Nonalcoholic fatty liver disease (NAFLD) has a high prevalence in the general population. Brachial artery flow-mediated dilation (FMD) is a surrogated marker of early atherosclerosis. Few data investigating the relation between FMD, NAFLD, and cardiovascular (CV) risk are available. We recruited 367 consecutive outpatients with cardiometabolic risk factors who underwent ultrasound scanning for liver steatosis and FMD. Mean age was 54.2 ± 12.2 years, and 37% were women. NAFLD was present in 281 patients (77%). Median FMD was 5.1%. FMD was significantly reduced in patients with NAFLD (p <0.001), diabetes (p = 0.001), history of coronary heart disease (p = 0.034), and metabolic syndrome (p = 0.050) and in those taking antihypertensive drugs (p = 0.022). Women disclosed greater FMD than males (p = 0.033). Moreover, FMD inversely correlated with age (Spearman rank correlation test [Rs], −0.171; p = 0.001), waist circumference (Rs, −0.127; p = 0.016), fasting blood glucose (Rs, −0.204; p <0.001), and gamma-glutamyl transpeptidase (Rs, −0.064; p = 0.234). At multivariate regression analysis, fasting blood glucose (β, −0.148; p = 0.008), age (β, −0.158; p = 0.005), and the presence of NAFLD (β, −0.132; p = 0.016) inversely correlated with FMD, whereas female gender predicted a better FMD (β, 0.125; p = 0.022). FMD and Framingham Risk Score (FRS) were inversely correlated (Rs, −0.183; p <0.001). After dividing patients into low (FRS <10; FMD, 5.5% [3.1% to 8.9%]), intermediate (FRS 10 to 20; FMD, 4.9% [2.7% to 7.5%]), and high (FRS >20; FMD, 3.3% [1.7% to 4.5%]) risk, FMD significantly decreased across risk classes of FRS (p = 0.003). At multivariate regression analysis, both FRS (β, −0.129; p = 0.016) and NAFLD (β, −0.218; p <0.001) were variables independently associated with FMD. In conclusion, the presence of NAFLD and FRS inversely correlated with FMD.
Nonalcoholic fatty liver disease (NAFLD) represents the most common and emerging liver disease worldwide. NAFLD includes a spectrum of diseases ranging from simple fatty liver to nonalcoholic steatohepatitis, which may progress to fibrosis and even cirrhosis and hepatocellular carcinoma. A growing number of evidence suggest that NAFLD is associated with greater overall mortality and that NAFLD is likely to be associated with increased risk of future cardiovascular (CV) diseases. Nevertheless, it is not clear if this increased CV risk is conferred by the presence of several CV risk factors associated to the metabolic syndrome, which are common in patients with NAFLD, or if NAFLD might be itself a mediator of atherosclerosis. It has been reported that patients with NAFLD show early signs of atherosclerosis. One of the earliest manifestation of atherosclerosis is the presence of the endothelial dysfunction, occurring even in the absence of angiographic evidence of disease. Brachial artery flow-mediated dilation (FMD) is the most often noninvasive test used for assessing endothelial function as the result of endothelial release of nitric oxide. An impaired FMD was found in a wide variety of CV and metabolic diseases associated to chronic low-grade inflammation, oxidative stress, and metabolic abnormalities. Moreover, FMD predicted future CV events in patients with and in those without CV diseases. Aim of the study was to investigate the association between endothelial dysfunction, evaluated by brachial FMD, and NAFLD in a cohort of patients presenting with cardiometabolic risk factors.
Methods
The study included 384 consecutive patients who were referred to the Day Service of Internal Medicine of the Policlinico Umberto I University Hospital in Rome, to undergo clinical evaluation for suspected cardiometabolic disorders. At baseline, all patients underwent a complete physical examination and ultrasonographic evaluation for liver steatosis (see in the following). History of previous coronary heart disease (CHD) including myocardial infarction and cardiac revascularization and anthropometric data were recorded; routine laboratory analyses were performed in the fasting condition including aspartate aminotransferase, alanine aminotransferase (ALT), gamma-glutamyl transpeptidase (γ-GT), blood glucose, and lipid profile.
Inclusion criteria for the study were no history of current or past excessive alcohol drinking as defined by an average daily consumption of alcohol >20 g; negative tests for the presence of hepatitis B surface antigen and antibody to hepatitis C virus; absence of history and clinical, biochemical, and ultrasound findings consistent with cirrhosis and other chronic liver diseases; and no current supplementation with vitamin E and other antioxidants.
The presence of arterial hypertension and diabetes mellitus was defined following international guidelines. Metabolic syndrome was defined according to modified ATP-III criteria. Framingham Risk Score (FRS) was calculated according to Wilson et al.
All patients provided signed written informed consent at entry. The study protocol was approved by the local ethical board of Sapienza University of Rome and was carried out according to the principles of the Declaration of Helsinki.
We performed liver ultrasound scanning (US) to assess the degree of steatosis. All US were performed by the same operator who was blinded to laboratory values using a GE VividS6 apparatus equipped with a convex 3.5-MHz probe. Liver steatosis was defined according to Hamaguchi criteria on the basis of the presence of abnormally intense, high-level echoes arising from the hepatic parenchyma, liver–kidney difference in echo amplitude, echo penetration into deep portion of the liver, and clarity of liver blood vessel structure.
Ultrasound assessment of endothelial-dependent FMD of brachial artery was investigated according to guidelines. Briefly, the study was performed in a temperature-controlled room (22°C) with the subjects in a resting, supine state between 8 and 10 A.M. after at least an 8-hour fasting. Brachial artery diameter was imaged using a 7.5-Mhz linear array transducer ultrasound system (Siemens) equipped with electronic calipers, vascular software for 2-dimensional imaging, color and spectral Doppler, and internal electrocardiogram; the brachial artery was imaged at a location 3 to 7 cm above the antecubital crease. To create a flow stimulus in the brachial artery, a sphygmomanometric cuff was placed on the forearm; the cuff was inflated at least 50 mm Hg above systolic pressure to occlude artery inflow for 5 minutes. All vasodilatation measurements were made at the end of diastole. FMD was expressed as a change in poststimulus diameter evaluated as a percentage of the baseline diameter.
Statistical analysis
Categorical variables were reported as counts (percentages) and continuous variables as means ± standard deviation (SD) or median and interquartile range (IQR) unless otherwise indicated. Independence of categorical variables was tested by chi-square test. Normal distribution of parameters was assessed by Kolmogorov–Smirnov test. The Student unpaired t test and the Pearson product–moment correlation analysis were used for normally distributed continuous variables. For group comparisons, we used the analysis of variance. Appropriate nonparametric tests (Mann–Whitney U test, Kruskal–Wallis test, and Spearman rank correlation test [Rs]) were used for all the other variables. Stepwise multivariable linear regression analysis was used to assess factors influencing FMD, and collinear variables were excluded from the multivariate analysis (tolerance ≤0.5 or variance inflation factor ≥2). We built a first model (model A) including presence of NAFLD, female gender, age, waist circumference, smoking, fasting blood glucose, insulin, systolic blood pressure, history of CHD, ALT, γ-GT, triglycerides, and high-density lipoprotein as covariates. Aspartate aminotransferase, diabetes, and use of antihypertensive drugs were not used for multivariate analysis because they were collinear with ALT, fasting blood glucose, and systolic blood pressure, respectively.
To better examine the relation between FMD and CV risk, we built a second model (model B) including the composite variable of FRS and the presence of NAFLD as covariates. In this second model, we excluded patients with previous coronary artery disease as FRS derived from a community-based cohort free from CV events. Only p values <0.05 were considered as statistically significant. All tests were 2 tailed and analyses performed using computer software packages (IBM SPSS Statistics v20.0, Armonk, NY). For the sample size determination, we estimated that 170 patients were required to have a 90% chance of detecting, as significant at the 5% level, a difference in FMD values of 1% between patients with and without NAFLD with an SD of 2%.
Results
Based on the previously listed exclusion criteria, 17 patients (4.4%) were excluded from the analysis and 367 were enrolled in the study. In the whole population, mean age was 54.2 ± 12.2 years, and 37% were women. US diagnosis of NAFLD was made in 281 patients (77%): mild steatosis in 19%, moderate steatosis in 43%, and severe steatosis in 38%. Table 1 reports clinical and biochemical characteristics of patients with and without NAFLD. Patients with NAFLD disclosed a greater prevalence of diabetes and elevation of liver function tests and were more likely to suffer from atherogenic dyslipidemia and central obesity ( Table 1 ). Systolic and diastolic blood pressure, clinical history of CHD, and FRS did not differ between the two groups.
| NAFLD (n=281) | Non-NAFLD (n=86) | p | |
|---|---|---|---|
| Age (years) | 53.7±11.9 | 54.9±13.2 | 0.444 |
| Women | 35(%) | 41(%) | 0.378 |
| Cigarette Smoker | 23(%) | 27(%) | 0.482 |
| Use of anti-hypertensive drugs | 61(%) | 61(%) | 1.000 |
| Systolic blood pressure (mmHg) | 130.7±13.0 | 129.7±14.3 | 0.563 |
| Diastolic blood pressure (mmHg) | 80.1±8.1 | 79.9±12.0 | 0.831 |
| Body Mass Index (kg/m 2 ) | 30.8±5.1 | 27.1±3.0 | <0.001 |
| Waist circumference (cm) | 108.4±11.8 | 96.7±9.2 | <0.001 |
| Metabolic syndrome | 67(%) | 36(%) | <0.001 |
| Diabetes Mellitus | 32(%) | 14(%) | 0.001 |
| History of Coronary Heart Disease | 8(%) | 9(%) | 0.826 |
| Flow-Mediated Dilation ∗ | 4.6 [2.6-8.1] (%) | 6.0 [4.0-10.3] (%) | <0.001 |
| Framingham Risk Score ∗ | 6.0 [2.0-12.0] | 5.0 [1.0-11.7] | 0.279 |
| Fasting plasma glucose (mg/dl) | 109.3±36.4 | 95.7±20.0 | 0.001 |
| Total cholesterol (mg/dl) | 198.3±40.6 | 202.0±43.8 | 0.462 |
| Low density lipoprotein (mg/dl) | 117.9±33.8 | 121.8±39.9 | 0.382 |
| High density lipoprotein (mg/dl) | 47.6±14.3 | 55.0±14.2 | <0.001 |
| Triglycerides (mg/dl) | 172.8±138.2 | 121.0±63.4 | 0.001 |
| Aspartate aminotransferase (U/l) | 26.0±15.1 | 20.4±7.0 | 0.001 |
| Alanine aminotransferase (U/l) | 36.6±24.2 | 23.2±21.4 | <0.001 |
| Gamma-glutamyl transpeptidase (U/l) ∗ | 28.0 [18.0-44.5] | 20.0 [14.0-25.5] | <0.001 |
Median FMD was 5.1% (2.9% to 8.4%). We found a significant reduction of median FMD in patients with NAFLD (p <0.001), diabetes (p = 0.001), metabolic syndrome (p = 0.050), and history of CHD (p = 0.034) and in those taking antihypertensive drugs (p = 0.022; Table 2 ). Women showed a significant greater median value of FMD than males (p = 0.033), and no differences were found in smokers than nonsmokers (p = 0.470). Moreover, age (Rs, −0.171; p = 0.001), waist circumference (Rs, −0.127; p = 0.016), fasting blood glucose (Rs, −0.204; p <0.001), and γ-GT (Rs, −0.064; p = 0.234) were significantly associated to FMD ( Table 3 ). FMD was not correlated with the degree of US liver steatosis (Rs, 0.047; p = 0.436).
| Yes | No | p | |
|---|---|---|---|
| Women | 5.6 [3.4-9.8] | 4.8 [2.8-7.9] | 0.033 |
| Metabolic Syndrome | 4.8 [2.6-8.2] | 5.6 [3.5-8.7] | 0.050 |
| Diabetes Mellitus | 4.3 [2.2-6.5] | 5.5 [3.3-8.9] | 0.001 |
| History of Coronary Heart Disease | 3.1 [2.3-5.8] | 5.2 [3.0-8.5] | 0.034 |
| Smoker | 5.2 [3.3-8.5] | 5.2 [2.8-8.4] | 0.470 |
| Rs | p | |
|---|---|---|
| Framingham Risk Score | -0.183 | <0.001 |
| Age | -0.171 | 0.001 |
| Waist circumference | -0.127 | 0.016 |
| Fasting blood glucose | -0.204 | <0.001 |
| Systolic blood pressure | -0.073 | 0.165 |
| Diastolic blood pressure | -0.013 | 0.811 |
| Aspartate aminotransferase | -0.007 | 0.892 |
| Alanine aminotransferase | -0.012 | 0.893 |
| Gamma-glutamyl transpeptidase | -0.064 | 0.234 |
A stepwise multiple linear regression analysis (model A, Table 4 ) found that fasting blood glucose (β, −0.148; p = 0.008), age (β, −0.158; p = 0.005), and the presence of NAFLD (β, −0.132; p = 0.016) were inversely associated with values of FMD, whereas female gender predicted a better FMD (β, 0.125; p = 0.022).
| Model A | ||||||
|---|---|---|---|---|---|---|
| B | SE | Beta | p | 95,0% C.I. for B | ||
| Fasting blood glucose | -0.016 | 0.006 | -0.148 | 0.008 | -0.027 | -0.004 |
| Age | -0.048 | 0.017 | -0.158 | 0.005 | -0.081 | -0.015 |
| Presence of Non-Alcoholic Fatty Liver Disease | -1.456 | 0.601 | -0.132 | 0.016 | -2.639 | -0.274 |
| Female | 0.930 | 0.404 | 0.125 | 0.022 | 0.135 | 1.726 |
| Waist circumference | -0.007 | 0.018 | -0.022 | 0.717 | -0.042 | 0.029 |
| Smoker | 0.214 | 0.469 | 0.026 | 0.648 | -0.708 | 1.137 |
| Insulin | 0.009 | 0.005 | 0.089 | 0.108 | -0.002 | 0.019 |
| Systolic blood pressure | 0.016 | 0.017 | 0.059 | 0.348 | -0.017 | 0.048 |
| Triglycerides | 0.001 | 0.002 | 0.055 | 0.342 | -0.002 | 0.005 |
| Gamma-glutamyl transpeptidase | 0.001 | 0.003 | 0.005 | 0.928 | -0.006 | 0.007 |
| High density lipoprotein | -0.003 | 0.014 | -0.015 | 0.812 | -0.031 | 0.025 |
| History of Coronary Heart Disease | -0.506 | 0.725 | -0.039 | 0.485 | -1.933 | 0.920 |
| Alanine aminotransferase | -0.004 | 0.009 | -0.027 | 0.662 | -0.022 | 0.014 |
| Model B | ||||||
| Presence of Non-Alcoholic Fatty Liver Disease | -1.855 | 0.452 | -0.218 | <0.001 | -2.744 | -0.966 |
| Framingham Risk Score | -0.067 | 0.028 | -0.129 | 0.016 | -0.121 | -0.013 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
