SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Disclosure
  8. REFERENCES

We assessed the differential contribution of nonalcoholic steatohepatitis (NASH) and visceral adiposity to nontraditional cardiovascular risk biomarkers in adult men. We enrolled 45 consecutive, overweight, male patients with biopsy-proven NASH, 45 overweight male patients without ultrasound-diagnosed hepatic steatosis, and 45 healthy male volunteers. All participants were matched for age; NASH and overweight patients were also matched for BMI and visceral adiposity (as estimated by abdominal ultrasonography). Nontraditional cardiovascular risk biomarkers were measured in all participants. Plasma concentrations of high-sensitivity C-reactive protein (hs-CRP), fibrinogen, plasminogen activator inhibitor-1 (PAI-1) activity, and adiponectin were markedly different among the groups; the lowest values (the highest for adiponectin) were in nonobese healthy subjects, intermediate in overweight nonsteatotic patients, and the highest (the lowest for adiponectin) in those with biopsy-proven NASH. The marked differences in these cardiovascular risk biomarkers that were observed between overweight and NASH patients were only slightly weakened after adjustment for age, BMI, smoking, plasma triglycerides, and insulin resistance (IR) as assessed by homeostasis model assessment (HOMA). In multivariate regression analysis, NASH and visceral adiposity predicted cardiovascular risk biomarkers independently of potential confounders. In conclusion, our results suggest that NASH can predict a more atherogenic risk profile in a manner that is partly independent from the contribution of visceral adiposity in adult men.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Disclosure
  8. REFERENCES

It is well known that visceral adiposity is strongly associated with the typical features of the metabolic syndrome (1,2) and chronic inflammation (3), which play a role in the development and progression of cardiovascular disease (CVD) (1,3,4,5).

Recent studies have shown that nonalcoholic fatty liver disease (NAFLD) and its necroinflammatory form—nonalcoholic steatohepatitis (NASH)—are associated with visceral adiposity, other cardiometabolic risk factors, and increased plasma inflammatory biomarkers (6,7,8,9,10,11,12,13), suggesting that NAFLD/NASH is an additional feature of the metabolic syndrome.

Recently, it has been reported that visceral adiposity and NAFLD independently predict glucose intolerance and atherogenic dyslipidemia in men (14,15). However, it remains still unclear/controversial whether NAFLD/NASH contributes to chronic inflammation, and the resulting increased CVD risk profile, independent from any effect of visceral adiposity. Recently, Kolak et al. have shown that adipose tissue inflammation is present only in individuals with hepatic steatosis independent of obesity (16), thus suggesting either that the presence of adipose tissue inflammation is a determinant of who gets NAFLD/NASH or that the release of mediators from the fatty liver is a cause of adipose tissue inflammation and therefore systemic inflammation, type 2 diabetes, and CVD.

It is conceivable that NASH is more important than steatosis alone, and that the presence of an inflamed liver adds further to the proinflammatory cytokine release beyond the effects of visceral adiposity and steatosis alone. Resolution of this controversy may help to clarify the underlying biological mechanisms of chronic inflammation in obesity, and may be of clinical importance in planning preventive and therapeutic strategies.

Aim of this study was to assess the differential contribution of NASH and visceral adiposity to plasma inflammatory biomarkers in adult men.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Disclosure
  8. REFERENCES

We enrolled 45 consecutive, overweight, male patients with biopsy-proven NASH and 45 matched, overweight, male patients without hepatic steatosis on ultrasound. Forty-five nonobese, healthy, male volunteers were also recruited from hospital staff members and relatives (none of them was working for or training with any of the authors). No participants had any clinical evidence of cancer, cirrhosis, overt nephropathy, or CVD. None of them was taking any medications known to affect plasma adiponectin and inflammatory biomarkers. The local ethics committee approved the study. All participants provided their written informed consent.

The sample size was not calculated a priori, but determined upon practical considerations. In fact, we included all of the 45 consecutive NASH participants (i.e., cases), who had been previously included in a pilot study assessing the relationship between plasma high-sensitivity C-reactive protein (hs-CRP) concentrations and NAFLD histology (12). Then, we enrolled 45 healthy male volunteers and 45 overweight nonsteatotic men, who were selected in a 1:1 ratio with the cases. All participants were selected for matching for age; the NASH, overweight, and nonsteatotic patients were also selected for matching for BMI and visceral adiposity. Healthy subjects and overweight nonsteatotic patients drank less than 20 g/day of alcohol, had seronegativity for viral hepatitis B and C, normal serum liver enzymes, and negative liver ultrasonography. Among NASH patients, the diagnosis was based on liver biopsy (17), which was scored by an experienced hepatopathologist blinded to subjects' details, and exclusion of other secondary causes of chronic liver disease (alcohol abuse, viral hepatitis, autoimmune hepatitis, medications). NASH was defined as the presence of steatosis plus lobular inflammation plus hepatocellular ballooning or steatosis plus any stage of fibrosis.

BMI was calculated by dividing weight in kilograms by height in meters squared. Waist circumference was measured at the level of the umbilicus. Blood pressure was measured in triplicate using a standard mercury manometer.

Venous blood was drawn in the morning after an overnight fast. Serum liver enzymes and other biochemical blood measurements were determined by standard laboratory procedures (DAX 96; Bayer Diagnostics, Milan, Italy). Serology for viral hepatitis B and C was assessed in all participants. LDL cholesterol was calculated according to Friedewald's equation. Insulin resistance was estimated by the homeostasis model assessment insulin resistance score (HOMA-IR score) (18). Plasma concentrations of fibrinogen (IL-test-PT-fibrinogen HS; Instrumentation Laboratory, Lexington, MA), hs-CRP (by a highly sensitive immunoturbidimetric assay; Roche Diagnostics, Milan, Italy), adiponectin (by an enzyme-linked immunosorbent assay method; B-Bridge International, San Jose, CA), and plasminogen activator inhibitor-1 activity (PAI-1) (by a chromogenic substrate assay; Spectrolyse/PL, Biopool, Umeå, Sweden) were measured in duplicate. Intra- and inter-assay coefficients of variation were 3.8 and 5.1%, 6.5 and 7.2%, 7.0 and 8.4%, and 3.5 and 7.2% for hs-CRP, fibrinogen, PAI-1 activity, and adiponectin, respectively.

Abdominal ultrasonography scanning was performed in all participants by an experienced radiologist, who was blinded to participants' details. The diagnosis of hepatic steatosis was made on the basis of characteristic sonographic features, i.e., evidence of diffuse hyperechogenicity of liver relative to kidneys, ultrasound beam attenuation, and poor visualization of intrahepatic structures (7,8). It is known that ultrasonography has a sensitivity of ∼90% and a specificity of ∼95% in detecting moderate and severe steatosis, but this sensitivity is reduced when hepatic fat infiltration upon liver biopsy is <33% (19). The intraobserver variability for the ultrasound diagnosis of steatosis was within 3%. Ultrasound intra-abdominal (visceral) fat was measured ∼5 cm from umbilicus on the xiphoumbilical line as the distance from the internal face of the rectus abdominis muscle and the anterior wall of the aorta. The intraindividual reproducibility of ultrasound measurements of visceral fat was within 1%. In accord with others (20,21), we have shown that abdominal ultrasound is a reliable and accurate method for estimating the amount of visceral adiposity. Ultrasound intra-abdominal fat was strongly correlated with visceral fat area as measured by computed tomography in adult men (r = 0.86; P < 0.0001) (22).

Statistical analysis

Data are means ± s.d. unless otherwise indicated. Skewed variables (hs-CRP, adiponectin, PAI-1, HOMA-IR score and triglycerides) were logarithmically transformed to improve normality before analysis and then back-transformed to their natural units for presentation in tables and figure. Statistical analyses included one-way ANOVA (followed by the Fisher's protected least significance difference test for multiple comparisons between groups), χ2 test (for categorical variables), univariate linear correlation, and analysis of covariance. Nonparametric statistical tests were also used, but because the results were identical to those obtained by parametric procedures, only the latter were presented. Multivariable linear regression analyses were also performed in pooled subjects (n = 135) to identify the independent predictors of plasma adiponectin and inflammatory biomarkers that were entered as dependent variables; the covariates included in these regression models, together with NASH and visceral adiposity, were age, BMI, smoking, IR, systolic blood pressure, and triglycerides. P < 0.05 was considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Disclosure
  8. REFERENCES

Among NASH patients, the liver histopathology results showed NASH with fibrosis score of 0 in 13 participants, NASH/fibrosis score 1 in 16 participants, NASH/fibrosis score 2 in 10 participants, and NASH/fibrosis score 3 in 6 participants. None had cirrhosis (a fibrosis score of 4).

The clinical and biochemical characteristics of participants are shown in Table 1. Because of the study design, the participants were males and almost identical in terms of age, the groups of NASH and overweight patients were also very similar for BMI, waist circumference, and ultrasound intra-abdominal fat. Overweight, nonsteatotic patients had remarkably higher BMI, waist circumference, ultrasound intra-abdominal fat, blood pressure, HOMA-IR score, triglycerides, and lower high-density lipoprotein cholesterol than healthy controls. Conversely, NASH patients had higher HOMA-IR score and plasma triglycerides than overweight nonsteatotic patients despite similar values of BMI and visceral adiposity. Most important, plasma hs-CRP, fibrinogen, PAI-1 activity, and adiponectin concentrations were markedly different among the groups (P < 0.001 for trends by one-way ANOVA; not shown); the lowest values (the highest for adiponectin) were in healthy controls, intermediate in overweight nonsteatotic patients, and the highest (the lowest for adiponectin) in those with NASH. In particular, the marked differences in these biomarkers that were observed between the NASH and overweight, nonsteatotic patients were little affected by adjustment for age, BMI, smoking, triglycerides, and HOMA-IR score (Table 1; last column); only the differences in plasma adiponectin concentrations were attenuated after adjustment for the above potential confounders.

Table 1.  Clinical and biochemical characteristics of nonobese healthy subjects, overweight, nonsteatotic patients, and those with biopsy-proven NASH
inline image

As shown in Figure 1, the severity of liver histology in NASH patients (i.e., NASH/fibrosis stage) was strongly associated with increasing plasma hs-CRP, PAI-1, and fibrinogen concentrations and decreasing plasma adiponectin concentrations. The results remained essentially unchanged after adjustment for age, BMI, ultrasound intra-abdominal fat, and IR (data not shown).

image

Figure 1. Means (±s.d.) of plasma high-sensitivity C-reactive protein (hs-CRP), plasminogen activator inhibitor-1 (PAI-1) activity, adiponectin and fibrinogen concentrations in relation to the histological severity of nonalcoholic steatohepatitis (i.e., NASH/fibrosis stage). P values for trends were assessed by one-way ANOVA. NASH, nonalcoholic steatohepatitis.

Download figure to PowerPoint

Table 2 shows the age-adjusted linear correlations of plasma adiponectin and inflammatory biomarkers with conventional vascular risk factors in pooled subjects. Plasma inflammatory biomarkers were positively correlated to BMI, waist circumference, ultrasound intra-abdominal fat, IR, triglycerides, and blood pressure, and negatively to high-density lipoprotein cholesterol. Opposite results were found for adiponectin concentrations. Importantly, the strength of relationships of plasma inflammatory biomarkers and adiponectin with BMI and waist circumference was generally weaker than that observed with ultrasound intra-abdominal fat.

Table 2.  Age-adjusted linear correlations of plasma adiponectin and biomarkers of inflammation with main vascular risk factors in the whole cohort of participants (n = 135)
inline image

In multivariate linear regression analysis (Table 3), NASH and ultrasound intra-abdominal fat were among the strongest predictors of measured cardiovascular risk biomarkers after adjusting for potential confounders. HOMA-IR score and triglycerides were also independently associated with plasma PAI-1 and adiponectin concentrations.

Table 3.  Multivariate linear regression analyses showing predictors of plasma adiponectin and inflammatory biomarkers (included as dependent variables in the regression models) in the whole cohort of participants (n = 135)
inline image

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Disclosure
  8. REFERENCES

Our results reinforce the importance of visceral adiposity as a strong predictor of typical components of the metabolic syndrome, chronic inflammation, and reduced adiponectin concentrations. These findings are consistent with the notion that visceral adipose tissue is not merely a fat storage and mobilization organ, but it is also an endocrine organ, which plays an active role in the development of NASH (7,8,23,24) and CVD (1,2,3), possibly through its multiple secreted factors, such as free fatty acids, hormones, adiponectin, interleukin-6, and other proinflammatory adipocytokines.

More important, our results indicate that NASH patients are more insulin resistant and have remarkably higher plasma hs-CRP, fibrinogen, and PAI-1 activity, and lower adiponectin concentrations than overweight nonsteatotic patients with comparable values of visceral adiposity, thus suggesting that NASH can predict a more atherogenic risk profile in a manner that is partly independent from the contribution of visceral adiposity. This finding is further supported by the strong, graded relationships of these inflammatory biomarkers with the histological severity of NASH independent of visceral adiposity, and by the results of multivariate regression analyses showing that NASH predicts plasma inflammatory biomarkers independent of visceral adiposity and other potential confounders.

Our findings extend the work of previous studies showing that NASH patients have higher plasma inflammatory parameters and lower adiponectin concentrations than nonsteatotic controls independently of underlying metabolic abnormalities (10,11,12,13,25). In these studies, however, visceral adiposity was estimated only by waist circumference, which is a crude and indirect estimate of visceral adiposity (2). So, we cannot be certain that visceral adiposity, when measured using more accurate imaging methods, could give a stronger independent contribution to the increased plasma inflammatory biomarkers seen in NASH. We have shown previously that the associations between ultrasound-diagnosed NAFLD and plasma inflammatory biomarkers were no longer apparent after adjusting for computed tomography-measured visceral fat in lean, healthy men (9). These apparently discrepant results can be explained by the larger sample size of this study and by differences in the study population (most participants of that study were lean, healthy men with normal serum liver enzymes and, therefore, more likely to have simple steatosis rather than NASH as in the present one) and study methodology (liver biopsy vs. ultrasound for diagnosing NAFLD). Moreover because in these previous studies (9,12,13,25) the independent contribution of visceral fat and fatty liver to increased plasma inflammatory biomarkers was inferred only by the results of multivariate regression analyses, we believe that some degree of caution should be exercised in interpreting these results. On the contrary in our study, we selected three groups of participants with different degrees of visceral adiposity and fatty liver.

Our findings provide further evidence that NASH is associated with an increased risk of developing CVD (26,27,28,29,30,31,32,33,34,35,36,37), and suggest possible biological mechanisms by which NASH may contribute to accelerated atherosclerosis. Notably, our observation that visceral adiposity and NASH were related to selected nontraditional cardiovascular risk biomarkers independently of each other argues against a common biological mechanism, and suggests that both fat depots may carry independent health risk. Thus, although it is not possible to draw conclusions about causality in our cross-sectional study, these findings suggest that NASH is not simply a marker of CVD but may also be involved in its pathogenesis. There is now growing evidence that NASH contributes to chronic inflammation through the systemic release of proinflammatory mediators from the liver, including CRP, fibrinogen, and PAI-1 (8,24,38,39). Our observation that NASH is associated with hypoadiponectinemia independent of visceral adiposity is not readily explained, but it could partly be linked to an increased production of tumor necrosis factor-α by the inflamed liver, which would inhibit adiponectin release from adipose tissue (24). Nevertheless, others have shown reduced plasma adiponectin concentrations in NASH patients independently of waist circumference and other metabolic abnormalities (40,41,42,43).

Our study has several strengths, including the large number of participants, the complete nature of the data set, the ability to adjust for multiple confounders, the histological diagnosis of NASH, and the measurement of abdominal visceral adiposity by ultrasonography, which is an accurate and reliable method for estimating the amount of visceral adipose tissue (20,21,22). Additionally, our patients are free of diagnosed CVD and cirrhosis; the evaluation of patients with such complications would almost certainly have confounded interpretation of the data.

Despite the comprehensive nature of the data set, there are some possible limitations to our study. First, the cross-sectional design of our study precludes the establishment of causal or temporal relationships between NASH, visceral adiposity, and nontraditional cardiovascular risk biomarkers. Prospective studies will be required to sort out the time sequence of events. Second, although our results have been adjusted for HOMA-IR score, a reliable method for estimating IR (18), we did not directly measure insulin sensitivity (by euglycemic clamp) in our population, so we cannot be certain that identical results could be obtained after adjustment for the clamp-measured IR. Third, the exclusion of NAFLD among healthy subjects and overweight patients was based on patient history, blood testing, and ultrasound imaging, but was not confirmed by liver biopsy. It is known that ultrasound has a good sensitivity and specificity in detecting moderate and severe steatosis, but this sensitivity is reduced when hepatic fat infiltration is <33% (19). However, although some nondifferential misclassification of NAFLD on the basis of ultrasound is likely (i.e., some of the controls or overweight patients could have underlying NAFLD, despite normal liver enzymes and a negative ultrasonography), this limitation would serve to attenuate the magnitude of our effect measures toward the null; thus, our results can probably be considered conservative estimates of the relationship between NASH and increased nontraditional cardiovascular risk biomarkers. Finally, whether these observations can also be extended to women remains to be determined.

In conclusion, our findings suggest that NASH can predict a more atherogenic risk profile beyond the contribution of visceral adiposity in adult men. Future studies are needed to elucidate the underlying molecular mechanisms before causality can be firmly established.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Disclosure
  8. REFERENCES