CERVICAL CANCER IS AN IMPORTANT public health issue in less developed countries, with over 300,000 incident cases per year in the world.1,2 In Mexico, malignant tumors of the uterine cervix are the third highest cause of cancer-related deaths in the general population and the first cause of cancer-related deaths among women. The incidence rate has increased from 9.72 cases per 100,000 women in 1975, to 42.9 cases per 100,000 women in 1992, and to 48 cases per 100,000 women in 2001.3–5
Over the past decade, research has expanded globally in search of a better understanding of the natural history of human papillomavirus (HPV) infection. Projects on this topic include studies of the detection of HPV infection, as well as studies looking for associations of HPV infection with different factors, including pregnancy, sexual behavior, diet, smoking, genetics, and immunosuppression. Approximately 50% of sexually active young women are infected with some type of HPV, and of these, 5% to 10% are estimated to be persistently infected with oncogenic HPV types that could progress to cervical cancer.6,7
Recent results suggest that parity increases a woman’s the risk of cervical cancer (3–4 pregnancies odds ratio [OR] = 2.5; 5–6 pregnancies OR = 2.8; 7 or more OR = 3.8; compared with fewer than 2 pregnancies) after accounting for the strong effect of HPV.8 This association was not found with a woman’s reported number of abortions, suggesting that relevant factors mediating malignant progression of HPV infection may occur during the second and third trimesters of pregnancy.9 Nonetheless, controversy exists over the effect of pregnancy on the natural history of HPV infection. Some studies suggest that the physiological process of pregnancy modifies cellular immunity, thereby increasing risk of HPV infection in young women and risk of oncogenic HPV persistence and progression to intraepithelial lesions in women older than 30 years of age. The association between pregnant status and HPV, however, is still not well established.10–12
Study Population and Enrollment
A cross-sectional study was conducted in a group of 278 pregnant women attending primary health care services at the Family Medicine Unit No. 1 of the Mexican Institute of Social Security (IMSS) in Cuernavaca, Morelos, Mexico, for a normal prenatal visit between April and August 2000. The IMSS has a large catchment population, and this antenatal population surveyed is representative of middle- and low-income workers and worker dependents from Morelos State. We included pregnant women who visited the unit for regular prenatal control, who were between the ages of 15 and 39 years, who had no medical or obstetric complications, who agreed to participate and respond to a questionnaire on reproductive health and practices, and who provided a self-collected vaginal swab for HPV DNA detection. Four women were excluded because they did not complete the questionnaire or had medical or obstetric complications.
Between May and October 1999, a total of 7868 nonpregnant women aged 15 to 85 years received cervical cancer screening at the IMSS in Cuernavaca, Mexico. This population is representative of middle- and low-income workers and worker dependents from Morelos State as previously described by Flores et al.13
In brief, women were ineligible to participate if they had had a previous hysterectomy or were currently pregnant. From this population, a random sample of 1060 women was selected in categories of 5-year age groups to constitute a frequency- and aged-matched control group to the pregnant women.
The study protocol for both pregnant and nonpregnant women was reviewed and cleared by the Institutional Review Boards of the National Institute of Public Health and IMSS in Cuernavaca, Mexico. After obtaining written informed consent from study participants, personal interviews were conducted using a structured questionnaire that included information on a woman’s age, age at first intercourse, and reproductive history.
Cervicovaginal Exfoliated Cell Specimen Self-Collection and Processing
Both pregnant and nonpregnant participants were asked to provide self-collected vaginal specimens with the same technique for HPV DNA testing. Nurses trained in the technique explained the self-collection procedure to study participants. Women were instructed to insert a 15-cm cotton-tipped sterile Dacron swab into their vagina until their fingers reached their labia, and then rotate the swab once to the left and once to the right. Women were also instructed to place the sample in a specimen transport medium (STM) test tube (Digene Corp., Gaithersburg, MD) immediately after removing the swab from their vagina. On the day of collection, vaginal specimens were sent to the laboratory of the Center for Infectious Diseases Research at the National Institute of Public Health, in Cuernavaca, Mexico, for storage at −20°C until processing.
Human Papillomavirus DNA Detection Techniques
Self-collected exfoliated vaginal specimens were tested for the presence of HPV high-risk DNA types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68 using the Digene Hybrid Capture 2 (HC2) microtiter assay, according to the manufacturer’s instructions.14 All HPV DNA testing was conducted masked to the woman’s pregnancy status. Specimens were classified as positive if the relative light units/positive control (RLU/PC) ratio was ≥1 using the 1-pg/mL positive controls supplied in the kit. Special precautions to minimize false-positive results in the HC2 assay were implemented, and selected positive and negative HPV samples were included in repeat test runs for quality control purposes.
Age-specific HPV DNA detection rates were computed for pregnant and nonpregnant women, as well as age-standardized rates using as standard the female population of Cuernavaca, Morelos, Mexico.15 Unconditional logistic regression models were conducted to evaluate the effect of socioeconomic, gynecologic, and sexual behavioral characteristics on HPV DNA infection rates among pregnant and nonpregnant women. In addition, conditional logistic regression was used to evaluate the effect of pregnancy status on the risk of HPV DNA infection. Statistical significance was calculated using the chi-squared and t tests (with significance set at P <0.05). Odds ratios (ORs) for HPV DNA detection and corresponding 95% confidence intervals (95% CIs) were computed. Tests for linear trend of categorical variables were performed giving an increasing score for each level of the categorized variable and fitting them in the model as continuous variables.
No differences in age among the 274 pregnant and 1060 nonpregnant women were observed as a result of age matching (25.7 vs 26.2 years, respectively) (P = 0.12). However, the average age at first sexual intercourse for pregnant women was modestly higher as compared with nonpregnant women: 20 and 18.8 years of age, respectively (P <0.001). The average number of deliveries as well as caesarean sections were lower in pregnant than in nonpregnant women (Table 1).
Human Papillomavirus Detection by Pregnancy Status
Pregnant women had a significantly higher proportion of high-risk HPV DNA infection (37.1%; 95% CI, 31–43%) than nonpregnant women (14.2%; 95% CI, 12–16%) (P <0.001). The pattern of differential detection rates, among pregnant and nonpregnant women, was relatively constant across age groups. The age-adjusted HPV DNA detection rate was higher in pregnant women (32.5%) than nonpregnant (11.5%) (P <0.001). Among those who were HPV DNA-positive, 25.7% of pregnant women had a viral load >10 RLU/PC compared with 9.5% of nonpregnant women (Table 1).
HPV DNA prevalence does not show a clear pattern with increasing trimester of a woman’s current pregnancy (Fig. 1). High-risk HPV DNA positivity was 41.2% in the first trimester, 28.3% in the second trimester, and increased again to 42.6% in the third trimester (Fig. 1).
Determinants of Human Papillomavirus Infection
Relative risk estimates are shown in Table 2 between different reproductive or sexual behavior characteristics and the risk for HPV DNA infection among pregnant and nonpregnant women. Results did not show a significant association between risk for HPV DNA infection and current age, age at first intercourse, number of lifetime pregnancies, or history of caesarean section. In the fully adjusted model, HPV DNA detection was higher in women aged 30 to 34 years, compared with older groups (30–39 years), but this association was not significant in either pregnant or nonpregnant women. No significant associations were observed between age at first intercourse and HPV DNA infection (pregnant: OR = 1.8; 95% CI, 0.6–5.4 and nonpregnant OR = 1.9; 95% CI, 1.0–3.8, respectively, for ≤16 years vs >21 years). Both pregnant and nonpregnant women with a higher number of lifetime deliveries had consistently lower HPV detection rates (pregnant: OR = 0.4; 95% CI, 0.2–0.9; nonpregnant: OR = 0.4; 95% CI, 0.2–0.7, for ≥2 vs no pregnancies). A woman’s history of cesarean sections was significant associated with a lower HPV DNA detection in both pregnant (OR = 0.4; 95% CI, 0.2–0.9) and nonpregnant women (OR = 0.5; 95% CI, 0.3–0.8) (Table 2). Conditional multivariate analysis combining both groups (pregnant and nonpregnant women) shows that pregnancy increased the risk of HPV DNA infection (OR = 3.0; CI, 2.1–4.3).
In addition to already established risk factors related to sexual behavior and gynecologic and obstetric status, various immunologic and hormonal factors have been shown to influence persistence of HPV infection and progression to intraepithelial neoplastic lesions. Some reports suggest that the process of pregnancy promotes the development of a persistent HPV infection. Nevertheless, the mechanisms by which pregnancy changes the patterns of HPV DNA and its persistence and progression to clinical lesions has not been well documented. Our study documented the principal finding of a higher detection of high-risk HPV DNA in pregnant women than nonpregnant women.
Our study found that the detection rate for 13 high-risk types of HPV DNA was almost 3 times greater in pregnant women compared with nonpregnant women (37.1% vs 14.1%), which may be attributed to immunosuppression that occurs during pregnancy. Similar differences were reported by Fife (24.9% in pregnant vs 11.4% in nonpregnant),16 Schneider (28% vs 12.5%);17 Gopalkrishna (44% vs 21%),18 Czegledy (35% vs 20%), and Watts (36% vs 7%).19 Chang-Claude reported similar HPV detection rates (pregnant 13.9% vs nonpregnant 15.1%), although on review of the groups by age, there was a higher detection rate in pregnant participants under 25 years compared with pregnant participants above the age of 35 years (31.3% vs 15%). This could represent a not clear pattern of a higher rate of infection in younger pregnant women compared with older pregnant women (Table 3). Beyond the laboratory techniques used in these studies, these differences may be attributable to age and time of gestation. It has been proposed that the higher prevalence of HPV observed in pregnant women is consistently found during the third trimester. For example, Fife et al. report an increase in prevalence of high-risk HPV from 31% in the first trimester to 35.6% in the third trimester. Morrison et al. reported an increase from 27.3% in the first trimester to 39.7% in the third trimester. In both cases, the mean age of study participants was younger than 25 years.20
Our study shows pregnancy to be an independent risk factor for high-risk HPV infection (OR = 3.5; CI, 2.7–4.9 vs nonpregnant women). Similar findings were reported by Fife (current pregnant OR 1.79; 95% CI, 1.11–2.89)16 and Morrison (current pregnant OR 2.2; 95% CI, 1.1–4.5).21 Differences in the sensitivity of the molecular hybridization techniques used, HPV genotypes studied, mean age of study participants, prevalence of abnormal cytology, and coexistence of other sexually transmitted infections make it difficult to compare study results. Nevertheless, it appears that during pregnancy, particularly in the third trimester, there is a higher prevalence of HPV DNA, particularly of oncogenic HPV types, and that this phenomenon is most likely related to hormonal and immunologic factors.22–24
HPV DNA positivity was inversely associated with a woman’s number of births. Previous IARC-conducted studies have similarly shown that HPV DNA positivity is lower in women with previous full-term deliveries compared with nulliparous women. Among parous women, no trend in HPV positivity has been seen by increasing number of previous deliveries in 2 of these studies as is found in our study. It is possible that our results of a lower HPV positivity among parous women with a greater number of previous deliveries may be the result of residual confounding by age or represent a true lower HPV prevalence as a result of increased immunity to infection.31–33
In relation with sexual behavior, we observed a higher risk of HPV infection in nonpregnant women who become sexually active before the age of 17 years, although this risk was not statistically significant (OR = 1.3; 95% CI, 0.7–2.4). Similar findings have been reported by Morrison (younger than 17 years OR = 1.6; 95% CI, 0.8–3.0) and Hagensee (≥6 years of sexual activity OR = 1.8; 95% CI, 1.2–2.8).21,22 A woman’s reported history of previous cesarean sections showed a protective effect in all women surveyed (OR = 0.4; 95% CI, 0.2–0.5), suggesting that the cervical trauma could be a risk factor for HPV viral acquisition or reactivation. Similar findings were reported by Brinton (cesarean ever OR = 0.75; 95% CI, 0.6–1.0).34
Regarding laboratory techniques and specimen collection procedures used, our study results are based on HPV DNA detection using the HC2 test, which detects over 13 high-risk HPV types, unlike previous studies with a limited range of HPV type detection. The HC test provided a higher level of test accuracy when compared with other research-based HPV DNA tests. Compared with polymerase chain reaction, HC2 was easier to learn and perform and exhibited a good laboratory validation profile for reproducible results for the detection of HPV. As for the collection procedure for the HPV DNA specimens, self-collection proved very acceptable to both pregnant and nonpregnant women. As a result of the lack of information on type-specific HPV DNA status, however, it is not possible to distinguish a single infection with high viral load from multiple HPV infections with lower viral loads.23–25 To our knowledge, there are no available data comparing the validity of self-collection versus physician collection for HPV sampling in pregnant women. Our study results of a higher HPV DNA positivity in cervicovaginal cells in pregnant women may thus possibly, although unlikely, be the result of an artifact of differential sampling using the self-collection technique, and further validation studies of HPV sampling in pregnant women are indicated.
The lack of information regarding a woman’s sexual behavior (i.e., number of sexual partners, use of condoms, other sexually transmitted infections), as well as lack of information regarding type-specific HPV infection, may result in some degree of information or confounding bias. Nonetheless, the measures of association between HPV and pregnant status did not significantly change after controlling for some potential confounders, including age at first intercourse and a woman’s reported reproductive history. Furthermore, results presented here, based on a large sample size, show plausibility and a consistently with other studies on this topic carried out in Mexico as well as in other countries.
In countries like Mexico where incidence of cervical cancer and HPV infection are among the highest in the world, it is important to continue studies on the epidemiology of HPV infection. We recommend that longitudinal studies be conducted that include periodic measurements including accurate HPV DNA typing starting in the first trimester of pregnancy until 6 months postpartum. This will allow for better documentation of the persistence and/or regression of high-risk HPV infection during the course of pregnancy.
1. Gerbase AC, Rowley JT, Mertens TE. Global epidemiology of sexually transmitted diseases. Lancet 1998; 351(suppl 3):2–4.
2. Galloway DA. Is vaccination against human papillomavirus a possibility? Lancet 1998; 351(suppl 3):22–24.
3. Dirección General de Epidemiología (DGE), Secretaría de Salud (SSA). Registro histopatológico de neoplasias en México; 1994.
4. Secretaría de Salud (SSA). Programa nacional de prevención y control del cáncer cervicouterino 1995–2000; México 1995.
5. SSA. Dirección General De Información y Evaluación del Desempeño. Secretaría de Salud. Salud, México; 2001: Cancer Cervicouterino. SSA 2002.
6. Restrepo HE, González J, Roberts E, Litvak J. Epidemiología y control del cáncer del cuello uterino en América Latina y el Caribe. Bol Oficina Sanit Panam 1987; 102:578–593.
7. Schiffman MH. Epidemiology of cervical human papillomavirus infections. Curr Top Microbiol Immunol 1994; 186:56–81.
8. Muñoz N, Bosch FX, de Sanjosé S, et al. El virus del papiloma humano en la etiología del cáncer cervicouterino. Bol Oficina Sanit Panam 1993; 115:301–309.
9. Bosch FX, Lorincz A, Muñoz N, Meijer CJ, Shah KV. The causal relation between human papillomavirus and cervical cancer. J Clin Pathol 2002; 55:244–265.
10. Ahdoot D, Van Nostrand KM, Nguyen NJ, et al. The effect of route of delivery on regression of abnormal cervical cytologic findings in the postpartum period. Am J Obstet Gynecol 1998; 178:1116–1120.
11. Autier P, Coibion M, Huet F, Grivegnee A. Transformation zone location and intraepithelial neoplasia of the cervix uteri. Br J Cancer 1996; 74:488–490.
12. Smith EM, Johnson SR, Cripe T, et al. Perinatal transmission and maternal risks of human papillomavirus infection. Cancer Detect Prev 1995; 19:196–205.
13. Flores Y, Shah K, Lazcano E, et al. Design and methods of the evaluation of an HPV-based cervical cancer screening strategy in Mexico: The Morelos HPV study. Salud Pública Mex 2002; 44:335–344.
14. Lorincz A. Hybrid Capture method for detection of human papillomavirus DNA in clinical specimens. Pap Report 1996; 7:1–5.
15. Instituto Nacional de Estadística Geográfia e Informática INEGI. XII Censo General de Población y Vivienda México; 2000.
16. Fife KH, Katz BP, Roush J, Handy VD, Brown DR, Hansell R. Cancer-associated human papillomavirus types are selectively increased in the cervix of women in the first trimester of pregnancy. Am J Obstet Gynecol 1996; 174:1487–1493.
17. Schneider A, Hotz M, Gissmann L. Increased prevalence of human papillomaviruses in the lower genital tract of pregnant women. Int J Cancer 1987; 40:198–201.
18. Gopalkrishna V, Murthy NS, Sharma JK, et al. Increased human papillomavirus infection with the increasing number of pregnancies in Indian women. J Infect Dis 1995; 171:254–255.
19. 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.
20. Chang-Claude J, Schneider A, Smith E, Blettner M, Wahrendorf J, Turek L. Longitudinal study of the effects of pregnancy and other factor on detection of HPV. Gynecol Oncol 1996; 60:355–362.
21. Morrison EA, Gammon MD, Goldberg GL, Vermund SH, Burk RD. Pregnancy and cervical infection with human papillomaviruses. Int J Gynaecol Obstet 1996; 54:125–130.
22. Hagensee ME, Slavinsky J 3rd, Gaffga CM, Suros J, Kissinger P, Martin DH. Seroprevalence of human papillomavirus type 16 in pregnant women. Obstet Gynecol 1999; 94:653–658.
23. Terry G, Ho L, Londesborough P, Cuzick J, Mielzynska-Lohnas I, Lorincz A. Detection of high-risk HPV types by the Hybrid Capture 2 test. J Med Virol 2001; 65:155–162.
24. Schiffman M, Herrero R, Hildesheim A, et al. HPV DNA testing in cervical cancer screening. Results from women in a high-risk province of Costa Rica. JAMA 2000; 283:87–93.
25. Czegledy J, Gergely L, Endrodi I. Detection of human papillomavirus deoxyribonucleic acid by filter in situ hybridization during pregnancy. J Med Virol 1989; 28:250–254.
26. De Roda Husman AM, Walboomers JM, Hopman E, et al. HPV prevalence in cytomorphologically normal cervical scrapes of pregnant women as determined by PCR: The age-related pattern. J Med Virol 1995; 46:97–102.
27. Smith EM, Johnson SR, Jiang D, et al. The association between pregnancy and human papilloma virus prevalence. Cancer Detect Prev 1991; 15:397–402.
28. Soares VR, Nieminen P, Aho M, Vesterinen E, Vaheri A, Paavonen J. Human papillomavirus DNA in unselected pregnant and non-pregnant women. Int J STD AIDS 1990; 1:276–278.
29. Chan PK, Chang AR, Tam WH, Cheung JL, Cheng AF. Prevalence and genotype distribution of cervical human papillomavirus infection: Comparison between pregnant women and non-pregnant controls. J Med Virol 2002; 67:583–588.
30. Tenti P, Zappatore R, Migliora P, et al. Latent human papillomavirus infection in pregnant women at term: A case–control study. J Infect Dis 1997; 176:277–280.
31. Muñoz N, Franceschi S, Bosetti C, et al., for the International Agency for Research on Cancer (IARC) Multicentric Cervical Cancer. Role of parity, human papillomavirus in cervical cancer: The IARC multicentric case–control study. Lancet 2002; 359:1093–1101.
32. Anh P, Hieu T, Herrero R, et al. Human papillomavirus infection among women in South and North Vietnam. Int J Cancer 2003; 104:213–220.
33. Shin H, Lee D, Herrero R, et al. Prevalence of human papillomavirus infections in women in Busan, South Korea. Int J Cancer 2003; 103:414–421.
34. Brinton L, Reeves W, Brenes M, et al. Parity as a risk factor for cervical cancer. Am J Epidemiol 1989; 130:486–495.