| |
Lipids are Lower in Chronic HCV
|
| |
| |
Serum lipid pattern in chronic hepatitis C: histological and virological correlations
Journal of Viral Hepatitis
Volume 13 Page 56 - January 2006
"....In this study, patients with chronic HCV infection were found to have significantly lower levels of serum TC, HDL-C and LDL-C than normal adults matched for age and sex....h
D. Siagris1, M. Christofidou2, G. J. Theocharis1, N. Pagoni1, C. Papadimitriou3, A. Lekkou1, K. Thomopoulos1, I. Starakis1, A. C. Tsamandas4 and C. Labropoulou-Karatza1
Summary.
Lipoproteins are closely connected to the process of hepatitis C virus (HCV) infection. The aim of this study was to evaluate the lipaemic profile in patients with chronic HCV infection, and to identify any association between serum lipid levels and viral load, HCV genotype or liver histology. Total cholesterol (TC), high-density lipoprotein-cholesterol (HDL-C), low-density lipoprotein-cholesterol (LDL-C) and triglycerides (TG) were measured in the sera of 155 patients with chronic HCV infection and 138 normal subjects, matched for age and sex. Viral parameters and liver histology were evaluated in HCV-infected patients.
Serum TC (P < 0.0005), HDL-C (P < 0.0005) and LDL-C (P < 0.0005) were lower in chronic hepatitis C patients compared with controls. Grading score was positively correlated with TC and LDL-C. Patients with HCV genotype 3a had significantly lower levels of TC, HDL-C, LDL-C, higher viral load and higher frequency of hepatic steatosis than those with other genotypes. Logistic regression analysis identified genotype 3a (OR, 6.96; 95% CI, 2.17-22.32, P = 0.0011) as the only significant predictive variable associated with low serum cholesterol concentration. HCV infection is associated with clinically significant lower cholesterol levels (TC, LDL and HDL) when compared with those of normal subjects. This finding is more pronounced in patients infected with HCV genotype 3a. Further studies are necessary to define the pathophysiology of the relationship between lipid metabolism and HCV infection.
DISCUSSION
In this study, patients with chronic HCV infection were found to have significantly lower levels of serum TC, HDL-C and LDL-C than normal adults matched for age and sex. The elimination of the genotype 3a patients from the whole group of chronic HCV cases did not significantly alter the observation that HCV of all subtypes groups lowers cholesterol levels. Relation of HCV infection to low LDL levels may be clinically relevant as it may reduce the risk of diseases associated with hyperlipidaemia, mainly atherosclerotic heart disease [13]. The mechanism by which HCV infection may lower serum cholesterol is not known. Liver disease resulting from HCV infection might impair VLDL synthesis in hepatocytes. In our population, hypocholesterolaemia was not the result of poor liver function, as cirrhotic patients were excluded. Moreover, no correlation was observed between serum cholesterol (TC, HDL-C and LDL-C) and staging score in our study. It is well established that chronic inflammation causes hypocholesterolaemia through a reduction of LDL, and, to some extent, HDL [14]. Cytokines, endogenously produced, have been related to change in blood lipid composition and to the outcome of infections [15]. The relationship between lipids and cytokines appears to change according to the kind of viral infection involved (HCV or HBV) [4]. However, a positive correlation (and not a negative as it was expected) was found between serum cholesterol (TC and LDL-C) and grading score. This finding is surprising and difficult to explain, as a higher grading score means more severe inflammation, elevated serum levels of cytokines and hypocholesterolaemia.
Binding of the HCV to lipoproteins LDL, VLDL, and HDL in plasma has been reported in many studies [5,6,16]. Thomssen et al. [17] described that heterogeneities in the density of hepatitis C virus RNA-carrying material found in human sera are attributed to the binding of low-density lipoproteins and/or of IgG. It has been postulated that the LDL may protect the virus from inactivation by antibodies [18]. Two additional studies give a clue to this hypothesis. First, the destruction of ƒÀ-lipoproteins enzymatically by incubating HCV-positive human sera with lipoprotein lipase causes strong reduction of HCV-RNA titers in the sera [19]. However, the authors of this study have not excluded as yet that the used preparation of lipase may have additionally enzyme activities like a phospholipase acting directly to a virus membrane, whether this have bound beta-lipoproteins or not. Secondly, LDL plasmapheresis of HCV-infected patients dramatically reduces HCV viral load [20]. On the contrary, other studies point to the opposite hypothesis, that high levels of LDL may protect the patients' cells from HCV. It has been reported that the LDL-receptor is one of the HCV receptors, and experiments in vitro showed competitive inhibition of binding between HCV and LDL-receptor by LDL [21]. Enjoji et al. [22] showed that HCV-Ag levels were negatively correlated with increased beta-lipoproteins in patients with chronic hepatitis C. Nevertheless, in our study there was not found any corellation between viral load and LDL-C. These findings, although contradictory, suggest that hypocholesterolemia is specifically related to HCV infection, and it is not a non-specific consequence of liver disease.
In our study, we detected predominantly genotypes 1b and 3a, which are the most prevalent in Greek patients [23]. Genotype 3a patients were found to have lower cholesterol levels (TC, HDL-C and LDL-C) than those with other genotypes. As shown by controlling for age and sex, and by logistic regression analysis, this difference was not connected to the younger age or to the male gender predominance of these patients, but only to the genotype 3a. Kono et al. [24] showed that there is an apparent difference in the lipoprotein-binding rate between different genotypes. He found that genotype 2a/2b HCV had a higher LDL binding rate and also a striking higher HDL binding rate than of genotype 1b HCV. Moreover, Moriya et al. [25] showed that genotype 1b HCV patients had lower serum cholesterol levels than those infected with genotype 2a. It is possible that this difference in serum cholesterol levels between genotypes may be connected to the different lipoprotein-binding rate between them. It is well known that, in Japan, the genotype 3 infection is very rare and the prevalent genotypes are 1b and 2a [26], in contrast to Greece where genotype 3a is prevalent.
In addition, genotype 3a patients presented higher frequency of hepatic steatosis than patients with other genotypes. Our results are in accordance with the report of Serfaty et al. [27] who suggested that in chronic hepatitis C, hypobetalipoproteinaemia is prevalent and associated with steatosis, especially in patients with genotype 3a. Genotype 3a could have stronger negative effect on beta-lipoprotein formation in hepatocyte through specific HCV core protein, which is able to inhibit VLDL liver secretion and to induce steatosis in transgenic mice [28].
Hepatitis C viral load was also found to be significantly higher in genotype 3a patients than those with other genotypes. This finding shows that decreased cholesterol level is associated with higher viral load and supports the hypothesis of Monazahian et al. [21] that beta-lipoproteins competitively inhibit the HCV infection. However, there was not noticed a negative correlation between serum HCV RNA levels and LDL-C in HCV-infected patients, as it was expected. This discrepancy could be attributed to the fragility of HCV-RNA and to that its concentration is easily influenced by the method of preserving the sera.
In conclusion, HCV infection is associated with clinically significant lower cholesterol levels (TC, LDL and HDL) than those of normal subjects. This finding is more pronounced in HCV patients infected by genotype 3a. Further studies are necessary to define the pathophysiology of the relationship between lipid metabolism and HCV infection.
Introduction
It is well known that liver plays a key role in serum lipoprotein synthesis and metabolism, and an impaired lipid metabolism is often found in patients with chronic liver diseases [1]. Decreasing levels of serum cholesterol, mainly low-density lipoprotein-cholesterol (LDL-C) has been related to increasing severity of liver disease [2]. Lower total cholesterol (TC) and LDL-C levels were described in hepatitis C virus (HCV)-infected patients soon after viral diagnosis. These lower levels are associated with HCV infection regardless of the degree of hepatic fibrosis [3]. Moreover, patients with chronic hepatitis C were found to have lower TC levels when compared with those with chronic hepatitis B [4]. In addition, an association between HCV infection and lipid metabolism has been reported. Particles of HCV exist in binding form with beta-lipoproteins [LDL and very low-density lipoproteins (VLDL)] and immunoglobulins in the sera of patients with chronic hepatitis C [5,6]. Complexing of the virus to VLDL or LDL could promote endocytosis of HCV via the LDL receptor [7]. These findings suggest that lipoproteins play an important role in the process of HCV infection.
The aim of this study was to determine the serum lipid profile in patients with chronic HCV infection, and to detect if there is any correlation between serum lipid levels and viral load, HCV genotype or liver histology.
Results
Patient lipid profile and demographic data
The main clinical and laboratory data are summarized in Table 1. Patients were absolutely comparable with controls according to age, sex and BMI. Serum TC (P < 0.0005), HDL-C (P < 0.0005) and LDL-C (P < 0.0005) were significantly lower in chronic hepatitis C patients compared with controls, whereas triglyceride levels were similar in the two groups.
Virological features of patients studied
Quantification of serum HCV-RNA was performed in 144 of 155 patients with chronic HCV infection (median, 340 000 IU/mL; range, 22 998 000 IU/mL). Using a Spearman's rank correlation analysis, no correlation was found between viral load and TC, TG, HDL-C or LDL-C.
The distribution of HCV genotypes was as follows: 1a [n = 15 (9.7%)], 1b [n = 43 (27.7%)], 2a [n = 20 (12.9%)], 3a [n = 65 (41.9%)], 4 [n = 6 (3.9%)], and unclassified [n = 6 (3.9%)]. The most prevalent genotypes were 3a and 1b.
Liver histology
The mean grading score of the 135 liver biopsy specimens from HCV-infected patients was 5.68 ± 2.26 and the mean staging score was 2.59 ± 1.30. A Pearson's rank correlation analysis of the lipid (TC, HDL-C, LDL-C and TG) and histological variables (grading and staging score), demonstrated that grading score was positively correlated with TC (r = 0.225, P = 0.009) and LDL-C (r = 0.234, P = 0.007). No correlation was found between staging score and TC, HDL-C, LDL-C or TG. Macrovesicular hepatic steatosis was found in 59 of 135 (43.7%) HCV patients: mild in 52 (38.8%), moderate in six (4.5%), and severe in one (0.7%). Mean serum levels of LDL-C were significantly lower in patients with hepatic steatosis (95.6 ± 44.2 mg/dL) compared with patients without steatosis (108.7 ± 27.7 mg/dL, P < 0.05).
Differences of blood lipid levels according to HCV genotype
The characteristics of the patients infected with genotype 3a, which is commonly encountered in young males, especially intravenous drug users, were also analysed [12]. When genotype 3a patients (n = 65) were compared with those with other genotypes (n = 89), we found, as it was expected, that genotype 3a patients were younger, more often males, intravenous drug users and had shorter duration of infection. Moreover, genotype 3a patients had lower levels of TC (152.3 ± 38.5 mg/dL vs 183.6 ± 41.7 mg/dL, P < 0.0005), HDL-C (38.9 ± 10.2 mg/dL vs 44.3 ± 12.6 mg/dL, P =0.006) and LDL-C (88.6 ± 32.3 mg/dL vs 115.6 ± 36 mg/dL, P < 0.0005) than those with other genotypes. No difference was found in the mean levels of TG between the two groups. Genotype 3a patients also presented higher frequency of hepatic steatosis than those with other genotypes (61.7%vs 29.3%, P < 0.0005) (Table 2).
As it is well known that age and gender may influence serum lipid levels [13]. We also compared 58 patients with genotype 3a with 53 patients with other genotypes, matched for age and sex (by excluding the HCV patients aged >53 years). The difference in TC (152.3 ± 39 mg/dL vs 177 ± 40.5 mg/dL, P = 0.001), HDL-C (38.6 ± 10.4 mg/dL vs 43.5 ± 11.2 mg/dL, P = 0.02) and LDL-C (88.2 ± 31.9 mg/dL vs 111.2 ± 37.9 mg/dL, P = 0.001) levels between the two groups was still significant, even after controlling for age and sex (Table 3). Serum HCV RNA levels were significantly higher in genotype 3a patients (median: 353 000 IU/mL; range: 10 616 000 IU/mL) than in the group with other genotypes matched for age and sex (median: 145 000 IU/mL; range: 22 998 000 IU/mL, P = 0.01).
Moreover, logistic regression analysis was performed to assess the independent variables (age, sex, BMI, duration of HCV infection, grading score, HCV-RNA levels and genotype) associated with low levels of serum cholesterol ≦127.3 (mean minus 1 SD of the patients with chronic HCV infection) in HCV group. This analysis identified genotype 3a (OR, 6.96; 95% CI, 2.17-22.32, P = 0.0011) as the only significant predictive variable associated with low serum cholesterol concentration.
To clarify that besides subtype 3a viruses, other subtypes may also be involved in the lipid-declining processes in HCV infections of the liver, we compared the serum TC, HDL-C and LDL-C of non-genotype 3a patients with those of normal subjects, and again serum TC (P < 0.0005), HDL-C (P < 0.0005) and LDL-C (P < 0.0005) were significantly lower in the non-genotype 3a group.
Patients
We studied 155 patients (94 males, 61 females, mean age 42 ± 14.6 years) with chronic HCV infection, who were being regularly followed up at the Hepatology Clinic of Patras University Hospital. Chronic HCV infection was defined by the presence of HCV-RNA in serum for >6 months. Thirty-four (21.9%) patients had been treated with alpha-interferon (IFN-a) in the past (4-11 years before) without achieving virological and biochemical response. Patients with other causes of liver dysfunction (chronic hepatitis B, primary biliary cirrhosis, autoimmune hepatitis, continued alcohol abuse), with known autoimmune diseases (systemic lupus erythematosus, rheumatoid arthritis, Sjogren's syndrome), with human immunodeficiency virus antibodies and under current treatment with IFN and ribavirin were excluded. In addition, patients with clinical signs of cirrhosis and/or definite cirrhosis (stage 6 on liver biopsy) were excluded. The probable causes of HCV infection were as follows: transfusion [n = 35 (22.6%)], intravenous drug addiction [n = 73 (47.1%)], sexual or household contacts [n = 13 (8.4%)], surgery [n = 13 (8.4%)] and unknown [n = 21 (13.5%)]. The duration of HCV infection was presumed by the date of transfusion or surgery or initial exposure to intravenous drugs (mean duration 14.3 ± 9 years).
The control group was made up of 138 normal subjects, matched for age and sex (84 males, 54 females, mean age 42.8 ± 10.6 years) from the volunteer blood donor programme of our hospital. Each individual had indicated the absence of significant illness. Physical examination, normal liver function test results and absence of hepatitis B surface antigen (HbsAg) and anti-HCV antibodies in their serum, excluded liver diseases in these controls. None of the enrolled subjects in these two groups had a history of diabetes mellitus or exposure to drugs influencing lipid metabolism (lipid-lowering agents, corticosteroids, non-steroid anti-inflammatory drugs).
Clinical and laboratory assessments
Serum lipid profile was determined in all participants after an overnight fast of 12 h. TC, TG, and HDL-C were measured enzymatically with commercial kits (Olympus System Reagent, Hamburg, Germany) by the use of an automated analyzer. LDL-C was calculated according to the formula of Friedewald et al. [8]. The body mass index (BMI) was calculated in accordance with the formula of weight (kg) divided by height2 (m2) [9].
Viral markers
Hepatitis C virus RNA was detected in patients' sera by reverse transcriptase-polymerase chain reaction (RT-PCR) (Hepatitis C virus test - version 3.0; Cobas Amplicor, Roche Diagnostics, Branchburg, NJ, USA). Quantification of serum HCV-RNA was performed by the branched DNA method [HCV RNA 3.0 Assay (b-dna); Versant-Bayer, New York, USA]. The detection limit of this assay is 615 IU/mL. HCV genotyping was performed by a second-generation, line-probe assay (INNO-LiPA HCV II; Innogenetics, Zwijndrecht, Belgium).
Histologic evaluation
Percutaneous liver biopsy was performed in 135 of 155 (87.1%) HCV-infected patients by Menghini needles. Each biopsy specimen was evaluated according to the grading and staging system described by Ishak et al. [10]. Necroinflammatory activity is graded on a scale of 0-18 (modified HAI grading) and the staging for liver fibrosis and architectural disturbances is performed using a scale of 0-6 (modified staging). All biopsy specimens were evaluated for hepatic steatosis. Steatosis was considered as absent when involving <1% of hepatocytes, mild when involving 1-33% of hepatocytes, moderate when involving 33-66% of hepatocytes and severe when involving >66% of hepatocytes [11].
Statistical analysis
Values were expressed as prevalence rates or as the mean ± 1 standard deviation (SD). Conventional chi-square and Fisher's exact test were used to analyse qualitative differences. The differences between parametric data were evaluated with Student's t-test. For non-parametric test values, a Mann-Whitney test was used. Relations between variables were explored by determining Pearson's or Spearman's rank correlation coefficients. Stepwise logistic regression analysis was performed in order to identify the predictor variables independently associated with a low level of serum cholesterol. Probability values of <0.05 were considered significant. Statistical analysis was performed with SPSS 8.0 statistical software. The study was approved by the Ethical Committee of Patras University Hospital, and informed consent to participate in the study was obtained from all patients.
|
|
| |
| |
|
|
|
