A different impact of HIV-1 infection according to gender is now being debated. Differences in the progression of HIV-1 infection in adults have been reported by some1,2 but are denied by other authors.3,4 These results may be biased by gender-related differences in compliance to therapy, behaviors, and lifestyle. These biases are less evident or absent in children, but only one study in perinatally infected children by the European Collaborative Study (ECS) is available showing that gender does not influence disease progression.5
To a lesser extent it is not known whether differences by gender exist in acquiring HIV-1 perinatal infection, because most published studies have not evaluated gender as a risk factor. Our preliminary findings6 and a subsequent study by Temmerman et al7 suggested that female gender was a significant risk factor for perinatal infection. Up to now, conflicting results have been reported. A paper by the ECS8 confirms the role of gender on mother-to-child HIV-1 transmission (MTCT) and shows that a larger proportion of girls are infected when born by elective cesarean section, thus suggesting that the increased risk for girls is limited to intrauterine transmission. Conversely, an individual patient data meta-analysis in breast-fed children has shown no differences in the proportion of prenatally infected children by gender, whereas a larger proportion of boys was infected postnatally.9 Recently, a large prospective study conducted in Africa showed an increased rate of transmission in girls both before and after birth.10
To provide further information on this intriguing issue, we evaluated the data emerging from the large prospective cohort of children born to HIV-1-positive mothers included in the Italian Register for HIV Infection in Children.
The Italian Register for HIV Infection in Children, instituted in 1985, is a nationwide multicenter study of children perinatally exposed to HIV-1. The Register's data come from a network of 106 participating pediatric centers located throughout Italy that aim to enroll all infants born to HIV-1-infected mothers and are representative of the overall population of exposed infants in Italy.11,12 All exposed or infected children are enrolled, including those not prospectively followed from birth, but only those identified at birth are included in studies on risk factors for MTCT.13
Data regarding mother-infant pairs were collected as previously described.11,12,14 In particular, baseline information was collected on the infant's demographics, age at first observation, mode of delivery, gestational age, and type of feeding. Starting in 1989, information on antiretroviral treatments during pregnancy was collected and initially coded as treated or not treated; from 1993 onward, data on type of therapy and gestational age at beginning and end of therapy were also included. Before the publication of the Pediatric AIDS Clinical Trials Group (PACTG) 076 protocol,15 zidovudine had been administered to pregnant women in selected centers in our country, for their own health, rather than for prophylaxis of MTCT.11 Thereafter, the 076 prophylactic protocol was usually followed, though not universally or simultaneously.11
Maternal viral load and CD4+ cell counts at the time of delivery are known in a few cases and exclusively after 1996.
Children were classified as infected, uninfected, or with indeterminate infection status as previously described.11 Only children identified at birth11 were considered for the present study. Infection is diagnosed by the persistence of HIV-1 antibodies after 18 months or, before 18 months, by detection of viral markers on at least 2 occasions.17 Proviral DNA or viral cultures done at birth are not used to assign infection status. Infants whose infection status is indeterminate are followed up at least until infection status can be ascertained. We defined children with indeterminate infection status as those who were either lost to follow-up, died prior to HIV-1 diagnosis, or had indeterminate HIV-1 status (because they were <18 months old and had not seroreverted and detection of viral markers was performed only once) when the study was closed. Elective cesarean section was defined as previously described.14 We defined as treated those mothers who had received an antiretroviral regimen in pregnancy for at least 1 month up till delivery. Moreover, to evaluate the effect on gender of type of antiretroviral regimen, we defined as monotherapy zidovudine included in the PACTG 076 protocol and as combined regimen those including ≥2 antiretroviral drugs.
Analyses were performed on data reported up to June 2002 on children born to HIV-1-positive mothers between June 1, 1985, and December 31, 2001. Children born between January 1, 2002, and June 30, 2002 were excluded from analysis because of the large percentage (137/312; 43.9%) of exposed children with undetermined status. Differences in proportions were evaluated by χ2 test.
To evaluate the association of gender with MTCT, we performed logistic regression models to estimate crude odds ratio (OR) and adjusted odds ratios (AOR) for factors potentially influencing MTCT. Analysis were separately conducted for the period 1985-1995 and the period 1996-2001, because up to 1996, the PACTG 076 protocol had not yet been widely introduced in Italy.11 The factors taken into consideration were infant gender, gestational age (≤36 weeks, 37-39 weeks, ≥40 weeks, unknown), mother's clinical condition at delivery (defined according to the Centers for Disease Control classification system for HIV-1-infected adults,16 as A vs. B/C vs. unknown), mode of delivery (ie, elective cesarean section vs. elective cesarean/unknown mode of delivery), and antiretroviral regimen received by the mother during pregnancy (ie, none vs. zidovudine monotherapy or combined maternal therapy).
To further evaluate whether the gender effect was present only in some specific subgroups, separate logistic regression models by mode of delivery (ie, elective cesarean and other type of delivery) and by maternal antiretroviral regimen during pregnancy (ie, no antiretroviral therapy, zidovudine monotherapy, combination therapy) were performed. Models were carried out joining maternal double and triple or more regimens because preliminary analyses showed no differences. Twins and children with the same mother were treated as independent, as preliminary analysis had shown that excluding them from the analysis did not change the results. Because of the small proportion of breast-fed children after 1995, these were not included in the stratified analyses. Similarly, because of the large percentage of missing data regarding viral load and CD4+ T-lymphocyte counts at time of delivery, these variables were not considered.
Infant's Gender and Mother-to-Child Transmission
Characteristics of mothers and children are reported in Table 1. Antiretroviral prophylaxis in the mothers was administered for a median time of 6 (range: 1-9) months. The total number of children born to HIV-1-positive mothers was 6526 children. Among these, 1324 (628/3400; 18.5%) boys and 696/3126 (22.3%) girls were infected. No differences in the characteristics of mothers and children, apart from children's infection status, were present.
A total of 4151 children (2166 boys and 1985 girls) were recorded. The overall transmission rate was 478/4151 (11.5%, 95% CI: 10.6-12.5). A total of 346 children were in an indeterminate infection status (180 boys and 166 girls). The proportion of children with indeterminate infection status who died prior to diagnosis was 9.8% (34/346). Twenty of 180 (11.1%) male and 14 of 166 (8.4%) female infants died prior to a definitive HIV-1 diagnosis, without a statistically significant difference (χ2 = 0.429, P = 0.512). The overall MTCT rate according to gender in children prospectively derived was 222/2166 (10.2%, 95% CI: 9.0-11.6) infected male and 256/1985 (12.9%, 95% CI: 11.5-14.5) female infants, with a significant difference between the 2 groups (P = 0.008).
Among children born in 1985-1995, the overall MTCT rate was 396/2374 (16.7%, 95% CI: 15.2-18.2); 194/1248 boys (15.5%, 95% CI: 13.6-17.7) and 202/1126 girls (17.9%, 95% CI: 15.7-20.3) were infected, and the difference was not significant (P = 0.1181).
After 1995, the overall MTCT rate was 82/1777 (4.6%, 95% CI: 3.7-5.7) and the proportion of infected children was significantly lower among male (28/918; 3.1%, 95% CI: 2.0-4.4) than among female infants (54/859; 6.3%; 95% CI: 4.8-8.1) (P = 0.002).
Infant's Gender and Mother-to-Child Transmission According to Type of Delivery and Maternal Antiretroviral Prophylaxis
Results of univariate and logistic regression analyses for all risk factors in the 2 periods considered are given in Table 2. In 1985-1995 the lower risk in boys was not statistically significant. For the period 1996-2001, at univariate analysis male infants were significantly less likely to be infected, and this finding persisted after having adjusted for other factors associated with changed MTCT. As expected,11 elective cesarean delivery (both in 1985-1995 and in 1996-2001) and maternal antiretroviral prophylaxis (in 1996-2001) were associated with a significant reduction in MTCT.
Separate analyses, performed according to mode of delivery and maternal antiretroviral regimen (Table 3), showed no significant difference in the MTCT rate in boys and girls in the period 1985-1995, even if a tendency to a lower risk of MTCT in boys was found in those born to mothers who received antiretroviral treatment with zidovudine. In the period 1996-2001, the estimated OR and AOR showed that the protective effect of male gender for MTCT persisted independently of type of delivery.
When separate analyses were performed by antiretroviral treatment administered to the mothers, a lower risk of infection in male vs. female infants was found when no antiretroviral prophylaxis was done. The estimated risk was even lower when the mother received zidovudine monotherapy during pregnancy. In contrast, no significant difference between genders was found when combined antiretroviral regimens were given to the mothers. However, analysis showed very large 95% CIs and the results are likely to be biased by the very low number of infected children in this group. Due to the strong overlapping among children born to treated mothers and those born by cesarean section, a further stratification combining maternal antiretroviral treatment and mode of delivery was performed (Table 4). As shown, male infants were at decreased risk of MTCT when the mother received zidovudine monotherapy, both in children born by elective cesarean section and in those born by other mode of delivery. In children born to mothers treated with combination regimens, the protective effect of male gender was not found either for cesarean section or for other-than-cesarean delivery, although the estimated OR and AOR were not large, due to the few numbers in these strata.
We observed a lower proportion of HIV-1-infected male infants compared with female infants in children born after 1995. The male/female ratios in infected children were 0.96 in those born in 1985-1995 and 0.52 in those born in 1996-2001. Ratios in infants born to HIV-1-infected mothers in 1985-1995 and 1996-2001 were 1.10 and 1.06, respectively. The ratios in overall children born in Italy during these periods were 1.10 and 1.06, respectively.18 A similar male/female ratio was observed in infants with indeterminate infection status born to mothers either receiving or not receiving antiretroviral treatment during pregnancy. Thus the different ratio in HIV-1-infected infants was unlikely to be due to a different male/female birth rate in HIV-1-infected mothers.
Our findings agree only in part with those from the ECS.8 Both studies found a lower MTCT in male infants, but the ECS concluded that gender differences were limited to children born by elective cesarean section (because of uncertain results concerning antiretroviral treatment), suggesting that an increased proportion of girls may be infected in utero.8 Recently, a large prospective study10 done in Africa in infants born to untreated mothers and enrolled at birth found similar results, showing that female gender had a significantly higher risk of prenatal infection. Taha et al10 have already discussed why several studies did not find differences in perinatal transmission according to infant's gender. Only recently has infant's gender been considered as an important biologic factor for MTCT to be included in multivariate analysis.
Our large data set and, consequently, the large number of infected children gave us the possibility of a more detailed analysis. First, we found that the effect of gender was present only in children born after 1995. Consequently, we investigated whether preventive interventions, such as elective cesarean section or maternal antiretroviral regimens in pregnancy (mainly introduced in Italy after this date), may be related to such an effect. Because of the strong overlapping of these 2 variables, the analyses were also stratified in children according to each combination of these variables. We found that the risk of infection in male was lower than in female infants independently of type of delivery and administration of antiretroviral regimen to the mother, even if this effect was not evident when combined regimens were administered, probably because there were very few infected children in this group.
Thus some factor(s) intrinsic to gender (rather than type of interventions) may be involved, but any explanation is speculative. The first hypothesis is that girls are more susceptible to acquiring perinatal HIV-1 infection. In another model of retrovirus MTCT, female infants acquire the human T-cell leukemia/lymphoma virus type I with a frequency that is significantly higher than that of male infants.19 In a different model of mother-to-infant virus transmission, Granovsky et al20 have shown that girls acquire hepatitis C virus infection about 3 times more frequently than boys.
The strong effect of gender as a risk factor in children born after 1995 is, however, an unexplained finding. Because both elective cesarean and antiretroviral prophylaxis prevent MTCT in the perinatal period, it is possible that, among infected fetuses, more girls survive up to this period and may take advantage of the benefits of preventive strategies. In general, boys are more vulnerable than girls to infectious agents19-21 and male fetuses have a higher rate of spontaneous abortion22 and postnatal mortality.23 Thus, a second hypothesis is a higher in utero mortality of infected boys. This hypothesis is biologically plausible, because experimental studies showed that relationships between hormones24,25 and immune mechanisms26,27 mediate a different susceptibility to intrauterine infections between boys and girls. Our data, moreover, showed a trend to a higher mortality in male infants with indeterminate infectious status, but differences were not significant and, in addition, deaths were reported to the register as non-HIV related. To explore this issue other analyses were performed assuming that those dead with indeterminate infectious status were HIV infected. However, results did not change.
It has been suggested that a prospective study monitoring reproductive losses among HIV-infected women could verify the hypothesis of an increased mortality in utero among male fetuses.10 However, such a study could be difficult to perform due to biologic and ethical difficulties in evaluating the infection status in fetuses. By contrast, a large meta-analysis study could assess the role of gender as a risk factor for MTCT.
1. Napravnik S, Poole C, Thomas JC, et al. Gender difference in HIV RNA levels: a meta-analysis of published studies. J Acquir Immune Defic Syndr
2. Gilad J, Walfisch A, Borer A, et al. Gender differences and sex-specific manifestations associated with human immunodeficiency virus infection in women. Eur J Obstet Gynecol Reprod Biol
3. Vanhems P, Routy JP, Hirschel B, et al. Clinical features of acute retroviral syndrome differ by route of infection but not by gender and age. J Acquir Immune Defic Syndr
4. Moore AL, Kirk O, Johnson AM, et al. Virologic, immunologic, and clinical response to highly active antiretroviral therapy: the gender issue revisited. J Acquir Immune Defic Syndr
5. European Collaborative Study. Gender and race do not alter early-life determinants of clinical disease progression in HIV-1 vertically infected children. AIDS
6. Gabiano C, Tovo P-A, de Martino M, et al. Mother-to-child HIV-1 transmission: risk of infection and correlates of transmission. Pediatrics
7. Temmerman M, Nyong'o AO, Byayo J, et al. Risk factors for mother-to-child transmission of human immunodeficiency virus-1 infection. Am J Obstet Gynecol
8. European Collaborative Study. Are girls more at risk of intrauterine-acquired HIV infection than boys? AIDS
9. The Breastfeeding and HIV International Transmission Study Group. Late postnatal transmission of HIV-1 in breast-fed children: an individual patient data meta-analysis. J Infect Dis
10. Taha TE, Nour S, Kumwenda NI, et al. Gender differences in perinatal HIV acquisition among African infants. Pediatrics
11. The Italian Register for HIV Infection in Children. Determinants of mother-to-infant HIV-1 transmission before and after the introduction of zidovudine prophylaxis. Arch Pediatr Adolesc Med
12. de Martino M, Tovo P-A, Balducci M, et al. Reduction in mortality with availability of antiretroviral therapy for children with perinatal HIV-1 infection. JAMA
13. Lindegren ML, Byers RH Jr, Thomas P, et al. Trends in perinatal transmission of HIV/AIDS in the United States. JAMA
14. Tovo P-A, de Martino M, Gabiano C, et al. Mode of delivery and gestational age influence perinatal HIV-1 transmission. Italian Register for HIV Infection in Children. J Acquir Immune Defic Syndr Hum Retrovirol
15. Connor EM, Sperling RS, Gelber R, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. Pediatric AIDS Clinical Trials Group Protocol 076 Study Group. N Engl J Med
16. Centers for Disease Control and Prevention. 1993 Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Morb Mortal Wkly Rep
17. 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
18. Istituto Nazionale di Statistica. Bollettino Mensile di Statistica
. Roma: Gennaio; 2001.
19. Ureta-Vidal A, Angelin-Duclos C, Tortevoye P, et al. Mother-to-child transmission of human T-cell-leukemia/lymphoma virus type I: implication of high antiviral antibody titer and high proviral load in carrier mothers. Int J Cancer
20. Granovsky MO, Minkoff HL, Tess BH, et al. Hepatitis C virus infection in the Mothers and Infants Cohort Study. Pediatrics
21. Han X, Lundberg P, Tanamachi B, et al. Gender influences herpes simplex virus type 1 infection in normal and gamma interferon-mutant mice. J Virol
22. Byrne J, Warburton D. Male excess among anatomically normal fetuses in spontaneous abortions. Am J Med Genet
23. Scheinfeld A. The mortality of men and women. Sci Am
24. Klein SL. Hormones and mating system affect sex species differences in immune function among vertebrates. Behav Processes
25. Falkenberg ER, Davis RO, DuBard M, et al. Effects of maternal infections on fetal adrenal steroid production. Endocr Res
26. Hofman-Lehmann R, Holznagel E, Lutz H. Female cats have lower rates of apoptosis in peripheral blood lymphocytes than male cats: correlation with estradiol-17beta, but not with progesterone blood levels. Vet Immunol Immunopathol
27. Romero R, Gomez R, Galasso M, et al. The natural interleukin-1 receptor antagonist in the fetal, maternal, and amniotic fluid compartments: the effect of gestational age, fetal gender and intrauterine infection. Am J Obstet Gynecol
Participants in the Italian Register for HIV Infection in Children: P. Osimani (Ancona); D. De Mattia, C. Di Bari, M. Manzionna (Bari); A. Cornelli, M. Ruggeri (Bergamo); F. Baldi, I. Bosi, M. Ciccia, A. Faldella, M. Lanari, M. Martelli, M. Masi (Bologna); M. Duse, R. Fausti, R. F. Schumacher (Brescia); G. Chiriacò (Brindisi); M. Dedoni, C. DessÌ, D. Gariel (Cagliari); E. Anastasio (Catanzaro); A. Grassi, M. Sticca (Como); A. Vivalda (Cuneo); T. Bezzi, E. Fiumana (Ferrara); A. Placido, P. Vitucci, S. Collini (Florence); A. Bigi, M.T. Cecchi (Forl); C. Gotta, G. di Siena, D. Lobello, D. Cosso, A. Timitilli (Genoa); P. Mussini (Mantova); G. Ferraris, C. De Giacomo, V. Giacomet, M. Giovannini, R. Lipreri, R. Pinzani, A. Plebani, F. Salvini, L. Schneider, A. Viganò, G.V. Zuccotti (Milan); C. Cano, M. Cellini (Modena); W. Buffolano, A. Guarino, L. Tarallo (Naples); R. D'Elia, C. Giaquinto, O. Rampon (Padua); E. R. Dalle Nogare, A. Romano, M. Saitta (Palermo); M. Bergamasco, A. Maccabruni (Pavia); R. Consolini, G. Palla (Pisa); A. Barone, I. Dodi (Parma); A. Rossi (Reggio Emilia); G. Anzidei, C. Ayassa, G. Bove, A. M. Casadei Pistilli, S. Bernardi, G. Castelli Gattinara, S. Cerilli, R. Chiodi, C. Facente, C. Fundarò, A. Krzysztofiak, P. Palma (Rome); A. Ganau (Sassari); L. Cristiano (Taranto); S. Garetto, C. Riva, C. Scolfaro (Turin); V. Portelli (Trapani); A. Mazza (Trento); L. Lepore, M. Rabusin (Trieste); A. Pellegatta (Varese); and P. Fortunati (Verona).