Intrafamilial Transmission of Hepatitis C in Egypt
Mostafa K. Mohamed,1,2Mohamed Abdel-Hamid,1,3Nabiel N. Mikhail,1,4Fatma Abdel-Aziz,5Ahmed Medhat,4
Laurence S. Magder,3Alan D. Fix,3and G. Thomas Strickland3
The incidence of hepatitis C (HCV) infection and associated risk factors were prospectively
to HCV (anti-HCV). Initial and follow-up sera were tested for anti-HCV by enzyme immu-
immunoassay (MEIA) and tested for HCV RNA. All follow-up serum samples converting
from negative to positive without detectable HCV-RNA were further tested by recombinant
immunoblot assay. Over an average of 1.6 years, asymptomatic anti-HCV seroconversion
Delta village (AES), where prevalence was 24% and 5 (0.8/1,000 PY) in the Upper Egypt
village (baseline prevalence of 9%). The strongest predictor of incident HCV was having an
anti-HCV–positive family member. Among those that did, incidence was 5.8/1,000 PY,
compared (P < .001) with 1.0/1,000 PY; 27 of 33 incident cases had an anti-HCV–positive
family member. Parenteral exposures increased the risk of HCV but were not statistically
significant; 67% of seroconverters were younger than 20 years of age, and the highest
incidence rate (14.1/1,000 PY) was in children younger than 10 who were living in AES
households with an anti-HCV–positive parent. In conclusion, young children would espe-
cially benefit from measures reducing exposures or preventing infection with HCV.
ies of risk factors for infection have focused on prevalent
infection, identifying historical risk factors that may no
longer contribute to HCV transmission, such as the mass
treatment campaigns for schistosomiasis with parenteral
tartar emetic3and blood transfusions.7,8Information on
he prevalence of antibodies to hepatitis C virus
(anti-HCV) in Egypt, 14% to 18% of the popu-
transmission of HCV in communities in Egypt and other
countries is scarce.
Studies of the incidence of HCV have generally been
restricted to special or high-risk populations, such as
blood donors and intravenous drug users (IVDU).2,9-11
alence, and thus a large reservoir of infection, the risk of
importance of, and opportunity for, the identification of
current risk factors for transmission. The purpose of this
report is to estimate a recent incidence of HCV infection
and to identify risk factors for such infections in 2 com-
munity-based cohorts with anti-HCV prevalence of 24%
Patients and Methods
Study Population. One half of households in villages
of Cairo, and all households in Upper Egypt, Sallam, 350
km south of Cairo, were systematically recruited in 1997
to obtain interview data and blood samples for epidemi-
ological studies of HCV. Details regarding the sampling
design, response rate, conduct of the survey, and risk fac-
Abbreviations: anti-HCV, antibodies to hepatitis C virus; EIA, enzyme immu-
polymerase chain reaction; RIBA, recombinant immunoblot assay; IR, incidence
rate; PY, person-years.
From the1National Hepatology & Tropical Medicine Research Institute;2Ain-
Shams University Faculty of Medicine, Cairo, Egypt;3University of Maryland
School of Medicine, Baltimore, Maryland;4Assiut University Faculty of Medicine,
Assiut, Egypt; and the5Center for Field and Applied Research, Qalubyia, Egypt.
Received October 5, 2004; accepted June 8, 2005.
Supported by U.S. Agency for International Development (263-G-00- 96-
00043), National Institute of Allergy and Infectious Diseases/National Institute of
Childhood Diseases (5U01A1058372-05) and Wellcome Trust-Burroughs Well-
come Fund (059113/Z/99/A & 059113/Z/99/Z) grants.
Address reprint requests to: G. Thomas Strickland, International Health Divi-
sion, Department of Epidemiology and Preventive Medicine, University of Mary-
land School of Medicine, 660 West Redwood Street, Suite 100, Baltimore, MD
21201. E-mail: firstname.lastname@example.org; fax: 410-706-8013.
Copyright © 2005 by the American Association for the Study of Liver Diseases.
Published online in Wiley InterScience (www.interscience.wiley.com).
Potential conflict of interest: Nothing to report.
tors for prevalent infection appear in previous re-
ports.4,5,7,8The subjects were asked to undergo annual
interviews and serosurveys, and herein we report results
from follow-up surveys conducted from 1998 to 2000.
survey were invited to join during subsequent surveys.
Informed consent was obtained from each subject, and
the study protocol conformed to the ethical guidelines of
the 1975 Declaration of Helsinki as reflected in a priori
approval by the Institutional Review Boards of the Egyp-
tian Ministry of Health and Population, Assiut Univer-
sity, and University of Maryland, Baltimore.
Laboratory Testing. Sera collected during the initial
survey were tested for anti-HCV by a second-generation
enzyme immunoassay (EIA) for anti-HCV immunoglob-
Delknheim, Germany) according to the manufacturer’s
instructions. Samples from subsequent years were tested
with the third-generation EIA from the same manufac-
turer. In situations in which individuals had negative ini-
were retested simultaneously with the third generation
EIA. Sera-confirmed EIA negative in the S1and positive
in the S2blood sample were again tested on the same
microtiter plates for anti-HCV using the Microparticle
Enzyme Immunoassay (MEIA, Abbott Laboratories,
Delknheim, Germany) according to the manufacturer’s
instructions. We previously reported the excellent reli-
ability of the combination of the third-generation EIA
and the MEIA.12All sera with anti-HCV were also tested
for HCV RNA by using a direct in-house reverse tran-
scriptase polymerase chain reaction (RT-PCR) generated
from the 5? untranslated region (5?UTR) of the HCV
genome,13and those S2samples negative by RT-PCR
were further tested for anti-HCV by the recombinant
immunoblot assay (RIBA; Chiron, Emeryville, CA).
Analysis. The 33 subjects who had EIA- and MEIA-
negative S1samples and EIA- and MEIA-positive S2sam-
ples and positive RT-PCR or RIBA were considered anti-
HCV incident cases. Because the intensity of HCV
exposures would differ between the 2 villages, incidence
ber of incident cases in the village by the number of per-
risk, we assumed seroconversion occurred halfway be-
tween the seroconvertor’s last negative and first positive
test. To identify risk factors for incident infection, we
compared the IR in subgroups defined based on informa-
tion collected in our comprehensive annual surveys of the
communities.4,5,7,8Confidence intervals (CI) for IR were
calculated using an exact approach based on the Poisson
distribution. To assess the statistical significance of ob-
Kleinbaum et al.14
Study Population. A seronegative cohort of 10,112
identified during the annual survey in 1997. Follow-up
was successfully performed in 6,738 (66.6%; 2,463 from
AES and 4,275 from Sallam). Participants providing S2
samples were similar to those of the 3,374 nonpartici-
pants, with the exception that the latter were more likely
to be young adult unmarried males, the group who were
more frequently absent from the villages. The subjects
were followed for an average of 1.6 years (total of 4,195
PY in AES and 6,720 PY in Sallam).
Incidence of Anti-HCV Conversion. Thirty-three
(0.5%; IR of 3.1/1,000 PY) of 6,738 subjects serocon-
verted using our stringent criteria for anti-HCV between
their S1and S2blood samples. Thirty-five among the 68
subjects who converted for anti-HCV by both third-gen-
a positive RIBA. The remaining 33 seroconverters (of
which 17 were HCV-RNA positive) included 28 cases
living in AES (anti-HCV conversion rate of 6.8/1,000
PY) and 5 cases from Sallam (anti-HCV conversion rate
of 0.8/1,000 PY; Table 1). None of these 33 seroconvert-
ing individuals were diagnosed as having acute viral hep-
atitis or gave a medical history compatible with viral
hepatitis during the interim between annual surveys. The
and women (3.2/1,000 PY; Tables 1 and 2). However, all
5 seroconvertors from Sallam were female.
Assessment of Risk Factors. Risk factors for infection
were similar in the 2 villages even though there were fewer
HCV exposures in Sallam than in AES, because of a lower
community reservoir of infection. Therefore, risk analysis
was performed on all 33 from both villages who converted
Table 1. Anti-HCV Conversion Rates per 1,000 Person-Years,
by Age, Sex, and Community
AES (Nile Delta)Sallam (Upper Egypt)
No. of Incident
No. of Incident
28 6.8 (4.5-9.9)50.8 (0.2-1.9)
684MOHAMED ET AL.HEPATOLOGY, September 2005
ever, young age was a powerful risk for HCV infection; 22
(67%) of those who seroconverted for anti-HCV were
rates for adolescents (2.8/1,000 PY; P ? .12), young adults
.20). In the Nile Delta village, AES, the IR was 12.0/1,000
in the second decade (Table 1).
Also, IR was significantly greater (P ? .001) among
those living with an anti-HCV–positive family member
(5.8/1,000 PY) than among those living in households
without an anti-HCV–positive person (1.0/1,000 PY);
27 (82%) of the 33 seroconvertors had antibody-positive
family members (Table 2). Men married to anti-HCV–
positive women (13.1/1,000 PY) had a higher incidence
(P ? .08) than men with an anti-HCV–negative wife
(1.9/1,000 PY). However, the anti-HCV status of the
husband had much less (P ? 1.00) impact on their wives’
anti-HCV IR (1.6 vs. 0.7).
Offspring of anti-HCV–positive fathers (P ? .01) and
mothers (P ? .002) were at high risk for HCV infection.
Incidence was slightly higher (IR of 8.7/1,000 PY) when
the mother was positive compared with when father had
anti-HCV (IR of 6.6/1,000 PY; Table 2). Those at great-
est risk, children younger than 10 years old living in AES
with either parent anti-HCV positive, had an IR 14.1/
Injections, reported by 41% of the sample, were the
most common potential parenteral risk for HCV expo-
sure, but were not significantly (P ? .48) associated with
seroconversion (Table 3). Only 7 men were among the
seroconvertors. None of the seroconvertors reported hav-
ing dental work. Although IR was higher in those having
wound sutures, surgery, intravenous infusions, and blood
Table 2. Assessment of Risk Factors for Incident Infection:
Demographic and Family Factors
Rate (95% CI)/
Education (those ? 20)
No school attendance
Marital status (those ? 18)
Anti-HCV-positive family member
Spousal anti-HCV status
For those with parental data available‡
Both parents negative
Father only positive
Mother only positive
Both parents positive
33/10,618 3.1 (2.2-4.3)
*Among married men. †Among married women.
‡For those living at home with their parents.
§Comparison between incidence rate in that row and incidence rate in the
reference category for that variable.
Table 3. Assessment of Risk Factors for Incident Infection:
Rate (95% CI)/1,000
Invasive hospital procedures*
All parenteral exposures†
*Invasive hospital procedures include wound sutures, surgery, intravenous
catheters, and blood transfusions.
†Parenteral risk factors include all invasive hospital procedures, dental work,
‡Comparison between incidence rate in that row and incidence rate in the
reference category for that variable.
HEPATOLOGY, Vol. 42, No. 3, 2005MOHAMED ET AL.685
significant risks (Table 3). Collective summarization of
invasive hospital procedures did not increase the risk of
having anti-HCV (P ? 1.00). Twenty-one seroconver-
tors had none of the usual parenteral risks for HCV, and
least 1 potential parenteral exposure.
During our cross-sectional surveys of these 2 villages,
we reported a prevalence of 63% and 65%, respectively,
of HCV-RNA in sera from subjects who were anti-HCV
positive.4,5This, and the lower proportion of RNA in our
centage of viremias in those having anti-HCV are
generally based on patients diagnosed with chronic
HCV.9Our results are from asymptomatic mostly young
subjects living in a community in which human immuno-
deficiency virus (HIV), a coinfection that favors HCV
persistence, is not present. Obviously, a higher propor-
cleared their HCV infection and become PCR negative
than patients who have sought medical care. Only ap-
women receiving HCV-contaminated Rh immune glob-
ulin in late pregnancy or early post-partum15,16and chil-
dren infected from blood transfusions during cardiac
surgery17had HCV-RNA. Similar to our incident cases,
young at the time of initial infection.
We believe HCV exposures in rural Egyptian commu-
nities are usually much less intense than those in individ-
uals infected by contaminated blood, either from
transfusion of blood or a blood product, or from abuse of
intravenous drugs. Exposures in (mostly) children to
small amounts of virus also may not result in persistent
seroconversion. Other data supporting this are included
in a recent publication in which we reported that 13
(18.3%) of 71 children who did not have anti-HCV or
HCV-RNA, and who were living with 2 or more HCV-
infected persons, had cellular immune responses to spe-
cific HCV peptides, a marker of exposure to the virus.18
A few community-based studies of HCV incidence
have been performed in Japan and Taiwan.19-23These
which included over 3,500 participants. These commu-
nities had a relatively high prevalence of HCV. Although
not uniformly presented in person-time, the IR in the
Japanese studies were 3 to 4 per 1,000 PY,19-21and in
Taiwan the estimates were 11.1 and 45 per 1,000 PY,22,23
with the higher rate estimate based on antibody conver-
sion in only 89 seronegative individuals.
Most community studies in rural Japan and Taiwan
either did not assess risk factors for infection21,22or failed
to identify specific risk factors.20One study identified
not testing for statistical significance, identified intercur-
rent medical exposures among all seroconvertors.21
Among those with incident infection included in the
analysis of risk factors, invasive hospital risks were rare.
We, like Okayama et al.,20noted that fewer than half our
subjects with incident infection had exposures known to
transmit blood-borne infections.
frequent injections, usually for health purposes and often
given at home by “injectionists” who sometimes reuse
syringes and needles in more than 1 person.7,8Dental
procedures, wound sutures, surgery, and intravenous in-
fusions were uncommon; blood transfusions, endoscopy,
and renal dialysis were very rare. Even though our sample
size is probably the largest for evaluating risk factors for
incident HCV infection in communities, exposure-spe-
cific seroconversion rates are based on small numbers of
incident cases, making achieving statistical significance
from any of these risks difficult. Although not statistically
significant, injections, wound sutures, surgery, and intra-
venous infusions all increased the chance of being anti-
HCV positive. Therefore, the absence of statistical
significance does not rule out injections and other paren-
teral exposures as a means of infecting some of the 2,612
subjects who reported receiving 1 or more injection dur-
ing the follow-up.
Having an anti-HCV–positive infected spouse in-
creased the risk of HCV infection. Men with anti-HCV–
with antibody-negative wives. However, the reverse was
less apparent; women with anti-HCV–positive husbands
antibody-negative husbands. Rao et al.24hypothesized
primarily in the 1960s and 1970s during mass treatment
campaigns for schistosomiasis with multiple intravenous
injections of tartar emetic. This practice certainly could
have caused multiple HCV infections among households
participating in these treatment campaigns. However, it
would not impact on incident infections in this report
other than place younger members of those households
having high HCV reservoirs of infection, that is, anti-
HCV–positive parents, at greater risk. For instance, 13 of
16 incident infections in children in whom anti-HCV
data on their parents are available had at least 1 anti-
HCV–positive parent. Children from the Nile Delta vil-
lage, AES, with a parent who had anti-HCV were at
686MOHAMED ET AL.HEPATOLOGY, September 2005
greatest risk of incident HCV infection, with an IR of Download full-text
14.1 per 1,000 PY.
a child to the presence of anti-HCV in their parents sug-
gests transmission of HCV is occurring between family
members but does not define the exact routes. This could
exposures to infectious blood or saliva, or by more tradi-
tional routes, such as sharing a needle within the house-
hold. However, as hypothesized by Rao et al.,24many
infections in the past were probably from external sources
in which more than 1 family member was infected at the
same time, such as during therapeutic injections. Al-
of exposures still occur and could be another explanation
for intrafamilial clustering. Learning the mechanisms by
which HCV transmission is occurring between family
members so that preventive measures can be initiated,
particularly in children having HCV-infected parents, is
for Field and Applied Research’s and Assiut University’s
community survey teams assisted in collecting data and
samples. The Egyptian Viral Hepatitis Research Labora-
Ismail, Director of the Hepatology and Tropical Medi-
siut University Faculty of Medicine, and Drs. Magda
Rakhe, Nasr El Sayed and others formerly and currently
in the Egyptian Ministry of Health and Population gave
their support. Dr. Robert Purcell of the National Insti-
tutes of Health provided invaluable advice. Careful
project management in Cairo of Mar-Jan Ostrowski and
Members of the Egyptian Center
1. Abdel-Wahab MF, Zakaria S, Kamel M, Abdel-Khaliq MK, Mabrouk
MA, Salama H, et al. High seroprevalence of hepatitis C infection among
risk groups in Egypt. Am J Trop Med Hyg 1994;51:563-567.
2. Arthur RR, Hassan NF, Abdallah MY, el-Sharkawy MS, Saad MD, Hack-
governorates in Egypt. Trans R Soc Trop Med Hyg 1997;91:271-274.
LS, et al. The role of parenteral antischistosomal therapy in the spread of
hepatitis C virus in Egypt. Lancet 2000;355:887-891.
Delta: population description and HCV prevalence. HEPATOLOGY 2000;
5. Nafeh MA, Medhat A, Shehata M, Mikhail NN, Swifee Y, Abdel-Hamid
M, et al. Hepatitis C in a community in Upper Egypt: I. Cross-sectional
survey. Am J Trop Med Hyg 2000;63:236-241.
6. Darwish MA, Faris R, Darwish N, Shouman A, Gadallah M, El-
Sharkawy MS, et al. Hepatitis C and cirrhotic liver disease in the Nile
delta of Egypt: a community-based study. Am J Trop Med Hyg 2001;
7. Habib M, Mohamed MK, Abdel-Aziz F, Magder LS, Abdel-Hamid M,
Gamil F, et al. Hepatitis C virus infection in a community in the Nile
Delta: risk factors for seropositivity. HEPATOLOGY 2001; 33:248-253.
al. Hepatitis C in a community in Upper Egypt: risk factors for infection.
Am J Trop Med Hyg 2002;66:633-638.
9. Villano SA, Vlahov D, Nelson KE, Lyles CM, Cohn S, Thomas DL.
Incidence and risk factors for hepatitis C among injection drug users in
Baltimore, Maryland. J Clin Microbiol 1997;35:3274-3277.
The incidence and risk factors of community-acquired hepatitis C in a
cohort of Italian blood donors. HEPATOLOGY 1997; 25:702-704.
11. Alter MJ. Epidemiology of hepatitis C. HEPATOLOGY 1997;26(Suppl 1):
12. Abdel-Hamid M, El-Daly M, El-Kafrawy S, Mikhail N, Strickland GT,
Fix AD. Comparison of second- and third-generation enzyme immunoas-
says for detecting antibodies to hepatitis C virus. J Clin Microbiol 2002;
13. Abdel-Hamid M, Edelman DC, Highsmith WE, Constantine NT. Opti-
tion-polymerase chain reaction protocol for the detection of hepatitis C
virus. J Hum Virol 1997;1:58-65.
14. Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic Research:
Principles and Quantitative Methods. Belmont, CA: Lifetime Learning
15. Fanning LJ, Levis J, Kenny-Walsh E, Wynne F, Whelton M, Shanahan F.
Viral clearance in hepatitis C (1b) infection: relationship with human
leukocyte antigen class II in a homogeneous population. HEPATOLOGY
in a hepatitis C (genotype 1b) single-source outbreak in Germany: a 20-
year multicenter study. HEPATOLOGY 2000;32:91-96.
and clinical outcome of hepatitis C infection in children who underwent
cardiac surgery before the implementation of blood-donor screening.
N Engl J Med 1999;341:866-870.
18. Al-Shirbiny M, Osman A, Mohamed N, Shata MT, Abdel-Aziz, F, Abdel-
Hamid M, et al. Exposure to hepatitis C virus induces cellular immune
responses without detectable viremia or seroconversion. Am J Trop Med
Hyg 2005; in press.
19. Sasaki F, Tanaka J, Moriya T, Katayama K, Hiraoka M, Ohishi K, et al.
Very low incidence rates of community-acquired hepatitis C virus infec-
tion in company employees, long-term inpatients, and blood donors in
Japan. J Epidemiol 1996;6:198-203.
20. Okayama A, Stuver SO, Tabor E, Tachibana N, Kohara M, Mueller NE,
in Japan. J Viral Hepatol 2002;9:43-51 [Japan]. Transfusion 1997;37:
21. Fukuizumi K, Sata M, Suzuki H, Nakano H, Tanikawa K. Hepatitis C
virus seroconversion rate in a hyperendemic area of HCV in Japan: a
prospective study. Scand J Infect Dis 1997;29:345-347.
of hepatitis C infection in a hepatitis C endemic township in southern
Taiwan. Kaohsiung J Med Sci 1997;13:605-608.
23. Sun CA, Chen HC, Lu SN, Chen CJ, Lu CF, You SL, et al. Persistent
hyperendemicity of hepatitis C virus infection in Taiwan: the important
role of iatrogenic risk factors. J Med Virol 2001;65:30-34.
JD, et al. Further evidence for association of hepatitis C infection with
HEPATOLOGY, Vol. 42, No. 3, 2005MOHAMED ET AL. 687