Main

Worldwide, HCV is the major cause of community-acquired and posttransfusion non-A, non-B hepatitis(1, 2). The HCV infection tends to persist and can induce chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma in the host(3, 4). An enigma of HCV infection, however, is that a defined route of infection, such as transfusion or illicit i.v. drugs, is not established in approximately one half of infected individuals(5).

Mother-to-baby transmission of HCV may fill in this gap, and has been reported in 0-33% of the babies born to infected mothers(6–15). HIV is proposed as an enhancing factor for the vertical transmission of HCV; the frequency varies widely from 0 to 100% in the babies born to mothers with antibody to HIV(6, 9, 13, 15–17). It is not certain how frequently HCV persists after mother-to-baby transmission. We followed babies with a prospective or retrospective diagnosis of mother-to-baby transmission and found that some of them rid HCV RNA from their serum.

METHODS

Subjects. Children with prospectively diagnosed vertical transmission. During August 1991 to May 1995, 2762 consecutive pregnant women were screened for anti-HCV at the Department of Obstetrics and Gynecology, Jichi Medical School. Babies born to the 19 mothers with anti-HCV were tested for anti-HCV and HCV RNA at birth (cord blood) and followed at 1, 3, 6, 9, and 12 mo thereafter. Thus mother-to-baby transmission of HCV was diagnosed prospectively in three of them.

Children with retrospectively diagnosed vertical transmission. Children with a family history of liver disease, as well as those who received an operation or who were otherwise admitted to the hospital, were tested for anti-HCV and serum transaminases. These tests were performed on 5010 (2.8%) of 177 229 children at the Department of Pediatrics of Jichi Medical School from January 1991 to August 1995, and 1021 (2.5%) of 40 076 children at Ibaraki Prefectural Central Hospital from January 1993 to December 1994. Mothers of children with HCV infection were recommended to receive tests for serologic markers of HCV infection. Mother-to-baby transmission of HCV was diagnosed retrospectively in five children who had not received transfusions or blood products and whose mothers possessed HCV RNA in their serum. Additionally, children of 63 mothers with anti-HCV who visited the Department of Obstetrics and Gynecology, Jichi Medical School, as well as Ibaraki Prefectural Central Hospital, were screened, and three children were found to be infected with HCV. None of them had received transfusions or blood products before, and none of their mothers had antibody to HIV.

Children with posttransfusion HCV infection. There were 14 children with anti-HCV and HCV RNA whose mothers were without markers of HCV infection, but who had received transfusions for various causes. They were considered to have contracted HCV infection by transfusion.

Children with HCV infection were followed for serologic markers of HCV infection and serum transaminases. HCV RNA was tested for and genotyped in sera separated within 1 h after blood drawing and stored frozen at -70 °C. The study was approved by the Ethics Committees of the institutions, and informed consent was obtained from parents of all children.

HCV-associated antibodies. Anti-HCV was determined by the first-generation ELISA (ELISA-I, Ortho Diagnostic Systems, Tokyo, Japan) until November 1992 with absorbance at 492 nm (A492) >0.40 considered reactive. Thereafter, it was tested for by the second-generation ELISA (ELISA-II, Ortho) with A492 readings >0.63 considered positive. Antibodies to HCV core were determined in sera diluted 100-fold by ELISA with synthetic oligopeptides deduced from the HCV core gene(18) (SMITEST HCV Core Ab ELISA, Sumitomo Metal Industries, Ltd., Tokyo, Japan), and absorbance at 450 nm(A450) was determined.

HCV RNA and genotypes. Nucleic acids were extracted from serum(100 μL), and HCV RNA was determined by reverse transcription PCR by the method described previously(19). Relative concentration of HCV RNA was expressed by the reciprocal of the highest dilution(10N) of extracted nucleic acids in which HCV RNA was detectable, which was converted to represent the titer per mL of serum.

HCV genotypes were determined by PCR with type-specific primers deduced from the HCV core gene(20, 21). Genotypes are designated I, II, III, IV, and V by Okamoto's classification(20, 21). In another classification system based on the phylogenetic relatedness by Simmonds et al.(22). HCV is grouped into types such as 1, 2 and 3, with each type further breaking down into subtypes like a, b and c. Genotype I corresponds to 1a, II to 1b, III to 2a, IV to 2b, and V to 3a. Both nomenclature systems were used in the present study, such as I/1a, II/1b, III/2a, IV/2b, and V/3a.

Statistical analyses. Frequencies between groups were compared using the χ2 test and Fisher's exact test. Group means were compared using the t test.

RESULTS

Follow-up of babies born to mothers infected with HCV. Anti-HCV was detected in sera from 19 (0.69%) of 2762 mothers. They included four(0.37%) of 1080 mothers who were tested by the first-generation ELISA before November 1992 and 15 (0.89%) of 1682 mothers tested by the second generation ELISA since then. Of the 19 mothers, one had a miscarriage and one gave birth to twins. In total, therefore, 19 babies were born to 18 mothers with anti-HCV who were aged 24-38 y at the delivery (mean 29.8 ± 4.2 y).

HCV RNA was detected in 14 (78%) of the 18 mothers with anti-HCV. Three(20%) of the 15 babies born to 14 viremic mothers developed HCV RNA in serum, whereas none of the four born to four mothers without HCV RNA did; twins born to a viremic mother were negative for HCV RNA. HCV genotypes were II/1b for all of the three mother-baby pairs. One of them tested positive for HCV RNA in low titers (101/mL) both at birth (cord blood) and 1 mo along with anti-HCV detectable by the second-generation ELISA with A492 values at 4.1 and 1.5, respectively. However, neither HCV RNA or anti-HCV was detected in his sera taken at 3 mo and after. In another baby, also, HCV RNA was detected at birth (102/mL) and 1 mo (101/mL) but turned negative in sera taken at 3 mo and thereafter. He had anti-HCV in high titer with A492 readings >4.5 up to 6 mo. Then, anti-HCV decreased gradually in titer and turned negative in sera taken 1 y and after. These two babies kept normal levels of ALT through the observation period up to 20 mo after birth.

In the remaining baby, HCV RNA of genotype II/1b persisted in high titers(≥104/mL) accompanied by high titered anti-HCV throughout the observation from the delivery to 16 mo. Thus, HCV carrier state developed in only one (7%) of 15 babies born to mothers with HCV RNA in serum. Seven babies of six mothers with HCV RNA in serum, including a set of twins, were delivered by cesarean section; none of them was infected with HCV.

HCV RNA in children with retrospectively diagnosed vertical transmission. Eight children were deduced to have acquired HCV infection from their mothers retrospectively (Table 1). They all had been born by natural delivery. HCV RNA of the same genotype was detected in each mother-child pair, and none of the children had received transfusions. They included four boys and four girls aged 0.8-13.6 y (mean 5.3 ± 4.7 y) at the first observation. Children were the index in five, and they were found anti-HCV positive in tests performed before minor operations or at admission for the treatment of nonhepatic disease. The remaining three children received tests because they had been born to mothers with anti-HCV. Although none of the eight children had clinically apparent disease, six (75%) had developed elevated ALT levels.

Table 1 Profiles of child-mother pairs with retrospective vertical transmission of HCV
Full size table

Some mothers had risk factors for HCV infection. A history of transfusion 5-7 y before the delivery was reported by two mothers, and two other mothers(cases 3 and 8) were nurses. One child had two siblings, five children had one sibling, and the remaining two children had none. None of the seven siblings of infected children was positive for anti-HCV.

The eight children with retrospective vertical HCV infection were followed for 1.4-5.0 y (mean 3.2 ± 1.3 y) until they were 3.3-16.7 y (mean 8.5± 4.3 y). HCV RNA disappeared in two children (cases 2 and 3), whereas it persisted in the remaining six. The two children who cleared HCV RNA had been reported previously when they retained HCV RNA in their serum(12). Their clinical courses are shown in Figs. 1 and 2, respectively.

Figure 1
figure 1

ALT and HCV markers in a child who cleared the virus after vertical infection (case 2 in Table 1). Fluctuating levels of ALT and antibodies to HCV core tested in sera diluted 100-fold are shown. Anti-HCV by ELISA of the second generation was detected throughout the observation. To convert values in U/L to μkat/L, multiply by 0.01667.

Full size image
Figure 2
figure 2

ALT and HCV markers in a child who cleared the virus after vertical infection (case 3 in Table 1). Antibodies to HCV core tested in sera diluted 100-fold. Anti-HCV by ELISA of the second generation was detected throughout the observation. To convert values in U/L to μkat/L, multiply by 0.01667.

Full size image

Case 2 (Fig. 1) was a 6-year-old girl born to a mother infected with hepatitis B virus and positive for antibody to hepatitis B e antigen in serum. The mother received massive transfusions at an operation for ectopic pregnancy 5 y before the delivery, and contracted posttransfusion non-A, non-B hepatitis. The baby (case 2) received hepatitis B immune globulin at birth and escaped hepatitis B virus infection. However, she had a bout of elevated ALT during the first year after birth and two such bouts during the second year. HCV RNA was detected with fluctuating titers from 1 y after birth when she was first tested, persisted until she became 3 y old, and then disappeared and stayed negative for 3 y at the last testing. She was positive for anti-HCV by the second-generation ELISA throughout the observation period. Her anti-HCV core antibodies increased in titer and peaked immediately after the last bout of elevated ALT, and then decreased gradually.

Case 3 was a 6-year-old boy (Fig. 2). While he was followed for ALT and HCV markers, he experienced a few small and one big bout of ALT elevation. He was found to have HCV RNA in his serum from the first test performed at the age of 1 y. He turned negative for HCV RNA 6 mo after the last and highest bout of elevated ALT, at the age of 3 y and 4 mo. He stayed positive for anti-HCV by the second-generation ELISA. Anti-HCV core antibodies increased in titer and reached the peak concurrently with the last bout of ALT elevation, and then decreased gradually.

HCV RNA in children who contracted infection by transfusions. Of the 280 children who had received transfusions, 17 (6.1%) tested positive for anti-HCV, and 14 of them had HCV RNA in the serum. The features of the 14 children with HCV infection are shown in Table 2. They included 10 boys and four girls aged 5.6-17.3 y (mean ± SD: 11.3± 3.8 y) at the first observation. None of their mothers was positive for anti-HCV or HCV RNA in serum.

Table 2 Profiles of children who contracted HCV infection by transfusion
Full size table

Transfusions were given to seven children with hematologic malignancy, five at the operation and one each because of hemorrhage due to vitamin K deficiency or premature birth. They first received transfusions 4-18 y before(mean ± SD: 10.1 ± 3.7 y), and 11 (79%) had a history of hepatitis 2-13 y previously (9.1 ± 3.5 y).

HCV RNA persisted in all of the 14 patients during the observation of 2.6-6.1 y (4.0 ± 1.1 y). They had the maximal ALT level of 535 ± 564 U/L (range: 42-1750 U/L), which was significantly higher (p < 0.05) than that of the eight children with retrospective vertical HCV infection at 130 ± 94 (13-314 U/L). The duration of HCV infection was not significantly different between the children receiving transfusions and those with vertical HCV infection (10.1 ± 3.7 versus 8.5± 4.3 y).

DISCUSSION

The efficacy of mother-to-baby transmission of HCV is, as yet, not certain. This is ascribable, at least in part, to the difficulty in diagnosing ongoing HCV infection in mothers. Early studies using anti-HCV by the first-generation ELISA as a marker of infection are subject to low sensitivity and specificity(23, 24). Anti-HCV by ELISA of the second generation would not be sufficient(25), because it can detect resolved infection and does not always represent concurrent HCV infection. Hence HCV RNA in mother's serum needs to be determined and correlated with the transmission to babies.

The frequence of transmission to babies from the mothers with serum HCV RNA varies widely(6–15), and the efficacy may depend on the viral load in maternal sera(12). Altogether, the frequency of perinatal HCV transmission would be infrequent. This view is supported by the present study in which an HCV carrier state was established in only one (7%) of the 15 babies born to mothers with serum HCV RNA and followed prospectively for longer than a year. Mother-to-baby transmission occurred in three of eight babies with natural delivery but not in any of seven babies delivered by cesarean section. However, the number of studied babies is too small to evaluate statistical significance, and further observations are necessary to corroborate this difference.

Infants are reported to have a prevalence of anti-HCV much lower than older populations(26, 27), which would reflect infrequent mother-to-baby transmission. Sexual transmission is not a major route of community-acquired infection(28, 29), except in long-time spouses of the patients with chronic hepatitis C(30). Hence, the main route of community-acquired HCV transmission remains unclear. A low efficiency of mother-to-baby transmission of HCV would be attributable to a low dose inoculum involved in it. Of many HCV quasispecies in maternal circulation, only a selected population of HCV is transmitted to their babies(12, 31). This view would be in line with a low risk of HCV infection after the exposure to contaminated needle sticks(32).

Two of the three babies born to infected mothers carried HCV RNA only transiently. One tested positive for HCV RNA in his cord serum and serum taken 1 mo after birth. It is not certain if HCV replicated in him. The lack of active immune responses to HCV through the observation, taken along with normal ALT levels, would support a passive transfer of HCV without replication. The other baby lost HCV RNA in 3 mo but kept high anti-HCV titers until 6 mo after birth. HCV RNA tests were performed in sera separated within 1 h after blood sampling and stored frozen at -70 °C in aliquots for all the babies and children studied. Sera were thawed only once before the test.

Infants perinatally infected with HCV may have spontaneous clearance within the first several months, and this may not represent true“infection” but transient viremia. This would have implications for how we counsel patients, and when we test infants born to anti-HCV seropositive mothers. Transient HCV viremia in babies born to infected mothers has been reported by others(6, 7, 9, 15, 17).

In adulthood, the natural history of HCV infection can be followed after the transmission by transfusions. The mother-to-baby transmission of HCV is another opportunity to follow the natural history, although immature immune responses of the infancy need to be taken into consideration. It would be of particular note that some babies with mother-to-baby transmission lost HCV after they kept it for a few years. The clearance of HCV infection in two(25%) of the eight children with retrospective vertical transmission stands in sharp contrast to the loss of circulating HCV RNA in none of the 14 children who had contracted infection by transfusions.

Whether large viral loads associated with transfusions may account for this difference is worth consideration. Blood units with a wide variety of HCV quasispecies would give an advantage for perpetuating infection, due to increased chances for them to evade immune surveillance of the host. The size of initial viral load might influence the clinical disease that develops later in life as well. The eight children with retrospective vertical HCV infection had peak ALT levels significantly lower than those in the 14 children with posttransfusion HCV infection, although the duration of HCV infection was comparable between them.