Visceral adiposity and insulin resistance are independent predictors of the presence of non-cirrhotic NAFLD-related portal hypertension



We previously demonstrated in an animal model that steatosis, in the absence of fibrosis, induces a significant rise in portal pressure, indicating substantial changes in liver hemodynamics. As assessment of portal pressure is an invasive procedure, non-invasive parameters are needed to identify patients at risk.


To study the portal pressure in nonalcoholic fatty liver disease patients and to identify factors that are possibly related to steatosis-induced changes in liver hemodynamics.

Materials and methods:

Patients presenting with a problem of overweight or obesity, and in whom non-invasive investigations showed signs of liver involvement, were proposed for transjugular hepatic vein catheterization and liver biopsy. The biopsy was scored according to the Nonalcoholic Steatohepatitis Clinical Research Network Scoring System.


A total of 50 consecutive patients were studied. Their mean age was 47.9±1.8 years; 31 (62%) were female. Hepatic venous pressure gradient was normal in 36 (72%) and elevated in 14 (28%) patients. The degree of steatosis was the only histological parameter that differed significantly between the two groups (P=0.016), and was a predictor of the presence of portal hypertension (PHT) in regression analysis (P=0.010). Comparing normal versus portal hypertensive patients, waist circumference (117±2 versus 128±4 cm, P=0.005), waist–hip ratio (0.96±0.06 versus 1.04±0.03, P=0.003), visceral fat (229±15 versus 292±35 cm2, P=0.022), fasting insulin (15.4±1.7 versus 21.8±2.4 μU ml−1, P=0.032), fasting c-peptide (1.22±0.06 versus 1.49±0.09 nmol l−1, P=0.035) and homeostasis model assessment–insulin resistance (HOMA IR) (3.28±0.29 versus 4.81±0.57, P=0.019) were significantly higher. Age, gender, liver enzymes, ferritin and high-sensitive C-reactive protein were not significantly different. In regression analysis, waist circumference (P=0.008) and HOMA IR (P=0.043) were independent predictors of PHT.


Estimates of both visceral adiposity and IR are predictors for the presence of PHT, related to the degree of steatosis, and may help in identifying patients who are at risk of developing steatosis-related complications.


Nonalcoholic steatosis is increasingly recognized as a cause of liver-related morbidity and mortality. It is frequently associated with features of cell damage and inflammation, leading to the definition of nonalcoholic steatohepatitis (NASH). NASH may lead to progressive fibrosis and cirrhosis. Evidence is increasing, however, that steatosis per se may also be harmful. Therefore, the term nonalcoholic fatty liver disease (NAFLD) covers a broader spectrum ranging from pure steatosis to steatohepatitis with fibrosis and cirrhosis and its complications, including hepatocellular carcinoma.1, 2

Steatotic livers seem to be more vulnerable to ischemia–reperfusion injury. This is illustrated by the high risk of primary graft non-function in case of severe steatosis of the donor liver in liver transplantation3 and by the increased mortality and complication rate (in terms of pre- and postoperative bleeding and liver dysfunction) in liver surgery on steatotic livers.3, 4 This increased vulnerability might be explained by the impact of steatosis on mechanisms of inflammation, fibrogenesis and regeneration.5, 6, 7 However, changes in liver blood flow, especially a reduction in sinusoidal flow,8, 9 were documented as well, both in animals9 and in humans.8

We recently demonstrated in the methionine–choline-deficient diet-induced rat model of steatosis that steatosis induces significant portal hypertension (PHT) in the absence of fibrosis and morphological signs of inflammation. This rise in portal pressure is associated with systemic vascular hyporeactivity, increased splanchnic blood flow and signs of a hyperdynamic circulation.10 These features resemble the hemodynamic changes that are well documented in cirrhotic PHT.11, 12 We subsequently examined the hepatic venous pressure gradient (HVPG) in a prospectively included series of patients with overweight and hence a risk of NAFLD. In this patient series, we show that steatosis was associated with PHT in the absence of extensive fibrosis or cirrhosis, confirming the previous experimental findings. This observation confirms the potential impact of steatosis on liver hemodynamics in NAFLD patients.

The assessment of portal pressure, however, is an invasive procedure. In the present analysis, we aim at identifying in our patient series clinical and biochemical parameters that are related to the presence of NAFLD-related PHT, in order to better characterize patients at risk for steatosis-related disease and its complications in an noninvasive manner.

Patients and methods

Patients presenting at the obesity clinic with a problem of overweight underwent a liver-specific program combined with a classical metabolic work-up. The classical metabolic work-up is approved by the Ethics Committee of the Antwerp University Hospital and has been used for many years at our center. It requires written informed consent of the patient and includes a detailed questionnaire on weight evolution, dietary habits, family history (with a focus on weight and diabetes), personal medical history, medication, alcohol and tobacco use. A clinical examination including anthropometry yields data on body weight and body fat distribution. Blood analysis includes blood cell count, white blood cell formula, coagulation tests, thrombophilic tests, electrolytes and kidney function tests, liver tests (aspartate aminotransferase, alanine aminotransferase (ALT), lactate dehydrogenase, γ-glutamyl transpeptidase, alkaline phosphatase, total bilirubin and fractions), creatinine kinase, total protein, protein electrophoresis, thyroid function, sex hormones, ferritin, vitamin B and folic acid. A 3-h oral glucose tolerance test (75 g of glucose, 3 h) including insulin and c-peptide analysis is carried out. Insulin resistance (IR) estimation using the homeostasis model assessment (HOMA) is calculated as (insulin (mU l−1) × glucose (mmol l−1))/22.5.13 Further examination includes visceral fat measurement by computed tomography.14 The liver specific program, approved by the Ethics Committee of the University Hospital Antwerp, includes additional blood analysis to exclude the classical etiologies of liver disease (for example, viral hepatitis and autoimmune disease); a Doppler ultrasound of the abdomen; a liver–spleen scintigraphy15 and an aminopyrine breath test as a measure for liver metabolic reserve.16

As liver steatosis seems to be closely related to the metabolic syndrome (MS),17 features of the MS were specifically recorded. According to the US Third Adult Treatment Panel of the National Cholesterol Education Program,18 the MS is defined as the presence of three or more out of the five known criteria. In 2005, the International Diabetes Federation accepted a slightly different definition, making visceral obesity a conditio sine qua non, and lowering the cut-off levels for waist circumference (waist circumference >94 cm in men and >80 cm in women) and fasting glucose (fasting glucose 100 mg per 100 ml).19

Patients were excluded from further analysis if they refused signing an informed consent, if they had a significant alcohol consumption (>20 g per day),20 if they had previously undergone bariatric surgery, if another liver disease was diagnosed or if they were known to have diabetes. The reason to exclude patients with known diabetes is explained by the fact that this study was part of a larger program studying MS in obese patients. In the design of this larger program, diabetic patients were excluded for methodological reasons. Patients <16 years of age were not included.

The possibility of liver involvement was defined by one or more of the following elements: abnormal liver tests (aspartate aminotransferase and/or ALT and/or γ-glutamyl transpeptidase and/or alkaline phosphatase), ultrasound abnormality of the liver (major criteria: enlarged liver,21 steatotic liver,22 splenomegaly;23 minor criteria: reduced portal flow (<800 ml min−1), abnormal resistance index in the hepatic artery, abnormal flow pattern (normally triphasic) in the hepatic veins 21), signs of parenchymal liver disease on liver–spleen scintigraphy15 and abnormal aminopyrine breath test.16 The cut-off levels for the most important parameters are listed in Table 1. For ALT, three different cut-off levels were used: the upper limit of normal set by the biochemistry laboratory (56 IU l−1), the classical cut-off of 40 IU l−124 and the limits proposed by Prati et al. 25 (30 IU l−1 in men, 19 IU l−1 in women).

Table 1 Basic characteristics of the patients undergoing a transjugular liver vein catheterization and liver biopsy

When one or more of these criteria were met, a liver biopsy was proposed.26, 27, 28, 29 A separate informed consent for liver biopsy was required. Patients were proposed for transjugular liver vein catheterization and biopsy whenever this was possible from a logistic point of view. Those in whom a transjugular biopsy could not be carried out in a reasonable time frame were referred for transparietal liver biopsy and excluded from further analysis.

Transjugular liver vein catheterization and biopsy30, 31, 32 were carried out as previously reported.14 Antihypertensive drugs were discontinued on the day of the procedure. HVPG was calculated by subtracting the average free pressure from the average wedge pressure in three different hepatic vein branches. A HVPG of >5 mm Hg was used to define PHT.33

A large part of the liver biopsy specimen (>2 cm, if available) was stored in formalin aldehyde. Hematoxylin-eosin stain, sirius red (Fouchet) stain, periodic acid-Schiff stain after diastase, reticulin stain (Gordon–Sweets) and Perl's iron stain were routinely carried out on all biopsy samples and subsequently analyzed by two different experienced pathologists, using the NASH Clinical Research Network Scoring System.34 Steatosis was graded as follows: <5% of liver parenchyma: 0, 5−33%: 1, >33−66%: 2, >66%: 3. Lobular inflammation was scored as no foci: 0, <2 foci per × 200 field: 1, 2–4 foci per × 200 field: 2, >4 foci per × 200 field: 3. Ballooning was scored as none: 0, few balloon cells: 1, many cells per prominent ballooning: 2. Fibrosis is staged as none: 0, perisinusoidal or periportal: 1, perisinusoidal and portal per periportal: 2, bridging fibrosis: 3, cirrhosis: 4. The NASH activity score was calculated as the sum of the scores for steatosis, ballooning and lobular inflammation.34 Furthermore the pathologist reported the length of the biospy and the number of portal tracts.

Values were expressed as mean±standard deviation (s.d.). The results were analyzed using an independent samples t-test (continuous variables), Mann–Whitney U-test and Somers’ D-tests (a chi-square variant for categorical ordinal values) (categorical variables, scores), chi-square test (prevalences) and with correlation analysis using SSPS 16.0 for Windows. Logistic regression was carried out to identify independent predictors of the presence of PHT. A P-value of <0.05 was considered statistically significant.


The patient selection process and the corresponding numbers are shown in Figure 1. In all, 50 patients underwent a transjugular liver vein catheterization and biopsy according to the above protocol between 1 September 2006 and 1 February 2009. In this time period, 652 new patients were seen at the obesity clinic. A total of 337 patients did not enter the protocol: 145 had significant alcohol use, 82 were already known to have diabetes, 45 had undergone bariatric surgery in the past, 35 refused to give informed consent, 30 were excluded for miscellaneous reasons. From the 315 patients included in the program, 16 patients were not included because the work-up showed an abnormal finding that needed specific attention (for example, significant thyroid dysfunction, a previously unknown liver lesion). Four had a formerly undiagnosed other liver disease (two had chronic hepatitis B, one had primary biliary cirrhosis, one had hemochromatosis). From the remaining 295, 38 had no arguments for liver involvement. From the 257 patients with a possible liver involvement according to the criteria of the protocol, 79 refused a biopsy or were lost to follow-up and 47 underwent surgery. From the remaining 131 patients, 50 underwent a transjugular biopsy. The other patients underwent a transparietal biopsy.

Figure 1

Flow chart of the patient selection process and numbers.

The main characteristics of the 50 patients who underwent a transjugular liver vein catheterization and liver biopsy are summarized in Table 1. Two patients had thyroid dysfunction. Five patients had suffered from a cardiovascular event, mainly ischemic coronary heart disease. A total of 12 patients were taking antiaggregant drugs, 7 as primary prevention. No patient was taking anticoagulant drugs. A total of 19 patients were taking antihypertensive drugs. From the 31 female patients, 7 were taking oral anticonceptives. Seven patients were on lipid-lowering drugs. No patient was on pharmacological treatment for overweight or diabetes.

The presence of signs of liver involvement, according to the protocol, is summarized in Table 2. If the criterion of ALT >40 U l−1 is used, 4% have two criteria, 30% three, 22% four, 22% five, 8% six, 8% seven and 6% eight criteria.

Table 2 Presence of criteria for liver involvement

Comparison of the main characteristics (age, gender distribution, body mass index (BMI), waist, ALT, γ-glutamyl transpeptidase, total cholesterol, high-density lipoprotein cholesterol, triglycerides, fasting glucose, fasting insulin and HOMA IR) of these 50 patients with those of the 81 patients who underwent a classical transparietal biopsy (see Figure 1), showed no significant differences (data not shown).

MS and its criteria

The presence of the five criteria for the MS, according to the National Cholesterol Education Program-Adult Treatment Panel III definition,18 is summarized in Table 3.

Table 3 Presence of the components of the metabolic syndrome

Of all patients, 12% met one criterion only, 26% two criteria, 34% three criteria, 26% four and 2% five criteria, meaning that 62% met the definition of the MS according to the Adult Treatment Panel III definition. According to the International Diabetes Federation criteria,19 64% were diagnosed with MS.

The two criteria with different cut-off values according to gender were compared. As could be expected, the absolute values for high-density lipoprotein cholesterol (but not for waist circumference) were significantly different between men and women, but in terms of % of abnormal values, there were no gender-related differences.

Mean BMI was 40.2±7.2 kg m−2 (39.9±5.9 kg m−2 in men versus 40.7±8.0 kg m−2 in women, P=0.684). All patients met the definition of obesity with a BMI >30 kg m−2 (range 31.3–69.1 kg m−2).35 Visceral adiposity as assessed by computed tomography was measured in 18 out of 19 (95%) men (criterion >100 cm2, with a mean of 252.2±122.9 cm2) and in 100% of women (criterion >80 cm2, with a mean of 207.7±93.2 cm2, P=0.119 compared with men).14 On the basis of oral glucose tolerance test, 40 out of 50 (80%) patients had a normal glucose tolerance, 9 out of 50 (18%) had Impaired Glucose Tolerance (defined as glycaemia 200 mg per 100 ml at 2 hours on oral glucose tolerance test), and 1 out of 50 (2%) had diabetes.36 Among the 10 patients with abnormal glucose tolerance, only 5 met the criterion of fasting glucose 100 mg per 100 ml and only 2 the former criterion of 110 mg per 100 ml. Mean fasting insulin was 18.27±9.97 μU ml−1. According to the criterion of fasting insulinemia 16 μU ml−1, 44% of patients had hyperinsulinism.37 On the basis of a HOMA IR of >1.0, only 4% had no hyperinsulinism.13

Hepatic venous pressure gradient

HVPG was normal in 36 out of 50 (72%) (group 1) patients and elevated in 14 out of 50 (28%). HVPG was 8.8±0.7 mm Hg in those with an elevated HVPG (group 2) versus 3.4±0.2 mm Hg in group 1.

In all, 14 out of 36 (39%) patients with a normal HVPG were taking antihypertensive drugs, and 5 out of 14 (36%) in the group had an elevated HVPG (P=0.766) (Table 5). The drugs were discontinued on the day of the HVPG measurement, as stated in the methods.


In all patients a liver biopsy was obtained. Mean length was 18.2±0.2 mm; the mean number of portal tracts was 8.4±0.8. The data for the NASH activity score and its three components and for the fibrosis stage compared between groups 1 and 2 are summarized in Table 4.

Table 4 Histological features in patients without and with portal hypertension

In both groups, only one patient had cirrhosis. In all, 9 out of 14 (65%) and 26 out of 36 (72%) patients had F0. Fibrosis score was not significantly different (Mann–Whitney U-test, P=0.794). Steatosis was the only histological feature that differed significantly between the two groups (Mann–Whitney U-test, P=0.016; Somers’ d–test, P=0.008). In regression analysis, only steatosis turned out to be a predictor of the presence of PHT (P=0.010).

Comparison of steatosis (Mann–Whitney U-test, P=0.273), NASH activity score (P=0.525) and fibrosis score (P=0.924) between the study group and the 81 patients who underwent a classical liver biopsy showed no significant differences.

In seven patients no steatosis was found on liver biopsy, although initial screening showed signs of liver disease. In two out of seven patients some sinusoidal dilatation was found. This can explain the presence of elevated ALT and a hyperechogenic liver parenchyma on ultrasound. Cardiac decompensation was excluded. The exact significance of this sinusoidal dilatation is unknown. It is probably related to hormonal disturbances. Elevated ALT (three out of five) and hyperechogenic liver parenchyma on ultrasound (three out of five) were the reasons to include the remaining five patients. They should be considered false-positive results of the screening, as liver tissue was normal on histological examination.

Clinical and biochemical parameters related to portal hypertension

The values (expressed as mean±s.d.) of some of the most relevant parameters related to liver disease and inflammation are listed in Table 5 for the patients without (group 1) and with (group 2) PHT, together with the P-value of the comparison between the two groups (Student's t-test). Waist circumference, waist–hip ratio, visceral fat, fasting insulin, fasting c-peptide and HOMA IR appeared to be significantly different. Age, liver enzymes, ferritin and high-sensitive C-reactive protein were not significantly different.

Table 5 Clinical and biochemical data of patients without and with PHT

Categorical variables such as gender and the presence of the MS criteria were compared using Mann–Whitney U-test (results not shown). Only the presence of the criterion for waist circumference appeared to be significantly different (P<0.0001).

In binary logistic regression analysis, waist circumference (P=0.004), waist–hip ratio (P=0.006), fasting c-peptide (P=0.025) and HOMA IR (P=0.024) were predictors of PHT. In a forward conditional analysis, waist circumference (P=0.008) and HOMA IR (P=0.043) appeared to be independent predictors of PHT.

Clinical and biochemical parameters related to steatosis

The clinical and biochemical parameters that are significantly correlated to the degree of steatosis (P<0.05) are listed in Table 6.

Table 6 Clinical and biochemical parameters that correlate significantly with the degree of steatosis

Steatosis is not an independent predictor of the presence of PHT after correction for waist–hip ratio or HOMA IR using a binary logistic regression analysis (forward conditional).


The main aim of this study was to identify clinical and biochemical factors that are associated with the PHT induced by steatosis. We conclude that both parameters of visceral adiposity and IR are related to the presence of PHT in severe steatosis.

We previously showed that steatosis per se induces an increase in portal pressure, regardless of the degree of fibrosis in an experimental models.10 In the present series we confirm the presence of PHT in patients with NAFLD, in the absence of significant fibrosis, confirming our experimental findings. As some patients were taking antihypertensive drugs, the results might even underestimate the problem. Although the exact clinical significance is still unclear, it is obvious that PHT is a marker of important changes in liver hemodynamics. These changes might significantly contribute to steatosis-related disease.38 Assessment of portal pressure, however, requires an invasive procedure. It is therefore of interest to identify noninvasive clinical and biochemical markers that predict the presence of PHT and that hence help in selecting patients at risk.

In this analysis, markers of visceral fat accumulation seem to be significantly correlated to PHT in steatosis. Waist circumference, a marker of visceral obesity,39 prove to be an independent predictor for the presence of PHT. Waist circumference, waist–hip ratio and visceral fat measured by computed tomography14 are significantly different between patients without and with PHT.

Data linking obesity and liver fat are numerous. The BMI is a well-known independent predictor of fat in the liver.40, 41 Regardless of BMI, central adiposity has been shown to be associated with NAFLD in normal weight, obese and diabetic individuals.17, 42, 43 As the presence of PHT is related to the degree of steatosis, and as the degree of steatosis is related to the amount of visceral fat, the predictive role of waist circumference can easily be understood. Even after correction for the degree of steatosis, however, waist circumference remains an independent predictor of PTH, indicating that the relation between waist and steatosis-associated PTH cannot be fully explained by steatosis alone. It is known that increasing obesity is also associated with an increased risk of steatosis-associated liver inflammation (NASH).17 The complex role of visceral adipocytes as an important source of cytokines, adipokines and para-endocrine substances may be involved.44, 45 Routine biochemical and histological parameters of inflammation, however, showed no significant differences or correlations in this study cohort.

A second factor that could be identified as related to PHT is insulin resistance. Fasting insulin, fasting c-peptide and HOMA IR are significantly different between patients with and without steatosis-induced PHT. Fasting c-peptide and HOMA IR are predictors of the presence of PHT in regression analysis. After correction for steatosis and waist circumference, HOMA IR seems to be an independent predictor of PHT.

The association between NAFLD and glucose homeostasis disturbances has been well established. Available data indicate that the onset of NAFLD is an early event in the development of IR and might predict the presence or future development of the MS.46 On the other hand, IR is often considered as a key etiological mechanism in the development of NASH, and is a possible therapeutic target.47, 48 In diabetic patients, NAFLD is an important source of morbidity and mortality.49 In patients with type 2 diabetes, cirrhosis and its complications are the second cause of disease-specific mortality.50 NAFLD, as diagnosed by liver ultrasound, is present in 69.5%, in an age-dependent manner.1 Patients with NAFLD and diabetes develop cirrhosis in 23.9% and experience liver-related death in 19% of cases, compared with 10.6 and 2% in NAFLD patients without diabetes, respectively.51 The role of IR as a predictor of steatosis-related PHT is clearly in line with these data.

Although both IR and visceral fat accumulation are independent predictors of PTH in our study, IR and visceral adiposity are intimately linked.46, 47, 52 Previous studies showed that that the severity of fatty liver was positively correlated with visceral fat accumulation and insulin resistance.42, 48 However, the exact causal relation between IR, visceral fat and steatosis remains unclear. Furthermore, there are patients with IR and NAFLD who are not obese, indicating that other factors besides visceral obesity contribute to the association between NAFLD and IR.17 This probably explains why the relationship of HOMA IR with the presence of PHT remains significant even after correction for waist circumference.

Finally, the selection of patients should be taken into account to correctly interpret the results. The selection for transjugular biopsy did not create a bias, as the main clinical, biochemical and histological characteristics were not significantly different from the other biopsy groups. However, all patients were selected from a larger program, studying the relationship between the MS and the liver, as described in the methods. This resulted in a series of patients who are obese and in whom there is a suspicion of NAFLD based on predefined criteria. This patient selection should be taken into account if we want to extrapolate the results to the standard obese and/or NAFLD patient seen in daily practice.


We were able to identify parameters of visceral fat accumulation and of IR as independent predictors of the presence of PHT associated with NAFLD. This illustrates the importance of liver fat, visceral fat and IR and their complex interplay in the pathogenesis of steatosis-related liver disease.


  1. 1

    Angulo P . Non-alcoholic fatty liver disease. N Engl J Med 2002; 346: 1221–1231.

    CAS  Article  Google Scholar 

  2. 2

    Bugianesi E, Leone N, Vanni E, Marchesini G, Brunello F, Carucci P et al. Expanding the natural history of non-alcoholic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology 2002; 123: 134–143.

    Article  Google Scholar 

  3. 3

    Selzner M, Clavien P . Fatty liver in liver transplantation and surgery. Semin Liver Dis 2001; 21: 105–113.

    CAS  Article  Google Scholar 

  4. 4

    Behrns K, Tsiotos G, DeSouza N, Krishna M, Ludwig J, Nagorney D . Hepatic steatosis as a potential risk factor for major hepatic resections. J Gastrointest Surg 1998; 2: 292–298.

    CAS  Article  Google Scholar 

  5. 5

    Rao M, Papreddy K, Abecassis M, Hashimoto T . Regeneration of liver with marked fatty change following partial hepatectomy in rats. Dig Dis Sci 2001; 46: 1821–1826.

    CAS  Article  Google Scholar 

  6. 6

    Selzner M, Rüdiger H, Sindram D, Madden J, Clavien . Mechanisms of ischemic injury are different in the steatotic and normal rat liver. Hepatology 2000; 32: 1281–1288.

    Article  Google Scholar 

  7. 7

    Choi S, Diehl AM . Role of inflammation in non-alcoholic steatohepatitis. Curr Opin Gastroenterol 2005; 21: 702–707.

    Article  Google Scholar 

  8. 8

    Seifalian A, Chidambaram V, Rolles K, Davidson B . In vivo demonstration of impaired microcirculation in steatotic human liver grafts. Liver Transpl Surg 1998; 4: 71–77.

    CAS  Article  Google Scholar 

  9. 9

    McCuskey RS, Ito Y, Robertson GR, McCuskey MK, Perry M, Farrell GC . Hepatic microvascular dysfunction during evolution of dietary steatohepatitis in mice. Hepatology 2004; 40: 386–393.

    Article  Google Scholar 

  10. 10

    Francque S, Wamutu S, Chatterjee S, Van Marck E, Herman A, Ramon A et al. Non-alcoholic steatohepatitis induces non-fibrosis related portal hypertension associated with splanchnic vasodilation and signs of a hyperdynamic circulation in vitro and in vivo in a rat model. Liver Int 2010; 30: 365–375.

    CAS  Article  Google Scholar 

  11. 11

    Lebrec D, Moreau R . Pathogenesis of portal hypertension. Eur J Gastroenterol Hepatol 2001; 13: 309–311.

    CAS  Article  Google Scholar 

  12. 12

    Laleman W, Van Landeghem L, Wilmer A, Fevery J, Nevens F . Portal hypertension: from pathophysiology to clinical practice. Liver Int 2005; 25: 1070–1090.

    Article  Google Scholar 

  13. 13

    Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC . Homeostasis model assessment: insulin resistance and β-cell functioning from fasting glucose and insulin concentrations in men. Diabetology 1985; 28: 412–419.

    CAS  Article  Google Scholar 

  14. 14

    van der Kooy K, Seidell JC . Techniques for the measurement of visceral fat: a practical guide. Int J Obes Relat Metab Disord 1993; 17: 187–196.

    CAS  Google Scholar 

  15. 15

    Bucombe JR, Hilson AJW . Radionuclide investigations of the liver. Oxford Textbook of Clinical Hepatology, 2nd edn. Oxford University Press: Oxford, UK, 1999. pp 579–588.

    Google Scholar 

  16. 16

    Merkel C, Bolognesi M, Bellon S, Bianca S, Honisch B, Lampe H et al. Aminopyrine breath test in the prognostic evaluation of patients with cirrhosis. Gut 1992; 33: 836–842.

    CAS  Article  Google Scholar 

  17. 17

    Marchesini G, Bugianesi E, Forlani G, Cerrelli F, Lenzi M, Manini R et al. Nonalcoholic fatty liver, steatohepatitis, and the metabolic syndrome. Hepatology 2003; 37: 917–923.

    Article  Google Scholar 

  18. 18

    Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001; 285: 2486–2497.

    Article  Google Scholar 

  19. 19

    Alberti KG, Zimmet P, Shaw J, IDF Epidemiology Task Force Consensus Group. The metabolic syndrome—a new worldwide definition. Lancet 2005; 366: 1059–1062.

    Article  Google Scholar 

  20. 20

    Farrell G, George J, de la M Hall P, McCullough AJ . An introduction to NASH and related fatty liver disorders. Fatty Liver Disease; NASH and related disorders. Blackwell Publishing: USA, 2005; 1–12.

    Google Scholar 

  21. 21

    Cosgrove DO . Liver Anatomy. In: Cosgrove D, Meire H, Dewbury K (eds). Abdominal and General Ultrasound vol. 1, 3rd edn. Churchill Livingstone: New York, 1994. pp 227–242.

    Google Scholar 

  22. 22

    Bellentani S, Saccoccio G, Masutti F, Croce LS, Brandi G, Sasso F et al. Prevalence of and risk factors for hepatic steatosis in northern Italy. Ann Intern Med 2000; 132: 112–117.

    CAS  Article  Google Scholar 

  23. 23

    Dardenne AN . The spleen. In: Cosgrove D, Meire H, Dewbury K (eds). Abdominal and General Ultrasound vol. 1, 3rd edn. Churchill Livingstone: New York, 1994. pp 351–366.

    Google Scholar 

  24. 24

    Pratt DS, Kaplan MM . Evaluation of abnormal liver-enzyme results in asymptomatic patients. N Engl J Med 2000; 342: 1266–1271.

    CAS  Article  Google Scholar 

  25. 25

    Prati D, Taioli E, Zanella A, Della Torre E, Buttelli S, Del Vecchio E et al. Updated definitions of healthy ranges for serum alanine aminotransferase levels. Ann Intern Med 2002; 137: 1–10.

    CAS  Article  Google Scholar 

  26. 26

    Joy D, Thava VR, Scott BB . Diagnosis of fatty liver disease: is biopsy necessary? Eur J Gastroenterol Hepatol 2003; 15: 539–543.

    PubMed  Google Scholar 

  27. 27

    Kichian K, McLean R, Gramlich LM, Bailey RJ, Bain VG . Nonalcoholic fatty liver disease in patients investigated for elevated liver enzymes. Can J Gastroenterol 2003; 17: 38–42.

    Article  Google Scholar 

  28. 28

    Tawalkar JA . Motion—all patients with NASH need to have a liver biopsy: arguments for the motion. Can J Gastroenterol 2002; 16: 718–721.

    Article  Google Scholar 

  29. 29

    Campbell MS, Reddy KR . The evolving role of liver biopsy. Aliment Pharmacol Ther 2004; 20: 249–259.

    CAS  Article  Google Scholar 

  30. 30

    Kalambokis G, Manousou P, Vibhakorn S, Marelli L, Cholongitas E, Senzolo M et al. Transjugular liver biopsy—indications, adequacy, quality of specimens, and complications—A systematic review. J Hepatology 2007; 47: 284–294.

    Article  Google Scholar 

  31. 31

    Lebrec D, Godfarb G, Degott C, Rueff B, Benhamou JP . Transvenous liver biopsy: an experience based on 1000 hepatic tissue samplings with this procedure. Gastroenterology 1982; 83: 338–340.

    CAS  PubMed  Google Scholar 

  32. 32

    Groszmann R, Vorobioff JD, Gao H . Measurement of portal pressure: when, how and why to do it. Clin Liver Dis 2006; 10: 499–512.

    Article  Google Scholar 

  33. 33

    Bosch J, Garcia-Pagan JC, Berzigotti A, Abraldes JG . Measurement of portal pressure and its role in the management of chronic liver disease. Semin Liv Dis 2006; 26: 348–362.

    Article  Google Scholar 

  34. 34

    Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, Cummings OW et al. for the nonalcoholic steatohepatitis clinical research network. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005; 41: 1313–1321.

    Article  Google Scholar 

  35. 35

    World Health Organisation (WHO). Obesity: preventing and managing the global epidemic report of a WHO consultation. WHO Technical Report Series 894. World Health Organisation: Geneva, 2000.

  36. 36

    Alberti KG, Zimmer P . Definition, diagnosis and classification of diabetes mellitus and its complications. Part I. Diagnosis and classification of diabetes mellitus. Provisional report of a WHO consultation. Diabet Med 1998; 15: 539–553.

    CAS  Article  Google Scholar 

  37. 37

    Ducimetiere P, Eschwege E, Papoz L, Richard JL, Claude JR, Rosselin G . Relationship of plasma insulin levels to the incidence of myocardial infarction and coronary heart disease mortality in a middle-aged population. Diabetologia 1980; 19: 205–210.

    CAS  Article  Google Scholar 

  38. 38

    Puoti C, Bellis L . Steatosis and portal hypertension. Eur Rev Med Pharmacol Sci 2005; 9: 285–290.

    CAS  PubMed  Google Scholar 

  39. 39

    Després JP, Lemieux I, Prud’homme D . Treatment of obesity: need to focus on high risk abdominally obese patients. Br Med J 2001; 322: 716–720.

    Article  Google Scholar 

  40. 40

    Clark JM, Brancati FL, Diehl AM . Nonalcoholic fatty liver disease. Gastroenterology 2002; 122: 1649–1657.

    Article  Google Scholar 

  41. 41

    Rinella M, Alonso E, Rao S, Whitington P, Fryer J, Abecassis M et al. Body mass index as a predictor of hepatic steatosis in living liver donors. Liver Transplant 2001; 7: 409–413.

    CAS  Article  Google Scholar 

  42. 42

    Egushi Y, Egushi T, Mizuka T, Ide Y, Yasutake T, Iwakini R et al. Visceral fat accumulation and insulin resistance are important factors in non-alcoholic liver disease. J Gastroenterol 2006; 41: 462–469.

    Article  Google Scholar 

  43. 43

    Stranges S, Dorn JM, Muti P, Freudenheim JL, Farinaro E, Russell M et al. Body fat distribution, relative weight, and liver enzyme levels: a population-based study. Hepatology 2004; 39: 754–763.

    Article  Google Scholar 

  44. 44

    Ahima RS, Flier JS . Adipose tissue as an endocrine organ. Trends Endocrinol Metab 2000; 11: 327–332.

    CAS  Article  Google Scholar 

  45. 45

    Jarrar MH, Baranova A, Collantes R, Ranard B, Stepanova M, Bennett C et al. Adipokines and cytokines in non-alcoholic fatty liver disease. Aliment Pharmacol Ther 2008; 27: 412–421.

    CAS  Article  Google Scholar 

  46. 46

    Neuschwander-Tetri BA . Fatty liver and the metabolic syndrome. Curr Opin Gastroenterol 2007; 23: 193–198.

    CAS  Article  Google Scholar 

  47. 47

    Bugianesi E, McCullough AJ, Marchesini G . Insulin resistance: a metabolic pathway to chronic liver disease. Hepatology 2005; 42: 987–1000.

    CAS  Article  Google Scholar 

  48. 48

    Verrijken A, Francque S, Mertens I, Talloen M, Peiffer F, Van Gaal L . Visceral adipose tissue and inflammation correlate with elevated liver tests in a cohort of overweight and obese patients. Int J Obes 2010; 34: 899–907.

    CAS  Article  Google Scholar 

  49. 49

    Francque S . Non-alcoholic fatty liver disease (NAFLD) and Non-alcoholic Steatohepatitis (NASH). In: Van Damme P, van Herck K, Michielsen P, Francque S, Shouval D (eds). Chronic Hepatitis & Other Liver Disease. Oxford Textbook of Public Health 5th edn. Oxford University Press: Oxford, UK, 2009. (Chapter 9) 16, pp 1249–1263.

    Google Scholar 

  50. 50

    de Marco R, Locatelli F, Zoppini G, Verlato G, Bonora E, Muggeo M . Cause-specific mortality in type 2 diabetes. The Verona Diabetes Study. Diabetes Care 1999; 22: 756–761.

    CAS  Article  Google Scholar 

  51. 51

    Abrams GA, Kunde SS, Lazenby AJ, Clements RH . Portal fibrosis and hepatic steatosis in morbidly obese subjects: a spectrum of non-alcoholic fatty liver disease. Hepatology 2004; 40: 475–483.

    Article  Google Scholar 

  52. 52

    Reaven GM . Role of insulin resistance in human diabetes. Diabetes 1988; 37: 1595–1607.

    CAS  Article  Google Scholar 

Download references


This work is part of the project ‘Hepatic and adipose tissue and functions in the metabolic syndrome’ (HEPADIP), which is supported by the European Commission as an Integrated Project under the 6th Framework Program (Contract LSHM-CT-2005-018734).

Author information



Corresponding author

Correspondence to S Francque.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Francque, S., Verrijken, A., Mertens, I. et al. Visceral adiposity and insulin resistance are independent predictors of the presence of non-cirrhotic NAFLD-related portal hypertension. Int J Obes 35, 270–278 (2011).

Download citation


  • non-alcoholic fatty liver
  • visceral fat
  • insulin resistance
  • metabolic syndrome
  • portal hypertension

Further reading