatments and future areas of research
Considering the primary role of hepatic stellate
cells and cytokines in the development and pro-
gression of liver brosis, this is likely to be a key
target area for drug treatment of NAFLD in the
future. Pentoxifylline inhibits TNF-a production,
and has been tested in 20 patients with biopsy-
conrmed NASH [37]. No serious adverse events oc-
curred, but nine patients withdrew from the study,
primarily because of nausea. Aminotransferase
levels improved with treatment, but no histology
data were presented. Also the renin-angiotensin
system has a role in hepatic brosis, and losartan
treatment has been tested in seven patients with
NASH and hypertension in a 48-week pilot study
[38]. Treatment signicantly reduced the blood
markers of hepatic brosis and aminotransferase
levels. At 1 year, hepatic necroinammation was
improved in ve patients and brosis in four pa-
tients, without any side effects reported. Angioten-
sin II receptor antagonists may thus become
indispensable in DM with liver disease, as they are
in the presence of renal impairment.
Diabetes and chronic hepatitis C (CHC)
HCV infection primarily causes liver disease, but
convincing evidence has been accumulating in the
last fewyears regarding its association with impaired
glucose metabolism and DM. A study from the US
reported a threefold greater risk of type 2 DMin HCV-
positive subjects [39], and the prevalence of HCVan-
tibody among DM patients ranged from 4.2 to 28%
[40]. Obesity and advanced histologic stage are asso-
ciated with DM in HCV patients [40], but a threefold
increase in the prevalence of glucose abnormalities
is present in HCV-positive cases much earlier than
the onset of cirrhosis [41], highlighting a specic
role of HCV in the action of insulin [42].
Steatosis might be the link between HCV, DM
and disease severity. Steatosis is present in 73% of
patients infected with genotype 3 and in 50% of
patients infected with genotypes other than 3 [43].
In a large meta-analysis on individual data from
3068 patients with histologically conrmed CHC
from ve countries [44], steatosis (50.9% of cases)
was independently associated with HCV genotype
3, liver brosis and inammation, and diabetes.
In genotype 1 cases [45], where steatosis is less
common than in genotype 3, moderate/severe
steatosis was independently associated with insu-
lin resistance, as assessed by the homoeostasis
model assessment (HOMA) score, and predicted
brosis. Notably, the presence of insulin resistance/
steatosis is a well known prognostic factor for
poor response to antiviral treatment [46,47]dan
additional cause of disease progression.
Therapeutic implications
The sequence of events from HCV infection, insulin
resistance, steatosis, brosis and disease progres-
sion is largely unexplained, and has been discussed
elsewhere [48]. The importance of insulin resis-
tance and steatosis in CHC is the pathological basis
for an ongoing Italian intervention trial with an
insulin sensitizer (metformin) before and during
interferon treatment, to improve sustained virologic
response to antiviral treatment.
Post-transplantation diabetes mellitus
Registry data show a 16% prevalence of pre-existing
diabetes among patients undergoing liver trans-
plantation [49]. This overall rate derives from a
balance of factors associated with DM (i.e. the
underlying advancedliver disease, previous cortico-
steroidtreatment for autoimmune hepatitis, the as-
sociation between liver disease and the metabolic
syndrome [49]), coupled with DM being a relative
or absolute contraindication to transplantation
[50]. Of note, the metabolic syndrome-associated
NAFLD recurs after liver transplantation [51].
Pre-existing DM affects outcome; in the United
Network for Organ Sharing Database, 5-year patient
and graft survival was signicantly lower for type 1
DM compared with type 2 DM recipients, and for
type 2 DM vs. non-diabetic recipients [52]. Liver
transplantation cures hepatogenous DM in 67% of
patients with pre-transplant DM by improving insu-
lin sensitivity, but in one-third of cases b-cell fail-
ure prevents recovery [53].
New-onset DM (NODM) may also occur, both with
liver and with other organ transplants. The exact
incidencedprobably around 15%dis difcult to
Diabetes and the liver 67
dene because as yet there is no world-wide
accepted criteria for assessment [54]. Also,
NODM increases the risk of mortality, and infection
is the most common complication [55].
Several risk factors have been associated with
the development of NODM. These are partly shared
with classical type 2 DM (age, family history of
diabetes, African-American or Hispanic race), with
hepatogenous diabetes (HCV-positive disease), and
with NAFLD (individual features of the metabolic
syndrome). Only the type of immunosuppressive
agent is a major modiable risk factor for NODM,
accounting for 74% of variability in the cumulative
incidence of NODM 12 months after liver transplant
[56]. Corticosteroids are associated with the great-
est risk of NODM, with an incidence related to both
the dose administration and the duration of the
therapy. Among calcineurin inhibitors, tacrolimus
is reported to be more diabetogenic than cyclo-
sporine (for a detailed analysis see [55]).
Therapeutic implications
Prevention of NODM after liver transplantation
requires an accurate screening of at-risk patients
and the implementation of preventive measures
to promote a healthy lifestyle. A correct approach
to diet and physical activity may be difcult to
achieve in subjects who experience a new life after
being so close to death. However, the prevention of
obesity is mandatory in all cases. Patients at high
risk for NODM should be given an immunosuppres-
sive regimen that is the least diabetogenic, whilst
being compatible with effective prevention of re-
jection; this includes limiting corticosteroids and
favoring cyclosporine treatment. In the presence
of NODM, switching fromtacrolimus to cyclosporine
should be considered, as well as a calcineurine-
free immunosuppressive therapy. There are no
evidence-based protocols for glucose control in
post-transplant DM, either in subjects with pre-
existing or new-onset DM, but every effort should
be made to normalize glucose proles through the
careful use of oral hypoglycemic agents or insulin,
coupled with a healthy lifestyle [55].
Conclusion
There is denite evidence for an ominous association
between DM and liver disease. The old concept of
considering fatty liver a trivial, occasional nding of
scarce clinical signicance has progressively turned
into a major concern regarding potential epidemics
of metabolic liver disease due to the growing
prevalence of obesity and DM. They are both prob-
ably incorrect estimates of the real risk. Pure fatty
liver, the most common nding in DM, remains a non-
aggressive metabolic condition, but in a few cases it
mayevolvetowards NASH, NAFLD-relatedcirrhosis or
primary liver cancer, or may increase the risk
associated with HCV infection, through largely un-
explained mechanisms (Fig. 1). The unsolved key
question is: what triggers disease progression to-
wards cardiovascular disease in the majority of DM
cases, but towards cirrhosis onlyinselectedpatients?
Extensive prospective monitoring of NAFLD cases,
molecular biology-based analysis of insulin signaling
in HCV-infected patients, and intervention studies,
will help to clarify the mechanisms of action of the
possible clinical strategies, the predictive value of
biochemical, histological, and clinical markers, and
the effectiveness of treatments available.
References
[1] Holstein A, Hinze S, Thiessen E, Plaschke A, Egberts EH.
Clinical implications of hepatogenous diabetes in liver
cirrhosis. J Gastroenterol Hepatol 2002;17:677e81.
Figure 1 Representationof theputativepathogenesis of
diabetes-associated non-alcoholic fatty liver disease and
its progression. In the pathogenesis of diabetes-associated
liver disease, insulin resistance plays a major role by in-
creasing the free fatty acid ux responsible for steatosis.
The presence of hepatitis C virus represents an additional
factor responsible for low insulin sensitivity, along with
lifestyleandgenetic factors. Thetransitionfrompurefatty
liver to steatohepatitis, and ultimately to cirrhosis and he-
patocellular carcinoma, is mediated by several factors in
NAFLD; diabetes per se increases the risk.
68 S. Moscatiello et al.
[2] Marchesini G, Ronchi M, Forlani G, Bugianesi E, Bianchi G,
Fabbri A, et al. Cardiovascular disease in cirrhosis e
a point-prevalence study in relation to glucose tolerance.
Am J Gastroenterol 1999;94:655e62.
[3] Bianchi G, Marchesini G, Zoli M, Bugianesi E, Fabbri A,
Pisi E. Prognostic signicance of diabetes in patients with
cirrhosis. Hepatology 1994;20:119e25.
[4] Toyoda H, Kumada T, Nakano S, Takeda I, Sugiyama K,
Kiriyama S, et al. Impact of diabetes mellitus on the prog-
nosis of patients with hepatocellular carcinoma. Cancer
2001;91:957e63.
[5] Huo TI, Lui WY, Huang YH, Chau GY, Wu JC, Lee PC, et al.
Diabetes mellitus is a risk factor for hepatic decompensa-
tion in patients with hepatocellular carcinoma undergoing
resection: a longitudinal study. Am J Gastroenterol 2003;
98:2293e8.
[6] Creutzfeldt W, Frerichs H, Kneer P. Liver disease, insulin
antagonism and diabetes mellitus. Horm Metab Res 1974;
(Suppl.):135e42.
[7] Booth AM, Caldwell SH, Iezzoni JC. Troglitazone-associated
hepatic failure. Am J Gastroenterol 2000;95:557e8.
[8] Gentile S, Turco S, Guarino G, Oliviero B, Annunziata S,
Cozzolino D, et al. Effect of treatment with acarbose and
insulin in patients with non-insulin-dependent diabetes
mellitus associated with non-alcoholic liver cirrhosis. Dia-
betes Obes Metab 2001;3:33e40.
[9] Gentile S, Guarino G, Romano M, Alagia IA, Fierro M,
Annunziata S, et al. A randomized controlled trial of acar-
bose in hepatic encephalopathy. Clin Gastroenterol Hepa-
tol 2005;3:184e91.
[10] Farrell GC, Larter CZ. Nonalcoholic fatty liver disease:
from steatosis to cirrhosis. Hepatology 2006;43:S99e112.
[11] Harrison SA, Torgerson S, Hayashi PH. The natural history
of nonalcoholic fatty liver disease: a clinical histopatholog-
ical study. Am J Gastroenterol 2003;98:2042e7.
[12] Adams LA, Sanderson S, Lindor KD, Angulo P. The histolog-
ical course of nonalcoholic fatty liver disease: a longitudi-
nal study of 103 patients with sequential liver biopsies.
J Hepatol 2005;42:132e8.
[13] Fassio E, Alvarez E, Dominguez N, Landeira G, Longo C. Nat-
ural history of nonalcoholic steatohepatitis: a longitudinal
study of repeat liver biopsies. Hepatology 2004;40:820e6.
[14] El-Serag HB, Tran T, Everhart JE. Diabetes increases the
risk of chronic liver disease and hepatocellular carcinoma.
Gastroenterology 2004;126:460e8.
[15] Adams LA, Lymp JF, St Sauver J, Sanderson SO, Lindor KD,
Feldstein A, et al. The natural history of nonalcoholic fatty
liver disease: a population-based cohort study. Gastroen-
terology 2005;129:113e21.
[16] Jepsen P, Vilstrup H, Mellemkjaer L, Thulstrup AM,
Olsen JH, Baron J, et al. Prognosis of patients with a diag-
nosis of fatty liver - a registry-based cohort study. Hepato-
gastroenterology 2003;50:2101e4.
[17] Dam-Larsen S, Franzmann M, Andersen IB,
Christoffersen P, Jensen LB, Sorensen TI, et al. Long
term prognosis of fatty liver: risk of chronic liver disease
and death. Gut 2004;53:750e5.
[18] Mofrad P, Contos MJ, Haque M, Sargeant C, Fisher RA,
Luketic VA, et al. Clinical and histologic spectrum of non-
alcoholic fatty liver disease associated with normal ALT
values. Hepatology 2003;37:1286e92.
[19] de Marco R, Locatelli F, Zoppini G, Verlato G, Bonora E,
Muggeo M. Cause-specic mortality in type 2 diabetes.
The Verona Diabetes Study. Diabetes Care 1999;22:
756e61.
[20] Trombetta M, Spiazzi G, Zoppini G, Muggeo M. Review arti-
cle: type 2 diabetes and chronic liver disease in the Verona
diabetes study. Aliment Pharmacol Ther 2005;22(Suppl. 2):
24e7.
[21] Marchesini G, Forlani G, Bugianesi E. Is liver disease
a threat to patients with metabolic disorders? Ann Med
2005;37:333e46.
[22] Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Over-
weight, obesity, and mortality from cancer in a prospec-
tively studied cohort of U.S. adults. N Engl J Med 2003;
348:1625e38.
[23] Westerbacka J, Lammi K, Hakkinen AM, Rissanen A,
Salminen I, Aro A, et al. Dietary fat content modies liver
fat in overweight non-diabetic subjects. J Clin Endocrinol
Metab 2005;90:2804e9.
[24] Huang MA, Greenson JK, Chao C, Anderson L, Peterman D,
Jacobson J, et al. One-year intense nutritional counseling
results in histological improvement in patients with non-
alcoholic steatohepatitis: a pilot study. Am J Gastroenterol
2005;100:1072e81.
[25] Suzuki A, Angulo P, Lymp J, St Sauver J, Muto A, Okada T,
et al. Chronological development of elevated aminotrans-
ferases in a nonalcoholic population. Hepatology 2005;41:
64e71.
[26] Suzuki A, Lindor K, St Sauver J, Lymp J, Mendes F, Muto A,
et al. Effect of changes on body weight and lifestyle in non-
alcoholic fatty liver disease. J Hepatol 2005;43:1060e6.
[27] Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF,
Lachin JM, Walker EA, et al. Reduction in the incidence
of type 2 diabetes with lifestyle intervention or metfor-
min. N Engl J Med 2002;346:393e403.
[28] Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT,
Hamalainen H, Ilanne-Parikka P, et al. Prevention of type
2 diabetes mellitus by changes in lifestyle among subjects
with impaired glucose tolerance. N Engl J Med 2001;344:
1343e50.
[29] Bugianesi E, Gentilcore E, Manini R, Natale S, Vanni E,
Villanova N, et al. A randomized controlled trial of metfor-
min versus vitamin E or prescriptive diet in nonalcoholic
fatty liver disease. Am J Gastroenterol 2005;100:1082e90.
[30] Marchesini G, Brizi M, Bianchi G, Tomassetti S, Zoli M,
Melchionda N. Metformin in non-alcoholic steatohepatitis.
Lancet 2001;358:893e4.
[31] Neuschwander-Tetri BA, Brunt EM, Wehmeier KR, Oliver D,
Bacon BR. Improved nonalcoholic steatohepatitis after 48
weeks of treatment with the PPAR-gamma ligand rosiglita-
zone. Hepatology 2003;38:1008e17.
[32] Promrat K, Lutchman G, Uwaifo GI, Freedman RJ, Soza A,
Heller T, et al. A pilot study of pioglitazone treatment for
nonalcoholic steatohepatitis. Hepatology 2004;39:188e96.
[33] Nair S, Diehl AM, Wiseman M, Farr Jr GH, Perrillo RP. Met-
formin in the treatment of non-alcoholic steatohepatitis:
a pilot open label trial. Aliment Pharmacol Ther 2004;20:
23e8.
[34] Bugianesi E, Gastaldelli A, Vanni E, Gambino R,
Cassader M, Baldi S, et al. Insulin resistance in non-dia-
betic patients with non-alcoholic fatty liver disease: sites
and mechanisms. Diabetologia 2005;48:634e42.
[35] Vajro P, Mandato C, Franzese A, Ciccimarra E, Lucariello S,
Savoia M, et al. Vitamin E treatment in pediatric obesity-
related liver disease: a randomized study. J Pediatr Gas-
troenterol Nutr 2004;38:48e55.
[36] Lindor KD, Kowdley KV, Heathcote EJ, Harrison ME,
Jorgensen R, Angulo P, et al. Ursodeoxycholic acid for
treatment of nonalcoholic steatohepatitis: results of a ran-
domized trial. Hepatology 2004;39:770e8.
[37] Adams LA, Zein CO, Angulo P, Lindor KD. A pilot trial of
pentoxifylline in nonalcoholic steatohepatitis. Am J Gas-
troenterol 2004;99:2365e8.
Diabetes and the liver 69
[38] Yokohama S, Yoneda M, Haneda M, Okamoto S, Okada M,
Aso K, et al. Therapeutic efcacy of an angiotensin II re-
ceptor antagonist in patients with nonalcoholic steatohe-
patitis. Hepatology 2004;40:1222e5.
[39] Mehta SH, Brancati FL, Sulkowski MS, Strathdee SA,
Szklo M, Thomas DL. Prevalence of type 2 diabetes mellitus
among persons with hepatitis C virus infection in the
United States. Ann Intern Med 2000;133:592e9.
[40] Harrison SA. Liver disease in patients with diabetes melli-
tus. J Clin Gastroenterol 2006;40:68e76.
[41] Lecube A, Hernandez C, Genesca J, Esteban JI, Jardi R,
Simo R. High prevalence of glucoseabnormalities inpatients
with hepatitis C virus infection: a multivariate analysis con-
sidering the liver injury. Diabetes Care 2004;27:1171e5.
[42] Aytug S, Reich D, Sapiro LE, Bernstein D, Begum N. Im-
paired IRS-1/PI3-kinase signaling in patients with HCV:
a mechanism for increased prevalence of type 2 diabetes.
Hepatology 2003;38:1384e92.
[43] Asselah T, Rubbia-Brandt L, Marcellin P, Negro F. Steatosis
in chronic hepatitis C: why does it really matter? Gut 2006;
55:123e30.
[44] Leandro G, Mangia A, Hui J, Fabris P, Rubbia-Brandt L,
ColloredoG, et al. Relationshipbetweensteatosis, inamma-
tion, and brosis in chronic Hepatitis C: a meta-analysis of in-
dividual patient data. Gastroenterology 2006;130:1346e62.
[45] Camma` C, Bruno S, Di Marco V, Di Bona D, Rumi M, Vinci M,
et al. Insulin resistance is associated with steatosis in non-
diabetic patients with genotype 1 chronic hepatitis C. Hep-
atology 2006;43:64e71.
[46] Lonardo A, Adinol LE, Loria P, Carulli N, Ruggiero G,
Day CP. Steatosis and hepatitis C virus: mechanisms and
signicance for hepatic and extrahepatic disease. Gastro-
enterology 2004;126:586e97.
[47] Romero-Gomez M, Del Mar Viloria M, Andrade RJ,
Salmeron J, Diago M, Fernandez-Rodriguez CM, et al.
Insulin resistance impairs sustained response rate to pegin-
terferon plus ribavirin in chronic hepatitis C patients. Gas-
troenterology 2005;128:636e41.
[48] Picardi A, DAvola D, Gentilucci UV, Galati G, Fiori E,
Spataro S, et al. Diabetes in chronic liver disease: from
old concepts to new evidence. Diabetes Metab Res Rev
2006;22:274e83.
[49] Yoo HY, Thuluvath PJ. The effect of insulin-dependent di-
abetes mellitus on outcome of liver transplantation. Trans-
plantation 2002;74:1007e12.
[50] Thuluvath PJ. When is diabetes mellitus a relative or abso-
lute contraindication to liver transplantation? Liver Transpl
2005;11:S25e9.
[51] Contos MJ, Cales W, Sterling RK, Luketic VA, Shiffman ML,
Mills AS, et al. Development of nonalcoholic fatty liver dis-
ease after orthotopic liver transplantation for cryptogenic
cirrhosis. Liver Transpl 2001;7:363e73.
[52] John PR, Thuluvath PJ. Outcome of liver transplantation in
patients with diabetes mellitus: a case-control study. Hep-
atology 2001;34:889e95.
[53] Perseghin G, Mazzaferro V, Sereni LP, Regalia E,
Benedini S, Bazzigaluppi E, et al. Contribution of reduced
insulin sensitivity and secretion to the pathogenesis of
hepatogenous diabetes: effect of liver transplantation.
Hepatology 2000;31:694e703.
[54] Davidson JA, Wilkinson A. New-Onset Diabetes After
Transplantation 2003 International Consensus Guidelines:
an endocrinologists view. Diabetes Care 2004;27:
805e12.
[55] Marchetti P. New-onset diabetes after liver transplanta-
tion: from pathogenesis to management. Liver Transpl
2005;11:612e20.
[56] Montori VM, Basu A, Erwin PJ, Velosa JA, Gabriel SE,
Kudva YC. Posttransplantation diabetes: a systematic re-
view of the literature. Diabetes Care 2002;25:583e92.
70 S. Moscatiello et al.