1121

Increased Risk of Maternal-Infant Hepatitis C Virus Transmission for Women

Coinfected with Human Immunodeficiency Virus Type 1
Pier-Angelo Tovo, Elvia Palomba, Gabriele Ferraris,
Nicola Principi, Ezia Ruga, Paola Dallacasa,
Anna Maccabruni, and the Italian Study Group
for HCV Infection in Children*
From the Department of Pediatrics, University of Turin, Turin; the
Maternal/Infant Center, USSL 75/I, Milan; the Pediatric Department IV,
University of Milan, Milan; the Department of Pediatrics, University of
Padua, Padua; the Institute of Infectious Diseases, IRCCS, University of
Pavia, Pavia; and the Department of Pediatrics, University of Bologna,
Bologna, Italy

To estimate the risk of mother-to-child transmission of hepatitis C virus (HCV) and identify
correlates of transmission, 245 perinatally exposed singleton children followed prospectively beyond
18 months of age were studied. Overall, 28 (11.4%) of the 245 children acquired HCV infection.
Transmission occurred in 3 of 80 children (3.7%) whose mothers had HCV infection alone and in
25 of 165 (15.1%; P .01) whose mothers had concurrent infection with human immunodeficiency
virus type 1 (HIV-1). The percentage of HIV-1-infected children was similar (22 of 165, 13.3%), but
each virus was transmitted independently; only six infants (3.6%) were coinfected with HCV and
HIV-1. The risk of HCV transmission was not associated with maternal HIV-1-related symptoms,

Downloaded from http://cid.oxfordjournals.org/ by guest on November 29, 2016

intravenous drug use, prematurity, low birth weight, or breast-feeding, whereas it was lower with
cesarean section than with vaginal delivery (5.6% vs. 13.9%, P .06). This suggests that transmission
occurs mainly around the time of delivery.

Mother-to-child transmission of hepatitis C virus (HCV) is
widely documented. However, in different investigations the
estimated risk of infection ranged from zero to 100% [1 15].
This variability derives from differences in methods used to
define the childs infection status, duration of follow-up, and
size and features of the populations studied.
Transmission of HCV may occur in utero via the transplacen-
tal route at any time during pregnancy, at the time of delivery,
and postnatally through breast-feeding. Several factors might
favor or hamper infection of the offspring. Identification of
these factors would allow the adoption of rational preventive
strategies and the offer of specific counseling, but little is
known about the timing and correlates of transmission. High
levels of viremia [5 7], certain HCV genotypes [8], and mater-
nal coinfection with HIV-1 [1, 2, 4, 7, 9, 15] have been associ-
ated with increased HCV transmission rates, although the re-
sults are controversial [10 13]; breast milk does not seem to
have a significant role [13 15].
The aim of this study was to quantitate the risk of HCV
infection in children born to mothers with or without
HIV-1 coinfection. The effects of other possible correlates
of transmission, such as maternal HIV-1 disease progression,
history of intravenous drug use (IVDU), length of pregnancy,
Received 10 January 1997; revised 21 May 1997.
Grant support: Ministero della Sanita-I.S.S.-Rome (AIDS Projects) and
Ministero dellUniversitae della Ricerca Scientifica, Italy.
* Other participants are listed at the end of the text.
Reprints or correspondence: Pier-Angelo Tovo, Department of Pediatrics,
Piazza Polonia 94, 10126 Turin, Italy.
Clinical Infectious Diseases 1997;25:11214
q 1997 by The University of Chicago. All rights reserved.
10584838/97/25050030$03.00
birth weight, mode of delivery, and type of feeding were also
investigated.
Patients and Methods
Patients. A cohort of 245 children born to HCV-infected
women were enrolled at 12 participating centers. In three cen-
ters all parturients were screened for HCV-seropositivity; in
the remaining centers, HCV testing was mostly performed only
for women who had a history of IVDU or were known to be
HIV-1-infected. Only singleton at-risk infants identified within
the first 2 weeks of life and followed up for at least 18 months
were included in this study.
Data collection. Specific information was collected by
questionnaire on maternal risk factors for HCV infection
(IVDU, transfusions, sexual contact with an infected partner,
other, unknown), mothers HIV-1 infection status at delivery
(presence or absence of specific antibody), length of pregnancy
(weeks), mode of delivery (vaginal, elective/emergency cesar-
ean), birth weight (grams), type of feeding (maternal, with
duration of breast-feeding, or formula only), and age at first
and last visit. The presence or absence of antibodies to HCV
and HIV-1 as well as of HCV RNA (with dates of the tests)
in the child was also reported.
Laboratory tests. Clinical examination and laboratory test-
ing were performed every 3 5 months over the first 18 months
of life; thereafter, only children with HCV and/or HIV-1 infec-
tion were regularly followed for a mean period of 28 months
(range, 19 42 months), while most uninfected children were
discharged or lost to follow-up. The mothers enrollment serum
samples and follow-up samples from their children were tested
for HCV-seropositivity by a second-generation EIA and a re-
combinant immunoblotting assay (RIBA II or RIBA III). Anti-

/ 9c3f$$no51 10-17-97 21:18:40 cidal UC: CID

1122 Tovo et al.
bodies to HIV-1 were measured by ELISA and confirmed by
western blotting.
The presence of HCV RNA was investigated by reverse
transcriptase PCR, performed by single participating centers.
Total RNA was extracted from serum samples of perinatally
exposed children; after a reverse transcription step, cDNA was
amplified with commercial primers of the 5* noncoding region
of the viral genome in a single PCR of 35 cycles. PCR product
was separated by agarose gel electrophoresis and visualized by
ethidium bromide staining.
Childs HCV (and HIV-1) infection status. Children were
considered to be infected with HCV and/or HIV-1 when spe-
cific antibodies persisted beyond 18 months of age. Children
with HCV RNA in at least two serum samples were also consid-
ered infected.
Risk factors for transmission. To assess the impact of ma-
ternal HIV-1 coinfection on transmission of HCV to the off-
spring, mother/child pairs were divided into two groups. Group
A included 80 women with HCV alone, and group B included
165 women coinfected with HIV-1. The effects of other factors
possibly related to HCV transmission were also evaluated.
These included a history of IVDU, mothers HIV-1 disease
state, length of pregnancy, birth weight, mode of delivery, and
type of feeding. The influence of the HIV-1-related maternal
clinical condition was analyzed to enable grouping of women
as asymptomatic or symptomatic at delivery; the latter included
those with any HIV-1-associated symptoms or signs. The ef-
fects of prematurity and low birth weight were assessed to
enable grouping of children according to gestational age and
birth weight (36 weeks vs. 37 weeks and 2,500 g vs.
2,500 g).
Results
In total, 28 of the 245 children (11.4%; 95% CI, 7.5 15.3)
acquired HCV infection. Of 80 group A children, 3 (3.7%;
95% CI, 0.4 7.9) were HCV-seropositive after 18 months of
age, while of 165 group B infants, 25 (15.1%; 95% CI, 9.6
20.6; P .01) contracted HCV. In particular, 22 (13.3%) were
seropositive beyond age 18 months. Of these 22, 20 (90.9%)
had transaminase levels more than twice the normal values in
several determinations. Another three HIV-1-uninfected chil-
dren (1.8%) seroreverted but had HCV RNA detected in at least
two separate determinations (one had 2 of 2 PCRs positive, one
had 3 of 3, and the third had 2 of 7), with enhanced transami-
nase values in two cases. Consistent increases in transaminase
levels were not observed among the remaining antibody-
negative children.
PCR was performed for another 89 healthy seroreverting
children (34 of group A and 55 of group B), with negative
findings for all but two group B children, who had one inconsis-
tently positive result and normal transaminase values (and were
thus judged to be uninfected). HCV RNA was detected in
16 (84.2%) of 19 antibody-positive children. No substantial
/ 9c3f$$no51 10-17-97 21:18:40
CID 1997;25 (November)
differences in HCV transmission were observed between the
centers.
Of 165 group B children, 22 (13.3%) acquired HIV-1, in-
cluding 6 (3.6%) coinfected with HCV.
Table 1 shows the influence of other possible risk factors
on HCV transmission. The percentage of HCV-infected chil-
dren was higher among those born by vaginal delivery than by
cesarean section, with differences at the limit of significance in
the cohort as a whole and in group B. No significant difference
emerged in comparison of elective and emergency cesarean
deliveries (2 of 50 [4%] vs. 2 of 21 [9.5%], respectively). Mode
of delivery was also associated with transmission of HIV-1,
with 20 (16.8%) of 119 HIV-1-infected children delivered vagi-
nally and 2 of 46 (4.3%; P .05) by cesarean section.
The presence of HIV-1-related symptoms in the mother was
not predictive of HCV infection (table 1). Likewise, the propor-
tion of children who acquired HIV-1 was comparable between
Downloaded from http://cid.oxfordjournals.org/ by guest on November 29, 2016
asymptomatic and symptomatic mothers (12 of 76 [15.8%] vs.
10 of 78 [12.8%], respectively).
All infants born to HIV-1-positive women were bottle-fed.
Thirty-three women with HCV infection alone breast-fed their
babies, for a mean duration of 2.4 months (range, 0.5 12
months), with no increase in the HCV transmission rate
(table 1).
A history of IVDU, prematurity, and low birth weight were
not significantly associated with HCV transmission. These pa-
rameters did not influence HIV-1 infection either (data not
shown).
Discussion
The results of this study highlight that vertical transmission
of HCV is a rare event when the mother is uninfected with
HIV-1; when she is also HIV-1-positive the risk of HCV infec-
tion of the infant increases significantly. High rates of HCV
transmission in children whose mothers had concurrent HIV-
1 infection have been reported [1, 2, 4, 7, 9, 15]. However, these
investigations were based on small sample sizes. Furthermore,
some included antibody-positive infants younger than 18
months of age as HCV-infected [7, 9, 15], and others had
an unusually large proportion of healthy HCV carriers among
seroreverting children [1, 2, 4].
The less-stringent criteria used in such studies to define HCV
infection status may account for their substantially higher trans-
mission rates in respect to that observed in the present analysis.
Passively acquired maternal anti-HCV antibody usually be-
comes undetectable by 6 12 months of age, but it may persist
longer in children born to mothers coinfected with HIV-1 [13].
In our setting, two group B children remained HCV-
seropositive up to age 17 months and then seroreverted. There-
fore, as established for HIV-1 [16], the diagnosis of HCV
infection in perinatally exposed infants cannot be based on
standard tests for IgG antibody to HCV before 18 months of
age.
cidal UC: CID
CID 1997;25 (November) Maternal-Infant HCV Transmission 1123

Table 1. Influence of maternal and infant factors on transmission of hepatitis C virus (HCV).

No. of infected children/total no. in category (%)

Variable Entire cohort Group A* Group B

Total 28/245 (11.4) 3/80 (3.7) 25/165 (15.1)

History of iv drug use
Yes 24/186 (12.9) 0/34 24/152 (15.6)
No 4/56 (7.2) 3/43 (7) 1/13 (7.7)
Clinical progression of HIV-1 infection
Asymptomatic mothers ... ... 13/78 (16.7)
Symptomatic mothers ... ... 11/76 (14.5)
Gestational age
36 w 2/38 (5.3) 0/13 2/25 (8)
37 w 26/207 (12.6) 3/67 (4.5) 23/140 (16.4)
Birth weight
2,500 g 3/51 (5.9) 0/16 3/35 (8.6)
2,500 g 25/194 (12.9) 3/64 (4.7) 22/130 (16.9)
Mode of delivery

Downloaded from http://cid.oxfordjournals.org/ by guest on November 29, 2016

Vaginal 24/172 (13.9) 2/53 (3.8) 22/119 (18.5)
Cesarean 4/71 (5.6) 1/25 (4) 3/46 (6.5)
Type of feeding
Breast-feeding ... 1/33 (3) ...
Formula-feeding ... 2/43 (4.6) ...

NOTE. Inconsistencies in total number of children for each variable are due to incomplete data.
* Children of HCV-positive, HIV-1-negative mothers.

Children of HCV-positive, HIV-1-positive mothers.

P .01.

P .06.

The presence of HCV RNA in some seroreverting children
is an intriguing phenomenon. As in other recent investigations
[5, 12, 13], we observed this phenomenon in few cases. PCR
is a very sensitive technique to detect HCV RNA in serum
samples from infected subjects, but the method has yet to be
standardized, and false-positive results cannot be ruled out. To
reduce this possibility, only children for whom results were
positive in two separate determinations were considered to be
HCV-infected.
On the other hand, not all seroreverters were tested by PCR,
and its sensitivity was not absolute in antibody-positive chil-
dren. In addition, results were not controlled in a central labora-
tory. With these limitations, the balance of evidence leads us
to conclude that the proportion of seronegative HCV carriers
among at-risk children is limited and thus should not bias the
results of this study.
In areas where the HIV-1 epidemic has spread mostly be-
cause of IVDU, many women are coinfected with HCV. In
Italy, about three-quarters of HIV-1-positive women are also
HCV-positive [13, 15]. Whereas worldwide informational cam-
paigns have been instituted for the prevention of mother-to-
child HIV-1 transmission, the risk connected with HCV has
not received adequate consideration. Our study points out that
women faced with child-bearing decisions and pregnant
women, particularly those with at-risk behaviors (e.g., IVDU),
should be encouraged to undergo antenatal testing for both
HIV-1 and HCV, and those with double infection should be
informed about the additional risk of transmitting HCV to their
offspring.
The greater HCV infectiousness of group B mothers may be
accounted for by HIV-1-induced immunosuppression leading
to impaired control of HCV, with consequent increased levels
of viremia. A direct correlation has been found between mater-
nal viral load and HCV transmission [5 7], although other
investigators failed to find this association [13, 14].
The higher the degree of immunosuppression, viral burden,
and disease progression in the mother, the greater the risk of
vertical transmission of HIV-1 [17 22]. In the present analysis,
HIV-1-related clinical manifestations were not predictive of
HCV or HIV-1 transmission. This might be due to the relatively
small number of mother/child pairs studied.
It is interesting that in women with HCV and HIV-1 coinfec-
tion, the transmission of the two viruses was similar. However,
each virus was acquired independently, attesting to the fact
that distinct mechanisms regulate their passage to the offspring.
The fact that a substantial fraction of at-risk children escaping
infection with HIV-1 are infected with HCV must be taken
into due account, particularly because perinatal HCV infection
can give rise to an indolent chronic infection with no clear
signs of hepatitis [23].
Another common feature in transmission of HCV and
HIV-1 was the lower percentage of infected children among
those delivered by cesarean section. Increasing evidence sup-
ports the association between mode of delivery and perinatal

/ 9c3f$$no51 10-17-97 21:18:40 cidal UC: CID

1124 Tovo et al.
HIV-1 infection [21, 22, 24]. For HCV, the difference was just
above the threshold of significance in the present study, while
a significant protective effect of cesarean section was observed
by Paccagnini et al. [15].
Abdominal surgical delivery can protect infants from infec-
tion in several ways, as exposure to contaminated maternal
blood during the passage through the birth canal is avoided as
well as additional risks linked to episiotomy and instrumental
delivery. Further protection may be derived from the prevention
of maternal-fetal blood transfusions during labor.
The preventive action of cesarean section suggests that a
large proportion of infants are infected around the time of
delivery. In fact, when tested for in the first days of life, HCV
RNA was undetectable in children who ultimately were shown
to be infected in this and other investigations [5, 15, 23]. No
difference in the HCV transmission rate was seen for children
delivered by elective vs. emergency cesarean section. This
could indicate that the infection occurs mainly during the pas-
sage through the birth canal, rather than at the onset of labor.
If intrapartum transmission is responsible for the majority of
perinatal HCV infections transmission then passive and/or ac-
tive immunization of newborns may be a successful preventive
intervention, analogous to hepatitis B virus.
Given the risk of HIV-1 transmission through breast-feeding,
this feeding method was adopted only for group A infants, with
no apparent increase in HCV infections. Other data support the
view that postnatal HCV transmission via maternal lactation
is unlikely [13 15]. Consequently, there is no evidence to
discourage breast-feeding by women who are HCV carriers.
Finally, it has been suggested that IVDU may facilitate the
passage of HCV to the fetus and newborn [25], but IVDU
was not associated with HCV transmission, nor was premature
delivery or low birth weight.
The Italian Study Group for HCV Infection in Children
Other participants: L. Rancilio, M.D.; A. Bucceri, M.D.;
A. Tagger, M.D. (Milan); C. Riva, M.D.; C. Scolfaro, M.D.;
E. Madon, M.D. (Turin); G. Zuin, M.D.; M. Boschetti, M.D. (Mi-
lan); L. Tessarotto, M.D. (Padua); I. Bosi, M.D. (Bologna);
G. Bossi, M.D. (Pavia); M. Stegagno, M.D.; I. Quinti, M.D.
(Rome); C. Fundaro, M.D. (Rome); M. Marcellini, M.D. (Rome);
R. Costa, M.D. (Parma); G. Busti, M.D. (Piacenza); and C. Salvi,
M.D. (Brescia).
References
1. Thaler MM, Park C-K, Landers DV, et al. Vertical transmission of hepatitis
C virus. Lancet 1991; 338:17 8.
2. Novati R, Thiers V, dArminio Monforte A, et al. Mother-to-child trans-
mission of hepatitis C virus detected by nested polymerase chain reac-
tion. J Infect Dis 1992; 165:720 3.
3. Roudot-Thoraval F, Pawlotsky J-M, Thiers V, et al. Lack of mother-to-
infant transmission of hepatitis C virus in human immunodeficiency
virus seronegative women: a prospective study with hepatitis C virus
RNA testing. Hepathology 1993; 17:772 7.
/ 9c3f$$no51 10-17-97 21:18:40
CID 1997;25 (November)
4. Kuroki T, Nishiguchi S, Fukuda K, et al. Vertical transmission of hepatitis
C virus (HCV) detected by HCV-RNA analysis. Gut 1993; 34(suppl 2):
S52 3.
5. Ohto H, Terazawa S, Sasaki N, et al. Transmission of hepatitis C virus
from mothers to infants. N Engl J Med 1994; 330:744 50.
6. Lin H-H, Kao J-H, Hsu H-Y, et al. Possible role of high-titer maternal
viremia in perinatal transmission of hepatitis C virus. J Infect Dis 1994;
169:638 41.
7. Zanetti AR, Tanzi E, Paccagnini S, et al. Mother-to-infant transmission
of hepatitis C virus. Lancet 1995; 345:289 91.
8. Zuccotti GV, Ribero ML, Giovannini M, et al. Effect of hepatitis C geno-
type on mother-to-infant transmission of virus. J Pediatr 1995; 127:
278 80.
9. Giovannini M, Tagger A, Ribero ML, et al. Maternal-infant transmission
of hepatitis C virus and HIV infections: a possible interaction [letter].
Lancet 1990; 335:1166.
10. Cilla G, Perez-Trallero E, Iturriza H, Carcedo A, Echeverria J. Maternal-
infant transmission of hepatitis C virus infection [letter]. Pediatr Infect
Dis J 1992; 11:417.
11. Perez Alvarez L, Gurbindo MD, Hernadez-Samplelayo T, et al. Mother-
Downloaded from http://cid.oxfordjournals.org/ by guest on November 29, 2016
to-infant transmission of HIV and hepatitis C infections in children born
to HIV-seropositive mothers [letter]. AIDS 1992; 6:427 8.
12. Lam JPH, McOmish F, Burns SM, Yap PL, Mok JYQ, Simmonds P.
Infrequent vertical transmission of hepatitis C virus. J Infect Dis 1993;
167:572 6.
13. Manzini P, Saracco G, Cerchier A, et al. Human immunodeficiency virus
infection as risk factor for mother-to-child hepatitis C virus transmis-
sion; persistence of anti-hepatitis C virus in children is associated with
the mothers anti-hepatitis C virus immunoblotting pattern. Hepatology
1995; 21:328 32.
14. Lin H-H, Kao J-H, HSU H-Y, et al. Absence of infection in breast-fed
infants born to hepatitis C virus infected mothers. J Pediatr 1995; 126:
589 91.
15. Paccagnini S, Principi N, Massironi E, et al. Perinatal transmission and
manifestation of hepatitis C virus infection in a high risk population.
Pediatr Infect Dis J 1995; 14:195 9.
16. Centers for Disease Control and Prevention. 1994 revised classification
system for human immunodeficiency virus infection in children less
than 13 years of age. MMWR Morb Mortal Wkly Rep 1994; 43(suppl
RR-12):1 13.
17. Gabiano C, Tovo P-A, de Martino M, et al. Mother-to-child transmission
of human immunodeficiency virus type 1: risk of infection and correlates
of transmission. Pediatrics 1992; 90:369 74.
18. European Collaborative Study. Risk factors for mother-to-child transmis-
sion of HIV-1. Lancet 1992; 339:1007 12.
19. Tibaldi C, Tovo PA, Ziarati N, et al. Asymptomatic women at high risk
of vertical HIV-1 transmission to their fetuses. Br J Obstet Gynecol
1993; 100:334 7.
20. Jackson JB, Kataaha P, Hom DL, et al. B2-microglobulin, HIV-1 p24
antibody and acid-dissociated HIV-1 p24 antigen levels: predictive
markers for vertical transmission of HIV-1 in pregnant Uganda women.
AIDS 1993; 7:1475 9.
21. European Collaborative Study. Caesarean section and risk of vertical trans-
mission of HIV-1 infection. Lancet 1994; 343:1464 7.
22. Tovo P-A, de Martino M, Gabiano C, et al. Mode of delivery and gesta-
tional age influence perinatal HIV-1 transmission. J Acquir Immune
Defic Syndr Hum Retrovirol 1996; 11:88 94.
23. Palomba E, Manzini P, Fiammengo P, Maderni P, Saracco G, Tovo P-A.
Natural history of perinatal hepatitis C virus infection. Clin Infect Dis
1996; 23:47 50.
24. Dunn DT, Newell ML, Mayaux MJ, et al. Mode of delivery and vertical
transmission of HIV-1: a review of prospective studies. J Acquir Im-
mune Defic Syndr 1994; 7:1064 6.
25. Weintrub PS, Veereman-Wauters G, Cowan MJ, Thaler MM. Hepatitis C
virus infection in infants whose mothers took street drugs intravenously.
J Pediatr 1991; 119:869 74.
cidal UC: CID