INFECTION WITH HUMAN PAPILLOMAVIRUSES (HPV) is causal in the development of genital and oral cancers. 1–3 Evidence suggests that the same HPV oncogenic types that cause cervical cancer also are associated with an increased risk of oral cancer. 2,3 HPV infection is considered a sexually transmitted disease among adolescents and adults. 1,4 Previous observations also suggest that HPV could be transmitted vertically during parturition. 5–9 Pregnant women can act as hosts for HPV transmission through cervical secretions, causing HPV infection in their newborns that could lead to cancer or other HPV-related diseases in their offspring years later. This proposition is particularly worrisome because the prevalence of detectable maternal HPV infection during pregnancy reportedly rises during the second and third trimester above that of nonpregnant women, thus increasing the probability of HPV exposure to perinatal transmission at the time of delivery. 6,9–11 Detection rates of HPV infection by polymerase chain reaction (PCR) have ranged widely between 1% and 20% in newborns of pregnant women without apparent infection in their cervix 6,7,11 and between 5% and 72% in women with HPV-related cervical diseases diagnosed during pregnancy. 8,9 Data on HPV in the oral cavity of parents and infection in their newborns have not been examined as another source of transmission early in life. Studies have also not been evaluated for parental/newborn concordance by sequencing HPV types to further clarify their likelihood as the source of transmission.
In this study, we compared the prevalence and types of HPV in the cervix and in the oral cavity of a large sample of pregnant women and in the oral cavity of male partners to those mucosal sites in their newborns. Furthermore, we evaluated whether there was concordance in HPV type and sequence to elicit information about potential modes of viral transmission between parents and their infants. Parental risk factors for perinatal transmission also were evaluated.
Patient Population and Specimen Collection
Pregnant women seeking routine obstetric care from the clinics at the University of Iowa Hospital Department of Obstetrics–Gynecology were solicited for inclusion in the study and administered a University approved Institutional Review Board form. All healthy patients were recruited on a consecutive basis between April 1998 and November 2001. They were evaluated for HPV DNA late in gestation and immediately before delivery. Among the enrolled women, 574 had live births and were recruited in the study. Another 83 pregnant women and 5 male partners refused to take part in the study, 7 women were too ill, 9 had complicated pregnancies and did not want to participate, 11 had a language barrier, 2 were mentally incompetent, and 13 moved from the area, transferred their obstetric care to another facility, and therefore were unavailable for study follow up, as were 5 who were giving up their babies for adoption. There were 574 newborns from whom oral and/or genital specimens were available for HPV assessment. All but 12 were singletons, 255 females and 319 males. Male partners (n = 68) who were present for the gestational appointment consented to provide an oral rinse to test for HPV infection. This timing allowed for a comparison in maternal/paternal HPV oral concordance. No same-sex couples or nonbiologic partners were identified during the study recruitment. A university-approved human subjects consent form was signed by the parents before enrollment.
Parental and Newborn Risk Factor Questionnaires
At the initial visit, expectant mothers and attending fathers were asked to complete separate questionnaires regarding demographic information, reproductive (females) and sexual history, and risk factors for HPV infection. Included were specific questions regarding age, education, income, marital status, age at first intercourse, number of partners, parity, oral contraceptive use and duration, smoking and alcohol history and duration, and HPV-related disease history. A medical form was completed for each newborn and included gestational age at birth, gender, type of delivery, bottle/breastfed, medical problems at birth, and number of hours after delivery when HPV specimen was performed. Characteristics of fathers included age, education, marital status, income, age at first intercourse, number of lifetime sex partners, smoking and alcohol history and duration, and history of HPV-related diseases.
Human Papillomavirus DNA Specimen Collection
Maternal cervical swabs were taken during pregnancy, primarily in the third trimester (median gestational age, 37 years) and immediately before delivery (median, 39 weeks). Only 44 women had gestational specimens retrieved before the third trimester. Their HPV infection rates and risks were not significantly different from those in late gestation and thus the data were pooled. An oral rinse was obtained during the gestational pregnancy appointment unless women reported to the clinic for a precipitous delivery in which case the oral rinse was collected at that time. To guard against a positive test of viral inoculation rather than infection, newborn specimens were not collected immediately after birth. Infants were evaluated for HPV at a mean of 65 hours after birth in their genital and oral regions. Newborns who returned to the university hospital for well baby care had oral samples obtained at 3- and 6-month follow-up periods for additional HPV testing.
The 2 maternal cervical samples were collected by the attending obstetricians and nurse practitioners. All were trained in the collection procedures for pregnant women. The maternal HPV DNA sample was obtained using a cotton swab around the transitional epithelium area of the cervix, the ectocervix, and pooled secretions in the posterior vaginal fornix. Cells from the swab then were placed in buffered saline and prepared for a cell count of nucleated squamous cells using a hemocytometer. All patient specimens were analyzed using a hemocytometer to count for adequacy of nucleated squamous cells, and this information was used in preparing the PCR reaction. Specimens were subsequently frozen at −80°C. The second HPV DNA maternal specimen was collected immediately before delivery using a sterile modified proctoswab to collect cells from the ectocervix only regardless of whether the woman had a vaginal or cesarean delivery. Women and their male sexual partner, the biologic father, were requested to swish 10 mL of normal saline for 30 seconds, expectorate into a pill cup, and specimens were immediately placed into a double snap-cap test tube. Specimens were immediately refrigerated until a cell count was obtained and then frozen as described for the cervical specimens. In newborns, a sterile cotton tip applicator dipped in normal saline was swabbed throughout the infant oral cavity and oropharynx. The genitals were swabbed with a sterile wet applicator on the vulvar and external genital area, and circumcised tissue was collected to evaluate infection in the newborn male genitals. All specimens were processed and frozen as described for the parental specimens.
These methods have been documented previously for sample preparation, PCR analyses, and DNA hybridization. 12 In processing the DNA from patient specimens, repeat HPV-negative oral cells were included as a control for contamination, and processing was conducted in a building separate from and before going to the research laboratory for the day. The DNA extraction was followed by PCR amplification, which included positive controls containing 0.1 pg and 1.0 pg of HPV plasmid DNA with the primers. Negative controls were 0.025 μg of human DNA from oral exfoliated squamous cells, normal saline, and PCR buffer. To confirm the presence of adequate cellular DNA in each specimen for HPV DNA testing, amplification with primers specific for β-globin were included. 13 Specimens repeatedly PCR-negative by gel electrophoresis of the amplified β-globin gene were excluded from the analyses. No maternal or paternal specimens were judged inadequate on the basis of the β-globin result.
Oligonucleotide primers used to identify HPV DNA in the samples were MY09 and MY11, 14 each a mix of primers that amplify a portion of the L1 region of the HPV genome. In addition, a primer to improve amplification of HPV-51 was included. Amplified samples that were band-negative with 10 μL of the amplification products after being electrophoresed through agarose gels were then analyzed by dot blot hybridization. Gel-negative, dot blot HPV-positive samples were reamplified with the GP5+ primer in a heminested fashion to generate adequate HPV DNA sequences. 15 When discordant HPV types were found in separate samples from an individual or in mother/baby pairs, we performed additional PCRs with type-specific primers that amplify portions of the E6 ORF. This allowed the presence of additional HPV types to be detected specifically in the presence of one or more other types. The procedure provides a more sensitive method for detecting specific multiple types when more than one infection could be present and in which the previously detected type could have a stronger detection signal that can prevent the observation of a signal of another HPV type. HPV types found in discordant mother/baby pairs included HPV 16, 39, 51, and 61. The primers used to reconcile the discordant pairs specifically amplified the E6 region in the following types: CAGGACCCACAGGAGCGACC and ATCGACCGGTCCACCGACCC amplifying nt 110 to 509 of HPV-16, TGCCAGACCTGTGCACAACG and GCACCGTCGACACTGRCCTG amplifying nt 147 to 510 of HPV-39, GGGAAACACCACGAACGCTG and GCGCATTGCCCGTCCAACG amplifying nt 113 to 506 of HPV-51, and GCAAGGACTACGAGGTGGAC and TATGCCTGTACCTGATGCTC amplifying nt 139 to 457 of HPV-61. Each PCR reaction contained 30,000 nucleated cells per sample. The dot blot detected less than 10 SiHa cells in a spiked sample, or less than 1 copy of HPV per 1000 squamous cells in sample specimens, and the type-specific primers had similar or greater sensitivity as the generic primers.
HPV types were determined using DNA sequencing and amplified DNA was evaluated at the DNA Core in the university medical facilities. This was performed by dye termination on a DNA sequencer (Applied Biosystems–PE Biosystems Group) and nucleotide sequences were compared with GenBank sequences using the BLAST program. 16 Specific HPV types were identified if there was a known type that matched by >90%. HPV types were defined as either high risk (HR) or low risk (LR) on the basis of types that have been associated with dysplasia, cancer, or benign lesions such as genital warts. The HPV-HR types detected in this study were 16, 18, 31, 33, 35, 39, 51, 54, 58, 59, 68, and 70. HPV-LR types identified included 6, 11, 13, 38, 44, 53, 56, 61, 66, 69, 83, 84, and 6 unnamed types (AF091451, AJ010822, L38388, U12480, U12489, and U12490). The unnamed types were all related to benign and low-risk cutaneous HPV types based on their sequence comparison using GenBank. Other samples were defined as “positive unclassified” when they were identified as HPV on a Southern blot but could not be typed because of insufficient DNA. There were no cervical and 0.3% of oral samples that could not be classified.
The Wilcoxon’s rank sum test was used to compare quantitative variables between HPV-positive and -negative samples. The chi-squared test or Fisher EXACT TEST was used to compare categorical variables by HPV status. Statistical significance was noted at 0.05 and all tests were 2-sided. As a result of the low HPV-positivity rates among newborn and paternal samples, exact logistic regression was used to examine the association between HPV detection and potential risk factors. Because of the potential lack of independence in HPV infection in the twin sets, HPV data from only the first-born were used in the analyses. All analyses were performed using the SAS statistical package. 17
Characteristics of the enrolled women are summarized in Table 1. The median age at delivery was 29 (range, 18–45 years). The mean age at first intercourse was 17.8 years, number of partners was 6 (range, 1–70), and parity was 2.5. Maternal HPV infection was statistically significantly associated with younger age, less education, lower income, unmarried status, earlier age at first intercourse, more partners, and smokers ( all P <0.0001). A current HPV infection also was significantly associated with a history of HPV-related diseases: condylomata, cervical dysplasia, and cervical cancer (Table 1).
Table 2 summarizes the HPV prevalence and types in mothers, newborns (specimens taken at baseline after birth), and attending fathers. In the maternal cervix and newborn genitals, there were 27 types identified and over one third were HPV-HR. There were 13 oral HPV types among these families, of which half were high risk. Among the pregnant women, 29% were infected with HPV in their cervix during pregnancy or labor. There was no statistically significant difference in the percent positive at either time period (26% vs. 25%). In the oral cavity, 2.4% were detected with HPV and almost two thirds were HR types. The overall prevalence at either site was 30% (n = 172) with 1% infected only in the oral cavity. The frequency in the cervix was 18% for HR types and 13% for LR types. The most common HR type in the cervix was HPV-16 and the most common LR type was AJ010822. Among the 68 biologic fathers who were enrolled, 6% had an oral infection (Table 2) of which two fourths were high-risk types. Surprisingly, none of the infected men had a pregnant partner with HPV detected at either mucosa site.
There were 571 newborn oral specimens and 403 genital specimens available to evaluate HPV DNA for baseline specimens taken after birth. The total incidence of HPV infection in newborns was 1.6% (Table 2). Less than 1% of oral and 1% of genital specimens contained HPV DNA and none of the infants was infected at both sites. Thus, the risk of HPV infection in newborns was low despite the relatively high positivity rate in the maternal cervix. In female newborns, 4 of 249 vulvar samples and 1 of 252 oral samples were HPV-positive. In the 319 male infants, none of the circumcision samples and 4 of 319 oral specimens were detected with HPV. Most of the HPV types detected in infants were HR types HPV-16 and HPV-51.
Based on infection identified in the cervix or oral cavity of the pregnant mothers, no maternal risk factors were statistically significantly associated with a higher incidence of HPV detection in their infants (data not shown). A nonsignificant higher rate of infection in newborns was noted in mothers with a history of cervical cancer compared with those without (14.3% vs. 1.6%, P = 0.1). There was no statistically significant difference in the rate of HPV detected in babies born by cesarean section or vaginally but the risk was low for either delivery type: 1.4% versus 1.6%. Other factors such as gestational age at birth, breastfeeding, and number of hours after birth at retrieval of newborn specimens also did not influence the likelihood of detecting HPV in the infants. None of the infected offspring was diagnosed with a genetic or other medical complication at birth. Paternal infection in the oral cavity also was not significantly associated with detection of the virus in the oral cavity of newborns.
Table 3 shows the concordance in HPV types between infected newborns and their parents. Only pregnant women whose offspring or male companion was infected were included in this table. There was no concordance in HPV positivity or type between the genital and oral cavity in newborns or between the cervical and oral specimens of their mothers. Among the 6 mother/baby pairs found to be infected, only one pair was concordant (no. 2375), with HPV-51 in the oral cavity of the newborn and in the cervix of the mother. They had the same sequenced type, thus type-specific concordance of vertical transmission was a very uncommon event: 0.2%. Among the 5 newborns positive for HPV in their oral cavity, only 1 included a mother who also tested positive in her oral sample (no. 2345). However, the HPV types were different between mother and newborn. The newborn was infected with HPV-61, a low-risk type, whereas the mother had HPV-39, a high-risk type. In the 3 other positive newborn oral specimens, 2 mothers were infected in their cervix but the types were different from their offspring (nos. 2021 and 2181) and the mother of the third infected infant was not detected with the virus (no. 2057). Of the 4 neonatal genital samples with detectable HPV, 1 mother tested positive in her oral sample (no. 2343). Her newborn had a LR type (HPV-61), whereas the mother had insufficient HPV DNA to determine the type in her oral cavity. Immediately before delivery she was detected with HPV-16 in the cervix. Another newborn (no. 2059) had a HR (HPV-51) type in her genitals compared with her mother who had a LR type (AJ010822) in the cervix. The mothers of the other 2 positive newborn genital specimens (nos. 2117 and 2294) were not detected with HPV.
Because it is possible that amplification with general primers could fail to detect additional HPV types when more than one is present, where there was discordance in HPV type between samples from the same individual or in samples from mother/baby pairs, type-specific PCR was performed to detect possible dual infections missed by the primary HPV detection method. In none of these cases was the discordance resolved by detection of additional HPV types. Thus, there were no double infections found in these individuals.
None of the newborns of infected fathers tested positive (Table 3). In one newborn who was infected with HPV-51 (no. 2375) in the oral cavity, the paternal specimen was HPV-negative. As discussed previously, the mother of this newborn was infected in the cervix with the same type before delivery. Thus, not only was vertical transmission limited to one mother/infant pair, but also there was no instance of early contact transmission seen between the newly born hospitalized infant and the parents based on findings from the oral mucosa.
Newborns were reassessed for HPV DNA in samples taken from the oral cavity at follow-up (FU) pediatric visits. There were 283 with a first FU, 117 with a second FU sample, and 115 with both a first and second FU sample. Among the newborns who were detected with HPV after birth, 3 were HPV-negative at FU and the other 6 did not return to the neonatal clinic and thus were not assessed. Two of the infants were positive in the oral cavity at birth, whereas the third was infected in the vulva. None of the uninfected newborns after birth was detected with HPV at FU. Unfortunately, genitals were not reexamined at FU as a result of the Human Subjects Consent Committee. Women whose infants were seen at FU did not differ in demographic or pregnancy characteristics from those who were not followed (data not shown).
This is the first study to evaluate concordance in HPV type between parents and newborns using DNA sequencing to verify the variant in their HPV infection. This is one of the largest studies of mothers and newborns, and the results indicate that type-specific concordance in HPV through vertical transmission is rare (less than 1%). Furthermore, postbirth transmission between the newly born hospitalized infant by parents does not seem to occur. This also is the first report to examine the issue of HPV transmission in the oral and genital mucosa among hospitalized newborns and that of their parents. Among families in which mother and offspring were infected, 5 of 6 infants showed discordant types to those of their mothers. Several other infants had genital infections in apparently uninfected mothers. Thus, there was no consistent pattern to the mode of vertical transmission. It is possible that leakage of amniotic fluid allowed viral infection in utero, which was not detected at the time of maternal cervical testings. Other possible modes of infection are less clear. This also is the only study that has evaluated fathers for HPV in their oral cavity and with specimens collected at the same time as the pregnant mother. Orally infected fathers also did not appear to transmit the virus to the oral mucosa of their infants after birth, although the sample size was small. Unlike genital infections, HPV detected in the oral cavity in these newborns could have been transmitted from the oral mucosa of untested relatives or family friends.
As found in our previous study 5 and by Puranen et al., 7 some infants with viral infection had HPV-uninfected mothers or fathers. Our laboratory methods that require a minimum number of squamous cells based on a cell count reduce the probability of false-negatives. In addition, use of DNA sequencing allows us to detect a wider variety of HPV types and to reduce the probability of false-positives because all viral detection is verified as HPV by sequencing. Because we use DNA sequencing, the possibility of HPV DNA contamination is reduced because there not only were different sequences among the specific types detected, but none of them matched that cloned in the laboratory. Furthermore, we reexamined the HPV type-specific primers to evaluate multiple infections in the discordant pairs of infected infants and mothers. We detected oncogenic types in both the oral and genital sites of very young infants, which suggests that if the virus does not clear, it could put them at increased risk of HPV-related diseases. Because specimens were retrieved from infants while still hospitalized after birth, we precluded the probability of sexual exposure. However, there was no way to prevent infants from being orally infected by other contacts. Nonetheless, there was no difference in the HPV rate associated with time of specimen retrieval after delivery and prevalence between oral and genital sites in newborns, arguing against other contacts as the source of HPV infection.
None of the previous studies of mother/newborn HPV infection has been based on DNA sequencing, thus none has evaluated type-specific concordance for the same variant. Some studies have either not been able to type infant specimens or have examined a limited number of HPV types. 6,19 Other large studies 5,6,18 support the evidence that vertical transmission or infection before sexual debut is rare, with low prevalence rates detected in the oral cavity, nasopharynx, external genitals, or anogenital region. We included women with a history of HPV-related diseases to determine whether their offspring were more likely to be infected at the time of delivery. Indeed, 3 of the 9 infected infants were born of mothers (n = 142) with prior cervical condylomata, dysplasia, or cancer. However, because 7 HPV-positive newborns had mothers with a current normal Pap smear, this test did not predict infants who were likely to be detected with viral infection. Five of the 7 mothers with a history of cervical cancer had viral infection during pregnancy and 4 were high-risk types. Vertical transmission was found in 1 of these 7 newborns (no. 2021).
Cesarean section often is considered to be protective against vertical HPV transmission, yet current and previous large studies do not support this belief. 5,6,18 In the current study, infection was found in the genitals of infants by either delivery type, and there was no difference in infection rates by infant gender, gestational age, newborn weight, or congenital or neonatal abnormalities. Tseng et al. 18 also did not find a difference in HPV status for these infant characteristics. Although this is one of the largest studies to date, the low HPV infection rate in newborns suggests that much larger studies (in the 1000s) will be required to have sufficient statistical power to examine these factors (eg, type of delivery) associated with newborn HPV positivity.
Although the oral infection rates were relatively low, parents were detected with a 2 to 6 times higher rate compared with infants. Our studies 2,12 (unpublished data) suggest that among those who have never had an HPV-related disease, the rate of oral infection is lowest in newborns, rises in children and adolescents, and is highest among middle and older aged healthy adults (11%). Furthermore, there was a 10-fold higher rate in the genital compared with the oral mucosal sites in adult women. The cervical and oral infection rates were well within the range described by others for pregnant or young adult women. 10,18,20,21 The difference between oral and genital infection rates suggest that there could be a high rate of transient virus that clears in a short timespan, explaining the different HPV types found in partners or in individual mucosal sites. Possible explanations for the low oral infection rate include the barrier effect of the heavily keratinized oral tissue, which makes it less permissive than the cervix to microbial infections that routinely exist in the oral cavity without producing disease or which prevents infection from invading the mucosa through microinvasive tears. Saliva also has been shown to protect against infections through a number of antimicrobial agents. 22–24
Future studies should evaluate pregnant women frequently during pregnancy to determine the level of HPV fluctuation that could occur during this period. This information could also clarify why some women who were not detected with infection nonetheless had infected infants, some of them on the genital area. Investigations also should focus on determining the biologic characteristics that explain differences between oral and genital prevalence rates. Because it is unclear whether age at initial sexual exposure or recency of exposure is an additional important factor predicting infection, these factors should be examined. These issues should be resolved to determine at what age or time in one’s sexual experience is best to vaccinate against HPV.
1. Munoz N. Human papillomavirus and cancer: the epidemiological evidence. J Clin Virol 2000; 19: 1–5.
2. Smith EM, Hoffman H, Summersgill KF, Kirchner HL, Turek LP, Haugen TH. Human papillomavirus and risk of oral cancer. Laryngoscope 1998; 108: 1098–1103.
3. Gillison ML, Koch WM, Capone RB, et al. Evidence for a causal association between human papillomavirus and a subset of head and neck cancers. J Natl Cancer Inst 2000; 92: 709–720.
4. Buckley JD, Harris RW, Doll R, Vessey MP, Williams PT. Case-control study of the husbands of women with dysplasia or carcinoma of the cervix uteri. Lancet 1981; 2: 1010–1015.
5. Smith EM, Johnson SR, Cripe T, et al. Perinatal transmission and maternal risk of human papillomavirus infection. Cancer Detect Prev 1995; 19: 196–205.
6. Watts DH, Koutsky LA, Holmes KK, et al. Low risk of perinatal transmission of human papillomavirus: results from a prospective cohort study. Am J Obstet Gynecol 1998; 178: 365–373.
7. Puranen M, Yliskoski M, Saarikoski S, Syrjanen K, Syrjanen S. Vertical transmission of human papillomavirus from infected mothers to their newborn babies and persistence of the virus in childhood. Am J Obstet Gynecol 1996; 174: 694–699.
8. Fredericks BD, Balkin A, Daniel HW, Schonrock J, Ward B, Fraxer IH. Transmission of human papillomavirus from mother to child. Aust N Z J Obstet Gynaecol 1993; 33: 30–32.
9. Smith EM, Johnson SR, Jiang D, et al. The association between pregnancy and human papillomavirus prevalence. Cancer Detect Prev 1991; 15: 397–402.
10. Morrison EA, Gammon MD, Goldberg GI, Vermund SH, Burk RD. Pregnancy and cervical infection with human papillomavirus. Int J Gynecol Obstet 1996; 54: 125–130.
11. Pakarian F, Kaye J, Cason J, et al. Cancer associated human papillomaviruses: perinatal transmission and persistence. Br J Obstet Gynaecol 1994; 101: 514–517.
12. Summersgill KS, Smith EM, Levy BT, Allen JM, Haugen TH, Turek LP. Human papillomavirus in the oral cavities of children and adolescents. Oral Surg Oral Med Oral Path Oral Rad Endod 2001; 91: 62–69.
13. Chehab FF, Doherty M, Cai SP, Kan YW, Cooper S, Rubin EM. Detection of sickle cell anaemia and thalassaemias [Letter]. Nature 1987; 329: 293–294.
14. Ting Y, Manos MM. Detection and typing of genital human papillomavirus. In: Innis M, Gelfand D, Sninsky J, White T, eds. PCR Protocols: A Guide to Methods and Applications. San Diego: Academic Press, Inc, 1990: 356–367.
15. De Roda Husman AM, Walboomers JM, van den Brule AJ, Meijer CJ, Snijers PJ. The use of general primers GP5 and GP6 elongated at their 3′ ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J Gen Virol 1995; 76: 1057–1062.
16. Altschul SF, Madden TL, Schaeffer AA, et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 1997; 25: 3389–3402.
17. SAS System for Windows, version 8.2. Cary, NC: SAS Institute, 2001.
18. Tseng C-J, Liang C-C, Soong Y-K, Pao C-C. Perinatal transmission of human papillomavirus in infants: relationship between infection rate and mode of delivery. Obstet Gynecol 1998; 91: 92–96.
19. Koch A, Hansen SV, Nielsen NM, Palevsky J, Melbye M. HPV detection in children prior to sexual debut. Int J Cancer 1997; 73: 621–624.
20. Smith EM, Ritchie JM, Galask R, Pugh EE, Jia J, Ricks-McGillan J. Case-control study of vulvar vestibulitis risk associated with genital infections. J Infect Dis 2002; 10: 193–202.
21. Jacobs MV, Walboomers JM, Snijders PJ, et al. Distribution of 37 mucosotropic HPV types in women with cytologically normal cervical smears: the age-related patterns for high-risk and low-risk types. Int J Cancer 2000; 87: 221–227.
22. Shugars DC, Watkins CA, Cowen HJ. Salivary concentration of secretory leukocyte protease inhibitor, an antimicrobial protein, is decreased with advanced age. Gerontologist 2001; 47: 246–253.
23. Miletic ID, Schiffman SS, Miletic VD, Satterly-Miller EA. Salivary IgA secretion rate in young and elderly persons. Physiol Behav 1996; 60: 243–248.
24. Steele C, Fidel PL Jr. Cytokine and chemokine production by human oral and vaginal epithelial cells in response to Candida albicans
. Infect Immunol 2002; 70: 577–583.