|
|
| |||||||||||||||||||||||||||||||||||||||
Background Although there are case reports of vertical transmission of hepatitis C virus (HCV), it remains uncertain to what extent infected mothers transmit this virus to their infants.
Methods We investigated the transmission of HCV from infected mothers to their babies by analyzing HCV RNA in the blood. Three independent studies were performed. First, 7698 parturient women were tested for anti-HCV antibodies; 53 were positive. Their 54 infants (including one set of twins) were followed prospectively for at least six months and tested for HCV infection. Second, the babies of six women with known HCV disease were prospectively studied. Third, the families of three HCV-infected infants were examined retrospectively.
Results Of the 53 antibody-positive mothers, 31 were also positive for serum HCV RNA. Three of the 54 babies born to these mothers (5.6 percent) became positive for HCV RNA during the follow-up period. None of the babies of the 22 women who were antibody-positive but HCV RNA-negative became positive for HCV RNA. In the second study, HCV RNA was detected in one of the six infants of infected mothers. In the third study, HCV RNA was detected in the mothers of the three HCV-infected infants. In each of the seven infected infants we studied, the genomic sequence of HCV was almost identical to that from the mother. These seven mothers had significantly higher titers of HCV RNA than did the mothers of infants with no evidence of infection (mean [±SD], 106.4 ±0.5 vs. 104.4 ±1.5 per milliliter; P<0.001).
Conclusions HCV is vertically transmitted from mother to infant, and the risk of transmission is correlated with the titer of HCV RNA in the mother. .
The presence of the HCV RNA genome10 or core-related antigen11 in the circulation during infection is a reliable marker for viremia, but the analytical methods used in the detection are not routine or convenient. We analyzed HCV RNA with the polymerase chain reaction (PCR) to assess maternal transmission of HCV in prospective and retrospective studies. To confirm transmission, regions of the HCV genome from mothers and their children were sequenced and compared.
Methods
Prospective Studies
Study 1
From May 1990 to December 1992, a total of 7698 parturient women at Fukushima Medical College, Ishiwata Obstetric and Gynecologic Hospital, Teikyo University Mizonokuchi Hospital, and Gifu Kouseiren Gihoku Hospital were tested for anti-HCV antibodies: c100-3 (Ortho),12 antibodies detectable by a second-generation enzyme immunoassay (Abbott HCV EIA 2.0), and antibodies against peptides derived from the HCV core region (CP9 and CP10)13,14. A total of 53 mothers were positive for at least one of these antibodies, and their 54 babies (including one set of twins) were then followed for at least six months after birth. None of these infants received blood or related products during follow-up. Serum or plasma was extracted from blood obtained from the mothers at the time of delivery and from the babies during follow-up. All specimens were stored at -20 °C or below for later testing for HCV RNA and anti-HCV antibodies.
Study 2
Ninety-three women with liver disease due to HCV who were treated at the Second Department of Internal Medicine, National Defense Medical College, from January 1991 to December 1992 were consecutively enrolled in this study. The diagnoses were based on histologic findings: 26 women had chronic persistent hepatitis, 42 had chronic active hepatitis, 18 had cirrhosis, and 7 had hepatocellular carcinoma. Six of the patients delivered babies during the study, and these infants were followed for at least six months after birth.
The Retrospective Study
Study 3
From February to October 1991 at Jichi Medical School, liver function was tested in 400 infants who were 12 months of age or younger. Eleven had serum alanine aminotransferase or aspartate aminotransferase levels of more than 60 U per liter. Two were positive for HCV RNA, although they had not received a blood transfusion and had no known risk factors for HCV infection. A two-year-old infant positive for HCV RNA was identified at Onomichi General Hospital through the screening of 700 children. Serum samples were obtained from the mothers and other family members of these three index children and examined for HCV RNA and anti-HCV antibodies.
Detection, Titration, Genotyping, and Sequencing of HCV RNA
Nucleic acids were extracted from 0.1 ml of frozen plasma or serum and analyzed for HCV RNA by nested PCR with primers derived from the 5' untranslated region as described previously15. Specimens positive for HCV RNA were further examined in duplicate, and the RNA analysis was performed in blinded fashion according to the following procedure: RNA extracted from 0.1 ml of serum was serially diluted in 10-fold increments with diethyl pyrocarbonate-treated distilled water containing 40 µg of poly(A) per milliliter and subjected to complementary DNA (cDNA) synthesis and amplification by PCR. We determined the lowest titer at which the amplified HCV sequence was detectable.
We determined the genotype of HCV by amplifying a C-region sequence with type-specific primers as described previously16. A partial sequence of the genome was also determined according to reported methods16,17. The regions sequenced are depicted in Figure 1. In addition, the sequence of the hypervariable region at the 5' terminal of the E2/NS1 region (HVR-1)17,18,19 was determined for several cDNA clones from each of the HCV-infected family members of Patient 6 according to a previously reported method17.
|
Anti-c100-3 antibody12 and anti-HCV antibodies detectable on a second-generation enzyme immunoassay were detected with commercial reagents (Ortho and Abbott). Anti-HCV core antibodies were detected with an enzyme-linked immunosorbent assay (ELISA) with synthetic peptides CP913 and CP1014 predicted from the sequence of the HCV C-region genome.
Statistical Analysis
The HCV RNA titers in the groups were calculated as means ±SD. We performed Student's or Welch's t-test for comparison, and a P value of less than 0.05 was considered to indicate statistical significance. Fisher's exact test was used to compare the frequencies of blood transfusion or chronic liver disease in the two groups.
Results
Prospective Study
In the first study, 54 infants born to the 53 antibody-positive mothers were followed for at least six months. Only three babies (Patients 1, 2, and 3) became positive for HCV RNA during the follow-up period. These babies had been born to 3 of the 31 mothers positive for HCV RNA, so about 10 percent of the mothers with viremia transmitted HCV to their babies. In the second study, one of six infants born during the study to women with chronic HCV disease was infected with HCV (Patient 4).
Patient 1
Patient 1 was initially negative for HCV RNA (cord-blood specimen) but became positive at one month of age, and the titers increased thereafter (Figure 2). Tests for anti-HCV core antibodies were initially positive, and then the titers gradually declined. The level of anti-c100-3 was always below the cutoff level on ELISA, but the optical density of anti-c100-3 gradually declined in a manner similar to that of anti-HCV core antibodies.
|
|
Patient 2 became positive for HCV RNA at one month of age and remained positive thereafter. She was positive for HCV antibodies until four months of age and again became positive at nine months of age. Her mother had viremia at delivery (titer, 106 per milliliter), and both mother and child had the same HCV genotype. The patient's mother was given a clinical diagnosis of chronic hepatitis; she had been given a diagnosis of chronic non-A, non-B hepatitis at 12 years of age. Her mother (the maternal grandmother of Patient 2) had also been given a diagnosis of chronic HCV infection; she had received a blood transfusion during surgery for an ovarian tumor several years before the birth of the patient's mother.
Patient 3
Unlike Patients 1 and 2, Patient 3 was negative for HCV RNA at one month of age, seroconverted at two months of age, and remained positive thereafter. She was positive for HCV antibodies throughout the follow-up period.
Her mother had viremia at delivery (titer, 107 per milliliter) and had the same HCV genotype as her baby. The mother was given a clinical diagnosis of chronic hepatitis. She had no history of blood transfusion, injection-drug use, or liver disease. Her three-year-old daughter was also positive for HCV RNA and HCV antibodies.
Patient 4
The mother of Patient 4 was an outpatient under medical care for chronic persistent hepatitis that had occurred after a blood transfusion five years before the birth of Patient 4 (her first child). She had viremia at delivery (titer, 107 per milliliter) and had the same HCV genotype as her baby. Her baby was positive for both HCV antibodies and HCV RNA throughout the follow-up period.
Retrospective Study
Family members of three children positive for HCV RNA (Patients 5, 6, and 7) in the retrospective study were examined for HCV RNA and HCV antibodies. In each of the three families, the mother was positive for HCV RNA, had the same HCV genotype as her infected child, and had a titer of at least 106 per milliliter. All siblings of these patients were negative for serum markers of HCV infection.
Family of Patient 5
The mother of Patient 5 had been given a diagnosis of an asymptomatic carrier of HBV before contracting chronic non-A, non-B hepatitis after a blood transfusion one year after the delivery of her first child. Because she was a carrier of HBV and positive for antibodies against hepatitis B e antigen, both of her babies received only HBV immunoglobulin, and not HBV vaccine, within 48 hours of birth. Both children were negative for hepatitis B surface antigen at the time of the study.
Family of Patient 6
The father of Patient 6 was positive for HCV RNA (titer, 106 per milliliter) on serologic testing, and the genotype of the virus was the same as that isolated from the mother and the infant. He had never received a blood transfusion and had no other risk factors, but he had been given a diagnosis of "liver disease" because of abnormal results of liver-function tests (aspartate aminotransferase, 30 U per liter; alanine aminotransferase, 57 U per liter) 15 years previously. The patient was delivered by cesarean section after 33 weeks of gestation because of placenta previa, and his mother received a blood transfusion after the operation; symptomatic hepatitis developed two months later.
Family of Patient 7
The mother of Patient 7 was a nurse in whom hepatitis developed after a needle stick during the period between her two deliveries. Her hepatitis was mild (alanine aminotransferase level, 100 to 150 U per liter) and transient. She was the only one among the seven mothers of infected infants who did not have chronic hepatitis at the time of delivery of the patient.
Comparison of HCV Sequences
More than 97 percent of the nucleotides within the sequenced regions a to d of the HCV genome were identical in each mother-infant pair (Table 2). The overall frequency of homology, as well as the frequency of homology for each sequenced region, in mother-infant pairs was significantly higher than that in pairs of unrelated controls.
|
Risk Factors for Maternal Transmission of HCV
To determine possible risk factors, the seven mothers of infants infected with HCV were compared with the 33 mothers who had HCV viremia but whose babies were negative for HCV RNA in the two prospective studies. All seven mothers had HCV RNA titers of at least 106 per milliliter. The mean (±SD) titer in these mothers was significantly higher than that in the other mothers (106.4 ±0.5 vs. 104.4 ±1.5 per milliliter, P<0.001) (Figure 3). Six of the seven mothers (86 percent) had been given a diagnosis of chronic hepatitis, whereas only 11 of the 33 other mothers (33 percent) had some type of chronic liver disease (P<0.04). The mothers of infants infected with HCV were more likely to have received a transfusion than were the other mothers (57 percent vs. 42 percent), but the difference between groups was not statistically significant.
|
Vertical transmission of HCV from mother to baby is said to be rare,20,21,22,23,24 whereas it is frequent for HBV25,26,27. Our results suggest, however, that when mothers have high titers of circulating HCV RNA, HCV is frequently transmitted to neonates. The increased association of HCV infection in infants and mothers with chronic liver disease, as compared with the incidence of such infection in the infants of mothers without chronic liver disease, may reflect the link between more advanced liver disease and higher circulating HCV RNA titers28. Since the PCR method used is sensitive (detecting 1 to 10 chimpanzee infectious units of HCV virions15), the mothers of infected infants had circulating levels of 106.4 to 107.4 infectious units of HCV per milliliter. This range is comparable to HBV titers of approximately 108 chimpanzee infectious units per milliliter in blood positive for HBV e antigen29. Mothers who are positive for e antigen transmit HBV to their babies nearly 100 percent of the time in the absence of immunoprophylactic measures to prevent transmission25. In our first prospective study, we found maternal transmission of HCV in only 3 of 32 babies (9 percent) born to mothers with viremia. This may reflect the fact that the circulating titer of HCV is generally lower than that of HBV. However, babies born to mothers with high levels of viremia ( 
106 per milliliter) had a high rate of infection (7 of 14, or 50 percent). It is unfortunate that no effective and convenient serologic markers, such as e antigen in HBV, are available to predict the circulating titer of HCV25,29,30.
The natural course of HCV infection appears to be very different from that of HBV infection. In the latter, seroconversion from e antigen to anti-e antibody usually occurs as viremia and infectivity decrease18. This may account for the observation that children born earlier in the natural course of their mothers' HBV infection are more frequently infected with the virus than children born later26. In our study of maternal transmission of HCV, only one of the five living elder siblings of the seven infected babies had evidence of HCV infection. The mother of Patient 1 probably had been infected with HCV by a blood transfusion long before she delivered her third child, who was negative for HCV markers at the time of our study. The mother's circulating HCV RNA titer was apparently too low to transmit the infection at the time of that third delivery.
The sister of Patient 3 was also infected with HCV, probably transmitted by her mother, in view of the selectively higher HCV sequence homology between mother and daughter. It is possible that the infant was infected by the sister, and not by the mother. However, the study of the nucleotide sequences from these three family members indicated that the baby was most likely infected by the mother. The small difference in the sequence similarities between the baby's mother and the baby's sister and between the baby and her mother certainly reflects the difference in the duration of infection (i.e., the age difference) between the baby and her sister.
The father of Patient 6 was also positive for HCV RNA, and the overall nucleotide sequence of the viral strain isolated from the father was almost identical to that of the strain isolated from his wife and son, suggesting that these family members were infected with HCV with a common origin. Paternal transmission can therefore be considered as an alternative explanation for the infection of this child. However, the analysis of the E2/NS1 HVR-1 sequences for these three subjects indicated that at least one of the three cDNA clones from the mother had a sequence that was homologous to that from the infant, and no such homology was found between viral sequences obtained from the father and the baby or from the father and the mother. This finding supports maternal transmission. In addition, the marked heterogeneity of the HVR-1 sequence between the father and the mother suggests that HCV had not been transmitted between them recently, if at all. Abnormal liver function had been detected in the father some 15 years earlier, and he was most likely infected with HCV at that time. One possibility is that the mother was infected with HCV by her husband long before she received the blood transfusion after the cesarean delivery of Patient 6. If this was true, then the patient was exposed to maternal HCV at the time of delivery, suggesting that transmission may have occurred at delivery, and not through lactation or other types of contact during the postpartum period. The fact that the infant had a homogeneous population of E2/NS1 HVR-1 sequences despite the heterogeneity of the sequences obtained from his mother suggests that the baby was exposed to a very limited amount of virus, as would be expected for an infection transmitted percutaneously at birth. Our experiments indicate that when extensively diluted infective material harboring a mixed population of the E2/NS1 HVR-1 quasispecies was injected into chimpanzees, the resulting infection was the product of a single strain31,32.
A highly related but not completely identical E2/NS1 HVR-1 sequence in mother-infant pairs has been reported, and this finding has been interpreted to suggest in utero transmission of HCV33. However, our studies do not support this route of transmission, since cord-blood samples from Patients 1, 2, and 3 in our prospective study were all negative for HCV RNA.
We are indebted to Dr. Lacy R. Overby and Ms. Carson Gleberman for careful reading of the manuscript.
Source Information
From the Blood Transfusion Service (H. Ohto), Department of Pediatrics (N.U.), and Department of Obstetrics and Gynecology (C.E.), Fukushima Medical College, Fukushima; the Division of Pediatrics, Gifu Kouseiren Gihoku Hospital, Gifu (S.T.); the Department of Pediatrics (Nobuhiko Sasaki, A.M.) and Immunology Division (H. Okamoto), Jichi Medical School, Tochigi; the Department of Pediatrics, Onomichi General Hospital, Hiroshima (Nobutaka Sasaki); the Second Department of Internal Medicine, National Defense Medical College, Saitama (K.H.); Ishiwata Obstetric and Gynecologic Hospital, Ibaragi (C.I.); the Fourth Department of Medicine, Teikyo University School of Medicine, Kanagawa (M.K.); and the Institute of Immunology, Tokyo (S.M.) -- all in Japan. The other members of the study group are as follows: Mineo Kojima, Kojima Clinic, Gifu; Tatsuya Aikawa, Aikawa Internal Hospital, Ibaragi; Kazumi Shimoda, National Defense Medical College, Saitama; Minoru Sakamoto and Yoshihiro Akahane, Yamanashi Medical College, Yamanashi; Hiroshi Yoshizawa, Hiroshima University School of Medicine, Hiroshima; Takeshi Tanaka, Japanese Red Cross Saitama Blood Center, Saitama; Hajime Tokita, Jichi Medical School, Tochigi; and Fumio Tsuda, the Viral Hepatitis Research Foundation of Japan, Tokyo -- all in Japan.
Address reprint requests to Dr. Mishiro at the Institute of Immunology, Bunkyo-ku Koraku 1-1-10, Tokyo 112, Japan.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Vertical Transmission of Hepatitis C Virus
Simon H. J., Gurakan B., Oran O., Yigit S., Tighe M. K., Miskovitz P., Leon J., Ohto H., Okamoto H., Mishiro S.
Extract |
Full Text
N Engl J Med 1994;
331:399-400, Aug 11, 1994.
Correspondence
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
