1. Hepatology Research 36 (2006) 100–106 Routes of infection and clinical outcome of Mexican women reactive to anti-hepatitis C virus antibodies Erwin Chiquete a,b , Laura V. S´ nchez a,b , Arturo Panduro a,b,∗ a aDepartment of Molecular Biology in Medicine, Civil Hospital of Guadalajara “Fray Antonio Alcalde”, Hospital 278, Guadalajara, Postal Code 44280, Jalisco, Mexico b Department of Physiology, University Center of Health Sciences (CUCS), University of Guadalajara, Guadalajara, Jalisco, Mexico Received 27 January 2006; received in revised form 12 June 2006; accepted 15 June 2006 Available online 1 August 2006AbstractBackground: Risk factors for the hepatitis C virus (HCV) infection inﬂuence both the frequency and the progression of the liver disease.Routes of transmission and severity of the liver damage may differ by gender. We aimed to describe the risk factors for HCV infection andfor the severity of the liver disease among women seroreactive to anti-HCV antibodies. Also, we tested the hypothesis that length of infectioninﬂuences the levels of anti-HCV, in transfusion-associated hepatitis C.Methods: Eighty-six interferon-naive women, repeatedly seroreactive to anti-HCV antibodies, aged > 20 years, were studied.Results: Surgeries (80%) and transfusion before 1993 (58%) were the main risk factors (52% cases had both). The main reason for practicingsurgery was obstetric/gynecologic (74%). The main indication for transfusion was also obstetric/gynecologic (68%). Fifty-ﬁve (64%) womenwere positive to HCV RNA in serum, of them, coinfection with the hepatitis B virus (HBV) occurred in three (5%) cases, being occulthepatitis B (i.e., positive to HBV DNA, but negative to hepatitis B surface antigen) in two (4%). In multivariate analysis, determining factorsof cirrhosis at histologic examination were age and the antecedent of transfusion before 1993. Anti-HCV levels correlated with the elapsedtime from transfusion to diagnosis, but not with age.Conclusion: An obstetric/gynecologic indication was the most frequent reason for both surgery and transfusion. Hepatitis B coinfection hada low prevalence and did not inﬂuence the severity of the liver disease, as age and the antecedent of transfusion certainly did. Infection lengthinﬂuenced the levels of anti-HCV antibodies in transfusion-associated hepatitis C.© 2006 Elsevier Ireland Ltd. All rights reserved.Keywords: Cirrhosis; Hepatitis B; Hepatitis C; Liver; Serology; Test; Transfusion1. Introduction tion of HCV [1,9]. Nevertheless, little is known about the indication for blood transfusion in women who underwent In countries in which the use of illicit intravenous drugs this procedure before the screening programs  and theis not frequent, the recipients of transfusion before the sys- relation of this antecedent with the progression of the livertematic screening for the hepatitis C virus (HCV) in blood disease.banks represent the largest group of persons with the infec- The severity of the liver ﬁbrosis in HCV infection hastion [1–5]. Several risk factors for HCV infection and for been associated with some clinical antecedents, like modethe progression of the liver disease may differ by gender of transmission [6,11–13], length of infection [6,7,13], age,[6–10]. Compared with men, in some countries women have gender, alcohol consumption, body mass index [6,7], hepati-transfusion of blood products as the main route of acquisi- tis B virus (HBV) coinfection , and other factors. Also, different levels of anti-HCV antibodies are observed among ∗ Corresponding author. Tel.: +52 33 3614 7743; fax: +52 33 3614 7743. people with distinct clinical outcomes [5,15]; however, the E-mail address: email@example.com (A. Panduro). exact signiﬁcance of this difference is scarcely known.1386-6346/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.hepres.2006.06.012
2. E. Chiquete et al. / Hepatology Research 36 (2006) 100–106 101 In some populations, women appear to be not only the Diagnostics, Chicago, IL, USA) was used in all sera sam-most affected gender with HCV infection [5,10,16–18], but ples, ﬁrst stored at −70 ◦ C. Immunoassay signal strength ofalso a homogeneous group with respect to risk factors and the sample to cut-off rate (S/CO) is the numeric value of theclinical outcome [10,19,20]. We aimed to describe the risk assay, which is directly proportional to the quantity of anti-factors associated with HCV infection and events related to bodies against HCV [15,23–26]. S/CO ratio > 1 is consideredblood transfusion prior to the screening program for HCV a reactive (i.e., positive) result, according to the manufacturer.in Mexican blood banks, in women currently seroreactive toanti-HCV antibodies. Also, we sought to test the hypothesis 2.3. Detection of HCV RNA in serumthat the mode of transmission and the length of infectioninﬂuence the clinical outcome and anti-HCV antibodies A home-made, qualitative nested reverse-transcriptionlevels. polymerase chain reaction (RT-PCR) was used to detect HCV RNA in all sera samples, ﬁrst stored at −70 ◦ C. Total RNA was extracted from each serum without pooling, using2. Methods QIAamp Viral RNA Mini Kit (QIAGEN, Chatsworth, CA) as indicated the manufacturer. Afterwards, RT was carried out This study was performed from March 2002 to June 2004 to obtain complementary DNA (cDNA) using M-MLV RTat the Department of Molecular Biology in Medicine, Hospi- kit (MMLV, GIBCO/BRL). PCR ampliﬁcation of cDNA andtal Civil de Guadalajara “Fray Antonio Alcalde”. The inter- later a nested-PCR were performed with two pairs of primersnal Committee of Ethics of our hospital approved the present that hybridize in a segment of the 5 non-coding region ofstudy. Informed consent was obtained from all patients. HCV genome, as is described elsewhere . Appropriate measures were taken to minimize the risk of sample cross-2.1. Study population contamination. These measures comprised the inclusion of serum samples from normal subjects and aliquots of water as Ambulatory patients were referred after our request from negative controls. Qualitative nested RT-PCR was performedblood bank and departments of infectology, gastroenterol- per duplicate in all samples. Cases with detectable HCV RNAogy, general internal medicine and hematology to conﬁrm in serum were referred to an external party to undergo liverHCV infection after at least two screening tests for anti- biopsy, if the patient accepted this procedure.HCV antibodies yielding reactive results. None of the patientswere referred to our research facility by the departments of 2.4. Assessment of HBV coinfectionobstetrics or gynecology. All patients had known their screen-ing test status before arrival to our department. Consecutive Infection with HBV was investigated by detection of thewomen older than 20 years and never treated for HCV infec- hepatitis B surface antigen (HBsAg) and anti-hepatitis Btion were eligible for the study. In cases with transfusion of core antigen (anti-HBc) antibodies, using an automated third-blood products, only those occurring before 1993 were con- generation microparticle enzyme immunoassay (MEIA, IMxsidered at signiﬁcant risk for HCV infection, since it was Version 3.0 Abbott Diagnostics, Chicago, IL, USA), as wellthe year in which blood bank screening for anti-HCV anti- as by detection of HBV DNA using a home-made qualitativebodies was introduced in Mexico [21,22]. In women who nested PCR. For HBV DNA assessment, an aliquot of 200 lreceived blood products in more than one occasion, the ear- of sera samples were obtained to isolate viral DNA using theliest transfusion was considered as the putative source of the QIAmp Blood Kit (Qiagen, Chatsworth, CA). HBV DNAinfection. Decompensated cirrhosis was considered to occur was ampliﬁed by standardized ﬁrst-round and nested PCR ofin a patient if jaundice, ascitis or collateral superﬁcial veins S-gene fragment using the primers and conditions describedwere present. Also, the antecedent of upper gastrointestinal previously . PCR assay was practiced per duplicate inbleeding, hepatic encephalopathy, tense ascitis, and sponta- all patients. The criterion for adjudication of a conﬁrmedneous peritonitis was considered as being part of the liver diagnosis and the measures taken to minimize the risk offailure syndrome. Patients were excluded if a new assay for cross-contamination were the same as in the assessment ofanti-HCV performed in our laboratory tested negative or if HCV RNA. Reactive sera to HBV DNA undergone directcomplete history and clinical data were not available. sequencing to determine HBV genotypes, as is described elsewhere .2.2. Detection of anti-HCV antibodies 2.5. Data analysis A new determination of anti-HCV was practiced in ourlaboratory for all patients, in order to conﬁrm the screen- Chi-square or Fisher exact test was used to assess nominaling tests performed prior to the arrival to our depart- variables in bivariate analyses, as corresponded. To comparement and to assess the quantity of anti-HCV antibodies. A quantitative variables between two groups, Student t-test andsemi-quantitative automated third-generation microparticle Mann–Whitney U-test were performed for parametric andenzyme immunoassay (MEIA, IMx HCV Version 3.0 Abbott non-parametric variables, respectively. Pearson correlation
3. 102 E. Chiquete et al. / Hepatology Research 36 (2006) 100–106Table 1Demographic and clinical characteristics of the 86 women analyzedVariable Total HCV RNA-positive (n = 55) HCV RNA-negative (n = 31) P-valueaAge Median (range) 53 (21–82) 54 (21–79) 51 (22–82) 0.18School education More than 6 years, n (%) 68 (79) 42 (76) 26 (84) 0.58Marital status Married, n (%) 64 (74) 41 (74) 23 (74) 0.98 Single, n (%) 10 (12) 4 (7) 6 (19) 0.11 Divorced, n (%) 7 (8) 6 (11) 1 (3) 0.23 Widower, n (%) 5 (6) 4 (7) 1 (3) 0.45Occupation Home (i.e., housewives), n (%) 43 (50) 31 (56) 12 (39) 0.17Liver enzymes activity ALT, mean (S.D.) 53.1 (48.8) 64.2 (63.6) 39.8 (15.7) 0.25 AST, mean (S.D.) 49.2 (39.3) 59.4 (47.9) 35.9 (18.9) 0.16Risk factors for HCV infection Major surgeries, n (%)b 69 (80) 46 (84) 23 (74) 0.40 Blood transfusion, n (%)c 50 (58) 34 (62) 16 (52) 0.37 None identiﬁed, n (%) 12 (14) 6 (11) 6 (19) 0.34ALT, alanine aminotransferase; AST, aspartate aminotransferase; HCV, hepatitis C virus; NA, not applicable; S.D., standard deviation. a P-value for differences between HCV RNA-positive and HCV RNA-negative women; Chi-square, Fisher exact test, t-test or Mann–Whitney U-test, ascorresponded. b Any major surgical procedure before the ﬁrst screening test for HCV infection yielded a reactive result. c Transfusion of blood products before 1993.was used in continuous distributions. To ﬁnd independent pre- Hosmer–Lemeshow goodness-of-ﬁt test, which was consid-dictors of the presence of cirrhosis at histologic assessment, ered as reliable if P > 0.20. Final statistical analyses werewe did a multivariate analysis by a binary logistic regres- regarded as signiﬁcant when P < 0.05. All P-values reportedsion model. Independent variables were chosen if P < 0.1 in are two-sided. SPSS v12.0 statistical package was used forselection step by bivariate analysis, subsequently a forward- all calculations.stepwise method was performed. Adjusted odds ratios (OR)with 95% conﬁdence intervals (CI) resulting in ﬁnal step ofthe model are provided. Corrected OR were calculated for 3. Resultscategorical predictors in order to approximate to the exact rel-ative risk in a small sample with a high frequency of a research In the period comprised by this study, 103 womenoutcome, with the formula proposed by Zhang and Yu  aged > 20 years who tested reactive to anti-HCV antibodiesas follows: corrected OR = multivariate OR/[(1 − incidence were studied. Of them, nine cases were excluded becauseof the outcome in the nonexposed group) + (incidence of the a new test for anti-HCV tested negative and eight casesoutcome in the nonexposed group × multivariate OR)]. The because complete data on medical records were not available.ﬁtness of the regression model was evaluated by using the Thus, 86 interferon-naive women, older than 20 years andTable 2Characteristics of the earliest transfusion in the 50 women who were recipients of blood products before 1993Variable Total HCV RNA-positive (n = 34) HCV RNA-negative (n = 16) P-valueaIndication for transfusion Obstetric/gynecologic, n (%) 34 (68) 24 (71) 10 (62) 0.75 Anemia, n (%) 7 (14) 3 (9) 4 (25) 0.19 Major surgery, n (%)b 5 (10) 4 (12) 1 (6) 0.98 Injury, n (%)c 4 (8) 3 (9) 1 (6) 0.75Time from transfusion to diagnosis Median (range) in years 21 (7–39) 23 (7–39) 20 (9–33) 0.11HCV, hepatitis C virus. a P-value for differences between HCV RNA-positive and HCV RNA-negative women; Chi-square, Fisher exact test or Mann–Whitney U-test, as corre-sponded. b Major surgeries related to blood loss that required transfusion, before the ﬁrst screening test for HCV infection yielded a reactive result. c Injuries related to blood loss that required transfusion.
4. E. Chiquete et al. / Hepatology Research 36 (2006) 100–106 103Table 3Multivariate analysis on factors predicting the presence of cirrhosis in histologic preparations of the liver (36 biopsy specimens of HCV RNA-positive cases)aVariable Regression coefﬁcient S.E. OR (95% CI) Corrected OR (95% CI) P-valueAge 0.194 0.075 1.21 (1.05–1.41) NA 0.01Transfusionb 0 = absent 1 = present 3.100 1.173 2.23 (1.20–221.14) 1.69 (1.14–3.80) 0.008Constant −11.214 4.328 NA NA <0.001CI, conﬁdence interval; HCV, hepatitis C virus; NA, not applicable; OR, odds ratio; S.E., standard error. a Adjusted for HBV coinfection, diabetes mellitus, and for anti-HCV antibodies levels. Hosmer–Lemeshow test for goodness of ﬁt in ﬁnal step of theregression model: χ2 = 4.70, 7 d.f., P = 0.70. b The antecedent of transfusion of blood products before 1993.repeatedly reactive to anti-HCV antibodies were analyzed 3.2. Coinfection with HBVfor the purposes of this study (Table 1). Mean age was 50.4years (median 53, range 21–82 years). Of the 86 women Three (3%) cases were positive to HBV DNA, two cor-studied, 55 (64%) had detectable HCV RNA in serum responding to genotype H and one to genotype A. No HCV(Table 1). RNA-negative cases with a positive result to HBV DNA in serum were identiﬁed. Hence, conﬁrmed coinfection (i.e., the presence of both HCV and HBV genomes in serum) occurred3.1. Risk factors for HCV infection in 3 (5%) of the 55 women who tested positive to HCV RNA. Occult hepatitis B (i.e., seronegative to HBsAg, but positive Any major surgical procedure and transfusion of blood to HBV DNA) occurred in two (4%) cases. One patient withproducts before 1993 were the most frequent risk factors, coinfection had transfusion of blood products and surgerieswith 74 (86%) cases having one or another, and 45 (52%) as risk factors, the other two had only surgeries.having both (Table 1). The main reason for practicing surgerywas obstetric/gynecologic (74%). There were no patientswho declared past or current illicit intravenous drug use, 3.3. Factors inﬂuencing liver diseasethe antecedent of a sex partner infected with HCV or sex-ual promiscuity. Twelve (14%) women had no identiﬁable Patients declaring signiﬁcant alcohol intake (i.e., >10 grisk factor for HCV infection. per day) were not identiﬁed. Thirty-six (42%) women had decompensated liver cirrhosis at inclusion in the study; 30 (83%) of them with detectable HCV RNA in serum. Among3.1.1. The antecedent of transfusion the 50 patients with the antecedent of transfusion, decom- Median age of the 50 (58%) women with the antecedent pensated cirrhosis was more frequent in those who had >20of transfusion was 54.5 years (range 21–82), which was dif- years of age at the moment of the transfusion (22 of 24, 92%)ferent from the median of 46 years (range 22–72) in the 36 than in younger women (17 of 26, 65%) (P = 0.02). Amongwomen without this antecedent (P = 0.009). The most fre- the 55 women with detectable HCV RNA in serum, 36 (65%)quent indication for transfusion was an obstetric/gynecologic underwent liver biopsy and histologic examination, with 13complication (e.g., miscarriage, abruptio placentae, dysfunc- (36%) cases having chronic liver inﬂammation with ﬁbrosistional uterine bleeding, and other conditions) in 34 (68%) and 23 (64%) cirrhosis. There were no cases with hepato-cases, followed by anemia in 7 (14%) (Table 2). The median cellular carcinoma. In a multivariate analysis adjusted foryear of the earliest transfusion was 1980 (range 1968–1992). diabetes mellitus, HBV coinfection, HBV genotypes and forThe median time from the earliest transfusion to the diagno- anti-HCV antibodies levels; determining factors of cirrho-sis of HCV infection by a reactive screening test was 21 years sis at histologic assessment were age and the antecedent of(Table 2). transfusion (Table 3).Table 4Median (minimum and maximum) of anti-HCV antibodies levels, expressed as sample-to-cut-off ratio (S/CO), in relation to several characteristicsHCV RNAa (n = 86) Transfusionb (n = 86) Surgeriesc (n = 86)Positive (n = 55) Negative (n = 31) Positive (n = 50) Negative (n = 36) Positive (n = 69) Negative (n = 17)43.3 (1.6–146.5) 2.0 (1–65.7) 41.1 (1–146.5) 34.9 (1.1–67.4) 40.0 (1–146.5) 38.0 (1.1–58.1) a P < 0.001, for differences between HCV RNA-positive and HCV RNA-negative women; Mann–Whitney U-test. b P = 0.08, for differences between women with and without transfusion of blood products before 1993; Mann–Whitney U-test. c P = 0.36, for differences between women with and without surgeries; Mann–Whitney U-test.
5. 104 E. Chiquete et al. / Hepatology Research 36 (2006) 100–106Fig. 1. Median and interquartile range of anti-HCV antibodies levels (expressed as sample-to-cut-off [S/CO] ratio) in different subgroups. Error bars indicateminimum and maximum, excluding two extreme cases. (A) Women with or without transfusion of blood products before 1993. (B) Women with or without apositive result to HCV RNA in serum. (C) Women with the antecedent of transfusion who had or had not detectable HCV RNA in serum. (D) Women withoutthe antecedent of transfusion who had or had not detectable HCV RNA in serum.3.4. Levels of anti-HCV antibodies found that the antecedent of surgeries was the most preva- lent risk factor for HCV infection, followed by transfusion Anti-HCV antibodies levels were higher in patients of blood products before 1993. However, a great propor-with decompensated cirrhosis than in asymptomatic women tion of patients who had the antecedent of surgeries also had(median S/CO value 41.2 versus 36.2, respectively; P = 0.03). transfusion, in most cases with an obstetric or gynecologicHowever, after considering only HCV RNA-positive cases indication. This ﬁnding had only been reported in a female(n = 55), anti-HCV levels were not different between patients population from Ireland . That most women with surg-with or without decompensated cirrhosis (median S/CO value eries or transfusion as the main risk factors for HCV infection41.2 versus 45.1, respectively; P = 0.28), and did not dif- had an obstetric/gynecologic indication for these procedures,fer between women who were or were not recipients of may help to explain the higher frequency of women with HCVtransfusion (Table 4; Fig. 1A). Nevertheless, HCV RNA- infection, as compared with that of men [5,16]; however, thispositive patients had higher anti-HCV levels than women issue needs further investigation.without detectable HCV RNA (Table 4; Fig. 1B), and this The levels of anti-HCV were higher in HCV RNA-difference remained regardless the antecedent of transfusion positive than HCV RNA-negative women. Furthermore, the(Fig. 1C and D). Anti-HCV levels directly correlated with the time elapsed from the acquisition of HCV to the diagnosiselapsed time from transfusion to diagnosis of HCV infection correlated with the levels of anti-HCV antibodies, in women(Fig. 2A), but not with age (Fig. 2B). After subgroup analysis, with transfusion as the putative route of infection. Hence,this correlation was higher in HCV RNA-positive (Fig. 2C) as it is shown by the determination coefﬁcient (r2 ) obtainedthan in negative women (Fig. 2D). from the correlation of anti-HCV levels and duration of the infection in HCV RNA-positive women, it is expected that about 14% of the variation in anti-HCV levels is determined4. Discussion by the state of chronicity. This is in agreement with a previous study that analyzed 43 recovered and 34 chronically infected In Mexico, the proportion of women with HCV infection women who were recipients of human Rhesus immunoglob-might be higher [5,16] than that reported in other countries ulin contaminated with HCV genotype 1b . This report[30–32], as is observed in several populations [10,17,18]. We demonstrated that in women with chronic hepatitis C,
6. E. Chiquete et al. / Hepatology Research 36 (2006) 100–106 105Fig. 2. Correlation between levels of anti-HCV antibodies (expressed as sample-to-cut-off [S/CO] ratio) and the elapsed time from the earliest transfusion tothe year of diagnosis (A), as well as age (B), in women who were recipients of blood products before 1993 (n = 50). Correlation between anti-HCV antibodieslevels and the elapsed time from the earliest transfusion to the year of diagnosis in the subgroup of HCV RNA-positive (C) (n = 34) and HCV RNA-negativewomen (D) (n = 16). Regression lines (ﬁt line, center) and the corresponding mean 95% conﬁdence intervals (outlying lines) are depicted.anti-HCV levels were higher whereas cellular and interferon vulnerability to oxidative stress  and the antecedent ofresponse were weaker than in patients with resolved infection transfusion before blood bank screening for HCV to a large. Thus, on one hand, there is a pattern favoring humoral inoculum that may imply a high number of viral genetic vari-response in the chronic state; and on the other hand, a strong ants , than other sources of infection .cellular immune response [15,33] and an initial low viral load In conclusion, the source of HCV infection inﬂuences the are associated with further resolved infection. It is possi- severity of the liver disease, and the duration of the infectionble that a long duration of the infection enables a continuous affects the levels of anti-HCV antibodies, in transfusion-reinforcement of the humoral response, thus affecting the associated hepatitis C. In most cases, surgery and transfusionlevels of anti-HCV antibodies. Nonetheless, certainly other had an obstetric or gynecologic indication, a ﬁnding that mayfactors as the cellular immune competence, viral replication explain why more women than men are observed with HCVreservoirs, and other complex characteristics inherent to infection in Mexico. Further studies are necessary for conﬁr-HCV may determine the rest of the variation of anti-HCV mation of these results.levels. Our ﬁndings conﬁrmed previous observations and add that, even in transfusion-associated hepatitis C, theduration of HCV infection inﬂuences the levels of anti-HCV Acknowledgementsantibodies. The frequency of occult HBV infection was lower than We thank to Dr. Jorge Segura, Dr. Miguel A Jim´ nez, Dr. ethat reported in other studies [35,36], and HBV coinfection Guadalupe Becerra, Dr. Angeles Quintero and Dr. Benjam´n ıwas not associated with the severity of the liver damage. The C´ rdenas for their clinical support and kind attention given aclinical importance of occult HBV infection has been a very to this work. We are also indebted to Montserrat Maldonadodebated topic, with studies that report a signiﬁcant correlation and Daniel Quezada for their friendship and assistance inwith the severity of the liver damage [14,37] and other stud- laboratory. CONACYT, Salud-2004-C01-025 to A.P.ies reporting no association at all [35,36], even in populationswith a high prevalence of occult HBV infection . Here,independent predictors of cirrhosis at histologic level were Referencesage and the antecedent of transfusion, a ﬁnding previouslyreported . Both age and transfusion may have a patho-  Vivas-Arceo C, Benavides SA, De Jesus Trujillo J, Panduro A, Rivas-physiologic connection with the chronic state, since age is Estilla AM. Hepatitis C virus: prevalence and routes of infection amongassociated with a longer persistence of the virus and a high blood donors of West Mexico. Hepatol Res 2003;25:115–23.
7. 106 E. Chiquete et al. / Hepatology Research 36 (2006) 100–106  Kim YS, Ahn YO, Lee HS. Risk factors for hepatitis C virus infection  Chiquete E, S´ nchez LV, Becerra G, Quintero A, Maldonado M, Pan- a among Koreans according to the hepatitis C virus genotype. J Korean duro A. Performance of the serologic and molecular screening of blood Med Sci 2002;17:187–92. donations for the hepatitis B and C viruses in a Mexican transfusion  S´ nchez JL, Sj¨ gren MH, Callahan JD, et al. Hepatitis C in Peru: risk a o center. Ann Hepatol 2005;4:275–8. factors for infection, potential iatrogenic transmission, and genotype  Sookoian S, Casta˜ o G. Evaluation of a third generation anti-HCV n distribution. Am J Trop Med Hyg 2000;63:242–8. assay in predicting viremia in patients with positive HCV antibodies.  Abdourakhmanov DT, Hasaev AS, Castro FJ, Guardia J. Epidemiolog- Ann Hepatol 2002;4:179–82. ical and clinical aspects of hepatitis C virus infection in the Russian  Dal Molin G, Tiribelli C, Campello C. A rational use of laboratory tests Republic of Daghestan. Eur J Epidemiol 1998;14:549–53. in the diagnosis and management of hepatitis C virus infection. Ann  Chiquete E, S´ nchez LV, Maldonado M, Quezada D, Panduro A. Pre- a Hepatol 2003;2:76–83. diction of the hepatitis C viremia using immunoassay data and clinical  Kiely P, Kay D, Parker S, Piscitelli L. The signiﬁcance of third- expertise. Ann Hepatol 2005;4:107–14. generation HCV RIBA-indeterminate. RNA-negative results in volun-  de Torres M, Poynard T. Risk factors for liver ﬁbrosis progression in tary blood donors screened with sequential third-generation immunoas- patients with chronic hepatitis C. Ann Hepatol 2003;2:5–11. says. Transfusion 2004;44:349–58.  Marcellin P, Asselah T, Boyer N. Fibrosis and disease progression in  Lu SN, Tung HD, Chen TM, et al. Is it possible to diagnose acute hepatitis C. Hepatology 2002;36:S47–56. hepatitis C virus (HCV) infection by a rising anti-HCV titre rather than  Flamm SL, Parker RA, Chopra S. Risk factors associated with chronic by seroconversion? J Viral Hepat 2004;11:563–70. hepatitis C virus infection: limited frequency of an unidentiﬁed source  Rivas-Estilla AM, S´ nchez LV, Matsui O, et al. Identiﬁcation of hep- a of transmission. Am J Gastroenterol 1998;93:597–600. atitis C virus (HCV) genotypes in infected patients from the west of  Dubois F, Desenclos JC, Mariotte N, Goudeau A, The Collaborative Mexico. Hepatol Res 1998;12:121–30. Study Group. Hepatitis C in a French population-based survey, 1994:  S´ nchez LV, Maldonado M, Bastidas-Ram´rez BE, Norder H, Pan- a ı seroprevalence, frequency of viremia, genotype distribution, and risk duro A. Genotypes and S-gene variability of Mexican hepatitis B virus factors. Hepatology 1997;25:1490–6. strains. J Med Virol 2002;68:22–34. Davoren A, Dillon AD, Power JP, et al. Outcome of an optional HCV  Zhang J, Yu KF. What’s the relative risk? A method of correct- screening program for blood transfusion recipients in Ireland. Transfu- ing the odds ratio in cohort studies of common outcomes. JAMA sion 2002;42:1501–6. 1998;280:1690–1. Harris DR, Gonin R, Alter HJ, et al. The relationship of acute  Alter MJ, Kruszon-Moran D, Nainan OV, et al. The prevalence of hep- transfusion-associated hepatitis to the development of cirrhosis in the atitis C virus infection in the United States, 1988 through 1994. N Engl presence of alcohol abuse. Ann Intern Med 2001;134:120–4. J Med 1999;341:556–62. Tong MJ, el-Farra NS, Reikes AR, Co RL. Clinical outcomes after  Dore GJ, Law M, MacDonald M, Kaldor JM. Epidemiology of transfusion-associated hepatitis C. N Engl J Med 1995;332:1463– hepatitis C virus infection in Australia. J Clin Virol 2003;26: 6. 171–84. Minola E, Prati D, Suter F, et al. Age at infection affects the long-  Chlabicz S, Bonifatiuk I, Radziwon P. Prevalence of hepatitis C virus term outcome of transfusion-associated chronic hepatitis C. Blood antibodies among blood donors in north-eastern Poland. Hepatol Res 2002;99:4588–91. 2005;33:206–10. Cacciola I, Pollicino T, Squadrito G, Cerenzia G, Orlando ME, Rai-  Semmo N, Barnes E, Taylor C, et al. T-cell responses and previous mondo G. Occult hepatitis B virus infection in patients with chronic exposure to hepatitis C virus in indeterminate blood donors. Lancet hepatitis C liver disease. N Engl J Med 1999;341:22–6. 2005;365:327–9. Takaki A, Wiese M, Maertens G, et al. Cellular immune responses  Uto H, Hayashi K, Kusumoto K, et al. Spontaneous elimination persist and humoral responses decrease two decades after recovery from of hepatitis C virus RNA in individuals with persistent infec- a single-source outbreak of hepatitis C. Nat Med 2000;6:578–82. tion in a hyperendemic area of Japan. Hepatol Res 2005;34: M´ ndez-S´ nchez N, Aguilar-Ram´rez JR, Reyes A, et al. Etiology of e a ı 28–34. liver cirrhosis in Mexico. Ann Hepatol 2004;3:30–3.  Kao JH, Chen PJ, Lai MY, Chen DS. Occult hepatitis B virus infec- Koerner K, Cardoso M, Dengler T, Kerowgan M, Kubanek B. Estimated tion and clinical outcomes of patients with chronic hepatitis C. J Clin risk of transmission of hepatitis C virus by blood transfusion. Vox Sang Microbiol 2002;40:4068–71. 1998;74:213–6.  Georgiadou SP, Zachou K, Rigopoulou E, et al. Occult hepatitis B virus Soza A, Arrese M, Gonzalez R, et al. Clinical and epidemiological infection in Greek patients with chronic hepatitis C and in patients with features of 147 Chilean patients with chronic hepatitis C. Ann Hepatol diverse nonviral hepatic diseases. J Viral Hepat 2004;11:358–65. 2004;3:146–51.  Raimondo G, Cacciamo G, Saitta C. Hepatitis B virus and hepatitis Barrett S, Goh J, Coughlan B, et al. The natural course of hepatitis C C virus co-infection: additive players in chronic liver disease? Ann virus infection after 22 years in a unique homogenous cohort: sponta- Hepatol 2005;4:100–6. neous viral clearance and chronic HCV infection. Gut 2001;49:423–30.  Koike K, Miyoshi H. Oxidative stress and hepatitis C viral infection. Bakr I, Rekacewicz C, El Hosseiny M, et al. Higher clearance of HCV Hepatol Res 2006;34:65–73. infection in females compared to males. Gut 2006, 24 January, [Epub  Zeuzem S, Herrmann E. Dynamics of hepatitis C virus infection. Ann ahead of print]. Hepatol 2002;1:56–63. Mar´n-L´ pez A. Bancos de Sangre. Rev Gastroenterol Mex ı o  Saito T, Watanabe H, Shao L, et al. Transmission of hepatitis C virus 2002;67:S11–2. quasispecies between human adults. Hepatol Res 2004;30:57–62.