Introduction

The human papillomavirus (HPV) causes the most common sexually transmitted infection (STI) in the world and can infect both skin and mucous membranes, comprising about 220 distinct subtypes [1, 2]. Of these, 40 subtypes have the potential to infect the anogenital tract, of which 12 are defined as high risk for the development of cervical cancer (CC), especially subtypes 16 and 18 [3]. However, CC has a good prognosis if diagnosed and treated early [4].

Worldwide, CC is the fourth most frequently diagnosed cancer and the fourth cause of cancer death in women, with an estimated 604,000 new cases and 342,000 deaths worldwide in 2020 [5]. In the state of Paraná, Brazil, the number of women who died due to cervical cancer was 336 in 2019 [6].

HPV infection occurs through skin microlesions or mucous membranes during sexual intercourse, favoring the entry of the virus into the basal layer of the epithelium cells [7]. Most infections are eliminated or suppressed by immune system-mediated cells between 1 and 2 years; however, persistent infections are strongly linked to the development of precancerous squamous intraepithelial lesions, typically classified as low and high grade according to the Bethesda system [8].

Although HPV infection is the main risk factor for CC, the infection alone is often an insufficient event for cancer development [8]. HPV can be detected in women with normal cytology and not every infected woman will present lesions in the cervix and evolve to cancer [9]. Many factors are associated with the HPV infection. Among them, studies cite the early onset of sexual activity, ethnicity, use of contraceptives, exposure to tobacco, immunological and genetic factors, coinfection by more than one viral subtype, and oxidative stress that may interfere with viral biology and disease progression [10,11,12,13,14].

A study conducted with adolescents showed that, in addition to smoking, passive exposure to tobacco increased the chances of HPV infection threefold in comparison to individuals not exposed to tobacco [15] (DU et al., 2020). Moreover, both Oh et al. [16] and Seo et al. [17] found that women exposed to secondhand smoke have increased likelihood of high-risk HPV persistence, especially when combined with alcohol consumption. According to Malacco et al. [18], alcohol is an important immunomodulator that can increase susceptibility to chronic diseases, including HPV infection. Regarding sexual behaviors, evidence shows that women with history of STIs and whom have had their first sexual intercourse in teenage years have 1.5 and 2.0 more chances to be infected by HPV, respectively [19].

There is a positive association between the number of sexual partners and the risk of HPV infection [20,21,22,23]. As pointed out by Winer et al. [24], the chances for HPV infection increase 25% for each new sexual partner. Not only new sexual partnerships but multiple and simultaneous partnerships are associated with increased chances of HPV infection (OR = 6.38) [25], including more serious subtypes (OR = 2.81) [26].

According to a multicenter study, the prevalence of the virus in Brazil ranges from 13.7 to 54.3%, depending on the population, geographical location, and other factors [27]. A recent study showed 25.41% as the mean prevalence in the national territory [9]. A multicenter study conducted between 2016 and 2017 detected 49.7% as the prevalence for HPV infection in women and men aged 16 to 25 years in the south of the country, the lowest prevalence among the regions evaluated; the same study reported a significant prevalence of HPV, ranging from 10.4 to 24.5%, in women without any alteration in the cervix [11].

Due to the importance of the theme for public and women’s health, the present study aimed to determine the prevalence of HPV and investigate factors associated with the infection in women in the city of Francisco Beltrão, southwest of the state of Paraná. The expectation with the results is to outline a general profile of the group with viral infection, thus contributing in the development of prevention strategies for STI, minimizing the pathological evolution of HPV infection.

Methods

Study population and ethics committee

This cross-sectional study included 324 women, aged between 18 and 65 years, who were attending public health services for routine gynecological consultation in the city of Francisco Beltrão, Paraná, Brazil. These services included Family Health Strategy (ESF) of Antonio de Paiva Cantelmo neighborhood; the Basic Health Unit (UBS) located in Cristo Rei, Industrial, and Pinheirinho neighborhoods; and a reference unit for outpatient gynecology and obstetrics care (Instituto da Mulher), located at the Cango neighborhood. The main inclusion criterion was to have had the first sexual intercourse. Pregnant women were excluded. The project was approved by the Ethics Committee in Human Research of Universidade Estadual do Paraná and the National Committee on Ethics in Research, with legal opinion number 2.254.450 and CAAE: 72,983,817.5.0000.0107. After receiving information about the research, those who agreed to participate signed the informed consent form.

Procedures

To characterize the population and to evaluate potential factors associated with HPV infection, the participants were interviewed with the aid of a structured questionnaire. Socioeconomic (age, schooling, marital status, ethnicity, and income), life habits (tobacco consumption and alcoholic beverage), and sexual and gynecological behavior (age of first sexual intercourse, number of sexual partners in life, pregnancy, contraceptive method, past and recent gynecological and vaginal infection, result of the last and current Pap test, immunization) were obtained with this instrument [28]. Subsequently, each participant went through the gynecological consultation with a specialized physician, where the material was collected for the Pap smear. The endocervical brush used in the procedure was preserved and packaged in a tube (Falcon type) with 2 ml of saline solution and kept in a freezer at − 20 °C until molecular analysis for viral detection. Storage of biological material, extraction of genetic material, and detection of HPV were processed in the Molecular Biology and Genetics Laboratory of Unioeste campus Francisco Beltrão. The results of the cytological analysis were provided by the public health services, following the Bethesda classification system (2001).

Molecular analysis

The total genetic material of the participants was isolated using a 200-µl aliquot from the original sample, following the protocol of extraction and purification of nucleic acids “Biological Fluid/blood Genomic DNA extraction kit” – “Purelink® Genomic DNA Mini Kit” (Invitrogen by Thermo Fisher Scientific, Carlsbad, California) according to the manufacturer’s instructions and stored at − 20 °C. For HPV detection, a polymerase chain reaction (PCR) technique was performed. The specific primers used for in vitro synthesis of a 450-bp fragment of the coding region of the virus L1 gene were MY09 (5′-CGTCCMAARGGAWACTGATC-3′) and MY11 (5′-GCMCAGGWCATAAYAATGG-3′). The final volume of each PCR reaction was 25 µl, containing, 10 mm Tris–HCl, 2 mm MgCl2, 0.1 mM dNTPs, 0.5 μm of each primer, and 1.25 U of Taq DNA polymerase (Ludwig Biotechnology Taq DNA Polymerase, Brazil), adding at the end 3.5 µl of the total DNA (50 ng/µl). To determine the quality of the extraction, all samples had a 268-bp segment of the human β-globin gene synthesized from primers GH20 (5′-GAAGCAGGAGGGTAC-3′) and PCO4 (5′-CAACTTCATCCTTCACCC-3′). The amplifications of both genomes were processed in applied Biosystems Veriti Thermal Cycler™ (ThermoFisher Scientific, Germany), following these steps: 10 min at 94 °C, followed by 37 cycles of 1 min at 94 °C; 1 min at 55 °C and 1 min at 72 °C; ending with extension for 10 min at 72 °C [29]. A sample of known genetic material from HeLLa cells, containing the HPV 16 genome, was included as positive control of viral detection.

All amplicons were fractioned by electrophoresis on agarose gel 2%, stained with ethidium bromide, under a potential difference of 150 V for 1 h, visualized under ultra-violet (UV) light, and photo documented.

Statistical analysis

The data were processed in the Statistical Package for the Social Sciences (SPSS) version 24.0 (IBM Corp., Armonk, NY, USA). Continuous variables had the normal distribution verified by the Kolmogorov–Smirnov test. The chi-square test (χ2), with Yates continuity correction, and Fisher’s exact test were used for comparisons between independent categorical variables and HPV infection. As effect size of bivariate analyses, Cramer’s V was used with the following classifications: weak effect > 0.05, moderate effect > 0.10, strong effect > 0.15, very strong effect > 0.25.

The independent variables with p-values < 0.20 in the bivariate analysis were included in the multivariate model of logistic regression, using the stepwise method. The results included odds ratios (OR) with robust standardized errors and 95% confidence intervals via bias-corrected and accelerated (BCa) bootstrap interval calculated with bootstrapping (1000 resamples). To select the best explanatory model of logistic regression, the Hosmer and Lemeshow test was employed (cut-off point > 0.05), together with the Omnibus test of model coefficients (cut-off point < 0.05). Moreover, a lower value of the Akaike information criterion (AIC) and increased of explained variance (R2 of Nagelkerke) were considered in the choice of the final multivariate model.

Results

The sociodemographic, gynecological, sexual, and life-related characteristics of the participants are presented in Table 1. The prevalence of HPV infection was 6.8%, that is, 22 out of the 324 women. Considering this study’s preliminary results, ethnicity and income were associated with HPV infection, with moderate effect size (according to Cramer’s V; Table 1). Smoking and alcohol consumption also were associated with STIs, although with weak effects. The number of sexual partners in life and new partners in the last year (p = 0.152 and 0.080, respectively) was the sexual behavior variables included in the analyses, with a moderate effect size for the latter. The same condition applies to gynecological variables, current cervical alterations (p < 0.001), recent vaginal infection (p = 0.029), and time of the last Papanicolaou exam (p = 0.149); among them, the first one presented the strongest effect for HPV infection, and the other two demonstrated moderate effect size (Table 1).

Table 1 Sociodemographic characteristics, life habits, and gynecological and sexual characteristics with effect for HPV infection in a group of women in southwest Paraná, Brazil
Full size table

Logistic regression analyses showed that that the presence of current cervical alterations can increase the chances of having HPV by almost 33 times (ORadj: 32.688; 95%CI: 8.508–125.589; p < 0.001). The presence of recent vaginal infection increased the chances of HPV infection by 2.7 times (ORadj: 2.773; 95%CI: 1.048–7.341; p = 0.040). The non-white ethnicity increased three times the odds of HPV infection (ORadj: 3.058; 95%CI: 1.056–8.857; p = 0.039) (Table 2).

Table 2 Multivariate analysis of factors associated with HPV infection in the group of women investigated
Full size table

Discussion

The present study found a prevalence of HPV infection of 6.8%. The main factors associated with the infection included the presence of current cervical alterations, recent vaginal infection, and non-white ethnicity. Evidence shows that the prevalence of HPV infection is higher in African and South American countries compared to other regions of the world [3]. However, generalizations should be avoided. For instance, in Ethiopia, a study showed that the prevalence of HPV in specific high-risk subtypes was relatively low in women (7.12%), being even lower among those living with HIV (3%) [19].

When comparing the prevalence of HPV infection found in this study with other locations, it was also possible to find a lower prevalence than those reported for the same age group in South America, including countries such as Chile, Paraguay, Bolivia, and Brazil (18.1%) [3]. Moreover, at a regional level, studies carried out in other cities from the same state found mixed results. For example, a prevalence of 2.2% was reported in the city of Ubiratã, whereas higher rates were found in Maringá (33.8%) and in Curitiba (48%) [30,31,32].

When comparing data with investigations carried out in southern Brazil, especially in municipalities with similar characteristics to Francisco Beltrão, variations on HPV prevalence were also found. In Cruz Alta (Rio Grande do Sul), the HPV prevalence was 34%, affecting around 15.7% women [1, 32]. In the city of Rio Grande (Rio Grande do Sul), data from 335 cervical smears indicated that HPV was detected in 24% of the samples [33]. Many factors can explain these variations, such as methodological, cultural, and sociodemographic differences.

Reinforcing the findings of this study, the prevalence of STIs was 6.7% in the municipality of Paiçandu, which is almost identical to what was found in the present study (6.8%), being equally similar to data from North American, European, and Asian countries (i.e., 6.2%) [34]. One tentative explanation is that economic development can play a strong role in the low prevalence of STIs [34].

It could be the case that Francisco Beltrão, the city with the seventh highest with human development index (HDI) in the state of Paraná, scoring above (0.774) than the capital (0.749), had lower HPV infection prevalence due to variations in education, health, and socioeconomical factors [35], thus suggesting that smaller cities might provide better access to health services when compared to major cities. Moreover, although the variable income did not remain a factor associated with HPV infection, most women who presented HPV (77.3%) belonged to the lower income group (Table 1). Indeed, poverty is a social determinant associated with women’s vulnerability and may influence the adoption of preventive measures against sexually transmitted infections, restricting access to both proper information and health services [36, 37]. However, in the present study, it was hypothesized that the low prevalence of HPV infection (6.8%) may reflect the structure and efficiency of the health care network offered by the city, allowing screenings and follow-ups of gynecological examinations.

Our results showed that only 12 women presented cervical alterations, and out of these, HPV was detected in seven (58.83%). This finding reflects those cervical alterations constitute the main risk factor associated with HPV infection (OR: 32.688) as the virus can cause severe modifications to the cervix epithelium [38, 39]. The International Agency for Research in Cancer states that about 10 to 30% of women with detectable HPV have cytological abnormalities [40]. A study observed cervical abnormalities in 3% (n = 10) of the women in their study; however, from these only one was not infected by the virus [32]. Another study found HPV in 49.7% of women with cervical alterations and 18.2% had high-grade lesions [41]. Therefore, the results of the present study reinforce those changes in the cervix have a strong relationship with HPV infection. Although cervical lesions frequently have viral presence as a leading cause, our results might suggest that tissue instability found in the lesions, along with the presence of other pathogens, could lead to cervical inflammation, hence increasing both the susceptibility of infection and persistence of HPV [42, 43]. Kumari et al. [44], in a review study, stated that a loss of the mutual relationship between the cervical microbiota and the host can lead to injuries and higher risks of HPV infection. In this context, the instability caused by parallel infections increased the chance of the presence of the virus in the group by 2.7 times. Vaginal infection, regardless of origin, may cause inflammation and rupture on the cervical epithelium, thus favoring cellular invasion by HPV [45].

Studies demonstrated that recent infections were potential risk factors for both viral and HPV infections, cervical changes, and carcinogenesis [31, 46]. Evidence exists suggesting that women with HIV might show an association between HPV infection and the vaginal microbiota composition; changes in vaginal microbiota, HPV infection, and the occurrence of bacterial vaginosis, especially Gardnerella, could cause mucosal damage, hence increasing the chances of viral infection [47, 48].

In addition to cervical alterations and vaginal infection, the multivariate model maintained ethnicity as a predictor of HPV infection. The sample investigated here was composed of women who self-declared as white, a common feature of the region that has strong Italian, German, and Polish background [49]. However, 31.8% of those who presented HPV reported other ethnicities, which revealed a threefold risk for HPV infection when compared with participants that considered themselves white. Racial or ethnic differences have been observed in incidence rates of invasive cervical cancer (up to 29% higher in black, Hispanic, and non-Hispanic women, compared to non-Hispanic and white ethnicities) [50]. Moreover, authors demonstrated a higher prevalence of HPV among the groups of non-Hispanic blacks, followed by Hispanic and non-Hispanic whites [51]. A study conducted with women with low income in the USA showed that Hispanics born in Mexico had a lower frequency of HPV (16%), followed by an increase between non-Hispanic white (29%), Hispanic born in the USA (35%), and non-Hispanic black (39%) [52]. The explanations for differences in HPV prevalence between and ethnic groups are unclear, although sociocultural conditions and genetic factors may be related [52, 53].

Although this study was the first to examine HPV prevalence in women treated in health services located in Francisco Beltrão (reference for health treatments to other localities), providing data to support preventive and interventional actions, some research limitations should be taken into account. For instance, participants’ memory lapses are cited when responding to the questionnaires, which might have interfered in the results. The limited number of women with HPV, although a positive result for the population investigated, may have neglected some risk factors for viral STI infection and also caused statistical bias. Therefore, an expansion of the sample is suggested for future studies, as well as the use of longitudinal designs to follow up the population, evaluating the perseverance of HPV infection.

Conclusion

The present study showed a low prevalence of HPV infection (6.8%) among women assisted by the Unified Health System compared to other municipalities in the state of Paraná and in other developed countries (6.2%). In summary, factors that may be linked to HPV infection were cervical alterations, recent vaginal infection, and women’s ethnicity. In this sense, the results indicate some warning signs to the service, especially pointing to the need for a better care for ethnic minorities, since their access to routine gynecological follow-up and regular cervical cancer screening may be limited.