Hepatitis C Virus Infection in Spouses of Patients with Type C Chronic Liver Disease

Methods

Spouses of consecutive patients at Yamanashi Medical College, Kojima Clinic, and Iwaki Kyoritsu General Hospital from October to December 1991 were studied. Most were referred from satellite medical facilities and screened for HCV-associated antibodies. Patients with antibodies were tested for HCV RNA. One hundred fifty-four patients were positive for both antibodies and HCV RNA, including 66 patients with chronic hepatitis, 49 with liver cirrhosis, and 39 with primary hepatocellular carcinoma, of whom 102 were men (age, 58 ±11 years) and 52 women (age, 57 ±8 years). It was not known how long the patients had had clinical hepatitis or were infected with HCV.

Their spouses, 102 women (age, 54 ±12 years) and 52 men (age, 60 ±9 years), were tested for HCV-associated antibodies and HCV RNA, and HCV genotypes were determined. They were asked to complete a questionnaire addressing the duration of present marriage, history of transfusion, occurrence of premarital non-A, non-B hepatitis, experiences of illicit intravenous drugs, previous marriages, and extramarital sexual relationships.

We also studied nine spouses who were excluded from the cohort because they had risk factors other than marriage to patients infected with HCV, such as a history of transfusion or premarital non-A, non-B hepatitis.

Hepatitis C Virus-Associated Antibodies

Antihepatitis C virus was screened using a second-generation enzyme immunoassay (EIA-II, Ortho Diagnostic Systems; Tokyo, Japan) with absorbance at 492 nm (A₄₉₂) values greater than 0.635 considered reactive. Antibody to a synthetic 36-mer peptide representing amino acids 39-74 of the product of the HCV core gene (anti-CP9) and antibody to a synthetic 19-mer peptide representing amino acids 5-23 (anti-CP10) were determined by EIA by methods described previously [17, 18] and A₄₉₂ values greater than 0.300 were considered positive.

Hepatitis C Virus RNA

The detection of HCV RNA by PCR was done using a slight modification of the method described previously [19], with primers deduced from the 5′-noncoding region of the HCV genome. Briefly, nucleic acids were extracted from 100 µL of serum, reverse-transcribed to cDNA using primer #299 (AC CCAACACTACTCGGCTAG, antisense, nucleotides [nt] 250-269) and Moloney murine leukemia virus reverse transcriptase (Superscript, GIBCO-BRL; Gaithersburg, Maryland), and amplified by a two-stage PCR with AmpliTaq DNA polymerase (Perkin-Elmer Cetus; Norwalk, Connecticut). The first PCR was done with primer pair #32 (CTGTGAGGAACTACT GTCTT, sense, nt 45-64)/#299 for 35 cycles and the second PCR with #33 (TTCACGCAGAAAGCGTCTAG nt 63-82)/#48 (GTTGATCCAAGAAAGGACCC, nt 188-207) for 25 cycles; nucleotides were numbered from the putative 5′-end of the HCV genome [20]. Each cycle included denaturation at 94 °C for 45 seconds, primer annealing at 55 °C for 45 seconds, and primer extension at 72 °C for 90 seconds. The HCV RNA assay was at least as sensitive as or 10³-fold more sensitive than that for detection of HCV in chimpanzee transmission experiments (Unpublished observations).

Hepatitis C Virus Genotypes

Hepatitis C virus RNA samples in spouses and those in corresponding patients were classified in terms of genotypes I, II, III, IV, and V [21-23]. A part of the HCV core gene spanning nt 480-751 (272 base pairs [bp]) was amplified on HCV cDNA with universal primers, #256 (CGCGCGACTAG GAAGACTTC nt 480-499) and #256V (CGCGCGACGCGTA AAACTTC, nt 480-499)/#186 (ATGTACCCCATGAGGT CGGC, nt 732-751). A portion of the product was amplifPCR with two universal sense primers, #104 (AGGAAGACT TCCGAGCGGTC nt 489-508) and #104V (CGTAAAACTTC TGAACGGTC, nt 489-508), and the mixture of five antisense primers deduced from sequences of the HCV core gene, #296 (GGATAGGCTGACGTCTACCT, nt 518-537), #133 (GAGC CATCCTGCCCACCCCA, nt 613-632), #134 (CCAAGAGG GACGGGAACCTC, nt 643-662), #135 (ACCCTCG TTTCCG TACAGAG, nt 592-611), and #339 (GCTGAGCCCAGGAC CGGTCT, nt 557-576), which were specific for genotypes I, II, III, IV, and V, respectively [20, 22]. The five genotypes were distinguished from one another by distinct sizes of PCR products: 49 bp for genotype I; 144 bp for II; 174 bp for III; 123 bp for IV; and 88 bp for V.

A typing method was proposed by Chan and colleagues [24] and by Simmonds and coworkers [25] that classifies HCV isolates into three major types—1, 2, and 3—based on the phylogenetic relatedness, with each type divided into distinct subtypes. Subtypes 1a and 1b in this classification correspond to I and II, respectively; 2a and 2b to III and IV; and 3a and 3b to V and VI. In another classification by Houghton and colleagues [26] and by Cha and coworkers [27], group I corresponds to genotype I (1a), group II to genotype II (1b), and group III to genotypes III (2a) and IV (2b); their group IV encompasses genotype V (3a) and group V includes a herd of variants reported only from South Africa so far.

Serologic Testing for the Other Viruses

Hepatitis B surface antigen was determined by passive hemagglutination with commercial assay kits (Mycell; Institute of Immunology, Co., Ltd., Tokyo, Japan). Antibody to human immunodeficiency virus type 1 was determined by passive agglutination of gelatin microparticles coated with viral antigens (SERODIA · HIV; Fuji Rebio, Tokyo, Japan).

Statistical Analysis

Frequencies between groups were compared using the chi-square test and the Fisher exact test. Exact confidence intervals were obtained by the standard method [28].

Results

Index patients with HCV-related chronic liver disease included 66 with chronic hepatitis, 49 with cirrhosis, and 39 with hepatocellular carcinoma. They all tested positive for HCV RNA as determined by PCR and for anti-HCV by EIA-II. None had hepatitis B surface antigen or antibody to human immunodeficiency virus type 1.

The patients' 154 spouses were tested for markers of HCV infection. The spouses included 52 husbands and 102 wives. None had been previously married, and all denied intravenous drug abuse or extramarital sexual contacts. None had received a transfusion or had a history of premarital non-A, non-B hepatitis.

Hepatitis C Virus Markers in Spouses of Patients

Anti-HCV as detected by EIA-II, anti-CP9, or anti-CP10 was detected in 42 (27%) of 154 spouses, of whom 25 (16%) tested positive for HCV RNA. Of the remaining 112 spouses without detectable HCV-associated antibodies, 2 (2%) were positive for HCV RNA. No appreciable differences were found in the diseases of the index patients married to spouses with HCV markers and to those without these markers. Hepatitis C virus markers were detected in 17 (33%) of 52 husbands and in 27 (26%) of 102 wives of index patients.

Prevalence of HCV antibodies and RNA increased in parallel with the duration of marriage (Figure 1). Hepatitis C virus markers were not detected in any of seven spouses who had been married to patients for fewer than 10 years. An increase therefore occurred in HCV-associated antibodies and HCV RNA detected in spouses proportional to the duration of marriage. Markers for HCV were detected in three (60%) of five spouses married longer than 50 years. Of 87 spouses married longer than 30 years, HCV RNA was detected in 21 (24%) compared with only 6 (9%) of 67 married less than 30 years (P < 0.05). Hepatitis C virus antibodies were also detected more frequently in the former group (28 of 87 [32%] compared with 14 of 67 [21%]). Logistic regression showed an increase in the odds of anti-HCV positivity of 50% per decade of marriage (odds ratio, 1.5; 95% CI, 1.05 to 2.2; P = 0.01), and the odds of HCV RNA positivity increased by 90% per decade (odds ratio, 1.9; 95% CI, 1.2 to 2.9; P = 0.004).

Of the 27 spouses with HCV RNA, 24 (89%) were infected with HCV of genotypes identical to those of index patients (Table 1). The remaining three spouses had HCV of genotypes different from those of patients. Thus, HCV RNA of the genotypes identical to those in the index patients was detected in 24 (16%) of 154 spouses, including 14 (14%) of 102 wives and 10 (19%) of 52 husbands.

Table 2 compares prevalences of HCV antibodies and HCV RNA in spouses of index patients classified by HCV genotypes. No statistically significant differences were found in the prevalence of HCV markers in spouses of patients infected with HCV of genotype II, III, or IV, or with mixed genotypes (II and III).

Liver Diseases in Spouses Infected with Hepatitis C Virus

Of 24 spouses infected with HCV of genotypes identical to patients, 8 had previously noted liver disease. Liver biopsy samples showed cirrhosis in one and chronic hepatitis in seven. Two additional spouses had elevated transaminase levels and were both found to have chronic hepatitis after liver biopsy. Thus, 10 spouses with apparent liver disease tested positive for HCV-associated antibodies. The remaining 14 did not have elevated alanine aminotransferase levels, but HCV-associated antibodies were detected in 12 (86%) of them. No appreciable differences were found between the 10 spouses with apparent liver disease and the 14 without liver disease in age (63 ±8 compared with 56 ±10 years) or duration of marriage (39 ±8 compared with 32 ±10 years).

Hepatitis C Virus Infection in Spouses with Other Risk Factors

Nine spouses with HCV antibodies were excluded from the cohort because they had received transfusions or had a history of non-A, non-B hepatitis before the marriage. Four were positive for HCV RNA. Hepatitis C virus genotypes were identical to those of patients in only one. The rate of identical HCV genotypes in these four patients (1 of 4 [25%]) was lower than that (24 of 27 [89%]) in spouses without risk factors other than marriage to patients (P < 0.04).

Discussion

Hepatitis C virus-associated antibodies were detected in 42 (27%) and HCV RNA in 27 (18%) of 154 spouses married to patients with viremia and HCV-related chronic liver disease for periods of 1 to 60 years. The prevalence of HCV antibodies among them was much higher than 1.5% in the general population [29] and probably in age- and sex-matched subpopulations. In 24 (89%) of the 27 spouses with viremia, genotypes of HCV were identical to those of the index patients: Nineteen were infected with HCV of genotype II, four with genotype III, and one with mixed genotypes (II and III). Although discordance of HCV genotypes may exclude the infection from suspected sources, concordance may not necessarily identify the route of infection. Restriction enzyme patterns and nucleotide sequences of HCV cDNA from patients and spouses with viremia would have to be compared to establish transmission between them. Such technical reservations notwithstanding, the 24 spouses in this study probably acquired HCV infection from their partners with HCV-related chronic liver disease. This view is supported by concordance of HCV genotypes in only one (25%) of four spouses with viremia who had risk factors other than the marriage.

The risk for HCV infection increased with duration of marriage, ranging from 1 to 60 years (mean, 30 years). None of the seven spouses married to patients for fewer than 10 years had HCV antibodies or HCV RNA. The prevalence of HCV markers gradually increased with the duration of marriage. Three (60%) of five spouses married for longer than 50 years had both anti-HCV and HCV RNA. Taken along with the lack of other risk factors for HCV infection in the studied spouses, such as transfusion, history of hepatitis, intravenous drug abuse, extramarital sexual contact, or previous marriages, the observed correlation between development of HCV markers and the duration of marriage over decades favors possible transmission of HCV infection from patients to spouses.

The correlation between the duration of marriage and the development of HCV infection in spouses presupposes that index patients were already infected at the time of marriage. However, because it is not known how long patients had clinically apparent disease or were infected with HCV, the actual exposure time cannot be determined. Prospective studies are needed to establish the correlation of HCV infection and the duration of sexual exposure but would require decades. A common source of infection for both patients and spouses is another possibility but would be rare considering the low rate of infectivity of HCV for community-acquired or household infection, except between spouses. Sharing needles for illicit intravenous drugs would be the only plausible explanation for a common source infection, but none of the spouses or patients studied were drug abusers.

Our results support those of previous studies in some ways and contradict them in others. It has been claimed that sexual transmission of HCV is infrequent based on screening of HCV-associated antibodies in homosexual men and patients seen in sexually transmitted disease clinics [8-10]. Hepatitis C virus is clearly less transmissible than hepatitis B virus or human immunodeficiency virus seen more frequently in such patients with sex-associated risk factors. In our study, HCV infection was not found in any of the seven spouses married for fewer than 10 years to patients with HCV-related chronic liver disease. Such a low rate of transmission to spouses is remarkable, considering that HCV RNA levels in patients with chronic liver disease probably would be much higher than those in symptom-free carriers having transient extramarital sexual relations. Substantial rates of transmission have not been observed after transient sexual contacts to inconsistent partners with unspecified HCV infection status [8-10] or among spouses married to patients with acute hepatitis C for 1 year or less [12]. Low estimates of risk have been based on studies documenting the lack or paucity of HCV-associated antibodies in spouses or sexual partners of patients with chronic hepatitis C [11-14] in which the risk is not correlated with the duration of marriage. Later studies have shown increased risk in family members of patients with chronic hepatitis C in which spouses have the highest risk [30, 31], in agreement with our results. The term “sexual contact” encompasses a wide range of relationships. It includes extramarital affairs with changing partners for limited periods, as well as lifetime commitment to only one partner over many years. Low rates of sexual transmission of HCV observed in some situations may not be extrapolated to others. Previous data therefore may not be applicable to spouses married to patients with HCV-related chronic liver disease for many decades. Our results corroborate substantial heterosexual transmission of HCV recently proposed by others [3, 30-32].

It is interesting that 10 (42%) of the 24 spouses with HCV infection, probably acquired after marriage, had overt clinical disease. Eight had been attending liver clinics for some time before the study. The remaining two were identified by screening for HCV markers and elevated transaminase levels in the course of the study, and both had chronic hepatitis diagnosed by liver biopsies. Because many HCV-infected persons may be asymptomatic and have normal aminotransferase levels, and because liver biopsies were not done on the remaining 14 asymptomatic spouses infected with HCV, we may have underestimated the extent of liver disease in these persons.

Spouses of patients with HCV viremia and chronic liver disease have an increased risk for developing HCV infection, and the risk increases over time. Spouses married to persons with demonstrable anti-HCV, whether symptom-free carriers or patients with documented liver disease, should be screened for HCV markers at regular intervals, perhaps twice a year, to receive prompt care if and when they become infected.