Perinatal transmission may occur: (1) directly, during the passage of the fetus through the birth canal and on coming into contact with infected maternal secretions[
13,
18]; in delivery by cesarean section by ascending infection from the vaginal canal, after a premature rupture of the amniotic membranes [
35]; in managing the mother with the baby (changing nappies, bathing)[
10]; (2) indirectly, during vaginal delivery from contaminated objects; and (3) intrauterine transmission at the time of fertilization from sperm carrying latent HPV[
36]; ascending infection from secretions of the maternal genital tract; and transplacental[
11,
19].
HPV DNA in pregnant women
HPV DNA was detected in 49 pregnant women (77.8%, n = 49/63). The percentage found was considered high compared to the existing literature. However, given the origin of the population studied, from outpatient clinics dealing with prenatal examinations and infectious diseases, these figures were already expected. The data regarding the prevalence of HPV infection in pregnancy are highly discordant: 5.4% reported by Tenti et al. (1997)[
37] and 68.8% mentioned by Cason et al. (1995)[
15]. The diversity of percentages observed is related to different factors that by themselves could influence the results, such as: diagnostic techniques, the characteristics of the samples and the inclusion criteria. Eppel et al. (2000)[
16] observed a 24.6% prevalence of HPV infection in the uterine cervix of pregnant women. Recently, Takakuwa et al. (2006)[
38], examining the cervical smears of 1.183 pregnant women for HPV DNA using the PCR-RFLP methods, observed a prevalence of 22.6% in pregnant women aged less than 25 years. This percentage was statistically significant (p < 0.0005) compared to the percentage obtained in pregnant women over the age of 25 years (11.3%), and it was concluded that the prevalence of HPV is considered high in young Japanese pregnant women.
Studying the type of lesion produced by HPV in the maternal genitalia, it was observed that 57.1% had genital warts, 24.5% low grade cervical intraepithelial lesions, and 18.4% high grade cervical intraepithelial lesions, results which could suggest a higher percentage of HPV DNA considered a low carcinogenic risk, which, however, was not observed. Of the HPV DNA types detected 54.9%, 1.2% and 40.2% were viral types considered a high carcinogenic risk, possible high risk and low risk, respectively. Genital infections produced by two or more types of HPV DNA were identified in 61.2% of the cases. Lu et al. (2003)[
39] studying the prevalence and viral type in pregnant women with a diagnosis of squamous atypias of the uterine cervix detected HPV DNA in 88.6% of the cases. Of the HPV positive cases, 79.6%, 4.3% and 5.4% were considered a high carcinogenic risk, probable high risk and low risk, respectively. The most frequent viral types detected were 52 (31.2%), 16 (15.1%), 39 (11.8%), 53 (10.8%), and 18 and 58 (9.7% each). Viral infection by multiple types was detected in 43% of the cases. Hernandez-Giron et al. (2005)[
40], in a population study in México detected high carcinogenic risk HPV DNA in 37.2% of 274 pregnant women and 14.2% of 1,060 non-pregnant women.
Infections by multiple types of HPV are considered relatively common among the population in general[
41]. Thomas et al. (2000)[
42] reported that infection by multiple types of HPV are acquired more frequently than expected. These authors suggested that populations with a specific sexual behavior of exposing themselves to an ensemble of different types of HPV, or else the preexistence of a type of HPV could make it easier to acquire a new type of virus through an as yet unknown mechanism. Other authors[
43] disagreed with the above statements and suggested that the risk factors are the same, both to acquire a single infection or a multiple one for HPV. A few authors suggested several hypotheses to account for the high rates of HPV infection observed in pregnant women, such as the immunosuppressive and hormonal states induced by pregnancy[
40,
44]. These hypotheses could also explain the rate of multiple HPV infection (61.2%) observed in this study.
HPV DNA in the placenta
The use of different methods to sample the placentas was determinant for a more accurate identification of HPV DNA in third trimester pregnancy placentas (24.5%, n = 12/49). The results show that the isolated use of scraping methods or biopsies, especially if applied only to one of the sides of the placental disk, would detect a smaller number than the total obtained in this study.
The viral types identified in the placentas were 6/11 (50%, n = 6/12), 16 (25%, n = 3/12), 18 (16.7%, n = 2/12), 42, 52 and 58 (8.3%, n = 1/12 – each). The HPV DNA identified in the placentas were 6/11, 16, 18, 52 and 58. Seven placentas (58.3%, n = 7/12) presented HPV considered a high carcinogenic risk (types 16, 18, 52 and 58) and two (16.7%, n = 2/12) presented two different types of HPV DNA. The presence of HPV DNA in the placenta indicates the possibility of transplacental exposure to viral infection and to the need of considering the possible consequences of this exposure during the period: (1) intrauterine, to miscarriages[
45] and possible malformations[
16]; (2) postnatal period to genital warts in childhood[
46], in adolescence juvenile-onset recurrent papillomatosis[
5]; (3) in lifetime, the possible transmission of the carcinogenic agent[
47,
48].
In addition, the concordance observed in the type specific HPV between the genital/placental samples (91.7%, n = 11/12), strongly suggests that the HPV DNA detected in the placenta comes from maternal viral infection. This placental infection could be the result of an ascending canalicular infection from genital secretions (transamniotic) or hematogenic. The difference found in the types of HPV DNA may be due to different causes, such as contamination of the samples (unlikely), infection from the semen at the time of fertilization, infection due to multiple types, or subtypes and/or variants of HPV. Eppel et al. (2000)[
16] in their study on HPV DNA detection in placentas, did not identify them in any of the 147 samples of chorionic vilosity collected by transabdominal amniocentesis. Even so, the authors suggested the possibility of transplacental viral transmission.
HPV DNA in newborns
As seen in the evaluation of methods to sample the placenta, the use of different sampling methods in the NB was determinant for a more precise identification of the percentage of vertical transmission of HPV (22.4%, n = 11/49). The results show that the isolated use of oral and bodily cavity scraping methods, or nasopharyngeal aspirates, or arterial cord blood, if applied individually for clinical screening would detect a smaller number of NB HPV DNA+ than the total obtained in this study.
The viral types identified in the NB were 6/11 (45.5%, n = 5/11), 42 (18.2%, n = 2/11), 52 (18.2%, n = 2/11), 18 and 59 (9.1%, n = 1/11 – each). Genital warts, which are caused by HPV types 6 or 11, are considered a frequent complication in pregnancy and clinically important due to the possibility of vertical transmission. Armstrong et al. (2000)[
49] considered juvenile recurrent respiratory papillomatosis a consequence of vertical transmission of HPV. However, the risk of developing this complication in a child born to a mother infected with HPV is one to several hundred exposures[
50]. Smith et al. (1995)[
51] showed a rate of only 1% of vertical transmission of HPV DNA. Other authors reported higher percentages, using the PCR methodology for HPV type 16 and 18 in genital scrapings and oral cavity of mother/NB pairs, respectively, and detected vertical transmission rates between 31% and 73%[
13,
15,
18,
52,
53].
Four NB (36.4%, n = 4/11) presented viral types considered a high carcinogenic risk (types 18, 52 and 59) and one presented two different types of HPV DNA (types 6/11 and 52). These data are sufficient evidence to confirm the perinatal transmission of HPV, considered a high carcinogenic risk. These findings require future studies to be able to establish: (1) the significance and consequences of infection in the child; (2) their relationship with the infections detected in adults; (3) the risk for the development of cancer in lifetime. The virus infects mainly the epithelial cells, where it may remain latent for a very long time, evolve to the subclinical form, and thus remain, or reactivate, with a resulting accumulation of chromosomal mutations in host cells. The next result after this accumulated latent carcinogenic potential of certain types of HPV during childhood would be the development of a neoplasm in lifetime. The natural history of papilloma infection is characterized by regression in a period that varies from months to years[
54].
In 5 cases (41.7%, n = 5/12) concordance of type specific HPV was observed between the genital/'placental/NB samples and in 1 case (8.3%, n = 1/12) between the genital/arterial cord blood samples, suggesting that there is often placental transmission (50%, n = 6/12). This was the first study in third trimester placentas to suggest the percentage of transplacental transmission of HPV DNA.
Several authors have focused special attention on the mode of HPV transmission. In 1992, Tseng et al.[
19] suggested transplacental transmission of the virus, after detecting the same viral genome (HPV type 16) in cervicovaginal smears, in mononuclear cells of peripheral blood of fifteen pregnant women and in the cord blood of seven newborns from these same mothers. Favre et al. (1998)[
11] showed the presence of several types of HPV DNA in amniotic liquid, placental cells and cervicovaginal smears of a mother and newborn with
epidermodisplasia verruciforme. Hermonat et al. (1997)[
25] recorded that the infection of HPV was three times more prevalent in specimens of spontaneous abortion in the first trimester. Later, in 1998, the same authors[
55], confirmed the presence of HPV DNA in placental tissue of spontaneous abortions and concluded that the predominant site for HPV DNA type 16 findings were the cells of the syncytiotrophoblast. Thus, they raised the hypotheses of viremia, not yet convincingly documented, and of contamination of placental cells by oocyte infection before or right after implantation, by ascending infection or by infection carried by sperm containing latent HPV.
This study pointed to five cases of NB (41.7%, n = 5/12) negative for the research of HPV DNA, while in their respective placentas HPV DNA was detected. Four of these placentas presented viral types considered a high carcinogenic risk (types 16, 18, 58), and 100% (n = 5/5) of type specific HPV concordance is observed among the genital/placental specimens. The results achieved show that transmission of the virus to the fetus is not a prerogative of every HPV DNA+ woman, or in all cases of HPV DNA+ placentas, pointing to the existence of other as yet unknown factors that could be involved in transplacental transmission. Sedlacek et al. (1989)[
56] detected the presence of HPV DNA in the oral cavity of 36.5% of the newborn, delivered vaginally to mothers with a diagnosis of HPV DNA+ for cervical cells. Kaye et al. (1994)[
57] showed that the pregnant women who transmitted the virus to their concepts had a higher viral load.
Among the eleven cases of RN HPV DNA+ (vertical transmission), four cases (36.4%, n = 4/11) did not present HPV DNA in their respective placentas. One case out of this total presented type-specific HPV concordance between the genital/arterial cord blood (HPV type 52) and two presented concordance of type specific HPV between the genital/NB samples (HPV types 11 and 42). These results emphasized the possibility that other HPV transmission routes exist during pregnancy (transamniotic ascending infection), or during labor (ascending infection after the amniotic membranes are ruptured), or during delivery (by the fetus passing through the contaminated birth canal).
HPV DNA in arterial cord blood and HPV DNA in maternal peripheral blood
Three cases of NB (6.1%, n = 3/49) who presented HPV DNA+ in arterial cord blood samples were seen. The three cases were considered transplacental transmission due to finding concordance of the type specific HPV among the genital/NB samples. In these three cases, as in all cases of vertical transmission, no HPV DNA was detected in maternal peripheral blood. On the other hand, three cases (6.1%, n = 3/49) were also observed of parturients who had HPV DNA in their peripheral blood, without HPV DNA in the respective placentas and NB. These results suggest that HPV infections in the placentas may have occurred by another route, which was not hematogenic, either by an infection that was already present before pregnancy in the endometrium, or by an ascending infection during the egg implantation period, or at the time of fertilization by sperm contaminated by the virus, or else during pregnancy facilitated by the uterine anatomy. The HPV predilection for tissues, apparently exclusive to the pavement epithelium of the skin and the mucosas, has been challenged in the last few years, since several studies demonstrated the capacity of HPV to infect different sites. The studies of Teseng et al. (1992)[
19] may be mentioned, who demonstrated the presence of HPV in the amniotic liquid of pregnant women before labor, and Hermonat et al. (1998)[
55] who described the presence of HPV in the syncytiotrophoblast of spontaneous abortions, proving the capacity of HPV to locate also in the uterine cavity. These studies showed the capacity of the virus to infect the uterine cavity, and therefore it is no surprise that the virus appears in the endometrium. Fedrizzi et al. (2004)[
58] found HPV DNA (types 16 and 18) in 10% of the women with a normal endometrium. The exception is the work by Lai et al. (1992)[
59] who found HPV DNA in 70% of the cases studied that had a normal endometrium or some benign disease. However, O'Leary et al. (1998)[
60] did not find HPV DNA in the normal endometrium.
The positive and significant correlation between placental HPV infection and the maternal epidemiological variable "history of immunodepression" (HIV, p = 0.011), may be related to the special characteristics of the gravid cycle, especially the changes in the hormonal and immunological balance prevailing during this period, which might favor placental HPV infection.
HPV detection and typing methods
Although the nested multiplex PCR methodology is used to identify and type only 9 types of HPV shown as the most prevalent in the city of Caxias do Sul, it performed very well in identifying maternal HPV DNA, and also considerably increased the number of pregnant women with multiple virus infections. In the newborn and placental samples the nested multiplex PCR method showed its great sensitivity and specificity to identify HPV. The use of that method was also crucial to evaluate the concordance of type specific HPV DNA among the maternal/placental/newborn samples, thus defining the vertical and transplacental transmission rates of the virus.
Concluding, the HPV DNA detection rate in the placenta was 24.5% (n = 12/49) and the transplacental transmission rate was 12.2% (n = 6/49). A transplacental transmission rate of 54.5% (n = 6/11) was observed when only the cases of vertical transmission were analyzed. These results were achieved in analyses of the placentas and newborns of mothers with genital warts or intraepithelial lesions of the uterine cervix. Thus, different forms of management can be adopted for each of these different stages (pre-gestational, gestation, delivery and the first few months after delivery), both from the diagnostic and therapeutic perspective. The mother and the newborn must be observed clinically and educational preventive measures must be established concerning the forms of HPV transmission, besides effective strategies for specific immunization.
In future, the HPV DNA rates in must be observed in the normal endometrium of women of reproductive age, in order to explain the possible route of infection by continuity and/or contiguity between the endometrium and conception products.