HCV infection in haemodialysed patients: A role for serum IL-10 and TGF-β1 in liver damage?

Results IL-10 serum levels were significantly higher in HD patients, both HCV+ (3.7 ± 0.4 pg/ml; p < 0.01) and HCV− (3.8 ± 0.8 pg/ml; p < 0.05) than in non-uremic HCV patients (2.3 ± 0.4 pg/ml). Among the HD/HCV+ patients, IL-10 serum levels were similar in HD/HCV-N and in HD/HCV-H patients. Among the HD/HCV+ patients, IL-10 serum levels were similar in those with moderate histological damage compared to those with mild damage. TGF-β 1 serum levels were significantly lower in HD patients, both HCV+ (4.6 ± 0.9 ng/ml) and HCV− (6.0 ± 0.9 ng/ml) than in non-uremic HCV+ patients (8.1 ± 1.1 ng/ml; p < 0.001 and p < 0.01, respectively), but similar to the values found in H (5.3 ± 0.9 ng/ml; p = n.s.). No correlation was seen between IL-10 and TGF-β 1 serum levels in any of the groups considered. Conclusions Patients on haemodialysis treatment to have high levels of IL-10, which remain high even when patients are anti-HCV+, whereas the opposite is true of TGF-β 1 . The cytokine pattern observed in HD patients is compatible with the hypothesis explaining the relatively benign evolution of HCV-related liver disease in HD patients, and has a pathophysiological role. Keywords HCV Haemodialysis IL-10 TGF-β 1 1 Introduction HCV infection is widespread, and 3% of the world population is reported to be persistently infected [1] . A European multicenter study [2] reported the trend of anti-HCV+ prevalence in HD patients in the last decade (1991–2000), which was from 28 to 16% in Italy. HCV infection correlates with the time on haemodialysis and the transfusion of blood products [3–5] , but the correlation between HCV infection and blood products has not been valid since 1992 [6] . In HD patients, HCV infection has certain differences compared to non-HD patients. It is frequently associated with normal or only minimally increased transaminase levels [7,8] and HCV-RNA may not be detectable in the blood of up to 80% of patients with histological evidence of chronic hepatitis [3,9] . A mild degree of histological liver damage has been reported in HD patients with no biochemical signs of hepatitis [10] , though the long-term evolution to cirrhosis seems to be similar to those with abnormal liver function tests [8] . On the other hand, anti-HCV positivity is an independent risk factor for mortality in patients on haemodialysis [11,12] . The reasons of this particular pattern of HCV infection are not known. It may be due to the different immune responses described in patients with end-stage renal disease, or it may be due to the haemodialysis treatment [13,14] . Uremic patients have an impaired cell-mediated immunity and phagocytic activity [15,16] , which accounts for their susceptibility to infection, malignancies and vaccination resistance [17,18] . This condition may disrupt the cytokine network by depressing Th1 response, enabling Th2-type mediators to prevail [19,13] . Among the Th2 mediators, IL-10 regulates inflammatory response and has a modulatory effect on hepatic fibrogenesis [20,21] , while TGF-β has numerous pro-fibrogenic, but also anti-inflammatory and immunosuppressive effects [22] . These actions must be balanced to maintain tissue homeostasis. In the liver, TGF-β suppresses hepatocyte proliferation and stimulates the production of extracellular matrix [23] . Tissue levels of TGF-β rise during fibrogenesis [23] . In experimental models using transgenic mice, overexpression of TGF-β gave rise to fibrosis [23] . The purpose of our study was to assess IL-10 and TGF-β serum levels in anti-HCV positive patients on haemodialysis treatment, and to correlate cytokine concentrations with the severity of liver disease. Cuprophane membranes [24,18] reportedly stimulate monocytes to produce high levels of pro-inflammatory cytokines, such as IL-1 and IL-6. The secretion of IL-6 is followed, within a few hours, by the induction of IL-10 [14] . IL-10 production by peripheral blood mononuclear cells may be the expression of persistent viral infection in HCV patients [25] . The polymorphisms in the promoter region of the IL-10 gene govern different immune functions of IL-10, giving rise to a variety of actions [26] . Human IL-10 has both anti-inflammatory and immunosuppressive properties [20] , and it down-regulates the production of pro-inflammatory cytokines by T cells. Endogenous IL-10 reduces the intrahepatic inflammatory response and curbs hepatotoxicity in several models of liver injury [14] . It is also expressed by hepatic stellate cells and may have a role in liver matrix remodelling by reducing collagen and enhancing collagenase production [27] . Nelson et al. reported that, though it had no antiviral activity, IL-10 could play a part in normalizing ALT serum levels, improving liver histology, and reducing liver fibrosis in a large proportion of patients treated with recombinant IL-10, supporting the hypothesis that IL-10 has an anti-fibrotic effect [28,29] . 2 Materials and methods 2.1 Patients The patients’ characteristics, liver function test results and virological data, at the baseline Gastroenterology outpatient clinic evaluation are given in Table 1 . Seventy-one consecutive HD anti-HCV positive patients (HD/HCV+) (47 males and 24 females, mean age 44.6, range 28–69), were seen between 1996 and 2000. Anti-HCV antibodies had been detected 3.91 ± 0.94 years before the patients’ enrollment in the study. Among the 71 HD/HCV+ patients were enrolled irrespective of their HCV-RNA status. The main risk factor for HCV infection was blood product transfusion, none of the patients had experienced episodes of acute hepatitis, and they had been on HD for a mean of 5.3 ± 0.68 years (5.59 ± 0.9 years for HD/HCV-N and 2.2 ± 0.8 years for HD/HCV-H p = n.s.) similar to the HCV-negative HD patients (3.2 ± 1.2 years p = n.s.). Antiviral therapy (IFN-α 3MUx3/week for 6 months) was given to 4/71 HD/HCV+ patients. One of the four was HCV-RNA negative 6 months after the end of the treatment and was grouped as HD/HCV+N. Blood sampling was performed at least 3 months after the end of antiviral treatment. 58/71 of HD/HCV+ patients were found to have ALT within the normal range [HD/HCV+N] and 13 ALT above the normal range [HD/HCV+H], ALT values are lower in HD patients than in healthy individuals [30,31] . The distribution of ALT values recorded in the study population was not normal. The HD/HCV+ patients were divided according to normal or high ALT serum levels based on the new range for HD patients recommended by Espinosa et al. [31] : the upper limit was considered as the 95th percentile of ALT in hepatitis-free HD patients. The 95th percentile of the ALT levels in our HD patients was 29 UI/L. During the same period, 56 HD anti-HCV-negative patients (HD) (34 males and 22 females, mean age 69 years, range 27–90) with normal liver function were randomly recruited from the Nephrology Unit's Haemodialysis Center, and 40 chronic hepatitis C (HCV) patients (27 males and 13 females, mean age 52.7 years, range 28–69) were randomly recruited from the Gastroenterology outpatient clinic. In patients with indication for antiviral therapy the sera were collected before the start of antiviral therapy. Twenty healthy volunteers (9 males and 11 females, mean age 39 years, range 24–61) were used as controls. All HD patients had a 4-h haemodialysis session three times a week using hemophan membranes. None of the dialysers were reused. Microbiological fluid quality was routinely tested. None of the patients were on immunosuppressive drugs, nor were any HIV-positive or hepatitis B surface antigen-positive. Haemoglobin (g/dl), AST (IU/l), ALT (IU/l), GGT (IU/l), ALP (IU/l), total bilirubin (mg/dl), prothrombin time (PT; %), platelet count (PLT; el/μl), serum iron (μg/dl), ferritin (μg/l) and alpha-fetoprotein (ng/ml) were measured in all HD patients. Anti-HCV antibody (EIA II-III or RIBA III) titer was assessed in all HCV positive patients, with qualitative and quantitative HCV-RNA (AMPLICOR). 2.1.1 Liver histology The indication for liver biopsy in HD/HCV+ patients was HCV-RNA positivity, irrespective of ALT levels, while in HCV patients it was more than twice the normal ALT in two consecutive tests with HCV-RNA positivity. Liver biopsies were taken with a 16–18 gauge needle and the tissue sample was fixed in formalin and embedded in paraffin, then stained with haematoxylin & eosin. Histological grading and staging were done according to Ishak's classification [32] . The time interval from blood sampling and liver biopsy ranged between 1 and 3 days. 2.1.2 Cytokine measurement Blood samples were drawn from all patients after overnight fasting and before haemodialysis in the HD and HD/HCV groups. The samples were centrifuged for 15 min at 1000 RPM and the sera were stored at −20 °C. IL-10 serum levels were assessed by ELISA (OptEia, Pharmingen). A 96-well plate (Nunc Maxisorb) was incubated overnight at 4 °C with 100 μl/well of primary antibody diluted (1:250) in carbonate buffer. Between incubations, the plate was washed three times in phosphate buffered saline with 0.005% Tween 20 (PBS/Tween) and blocked with 10% fetal bovine serum in PBS (FBS/PBS). Standard curve dilutions and 100 μl of the sample were incubated in duplicate for 3 h at room temperature (RT). The plate was then incubated for 1 h at RT with 100 μl/well of biotinylated secondary antibody (diluted 1:500 in FBS/PBS), followed by 1 h of incubation at RT with streptavidin–avidin–peroxidase (1:250 in FBS/PBS). The reaction with tetramethylbenzidine and hydrogen peroxide substrate solution (30 min incubation at RT) was stopped by adding 50 μl/well of 2N sulphuric acid. Plates were read by the spectrophotometer at 495 nm, and the correlation coefficient was linear in a concentration range between 3.9 and 250 pg/ml for IL-10 ( r = 0.989). The intra-test variability for IL-10 dosage by evaluating 11 non-selected samples and it was found to be 10% (E.S. ± 0.003); the inter-test variability, measured in 27 non-selected samples, was 10% (E.S. ± 0.5). TGF-β serum levels were assessed by ELISA (OptEia, Pharmingen) according to the manufacturer's instructions. The intra-test variability was performed among eight non-selected patients and it was found to be 6.1% (E.S. ± 6). 2.2 Statistical analysis All data are expressed as means ± S.E. Comparisons between groups were performed using non-parametric tests (Mann–Whitney). Correlations between variables were estimated using Pearson's correlation coefficients. A p value <0.05 was considered significant. 3 Results 3.1 Liver function tests ALT was within the normal range in three consecutive assays performed within 6 months of one another (in accordance with the Italian Association for the Study of the Liver guidelines [33] ) in 58/71 HD/HCV+ patients (81%). HCV-RNA was positive in 37/58 patients (64%) with normal ALT (mean viremia 9.8 ± 0.8 × 10 6 copies/ml), and in 13/13 patients (100%) with high ALT (mean viremia 1.5 ± 0.9 × 10 6 copies/ml, not significantly different between the two groups, p = 0.67). Genotype 1b was found in 63% of the HD/HCV RNA (+) patients. No clinical, endoscopic or ultrasonographic signs of portal hypertension were seen in HD/HCV+ patients; mean prothrombin time was 95.9 ± 1.9%, and mean platelet count was 196 × 10 3 ± 0.9 el/μl (there was no difference between HCV-positive or HCV-negative HD patients). In the 40 HCV+ patients, mean ALT was 125 ± 15.4 UI/L and HCV-RNA was positive in 40/40. Antiviral therapy was clinically indicated in 25/40. 3.2 Liver histology The clinical indication for liver biopsy in HD/HCV+ patients was HCV-RNA positivity—seen in 50 cases (37 HD/HCV+N and 13 HD/HCV+H), but antiviral therapy was contraindicated in 7 patients, 10 refused liver biopsy, 4 had clotting which was too deranged and 12 patients were unavailable for logistic reasons, so 17 HD/HCV+ HCV-RNA positive patients (6/17 HD/HCV+H and 11/17 HD/HCV+N) underwent liver biopsy. When patients were grouped as having a liver biopsy or not, no differences were observed in their age ( p = 0.63), time on haemodialysis ( p = 0.75), time of HCV infection ( p = 0.61), prothrombin time ( p = 0.12). Ishak's score was grade <5 and stages 1–2 in 12/17 patients, classified as mild chronic hepatitis, and grades 6–9 and a stage <2 in 5/17 patients classified as moderate chronic hepatitis. Clinical indication for liver biopsy was present in 25/40 HCV patients, 10 refused biopsy or could not have a biopsy, 15 had a biopsy. The Ishak score was grade <5 and stages 1–2 in 7/15 patients (mild chronic hepatitis), grades 6–9 and a stage <2 in 4/15 patients (moderate chronic hepatitis), and a grade >9 and a stage >2 in 4/15 patients (severe chronic hepatitis). When patients were grouped as having a liver biopsy or not, no differences were observed in their age or history of HCV positivity. Neither grade nor stage correlated with the IL-10 or TGF-β 1 serum concentrations, nor were any correlations found between quantitative HCV-RNA, and ALT, IL-10 or TGF-β 1 serum concentrations. 3.3 IL-10 serum concentration IL-10 serum concentrations were significantly higher in HD patients, both HCV+, irrespective to ALT levels (3.7 ± 0.4 pg/ml; p < 0.01) and HCV− (3.8 ± 0.8 pg/ml; p < 0.05) than in non-uremic HCV patients (2.3 ± 0.4 pg/ml) and controls (1.2 ± 0.1 pg/ml). Among the HD/HCV+ patients, IL-10 serum concentrations were similar in HD/HCV-N (3.9 ± 0.5 pg/ml) and in HD/HCV-H (1.9 ± 0.5 pg/ml; p = n.s.) although increased IL-10 values appeared to be associated with a lower transaminase values ( Fig. 1 ). HD/HCV-N patients were split into two groups, i.e. those with detectable HCV-RNA and those without HCV-RNA, but no significant difference in serum levels of IL-10 was seen (3.3 ± 0.6 pg/ml and 4.2 ± 1 pg/ml, respectively; p = n.s.). Among HD/HCV patients who underwent liver biopsy, IL-10 was similar in HD/HCV+ patients with mild or moderate histological damage (3.1 ± 0.7 pg/ml vs. 1.1 ± 0.4 pg/ml; p = n.s.) ( Fig. 3 ). 3.4 TGF-β 1 serum concentrations TGF-β 1 serum concentrations were significantly lower in HD patients, both HCV+ (4.6 ± 0.9 ng/ml) and HCV− (6.0 ± 0.9 ng/ml), than in non-uremic HCV+ patients (8.1 ± 1.1 ng/ml; p < 0.001 and p < 0.01, respectively), but similar to the values found in H (5.3 ± 0.9 ng/ml; p = n.s.). TGF-β 1 serum levels were higher in HCV+ patients than in controls ( p = 0.018) ( Fig. 2 ). Among HD/HCV patients who underwent liver biopsy TGF-β 1 did not differ in patients with mild versus moderate liver damage (4.5 ± 0.6 ng/ml vs. 5.7 ± 1.2 ng/ml; p = n.s.) ( Fig. 4 ). No correlation was seen between IL-10 and TGF-β 1 serum concentrations in any of the groups considered and neither with age. No correlations were found between quantitative HCV-RNA, ALT, HCV genotype, IL-10 and TGF-β 1 serum concentrations among HD/HCV+ patients. 4 Discussion HCV infection is relatively common in HD patients, the current risk factors being time on HD and type of renal replacement [3–5] , and before 1992 blood transfusions. Chronic HCV infection has a relatively benign course in this setting, with only mild biochemical and histological signs of liver damage [9,34] , possibly due to these patients’ higher serum levels of anti-fibrosis [13] and liver-protecting mediators [35] . However, in some cases, histological damage can occur even despite HCV-RNA negativity [4] . Many cytokines secreted by Th1 and Th2 cells are involved in the immune response to HCV infection and progression of HCV-related liver disease. Th1 cells release TNF-α, INF-γ and IL-2, causing inflammation and necrosis [36] , and Th2 release IL-4 and IL-10, which modulate hepatic injury by suppressing the Th1 response and counteracting the fibrogenic effects of TNF-α, INF-γ and IL-2 [37] . Long-term haemodialysis results in changes in cellular immunity [17] with a decreased production of cytokines such as TNF-α and INF-γ, suggesting a disturbance of the Th1, but not of the Th2 response [38] . Rampino et al. [34] reported that haemodialysis treatment stimulates the release of HGF, which was shown to protect against HCV-related liver damage. This study investigated whether IL-10 and TGF-β 1 influence the impact of HCV infection in HD patients. HD patients, with or without HCV infection, have higher IL-10 levels than non-uremic patients with chronic hepatitis C, suggesting that either renal insufficiency or haemodialysis stimulates IL-10 release. Previous studies reported that IL-10 levels were higher in CAPD and HD patients than in controls [39] , possibly due to a counter-regulatory mechanism to control uremia and dialysis-induced activation of inflammatory events in HD patients [40] . IL-10 concentrations were also related to renal function indices and nutritional markers in patients on long-term HD [41] . Other factors such as age may influence the serum levels of cytokines. We found that HCV-negative HD patients were older than the HCV-positive HD patients. A recent study [42] demonstrated that older mice produce more IL-10 than younger mice. Age-related changes in immune function may become measurable in frail elderly people with a markedly higher risk of infection [43] , such as HD patients. Thus, IL-10 production in HCV-negative HD patients may be influenced by the patient's age. We did not find this in our study when age was not associated with IL-10 irrespective of HCV positivity or negativity. To the best of our knowledge, no data are available on the influence of age on TGF-β 1 serum levels, but we found no association with age like of IL-10. In our study HD/HCV+ patients with normal ALT levels had a longer history of dialysis compared to patients with abnormal ALT levels: the data seem to reinforce the hypothesis [34] that long-term dialysis protects against liver necrosis. TGF-β is a superfamily of proteins including three isoforms (beta1, beta2, beta3), produced mainly by T-lymphocytes, but also by neutrophils, human hepatic stellate cells and Kupffer cells [23] . It is a fibrogenic cytokine for which we found lower concentrations in HCV-positive HD patients than in chronic HCV patients, and higher concentrations in chronic HCV patients compared to controls. The severity of liver disease is not usually assessed on the basis of transaminase serum levels [3,5] because they usually tend to be low in HD patients [7] . However, Espinosa et al. described that high ALT serum levels correlate with positive HCV viremia [31] . This was confirmed in our series, since in all HD patients with hypertransaminasemia HCV-RNA was positive, but when the HD/HCV-N patients were splitted into two groups, according to HCV-RNA status, we found no significant difference in serum IL-10 concentration between HD/HCV+ patients with or without HCV-RNA positive. Since only HCV-RNA positive patients underwent liver biopsy, we cannot postulate any difference in term of histological liver damage between HCV-RNA positive and negative patients. This means that, with this study, we are unable to confirm whether liver damage can be seen in HD/HCV-positive patients with undetectable HCV-RNA. To test whether hypertransaminasemia was associated with IL-10 serum levels or not, we divided the HD/HCV+ patients according to ALT levels. HCV patients on HD had high ALT serum concentrations associated with lower levels of IL-10 compared to HD patients with normal transaminases, although this difference does not reach statistical significance, probably due to the small number of patients in each group. Liver biopsy is the gold standard method for assessing the severity of liver disease, but it could only performed in a subgroup of our HD/HCV+ patients (since many of them refused or had logistic problems). However, the limited number of liver biopsies could explain why we found no correlation between the histological damage and IL-10 serum levels. There appeared to be no bias for biopsying patients when the clinical characteristics were taken into account. Moreover, assessing tissue cytokine expression would help define their role in determining liver damage and its progression. Our results would justify further studies in biopsy specimens. Overall, HD/HCV+ patients had lower concentrations of TGF-β 1 and higher concentrations of IL-10 compared to HCV+ patients and healthy controls. Nevertheless no correlations were seen between these two cytokine levels probably due to the number and complexity of factors that can influence both production and serum concentrations of this cytokines. However, the cytokine pattern that we observed in HD patients is compatible with a hypothesis explaining the relatively benign evolution of HCV-related liver disease in HD patients and may have a pathophysiological role. Practice points • HD patients have high levels of IL-10 even in HD anti-HCV positive patients. • The hemodialised patients have a relatively benign evolution of HCV-related liver disease. • IL-10 can play a role in mediating HCV-related liver damage in this setting. Research agenda • Liver damage and hypertransaminasemia among HD/HCV+ reach significance toward IL-10 levels: a higher number of cases should be enrolled in a successive study to confirm this trend. • It would be interesting to analyse the effect of antiviral therapy on cytokine production among HCV patients. Conflict of interest statement None declared. List of abbreviations H, healthy volunteers; HCV, hepatitis C virus; HCV patients: non-uremic anti-HCV positive patients; HD patients, haemodialysed patients; HD/HCV− patients, haemodialysed anti-HCV negative patients; HD/HCV+ patients, haemodialysed anti-HCV positive patients; HD/HCV-H patients, haemodialysed anti-HCV positive patients with high serum transaminases; HD/HCV-N patients, haemodialysed anti-HCV positive patients with normal serum transaminases; IL-10, interleukin 10; TGF-β 1 , transforming growth factor beta 1; Th, T helper lymphocytes. References [1] M.J. Alter Epidemiology of hepatitis C Hepatology 26 1997 S62 S65 [2] M. Jadoul J.L. Poignet C. Geddes F. Locatelli C. Medin M. Krajewska The changing epidemiology of hepatitis C (HCV) infection in haemodialysis: European Multicentre Study Nephrol Dial Transpl 19 2004 904 909 [3] S. Pol A. Vallet-Pichard H. Fontaine P. Lebray HCV infection and hemodialysis Sem Nephrol 22 2002 331 339 [4] S.N. Natov J.Y. Lau B.A. Bouthot B.V. Murthy R. Ruthazer C.H. Schmid Serologic and virologic profiles of hepatitis C infection in renal transplant candidates. New England Organ Bank Hepatitis C Study Group Am J Kidney Dis 31 1998 920 927 [5] D. Roth Hepatitis C virus: the nephrologist's view Am J Kidney Dis 25 1995 3 16 [6] S. Bergman N. Accortt A. Turner J. Glaze Hepatitis C infection is acquired pre-ESRD Am J Kidney Dis 45 2005 684 689 [7] S.K. Herrine B. Michael W.L. Ma S. Rossi S.R. Dunn T. Hyslop Development of an HCV infection risk stratification algorithm for patients on chronic hemodialysis Am J Gastroenterol 97 2002 2619 2622 [8] M. Espinosa A. Martin-Malo M.A. Alvarez de Lara R. Gonzales M. Rodriguez P. Aljama Natural history of acute HCV infection in haemodialysed patients Clin Nephrol 58 2002 143 214 [9] R.K. Sterling Chronic hepatitis C infection in patients with end stage renal disease: characterization of liver histology and viral load in patients awaiting renal transplantation Am J Gastroenterol 94 1999 3576 3582 [10] C. Caramelo A. Ortiz B. Aguilera J.C. Porres S. Navas E. Marriott Liver disease pattern in haemodialysis patients with antibodies to hepatitis C virus Am J Kidney Dis 22 1993 822 828 [11] C.O. Stehman-Breen S. Emerson D. Gretch R.J. Johnson Risk of death among chronic dialysis patients infected with hepatitis C virus Am J Kidney Dis 32 1998 629 634 [12] M. Espinosa A. Martin-Malo M.A. Alvarez de Lara Risk of death and liver cirrhosis in anti-HCV positive long-term haemodialysis patients Nephrol Dial Transpl 16 2001 1669 1674 [13] C. Libetta T. Rampino A. Dal Canton Polarization of T helper lymphocytes toward the Th2 phenotype in uremic patients Am J Kidney Dis 38 2001 286 295 [14] Y. Rousseau N. Haeffner-Caravaillon J.L. Poignet A. Meyrier M.P. Carreno In vivo intracellular cytokine production by leucocytes during haemodialysis Cytokines 12 2000 506 517 [15] G. Cohen M. Haag-Weber W.H. Horl Immune dysfunction in uremia Kidney Int 62 1997 S79 S82 [16] M.C. Amoureux N. Rajapakse E. Hegyi D. Le P. Grandics S. Szathmary Endotoxin removal from whole blood by a novel adsorption resin: efficiency and hemocompatibility Int J Artif Organs 27 2004 480 487 [17] B. Descamps-Latscha A. Herbelin Long-term dialysis and cellular immunity: a critical survey Kidney Int 43 1993 S135 S142 [18] M. Girndt H. Kaul C.K. Leitnaker M. Sester U. Sester H. Köhler Selective sequestration of cytokine-producing monocytes during hemodialysis treatment Am J Kidney Dis 37 2001 954 963 [19] M. Girndt M. Sester U. Sester H. Kaul H. Kohler Molecular aspects of T and B-cell function in uremia Kidney Int 59 2001 S206 S211 [20] T.R. Mosmann Properties and function of interleukin-10 Adv Immunol 56 1994 1 26 [21] D. Rennick D. Berg G. Holland Interleukin 10: an overview Prog Growth Factor Res 4 1992 207 227 [22] A.M. Gressner R. Weiskirchen K. Breitkopf S. Dooley Roles of TGF-beta in hepatic fibrosis Front Biosci 7 2002 793 807 [23] D.M. Bissel D. Roulot J. George Transforming growth factor beta and the liver Hepatology 34 2001 859 867 [24] S. Pol R. Romeo B. Zins F. Driss B. Lebkiri F. Carnot Hepatitis C virus RNA in anti-HCV positive hemodialysed patients: significance and therapeutic implications Kidney Int 44 1993 1097 1100 [25] M.T. Brady A.J. Mac Donald A.G. Rowan K.H. Mills Hepatitis C virus non-structural protein 4 suppresses Th1 responses by stimulating IL-10 production from monocytes Eur J Immunol 33 2003 3448 3457 [26] M. Girndt U. Sester M. Sester E. Deman C. Ulrich H. Kaul The interleukin-10 promoter genotype determines clinical immune function in haemodialysis patients Kidney Int 60 2001 2385 2391 [27] S.C. Wang M. Ohata L. Schrum R.A. Rippe H. Tsukamoto Expression of IL-10 by in vitro and in vivo activated hepatic stellate cells J Biol Chem 273 1998 302 308 [28] H. Tsukamoto Is interleukin10 antifibrogenic in chronic liver injury? Hepatology 28 1998 1707 1709 [29] D.R. Nelson G.Y. Lauwers J.Y. Lau G.L. Davis Interleukin 10 treatment reduces fibrosis in patients with chronic hepatitis C: a pilot trial of interferon nonresponders Gastroenterology 118 2000 655 660 [30] F. Fabrizi G. Lunghi S. Andrulli B. Pagliari S. Mangano P. Faranna Influence of hepatitis C virus (HCV) viremia upon serum aminotransferase activity in chronic dialysis patients Nephrol Dial Transpl 12 1997 1394 1398 [31] M. Espinosa A. Martin-Malo M.A. Alvarez de Lara S. Soriano P. Aljama High ALT levels predict viremia in anti-HCV-positive HD patients if a modified normal range of ALT is applied Clin Nephrol 54 2000 151 156 [32] K. Ishak A. Baptista L. Bianchi F. Callea J. De Groote F. Gudat Histological grading and staging of chronic hepatitis J Hepatol 22 1995 696 699 [33] Commissione AISF. Il portatore di HCV con transaminasi normali; 2002. [34] T. Rampino E. Arbustini M. Gregorini P. Guallini C. Libetta M. Maggio Hemodialysis prevents liver disease caused by hepatitis C virus: role of hepatocyte growth factor Kidney Int 56 1999 2286 2291 [35] M. Girndt H. Kolher E. Schiedhelm-Weick J.F. Schlaak K.H. Meyer zum Büschenfeldeh B. Fleischer Production of interleukin-6, tumor necrosis factorα and interleukin-10 in vitro correlates with the clinical immune defect in chronic hemodialysis patients Kidney Int 47 1995 559 565 [36] M.J. Koziel Cytokines in viral hepatitis Sem Liv Dis 19 1999 157 169 [37] P.M. Brown M.G. Neuman Immunopathogenesis of hepatitis C viral infection: Th1/Th2 responses and the role of cytokines Clin Biochem 34 2001 167 171 [38] J. Martín P. de Sequera J.A. Quiroga M. Rico M. Fernández C. Arocena Role of haemodialysis and hepatitis C virus infection in spontaneous and induced cytokine production of patients with chronic renal disease Cytokine 12 2000 1248 1252 [39] A. Borazan H. Ustün Y. Ustundag S. Aydemir T. Bayraktaroglu M. Sert The effects of peritoneal dialysis and hemodialysis on serum tumor necrosis factor-alpha, interleukin-6, interleukin-10 and C-reactive-protein levels Mediat Inflamm 13 2004 201 204 [40] M. Girndt C. Ulrich H. Kaul U. Sester M. Sester Uremia-associated immune defect: the IL-10-CRP axis Kidney Int 63 2003 S76 S79 [41] V.S. Balakrishnan D. Guo M. Rao B.L. Jaber H. Tighiouart R.L. Freeman Cytokine gene polymorphisms in hemodialysis patients: association with comorbidity, functionality, and serum albumin Kidney Int 65 2004 1449 1460 [42] E.D. Boehmer M.J. Meehan B.T. Cutro E.J. Kovacs Aging negatively skews macrophage TLR2- and TLR4-mediated pro-inflammatory responses without affecting the IL-2-stimulated pathway Mech Ageing Dev 126 2005 1305 1313 [43] K. Uyemura S.C. Castle T. Makinodan The frail elderly: role of dendritic cells in the susceptibility of infection Mech Ageing Dev 123 2002 955 962