Introduction

Cancer of the uterine cervix is the fourth most common gynecologic cancer worldwide.1 In almost all cases, cervical cancer is attributed to persistent high-risk human papillomavirus (hr-HPV) infection, with co-risk factors including early onset of sexual activity, multiple sexual partners, a history of sexually-transmitted infections, early age at first childbirth, immunosuppression, cigarette smoking, and exposure to exogenous estrogens.

According to the Regulation of the Polish Minister of Health on the organizational standards of perinatal care,2 in Poland, cervical cytology must be performed by the 10th week of pregnancy or during the first prepartal visit, if it has not been performed in the past 6 months. This Regulation conflicts with the national cervical cancer screening program which targets women aged 25–64 years who have not undergone cervical cytology within the past 3 years,3 potentially leading to overscreening in certain patients. However, for women who are infrequently screened or unable to complete appropriate follow-up, pregnancy represents an opportunity for an incidental detection of cervical cancer precursors or invasive disease.4,5

Colposcopy is employed as a diagnostic follow-up procedure to further evaluate abnormal results of cervical cancer screening tests, including cytology and / or hr-HPV testing. The primary aim of colposcopy during pregnancy is to exclude invasive disease and, for patients with preinvasive conditions, defer treatment until after delivery.6 Pregnant patients with suspected or confirmed high-grade squamous intraepithelial lesions (HSILs) corresponding to cervical intraepithelial neoplasia grade 2 or worse (CIN2+)are not candidates for expedited treatment. Instead, they undergo surveillance with colposcopy and age-based testing every 12–24 weeks.4 Surveillance interval is individualized based on gestational age, the experience of the colposcopist, who should be trained and certified by a recognized body, and the risk of loss to follow-up.7 Colposcopic features of invasive cancer are consistent between pregnant and nonpregnant women and may include abnormal vessels, irregular surface contour, mosaicism, and punctuation.6,8 However, after 20 weeks of gestation, visualizing the cervix becomes more challenging to examine due to physiological changes, such as hyperemia, making it harder to identify cancerous lesions and increasing the risk of missing a diagnosis.4,9

The primary objective of our study was to evaluate the association of potential risk factors with the likelihood of abnormal cervical cytology findings and / or hr-HPV positivity during pregnancy. Additionally, we aimed to present a possible approach to perinatal care for women on the basis of the experience of our colposcopy unit.

Methods

We collected data from medical records of pregnant women referred to our colposcopy unit between June 2021 and May 2024. The study group comprised pregnant women over 18 years of age who were receiving care in our hospital. These women were referred for colposcopy due to a cytological diagnosis of a squamous intraepithelial lesion and / or a positive hr-HPV test result. The abnormal cervical cytologies, classified according to the 2014 Bethesda System,10 included atypical squamous cells of undetermined significance (ASCUS), low-grade squamous intraepithelial lesions (LSILs), atypical squamous cells–cannot rule out HSIL, HSIL, and atypical glandular cells. A key inclusion criterion was the availability of complete electronic medical records within the Asseco Medical Management Solutions (AMMS) system (Asseco Poland S.A., Rzeszów, Poland). Patients under 18 years of age and those with incomplete electronic medical records in the AMMS system were excluded from the study.

This study was purely retrospective and utilized electronic hospital records. The data collected included patient age at the first visit, concomitant diseases documented in patient’s medical files and grouped as autoimmune and lifestyle-related, tobacco use, hormonal contraception use, HPV vaccination status, results of cytology / hr-HPV / histopathologic tests, pregnancy details, follow-up visit data, and date of delivery. The collected data were entered into a spreadsheet via Microsoft Excel, version 16.16 27 (Microsoft Corporation, Redmond, Washington, United States).

Statistical analysis

In this study, a P value below 0.05 was deemed significant. Data normality was assessed with the Shapiro‒Wilk test. For continuous variables, median with interquartile range (IQR) is reported. Categorical variables are presented as numbers and percentages. The Wilcoxon rank-sum test was used to compare continuous variables between 2 independent groups with non-normal distributions. Categorical variables were compared using the Fisher exact test, which is appropriate for small sample sizes. The analyses were performed with R software, version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria).

Ethics

This research was conducted in compliance with the Declaration of Helsinki and received approval from the research ethics committee of Jagiellonian University Medical College (118.0043.1.164.2024). All study participants provided informed consent to participate in the study.

Results

The study analyzed medical data of 59 pregnant women who were referred to our colposcopy unit mainly due to an abnormal cervical cytology result alone (49.15%), and in 40.67%, positive co-testing (cytology and HPV test). Three women (5.09%) were referred solely on the basis of a positive hr-HPV test result. Two came based on their medical history of large loop excision of the transformation zone (LLETZ) and histopathological results of LSIL (CIN1) or HSIL (CIN2+), and 1 was referred because of abnormal cervical findings caused by a treatment-resistant vaginal infection (altogether 5.09%). Median (IQR) age of the patients was 29 (18–39) years. Women in their first pregnancy were younger than those who had given birth previously. Tobacco use was reported by approximately 41% of the patients. These women smoked cigarettes or their substitutes before pregnancy, and 1 continued to smoke despite becoming pregnant. Autoimmune diseases (endocrine-like Hashimoto thyroiditis, rheumatoid arthritis, and systemic lupus erythematosus) were identified in approximately 17% of the patients. The analysis also included an examination of the HPV vaccination status; however, none of the participants had been vaccinated. Information about the hr-HPV status was available for approximately 46% of the women. The majority of those tested were positive for more than 1 HPV type, including both the 16/18 and non-16/18 groups. Four patients had a history of vaginal infection during pregnancy. Table 1 presents the indications for referral to the colposcopy unit along with demographic and clinical data.

Table 1. Indication for referral, demographic characteristics, lifestyle habits, and comorbidities of the pregnant patients attending the colposcopy unit

Parameter

Value

Indication for referral

Abnormal cytology

29 (49.15)

Abnormal cytology and positive hr-HPV test

24 (40.67)

Positive hr-HPV test

3 (5.09)

Other

3 (5.09)

Demographic characteristics

Age, y, median (IQR)

29 (18–39)

Age of primigravidae (n = 30), y, median (IQR)

28 (18–39)

Age of multigravidae (n = 29), y, median (IQR)

30 (20–37)

Cigarrette smoking

Yes

24 (40.68)

Before pregnancy

23 (38.98)

During pregnancy

1 (1.7)

Autoimmune diseases

Endocrine

8 (13.56)

Rheumatological

2 (3.39)

None

49 (83.05)

HPV vaccination

No

59 (100)

HPV status

Negative

3 (5.08)

16/18

6 (10.17)

Non-16/18

8 (13.56)

16/18 and non-16/18

10 (16.95)

Not performed or unknown

32 (54.24)

Vaginal infections

None reported

55 (93.24)

Ureaplasma urealyticum

1 (1.69)

Candida albicans

1 (1.69)

Klebsiella pneumoniae and Escherichia coli

1 (1.69)

Bacterial vaginosis

1 (1.69)

Data are presented as numbers (percentages) unless indicated otherwise.

Abbreviations: IQR, interquartile range; hr-HPV, high-risk human papillomavirus

In our cohort, 22 patients (37.3%) attended only 1 visit, 31 (52.5%) had 2 visits, and 6 (10.2%) came to 3 visits (Figure 1). Median (IQR) interval between visits 1 and 2 was 11 (5.86–24.71) weeks. For the 6 patients who attended visit 3, median (IQR) interval between visit 2 and visit 3 was 10.43 (9–15.71) weeks. Although all patients were advised to attend the postpartum visit, only 23 (38.9%) adhered to the recommendation. Median (IQR) interval between the delivery and follow-up colposcopy was 15.43 (6.71–41.71) weeks.

Figure 1. Flow diagram showing the number of women included at each stage of the study

We decided to verify whether there was an association between pregnancy and cervical cytology outcomes, in the context of the Regulation of the Polish Minister of Health on the organizational standards of perinatal care. Table 2 presents cytology results of the patients, considering 2 different periods of sample collection: before pregnancy (not later than 6 months prior) and during pregnancy (by the 10th week of pregnancy or during the first prepartal visit). In the group of women whose cervical cytology was collected during pregnancy, mostly performed in the first trimester, 90.74% had abnormal results, whereas in the nonpregnant group, this value reached 80%. No significant association was found between pregnancy and cervical cytology results (P = 0.43).

Table 2. Cytology results by sample collection period

Cytology results

Pregnancy

P value

Yes (n = 54)

No (n = 5)

Abnormal

49 (90.74)

4 (80)

0.43

Normal

5 (9.26)

1 (20)

Data are presented as numbers (percentages).

As shown in Table 3, a significant association was identified between abnormal cervical cytology results and the use of hormonal contraception prior to pregnancy (P = 0.01). All 13 women in this group exclusively used combined oral contraceptives (COCs) and exhibited better cytology outcomes than those without such a history. We also assessed the association between abnormal cervical cytology results and autoimmune disease (Table 3). The patients with autoimmune diseases had better cytology outcomes than those without them (P = 0.03).

Table 3. Cytology results according to the use of combined oral contraceptives before pregnancy and the presence of an autoimmune disease

Cytology results

Combined oral contraceptives before pregnancy

Autoimmune disease

No (n = 46)

Yes (n = 13)

No (n = 49)

Yes (n = 10)

NILM

2 (4.35)

4 (30.77)

2 (4.08)

4 (40)

AGC

0

1 (7.69)

1 (2.04)

0

ASCUS

4 (8.7)

0

4 (8.16)

0

ASC-H

20 (43.48)

3 (23.08)

5 (10.2)

1 (10)

LSIL

29 (49.15)

3 (23.08)

19 (38.78)

4 (40)

HSIL

17 (36.96)

2 (15.38)

18 (36.73)

10 (10)

P value

0.01

0.03

Data are presented as numbers (percentages).

Abbreviations: AGC, atypical glandular cell; ASCUS, atypical squamous cell of undetermined significance; ASC-H, atypical squamous cell–cannot rule out HSIL; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy

The Wilcoxon rank-sum test was used to analyze the relationships between patient age and the cervical cytology and HPV test results, which led to their referral to the colposcopy unit. There were no age differences between the patients with normal (median [IQR], 31 [25–37] years) and abnormal cytology results (median [IQR], 29 [18–39] years; P = 0.33), indicating no clear correlation between age and cervical cytology outcomes in this sample.

Among the patients with a positive HPV result, median (IQR) age was significantly lower than that in the group with a negative HPV result (28.5 [18–24] vs 35 [32–35] years, respectively; P = 0.02).

We observed considerable variations in the context of the number of pregnancies and their impact on cervical cytology and HPV test results (Table 4). The majority of patients, regardless of the number of pregnancies, had abnormal cervical cytology results, indicating a high frequency of cytologic abnormalities in the study group. These differences were insignificant (P = 0.67). In contrast, the analysis of the HPV test results showed significant differences (P = 0.02) depending on the number of pregnancies. In the group of patients with a single pregnancy, all tested individuals had a positive HPV result, whereas in the group with a greater number of pregnancies, a smaller percentage (62.5%) had a positive result.

Table 4. Comparative analysis of the number of pregnancies in relation to cytology and human papillomavirus test results

Characteristics

Number of pregnancies

P value

One (n = 30)

Two and more (n = 29)

Cytology

Abnormal

26 (86.67)

27 (93.1)

0.67

Normal

4 (13.33)

2 (6.9)

HPV results

Positive

19 (63.33)

5 (17.24)

0.02

Negative

0

3 (10.34)

HPV test not performed

11 (36.67)

21 (72.41)

–

Data are presented as numbers (percentages).

Abbreviations: see Table 1

In the smoking group (n = 24), 87.5% of the women had abnormal cytology results, whereas in the nonsmoking group (n = 35), this value was 91.43% (Table 5). Although the rates were relatively high in both groups, there was a trend toward a greater frequency of abnormal cytology findings among the nonsmokers, suggesting that other factors beyond smoking influenced cytology outcomes. The analysis of the HPV test results demonstrated that all smoking patients who were tested had positive results (Table 5). Notably, negative HPV test results were found exclusively in the nonsmoking women. This difference was insignificant (P = 0.27).

Table 5. Correlations between prenatal cytology, human papillomavirus test results, and smoking status

Characteristics

Cigarrette smoking

P value

Yes (n = 24)

No (n = 35)

Cytology

Abnormal

21 (87.5)

32 (91.43)

0.68

Normal

3 (12.5)

3 (8.57)

HPV results

Positive

10 (41.67)

14 (40)

0.27

Negative

0

3 (8.57)

HPV test not performed

14 (58.33)

18 (51.43)

–

Data are presented as numbers (percentages).

Abbreviations: see Table 1

Ultimately, the analysis focused on recommendations after delivery. The analyzed data were collected during the postpartum follow-up visit (Figure 2). During that visit, 17 patients had cervical biopsies performed, whereas only 1 woman, who obtained cervical biopsy results outside our colposcopy unit, was directly advised to undergo the LLETZ procedure. Recommendations for a follow-up visit 6 months after delivery were given to 3 patients, and 2 women were advised to attend a follow-up visit 12 months after delivery. Histopathology showed a range of cervical abnormalities. Notably, 6 patients presented with HSIL CIN3, 3 had HSIL CIN2, and LSIL was identified in 4 patients. Five patients had no dysplasia, suggesting a complete regression of the primary abnormalities.

Figure 2. Flow diagram showing the recommendations from the postpartum follow-up visit and histopathology of specimens taken postpartum

Abbreviations: CIN, cervical intraepithelial neoplasia; LLETZ, large loop excision of the transformation zone; others, see Table 3

Discussion

Pregnancy presents a unique clinical scenario where management and treatment decisions must carefully balance the risks to both the fetus and the mother against the possibility of undiagnosed cancer. The rate of progression to cervical cancer during pregnancy is not believed to differ from that in nonpregnant patients.4 In particular, HSIL during pregnancy is estimated to have a less than 2% risk of progressing to invasive carcinoma.8

In our cohort, no differences in cytology results were observed across sample collection periods; however, the group in which the smear was collected prior to pregnancy was relatively small. Pregnancy does not significantly influence the biology of cervical cancer; however, given the proposed association between estradiol and the development of HPV-related tumors, the possibility of hormone-dependent carcinogenesis cannot be entirely excluded.11 Its role in the pathogenesis of cervical adenocarcinomas and the expression of estrogen receptors in nearly all cervical cancers diagnosed during pregnancy further support this hypothesis.12-14 Additionally, alterations in maternal immunity during pregnancy, characterized by suppression of the T helper 1 cell pathway and enhancement of the regulatory T cell pathway to promote immune tolerance, could contribute to the progression of intraepithelial neoplasia.15

A review by Bowden et al16 confirmed the associations of HIV infection, immunosuppressive medications, altered vaginal microbiota, smoking, and young age at first pregnancy with cervical cancer. The majority of epithelial cell abnormalities are found in women aged 31–50 years, particularly in those who are multiparous.17 Our study group consisted of relatively young women, and we found no correlation between age and abnormal cytology results or between the number of pregnancies and abnormal cervical cytology results. However, the analysis of the relationship between HPV and age demonstrated that the average age of patients with a positive result was markedly lower than that of patients with a negative result. This finding may suggest that younger women have a greater likelihood of detectable infection, which is consistent with previous studies indicating a higher prevalence of HPV in younger age groups.18 CIN is frequently detected in young women; however, the rate of spontaneous regression is considerably greater than that in older women.19 Furthermore, in the group of women in their first pregnancy, all tested individuals had a positive HPV result, whereas in the women in their second and subsequent pregnancies, this percentage was lower. This difference may also be due to the fact that women in their first pregnancy are younger, and transient HPV infections are more common. Although primary prevention of HPV-related diseases became available in 2006, none of our patients had been vaccinated against HPV.20

The relationship between COC use and cervical oncogenesis remains a subject of debate within the literature.21,22 The heterogeneity in the active ingredients, formulations, and dosages of COCs complicates the ability to generate consistent and uniform data for robust analysis.23 Contrary to our findings, evidence indicates that an elevated risk is associated with prolonged use of COCs exceeding 5 years, whereas data concerning injectable contraceptives remain inconsistent and inconclusive.24 Kamani et al25 suggested that the impact of COCs on the lifetime incidence of cervical malignancy is primarily observed later in life, when the majority of individuals are former users. Our cohort included young women, which is why this effect may not have been observed yet.

Over 40% of our patients reported tobacco use. Surprisingly, the percentage of abnormal cervical cytology results was slightly greater in the nonsmoking group, confirming the influence of other factors, such as hr-HPV infection and, as in our cohort, the use of COCs, on abnormal cytology results. Interestingly, although insignificant, the analysis of the HPV test results showed that all tested smoking patients were HPV-positive, whereas negative HPV test results were found exclusively in the nonsmoking women. Smoking, including passive exposure, has been identified as a major risk factor for cervical cancer and its precursor lesions, second only to HPV infection.26 A recent study27 demonstrated that women who smoked had a greater risk of persistent hr-HPV infection than nonsmokers during follow-up, suggesting that smoking influences the natural history of HPV infections. Although the precise mechanism by which smoking impacts HPV infection persistence remains unclear, it is speculated that this correlation is more likely due to the effect of smoking on the immune system rather than on DNA damage, as persistence is strongly influenced by host immunity.

Although autoimmune conditions, particularly systemic lupus erythematosus, mixed connective tissue disease, and HIV infection, have been shown to increase the risk of cervical HPV infection and related diseases,28 the cervical cytology results in our cohort were worse in the patients without autoimmune diseases than those with autoimmune diseases. This may be explained by the disease type, as our group consisted primarily of women with autoimmune thyroid dysfunction rather than systemic diseases.

The indications for colposcopy during pregnancy are the same as those for other populations.4,6-8 In our group, the most common indication for a referral to a colposcopy unit was abnormal cervical cytology, ranging from ASCUS to HSIL, followed by a positive hr-HPV test results and a medical history requiring an expert check-up. Pregnant individuals with cervical dysplasia present rates of spontaneous regression or persistence of lesions similar to those of nonpregnant individuals, and the progression from cervical dysplasia to cervical cancer during pregnancy is rare.29 In our study, all women who had more than 1 visit to the colposcopy unit were advised to attend a check-up every 10–12 weeks. The follow-up intervals vary, considering the colposcopist’s experience, gestational age, and risk of loss to follow-up.4 We performed cervical biopsies during pregnancy only when colposcopy results were unclear or inconclusive, or when there was a suspicion of invasive disease. Even though adverse obstetric outcomes of cervical biopsies are rare, recommendations for performing biopsies during pregnancy vary depending on local policies. They are indicated when there is a concern for HSIL or cancer.4,5,29 Other authors recommend biopsies only if there is suspicion of invasion at colposcopy.6-8,30

We acknowledge several limitations of our study. First, the small sample size, coupled with significant loss to follow-up, limits the ability of the study to accurately capture the natural course of squamous intraepithelial lesions and HPV infection and negatively affects its statistical power.

Between June 2021 and May 2024, approximately 90 000 children were born in the province where our colposcopy unit operates.31 Given that an estimated 5% of cervical cytologies are abnormal, approximately 4500 pregnant women in the region would require further diagnostic follow-up.32-34 However, only 1.3% (59/4500) were referred to our colposcopy unit.

Second, liquid-based cytologies (LBCs) are available under public health insurance only in a limited number of centers, requiring most women to pay out-of-pocket for this service. Consequently, our study included both conventional cervical cytology (Pap smear) and LBC. Similarly, HPV testing is not part of the national screening program and is often prohibitively expensive for patients, further limiting its routine use.

Conclusions

Pregnancy presents a valuable opportunity for the detection of cervical abnormalities. Despite the lack of increased risk of progression to cancer during pregnancy, vigilant colposcopy and subsequent histopathological evaluation remain crucial, especially for women with identified risk factors.