ABSTRACT
Objective
To assess the impact of thermal ablation (TA) for the treatment of cervical dysplasia on fertility and pregnancy outcomes among women screened for cervical cancer in Cameroon.
Design
Retrospective cohort study.
Setting
Dschang health district, Cameroon.
Population
Participants aged 30–45 years from two screening trials conducted between 2015 and 2020 in Dschang District Hospital.
Methods
Participants were primarily screened for human papillomavirus infection, triaged by visual inspection and treated by TA if needed. Between October 2021 and March 2022, interviews on subsequent fertility were conducted with participants treated by TA and a control group of untreated women.
Main Outcome Measures
Pregnancy and miscarriage after screening/treatment.
Results
A total of 760 participants (219 treated and 541 untreated) completed the survey, with a mean follow‐up time of 1297 days. Sixty‐two women (28.3%) treated by TA reported a pregnancy postscreening versus 165 (30.5%) in the control group (p = 0.550). Adjusted for potential confounders, the hazard ratio of pregnancy for treated compared with untreated women was 0.82 (0.54–1.24, p = 0.350). Among women reporting pregnancies with a known outcome and which were not voluntarily terminated, 18 (35.3%) treated participants had a miscarriage versus 31 (21.4%) in the control group (p = 0.048). In the adjusted model, no association remained between TA and miscarriage (1.04, 0.39–2.78, p = 0.935).
Conclusions
In our study population, TA did not significantly impact fertility nor miscarriage risk. Our results support the widespread use of TA as a treatment of choice for precancerous cervical lesions in low‐income settings.
Keywords: cervical cancer, fertility, miscarriage, pregnancy, screening, thermal ablation
1. Introduction
Cervical cancer (CC) is the second most frequent cancer in sub‐Saharan Africa and the most deadly, despite being largely preventable [1]. Significant global disparities in this preventable cancer prompted the World Health Organization (WHO) to launch a global initiative in 2020 aiming to eliminate CC as a public health issue [2]. The WHO initiative was designed to meet the following targets by 2030: (i) 90% of girls vaccinated against HPV, (ii) 70% of women screened with a high‐performance test and (iii) 90% of women with precancerous/cancerous lesions receiving treatment.
WHO guidelines for CC prevention recommend several suitable options for treating precancerous lesions in low‐ and middle‐income countries (LMICs). Recommended therapies include ablative or excisional methods, depending on the extent and visibility of the lesion. Among available ablative therapies, thermal ablation (TA) with a heated metallic probe is recommended as an alternative to cryotherapy since 2019 [3]. TA has the advantage of being particularly adapted to low‐resource settings, as it requires simple reusable equipment and little training, while achieving high cure rates of precancerous lesions [4, 5], with mainly minimal side effects [6]. However, the safety of TA for future fertility and pregnancy outcomes is still unclear.
Assessing the effect of cervical treatment on future fertility is essential in LMICs, as the primary target population for CC screening are women aged 30–49 years with reproductive potential [7]. In this population, adherence to treatment and screening coverage may be reduced by fear of infertility [8, 9]. The safety of cervical treatment is especially relevant in resource‐constrained settings where international screening guidelines recommend a ‘screen‐and‐treat approach’ to avoid loss to follow‐up, leading to a high proportion of women of reproductive age receiving treatment [7]. The drawback of this strategy is that most women with a positive primary HPV test are overtreated, since 90% of HPV‐positive women do not have high‐grade cervical lesions, although they could develop them in the next years [10, 11].
The risk of adverse effects on fertility and early pregnancy outcomes after local cervical treatment has mostly been studied for excisional therapies and did not show evidence for affected fertility [12]. The effect on fertility of ablative therapies such as cryotherapy or laser vaporisation was assessed only in three studies in high‐income countries [13, 14, 15]. To date, long‐term post‐operative fertility and pregnancy outcomes following TA have not been evaluated in resource‐constrained settings. Our objective was to assess the impact of TA on fertility and pregnancy outcomes among women screened for CC in Cameroon.
2. Methods
2.1. Study Design
The TaCCare (‘Evaluation of Thermal ablation in Cervical Cancer screening on reproductive outcomes’) study was a retrospective cohort study conducted with participants from two previous CC screening trials taking place at the Dschang District Hospital (now renamed Dschang Regional Hospital Annex) since 2015.
2.2. Setting and Screening Procedure
Since 2015, a collaboration between the Geneva University Hospitals (HUG), Dschang District Hospital, the University of Dschang and the Cameroonian Ministry of Health, has given rise to a CC screening programme in the health district of Dschang, in the West Region of Cameroon. The screening program has been part of two clinical trials evaluating the real‐life effectiveness of a same‐day test‐triage‐and‐treat approach to CC screening in Cameroon. The first trial was conducted between 2015 and 2016 [16], and the second trial started in 2018 and was still ongoing at the time of the TaCCare study [17]. Participants aged 30–49 years were primarily screened for HPV infection; HPV‐positive women were triaged by visual inspection after application of acetic acid (VIA) and Lugol's iodine (VILI), and treated on the same day by TA if VIA/VILI positive. VIA/VILI was assessed by naked eye enhanced by smartphone digital photography, using the previously developed ABCD criteria for VIA evaluation [18]. HPV status was determined by a point‐of‐care HPV assay (Xpert HPV, GeneXpert technology, Cepheid, Sunnyvale, CA), which identifies 14 high‐risk genotypes. PAP‐tests, endocervical brushing and cervical biopsies were performed on all HPV‐positive participants for quality control and were analysed at the Pathology Division of Geneva University Hospitals, Switzerland. Participants with suspected cancer were referred to a gynaecologist or gynaecologic oncologist for further evaluation. HPV‐negative women were advised to repeat screening after 5 years, respectively, 3 years if HIV‐positive, while HPV‐positive women were invited for a follow‐up visit after 1 year. Following local guidelines, HPV‐positive participants were advised to avoid pregnancy until their 12‐month follow‐up visit, and those treated by TA were additionally advised to refrain from penetrative intercourse for 4 weeks. If the laboratory analyses identified a high‐grade lesion that was missed upon visual examination, patients were called back for treatment by TA or large loop excision of the transformation zone (LLETZ), depending on the site of the lesion. Patients were eligible for TA if the lesion did not extend into the endocervix beyond reach of the thermal probe. TA was performed by trained midwives using a WISAP probe (Medical Technology GmbH, Brunnthal/Hofolding, Germany) heated at 100°C and applied for 60 s on the cervical transformation zone and suspected lesion when present. The treatment depth achieved in these conditions is between 3 and 4 mm [19, 20]. If necessary, the application was repeated several times until the entire abnormal area and transformation zone were covered.
2.3. Recruitment Procedure and Data Collection
Participants were eligible for the TaCCare study if they had been screened in one of the trials conducted in Dschang District Hospital between 2015 and 2020 and were aged 30–45 years at the time of screening. Women having received cervical treatment other than TA, as well as women having undergone a hysterectomy, were excluded from the study. Women constituting the ‘treatment group’ were all participants treated by thermal ablation within one of the two source trials, while participants of the control group were untreated women from the source trials, both HPV‐positive and ‐negative.
Data collection took place between October 2021 and March 2022, during which women from both original screening trials were invited to answer a phone survey (Figure 1). Interviews were led by two Cameroonian medical anthropologists trained in CC screening and a Cameroonian physician, based on a paper questionnaire structured along the following sections: (i) update of CC screening and treatment, (ii) medical history and risk factors, (iii) obstetrical history since CC screening and (iv) obstetrical history prior to screening. The questionnaire was designed by a team of physicians specialised in gynaecology‐obstetrics and public health, and medical anthropologists, and was further adapted after being pretested on 10 participants. Sociodemographic data, previous screening results and treatment received, contraception use, tobacco consumption and human immunodeficiency virus (HIV) status, were retrieved from the data sets of the original screening trials. Participants who did not answer their phone on the first try were called again at least twice, at different times of the day and week. If no answer was received after three tries, a text message was sent inviting the participant to call back, and a relative indicated by the participant as ‘additional contact’ was called to update the participant's contact information. All participants provided informed oral consent over the phone to participate in the study.
FIGURE 1.
Kaplan–Meier curve for pregnancy probability by treatment received.
2.4. Statistical Analyses
The primary outcome was pregnancy since screening/treatment, and the secondary outcome was miscarriage. Only the first pregnancy occurring after screening/treatment for each participant was considered in the analysis. Core outcome sets were not considered at the time of trial design.
The sample size was determined based on the number of eligible participants treated by TA (N = 296) with an estimated participation rate of 80%. The size of the control group was calculated using Pearson's chi‐squared test to obtain 80% power with a 95% confidence level (95% CI) for an estimated 10% difference in pregnancy rates between both groups (30% in the control group vs. 20% in the TA group) over the full study period (estimated at an average of 3 years), giving a necessary sample size for the control group of 441 participants. Considering a slightly lower participation rate of 70% in the control group, approximately 670 participants were considered necessary to be recruited. To reach this sample size, all eligible untreated HPV‐positive participants (N = 159) were included, as well as 505 additional HPV‐negative participants (rounded to obtain a total sample of 960 participants).
Descriptive statistics summarised categorical variables as percentages stratified by exposure group and compared using the chi‐squared test, and numerical variables as means with their standard deviations and compared using the t‐test. Because of the high number of participants who did not know their HIV status and body mass index (BMI) and their unequal distribution across exposure groups, unknown HIV status and BMI were kept in the analyses as distinct categories. p values were considered statistically significant at <0.05.
Reverse Kaplan–Meier survival curves were plotted to graph the probability of pregnancy by exposure group and compared using the log rank test. Having a cervical treatment other than TA after initial screening was treated as censored data. Pregnancy incidence rates and their 95% CI were calculated per 100 person‐years. Cox regression models were used to examine the effect of TA on pregnancy rates, expressed as hazard ratios (HR). Potential confounders of the association between TA and subsequent pregnancy were identified a priori based on published literature and clinical experience, and included age, gravidity, parity, relationship status, tobacco consumption, reported infertility prior to initial screening, contraception use, BMI, suffering from a chronic disease, history of gynaecologic disease/surgery, HIV status, HPV status and presence of cervical intraepithelial neoplasia grade 2 or higher (CIN2+) at initial screening. Among these, variables showing no strong association (p > 0.1) with TA or with pregnancy rates were excluded from the adjusted model. CIN2+ at initial screening was excluded from the adjusted model due to low counts across categories. Gravidity was also excluded due to collinearity with parity. As the primary analysis did not consider participants' desire for pregnancy, a subanalysis was conducted excluding participants without any reported pregnancies and who did not attempt to get pregnant during the follow‐up period.
For the secondary outcome of miscarriage, analyses included participants reporting at least one pregnancy after initial screening/treatment. Participants still pregnant at the time of the survey and those reporting voluntary abortion were excluded from the analysis. Considering that miscarriage was a common outcome in our study population (>10%), risk ratios (RR) were calculated for the crude analysis. Logistic regression models were used to examine the association between TA and miscarriage adjusted for potential confounders. As odds ratios (OR) are known to overestimate (if OR > 1) or underestimate (if OR < 1) the effect when the incidence of an outcome is common, adjusted ORs of more than 2.5 or less than 0.5 were corrected based on the formula proposed by Zhang et al. [21] A priori potential confounders of the association between TA and miscarriage included age, tobacco consumption, BMI, suffering from a chronic disease, history of gynaecologic disease/surgery, history of miscarriage, HIV status, HPV status and presence of CIN2+ at time of initial screening. As for the primary analysis, the presence of CIN2+ and variables showing no strong association (p > 0.1) with TA or with subsequent miscarriage were excluded from the adjusted model.
2.5. Patient and Public Involvement
Patients and the general public did not play a direct role in shaping the study's design or its implementation. Nevertheless, the research question took into consideration patient priorities and experiences. Notably, in both source studies, over 50% of women who underwent thermal ablation expressed a desire for future pregnancies.
To ensure that participants felt at ease when responding to the study survey, the questions were administered by a local anthropologist and physician who were well‐versed in the local culture and vocabulary. The survey was also collaboratively developed with local researchers, ensuring that the questions were culturally appropriate.
The findings of the study will be conveyed to community members in a simplified manner through a biannual newsletter published by our research group. This newsletter is a result of collaboration between our Swiss and Cameroonian teams. It is distributed to local partners within the healthcare system and throughout the community with the assistance of community health workers and women's associations.
3. Results
Out of 2508 participants from both source trials aged 30–45 years, 16 patients meeting exclusion criteria, 11 included in the pretest and two with no contact information were removed from the sampling frame (Figure S1). Of the remaining eligible participants, 296 had received treatment by TA and 2183 had not, among which all 159 HPV‐positive women were included in the study and 505 HPV‐negative women were randomly selected. A total of 760 participants (219 treated by TA) completed the survey and were followed up for an average of 1297 days (range 244–2432). Follow‐up duration was similar between treated (mean 1273 days, SD ± 683), untreated HPV‐positive (mean 1113 ± 637) and untreated HPV‐negative (mean 1366 ± 696) women. Baseline characteristics of responders versus nonresponders are available in the supplement (Table S1).
Participants had a mean age of 36.9 years (SD ± 4.9) in the treated group and 37.5 (±4.8) in the control group (p = 0.102) (Table 1). Treated participants were more frequently without a partner at the time of initial screening (16.0% vs. 10.2%, p = 0.026). There was a larger proportion of HIV‐positive participants in the treated group (8.2% vs. 3.9%, p < 0.001), as well as participants with a BMI <25 kg/m2 (18.7% vs. 10.7%), and a lower proportion with a BMI >30 kg/m2 (27.4% vs. 37.2%, p = 0.001). Treated participants had had less pregnancies prior to screening (mean 4.8 vs. 5.3, p = 0.007) and less children (mean 3.8 vs. 4.2, p = 0.023). A higher proportion of women in the treated group reported past infertility (22.9% vs. 15.4%, p = 0.014), whereas past caesarean sections were less frequent (9.2% vs. 14.4%, p = 0.050).
TABLE 1.
Participants' characteristics stratified by exposure to thermal ablation.
| Treated (n = 219), N (%) | Not treated (n = 541), N (%) | p | |
|---|---|---|---|
| Age a (mean ± SD) | 36.9 ± 4.9 | 37.5 ± 4.8 | 0.102 |
| Relationship status | 0.026 | ||
| With partner | 184 (84.0) | 483 (89.8) | |
| Without partner | 35 (16.0) | 55 (10.2) | |
| Tobacco consumption | 0.626 | ||
| Yes b | 8 (3.7) | 24 (4.4) | |
| No | 211 (96.3) | 517 (95.6) | |
| Chronic disease | 0.299 | ||
| Yes | 33 (15.2) | 67 (12.4) | |
| No | 184 (84.8) | 474 (87.6) | |
| HIV status | <0.001 | ||
| Positive | 18 (8.2) | 21 (3.9) | |
| Negative | 112 (51.1) | 209 (38.6) | |
| Unknown | 89 (40.6) | 311 (57.5) | |
| Gynaecological disease | 0.801 | ||
| Yes | 74 (34.4) | 180 (33.5) | |
| No | 141 (65.6) | 358 (66.5) | |
| BMI | 0.001 | ||
| <25 | 41 (18.7) | 58 (10.7) | |
| 25–30 | 50 (22.8) | 149 (27.5) | |
| >30 | 60 (27.4) | 201 (37.2) | |
| Unknown | 68 (31.1) | 133 (24.6) | |
| Contraception use c | 0.157 | ||
| No | 92 (42.0) | 196 (36.5) | |
| Yes | 127 (58.0) | 341 (63.5) | |
| Obstetrical history at initial screening | |||
| Gravidity a (mean ± SD) | 4.8 ± 2.3 | 5.3 ± 2.5 | 0.007 |
| 0–4 | 93 (42.5) | 191 (36.0) | 0.099 |
| ≥5 | 126 (57.5) | 339 (64.0) | |
| Parity a (mean ± SD) | 3.8 ± 2.0 | 4.2 ± 2.0 | 0.023 |
| 0–4 | 128 (58.5) | 282 (53.3) | 0.199 |
| ≥5 | 91 (41.5) | 247 (46.7) | |
| Past miscarriage(s) | 0.738 | ||
| Yes | 84 (38.7) | 202 (37.4) | |
| No | 133 (62.2) | 338 (62.6) | |
| Past infertility d | 0.014 | ||
| Yes | 49 (22.9) | 83 (15.4) | |
| No | 165 (77.1) | 457 (84.6) | |
| Premature birth | 0.525 | ||
| Yes | 13 (6.0) | 39 (7.3) | |
| No | 204 (94.0) | 496 (92.7) | |
| Caesarean section | 0.050 | ||
| Yes | 20 (9.2) | 78 (14.4) | |
| No | 198 (90.8) | 462 (85.6) | |
| Adverse obstetrical outcomes | 0.100 | ||
| No | 201 (92.2) | 475 (88.1) | |
| Yes | 17 (7.8) | 64 (11.9) | |
| Stillbirth | 10 (4.6) | 54 (10.0) | |
| Peripartum haemorrhage | 2 (0.9) | 6 (1.1) | |
| Other e | 5 (2.3) | 4 (0.7) | |
At the first screening visit (or at the time of thermal ablation for age).
Includes past consumption (n = 3).
Current or in the period of time since the first screening visit or date of thermal ablation.
Defined as self‐reported failure to achieve pregnancy after 12 months of unprotected sexual intercourse.
Cervical tear (n = 3), uterine rupture (n = 1) or not specified (n = 5).
Two‐hundred and twenty‐seven participants reported having at least one pregnancy since screening or treatment. Twenty‐eight per cent (n = 62) of treated women reported a pregnancy compared with 30.5% (n = 165) in the control group (p = 0.550) (Table S2). Time to first pregnancy postscreening/treatment ranged between 0 and 65 months, with a median of 12 months (interquartile range 6–18). Distribution of time to pregnancy varied according to HPV status and treatment group (Figure S2), with a peak of pregnancies observed after 1 year among untreated HPV‐positive women, mirroring the recommendations given by healthcare providers to this group of participants to avoid pregnancy until their follow‐up visit.
The Kaplan–Meier curves showed no significant difference in probability of pregnancy between treatment groups (p = 0.543) (Figure 1). The adjusted hazard ratio of pregnancy for treated women was 0.82 (95% CI: 0.54–1.24, p = 0.350) (Table 2). When excluding women who did not attempt pregnancy, the probability of pregnancy appeared to be slightly higher in the control group on the Kaplan–Meier curves with borderline significance (p = 0.072) (Figure S3). However, the adjusted analysis showed no significant association (aHR 0.92 [0.60–1.40], p = 0.689) (Table S3). To account for the high proportion of missing values for BMI and HIV status, a sensitivity analysis was conducted excluding participants with missing values, which did not significantly affect the results (Table S4).
TABLE 2.
Incidence rates and hazard ratios of pregnancy, crude and adjusted for participants' sociodemographic and medical characteristics.
| Variables | Number of pregnancies (% of category) | Person‐years | Pregnancy IR a (95% CI) | Crude HR (95% CI) | p | Adjusted HR (95% CI) | p |
|---|---|---|---|---|---|---|---|
| Thermal ablation | |||||||
| No | 165 (30.5) | 1509 | 10.9 (9.4–12.7) | Ref | Ref | ||
| Yes | 62 (28.3) | 618 | 10.0 (7.8–12.9) | 0.91 (0.68–1.22) | 0.547 | 0.82 (0.54–1.24) | 0.350 |
| Age (years) b | |||||||
| <35 | 147 (58.8) | 484 | 30.3 (25.8–35.7) | ||||
| 35–40 | 64 (23.5) | 807 | 7.9 (6.2–10.1) | 0.81 (0.78–0.84) | <0.001 | 0.82 (0.79–0.85) | <0.001 |
| >40 | 16 (6.6) | 843 | 1.9 (1.2–3.1) | ||||
| Relationship status | |||||||
| Without partner | 25 (27.8) | 196 | 12.7 (8.6–18.9) | Ref | Ref | ||
| With partner | 201 (30.1) | 1921 | 10.5 (9.1–12.0) | 1.03 (0.68–1.56) | 0.886 | 1.10 (0.71–1.72) | 0.668 |
| HPV status b | |||||||
| Negative | 124 (29.4) | 1216 | 10.2 (8.5–12.2) | Ref | Ref | ||
| Positive | 103 (30.4) | 910 | 11.3 (9.3–13.7) | 1.04 (0.80–1.35) | 0.768 | 0.96 (0.66–1.42) | 0.852 |
| CIN2+ b | |||||||
| No | 223 (30.3) | 2057 | 10.8 (9.5–12.4) | Ref | — | ||
| Yes | 4 (16.7) | 69 | 5.8 (2.2–15.4) | 0.55 (0.20–1.48) | 0.235 | ||
| HIV status | |||||||
| Negative | 129 (40.2) | 803 | 16.1 (13.5–19.1) | Ref | Ref | ||
| Positive | 9 (23.1) | 96 | 9.4 (4.9–18.0) | 0.58 (0.29–1.13) | 0.110 | 1.71 (0.75–3.91) | 0.204 |
| Unknown | 89 (22.2) | 1227 | 7.3 (5.9–8.9) | 0.50 (0.38–0.66) | <0.001 | 0.80 (0.60–1.07) | 0.129 |
| Gravidity b | |||||||
| 0–4 | 116 (40.9) | 725 | 16.0 (13.3–19.2) | Ref | — | ||
| ≥5 | 108 (23.2) | 1390 | 7.8 (6.4–9.4) | 0.51 (0.39–0.66) | <0.001 | ||
| Parity b | |||||||
| 0–4 | 158 (38.5) | 1063 | 14.9 (12.7–17.4) | Ref | Ref | ||
| ≥5 | 65 (19.2) | 1049 | 6.2 (4.9–7.9) | 0.44 (0.33–0.59) | <0.001 | 0.78 (0.57–1.08) | 0.137 |
| Past infertility b , c | |||||||
| No | 191 (30.7) | 1733 | 11.0 (9.6–12.7) | Ref | Ref | ||
| Yes | 36 (27.3) | 378 | 9.5 (6.9–13.2) | 0.87 (0.61–1.25) | 0.457 | 0.71 (0.47–1.06) | 0.095 |
| Chronic disease | |||||||
| No | 204 (31.0) | 1783 | 11.4 (10.0–13.1) | Ref | Ref | ||
| Yes | 23 (23.0) | 339 | 6.8 (4.5–10.2) | 0.67 (0.44–1.03) | 0.070 | 0.80 (0.48–1.35) | 0.412 |
| Gynaecological disease | |||||||
| No | 134 (26.9) | 1426 | 9.4 (7.9–11.1) | Ref | Ref | ||
| Yes | 90 (35.4) | 686 | 13.1 (10.7–16.1) | 1.37 (1.05–1.79) | 0.021 | 1.06 (0.80–1.41) | 0.675 |
| Contraception use d | |||||||
| No | 71 (24.6) | 888 | 8.0 (6.3–10.1) | Ref | Ref | ||
| Yes | 156 (33.3) | 1234 | 12.6 (10.8–14.8) | 1.44 (1.09–1.91) | 0.011 | 0.99 (0.73–1.33) | 0.945 |
| BMI (kg/m2) | |||||||
| <25 | 33 (33.3) | 239 | 13.8 (9.8–19.4) | Ref | Ref | ||
| 25–30 | 66 (33.2) | 538 | 12.3 (9.6–15.6) | 0.97 (0.64–1.48) | 0.895 | 1.19 (0.77–1.83) | 0.433 |
| >30 | 81 (31.0) | 811 | 10.0 (8.0–12.4) | 0.87 (0.58–1.30) | 0.496 | 1.10 (0.72–1.68) | 0.658 |
| Unknown | 47 (23.4) | 539 | 8.7 (6.6–11.6) | 0.66 (0.42–1.03) | 0.068 | 1.04 (0.65–1.66) | 0.868 |
| Tobacco consumption | |||||||
| No | 221 (30.4) | 2026 | 10.9 (9.6–12.4) | Ref | — | ||
| Yes e | 6 (18.8) | 100 | 6.0 (2.7–13.3) | 0.55 (0.25–1.24) | 0.152 | ||
Note: Significant results at the p < 0.05 level in bold.
Abbreviations: BMI, body mass index; CI, confidence interval; CIN2+, cervical intraepithelial neoplasia of grade 2 or more; HIV, human immunodeficiency virus; HPV, human papillomavirus; HR, hazard ratio; IR, incidence rate; Ref, reference.
Per 100 person‐years.
At first screening visit.
Defined as failure to achieve pregnancy after 12 months of unprotected sexual intercourse.
Current or past, in the period of time since the first screening visit or date of thermal ablation.
Includes past tobacco consumption.
Of the 227 first postscreening pregnancies reported, 11 were voluntarily terminated, 49 ended in spontaneous abortion, 147 were delivered (livebirth or stillbirth), and 20 were ongoing at the time of data collection (Table S2). In the crude analysis, the proportion of miscarriages was higher among pregnant women who had been treated (35.3%) compared with the control group (21.4%, p = 0.048) (Figure 2A). However, when further stratified by HPV status, the proportion of miscarriages did not differ between HPV‐positive women treated or not (Figure 2). After adjusting for potential confounders including HPV status, there remained no association between treatment by TA and subsequent miscarriage (aOR 1.04 [0.39–2.78], p = 0.935) (Table 3).
FIGURE 2.
Proportion of miscarriages in post‐screening pregnancies stratified by (A) treatment group and (B) treatment group and HPV status. Error bars represent 95% confidence intervals. TA, thermal ablation; HPV+, Human papillomavirus‐positive participants, HPV−, HPV‐negative participants.
TABLE 3.
Odds ratios of miscarriage according to pregnant participants' sociodemographic and medical characteristics (N = 196).
| Characteristics | Miscarriage a | Risk ratio (95% CI) | p | Adjusted OR b (95% CI) | p | |
|---|---|---|---|---|---|---|
| No, n (%) | Yes, n (%) | |||||
| Thermal ablation | ||||||
| No | 114 (78.6) | 31 (21.4) | Ref | Ref | ||
| Yes | 33 (64.7) | 18 (35.3) | 1.65 (1.02–2.68) | 0.048 | 1.04 (0.39–2.78) | 0.935 |
| Age at pregnancy | ||||||
| <35 | 105 (81.4) | 24 (18.6) | ||||
| 35–40 | 31 (62.0) | 19 (38.0) | 1.10 (1.02–1.20) | 0.018 | 1.11 (1.02–1.21) | 0.019 |
| >40 | 11 (64.7) | 6 (35.3) | ||||
| HPV status c | ||||||
| Negative | 93 (82.3) | 20 (17.7) | Ref | Ref | ||
| Positive | 54 (65.1) | 29 (34.9) | 1.97 (1.20–3.24) | 0.006 | 2.26 (1.15–3.58) d | 0.021 |
| Presence of CIN2+ c | ||||||
| No | 145 (75.1) | 48 (24.9) | Ref | — | ||
| Yes | 2 (66.7) | 1 (33.3) | 1.34 (0.27–6.77) | 0.737 | ||
| HIV status | ||||||
| Negative | 83 (76.9) | 25 (23.1) | Ref | Ref | ||
| Positive | 7 (87.5) | 1 (12.5) | 0.54 (0.08–3.49) | 0.486 | 0.44 (0.05–2.22) d | 0.392 |
| Unknown | 57 (71.3) | 23 (28.7) | 1.24 (0.76–2.02) | 0.384 | 1.41 (0.69–2.87) | 0.350 |
| Parity c | ||||||
| 0–4 | 103 (74.1) | 36 (25.9) | Ref | — | ||
| ≥5 | 44 (80.0) | 11 (20.0) | 0.77 (0.42–1.41) | 0.387 | ||
| Previous miscarriage | ||||||
| No | 93 (75.6) | 30 (24.4) | Ref | — | ||
| Yes | 53 (73.6) | 19 (26.4) | 1.08 (0.66–1.78) | 0.756 | ||
| Chronic disease | ||||||
| No | 132 (75.9) | 42 (24.1) | Ref | |||
| Yes | 15 (68.2) | 7 (31.8) | 1.32 (0.68–2.57) | 0.433 | — | |
| Gynaecological disease | ||||||
| No | 92 (79.3) | 24 (20.7) | Ref | — | ||
| Yes | 54 (70.1) | 23 (29.9) | 1.44 (0.88–2.37) | 0.146 | ||
| BMI (kg/m2) | ||||||
| <25 | 19 (67.9) | 9 (32.1) | Ref | Ref | ||
| 25–30 | 42 (75.0) | 14 (25.0) | 0.78 (0.38–1.57) | 0.489 | 0.73 (0.25–2.14) | 0.568 |
| >30 | 54 (76.1) | 17 (23.9) | 0.74 (0.38–1.47) | 0.404 | 0.60 (0.22–1.67) | 0.331 |
| Unknown | 32 (78.0) | 9 (22.0) | 0.68 (0.31–1.50) | 0.344 | 0.51 (0.18–1.19) d | 0.132 |
| Tobacco consumption | ||||||
| No | 145 (75.9) | 46 (24.1) | Ref | Ref | ||
| Yes e | 2 (40.0) | 3 (60.0) | 2.49 (1.17–5.32) | 0.067 | 2.53 (0.75–3.81) d | 0.110 |
Note: Data presented does not include participants pregnant at the time of the survey or voluntary abortions.
Abbreviations: BMI, body mass index; CIN2+, cervical intraepithelial neoplasia grade 2 or worse; HIV, human immunodeficiency virus; HPV, human papillomavirus; OR, odds ratio; Ref, reference.
Includes intrauterine fetal death (>24 gestational weeks).
OR adjusted for thermal ablation, age, HPV status, HIV status, BMI category and tobacco consumption.
At first screening visit.
Adjusted OR corrected for common outcome (only for original adjusted ORs of >2.5 or <0.5).
Includes past tobacco consumption.
4. Discussion
4.1. Main Findings
In this cohort of women screened for CC in the West Region of Cameroon, TA did not significantly impact fertility (aHR 0.82 [0.54–1.24], p = 0.350) or risk of miscarriage (aOR 1.04 [0.39–2.78], p = 0.935) after adjusting for potential confounders. Interestingly, HPV‐positive women were twice as likely to have a miscarriage than HPV‐negative women.
4.2. Strengths and Limitations
Strengths of this study include the availability of reliable data from previous clinical trials on cervical treatment received and HPV status, therefore minimising the risk of misclassification. Further, a high participation rate of nearly 80% was achieved thanks to a rigorous recall procedure, limiting the risk of selection bias. Although our study fills an important evidence gap on the safety of TA for women's fertility and pregnancy outcomes, several study limitations should be acknowledged. First, women were included in the analysis irrespective of their desire of pregnancy at the time of screening, which may have confounded the association with fertility due to potentially unequal reproductive behaviours across study groups. To mitigate this, a sensitivity analysis was conducted considering retrospectively reported desire for pregnancy, which did not differ substantially from the main analysis. However, potential reporting bias due to stigma around infertility may have led to misclassification of self‐reported desire for pregnancy. The study population also varied considerably in the duration of follow‐up, due to the 5‐year screening period and censoring of a limited number of participants having received a different cervical treatment. These variations, however, were accounted for in the survival analysis. Second, differing recommendations regarding reproductive behaviour were given to HPV‐negative and HPV‐positive women. Thus, behaviour was likely to vary across study groups during the first year of follow‐up, which may have further confounded the association with fertility. We have attempted to account for this potential imbalance by adjusting for HPV status at the initial screening visit, although residual confounding cannot be excluded. Third, there were some differences in baseline characteristics between the study groups, most notably the roughly 50% higher prevalence of reported past infertility among women in the thermal ablation group. These baseline disparities, however, were included as potential confounders in the adjusted analysis. Fourth, generalisability of our results cannot be assumed as the study was conducted in a specific setting in the West Region of Cameroon where population characteristics such as HPV and HIV prevalence, as well as reproductive behaviour, may impact fertility outcomes. Finally, our study sample did not provide sufficient power to assess absolute differences in fertility of less than 10%, as well as less frequent pregnancy outcomes such as preterm birth, late miscarriages, premature membrane rupture, low birthweight or ectopic pregnancy, which have been associated with excisional cervical treatment [12, 22].
4.3. Interpretation
To the best of our knowledge, this is the first study to specifically assess the effect of TA on subsequent fertility. A meta‐analysis published in 2014 [12] assessing fertility and early pregnancy outcomes after treatment of patients with suspected cervical precancerous disease identified only three studies assessing ablative methods [13, 14, 23], none of which included patients treated by TA. Pregnancy rate was higher among patients treated by laser ablation than an untreated control group, and time to conception was not significantly associated with any ablative treatment. More recently, two cohort studies in North America assessed pregnancy rates following ablative treatment, showing either increased pregnancy rates for grouped ablative procedures, including thermal cautery [24], or no significant effect for laser treatment and cryotherapy [15]. However, the effect of TA was not individually assessed in either study. The safety of cryotherapy on fertility and early pregnancy outcomes has also been shown in older studies, albeit with low‐grade quality of evidence, contributing to its widespread use as a screen‐and‐treat approach in low‐resource settings before the advent of TA [25, 26]. Despite some differences across studies, which may be explained by heterogeneous study populations and adjustment for different variables, none of the published research has shown significantly reduced fertility following ablative treatment, which is consistent with our findings. Our study is also the first to assess the effect of an ablative treatment on fertility in a low‐income setting.
Although pregnant women treated by TA in our study had a higher proportion of miscarriages, no association was observed after adjusting for potential confounders. Previous literature published on the topic is scarce and has provided conflicting results. Similar to our findings, a study led in the United Kingdom [27] observed a higher overall rate of miscarriages in women treated by TA than in untreated women, but the analysis was not adjusted for potential confounders. On the contrary, an earlier cohort study of women treated by TA between 1975 and 1989 reported the outcomes of 226 pregnancies without detecting an increase in miscarriage rates compared to expected figures [28]. However, the study did not include a control group. Another study conducted in the United States assessed the effect of laser vaporisation on pregnancy outcomes and found no significant effect on the rate of spontaneous abortions compared to a control group of untreated women [23]. The association between HPV infection and spontaneous abortions has rarely been investigated in the literature to date, with existing studies being limited by small sample sizes and showing conflicting results [29, 30]. To the best of our knowledge, ours is the first study to provide insight into the association between vaginal HPV infection and risk of miscarriage in a sub‐Saharan African population.
5. Conclusion
Our findings provide essential information on the safety of TA on subsequent fertility and pregnancy outcomes in a sub‐Saharan African population. This is particularly relevant in low‐income contexts where TA is one of the main recommended methods by the WHO for the treatment of cervical precancerous lesions, due to its accessibility, affordability and effectiveness [3, 6]. Safety of TA is especially important in low‐resource contexts using a ‘see and treat’ approach where the risk of overtreatment is considerable. Reassurance on the lack of negative impact of cervical treatment on future fertility is paramount to improve screening coverage in a female target population of childbearing age, as deleterious effects on fertility may lead to major social and psychological impacts. Although this study was not designed to assess the impact of HPV infection on pregnancy outcomes, our results further indicate a strong association between HPV positivity and miscarriage rate. This unexpected finding may prove useful to advocate for HPV vaccination in sub‐Saharan African countries where coverage remains very low [31], as it highlights a potential additional benefit of HPV vaccination in reducing the risk of miscarriage.
Further prospective studies with larger sample sizes should be conducted in different settings, among women with a desire for pregnancy, to confirm the lack of association between TA and subsequent fertility, and assess whether TA increases the risk of less frequent adverse obstetrical outcomes. Multicenter prospective studies comparing ablative to excisional methods may also offer better insight into the differential effect on future pregnancies.
In conclusion, treatment of suspected cervical precancerous lesions by TA did not significantly impact subsequent fertility and risk of miscarriage in a population of Cameroonian women. Our results support the widespread use of TA as a treatment of choice for precancerous cervical lesions in low‐income settings.
Author Contributions
A.W., P.V., B.K. and P.P. designed the study. Data were collected by V.Y., S.E.B. and A.M. Data analysis was done by A.W. Data interpretation was done by A.W., V.Y., J.S., P.V. and P.P. Data were accessed and verified by A.W. and J.S. This report was written by A.W. and was reviewed and approved for publication by all co‐authors. All authors had full access to all data in the study and had final responsibility for the decision to submit for publication.
Ethics Statement
The protocol of the original clinical trial in which participants underwent CC screening was amended to include TaCCare as a follow‐up study and was approved by the Cantonal Ethics Board of Geneva, Switzerland (Commission cantonale d'éthique de la recherche, No. 2017‐0110, amendment No. 4) and the Cameroonian National Ethics Committee for Human Health Research (No. 2018/07/1083/CE/CNERSH/SP and 2022/12/1518/CE/CNERSH/SP). Protocols were carried out in accordance with the principles embodied in the Declaration of Helsinki.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Figure S1.
Figure S2.
Figure S3.
Table S1.
Table S2.
Table S3.
Table S4.
Acknowledgements
We thank Prof. Nicole Schmidt for her insights into the course of the study; Dr. Thomas Perneger for his expertise in statistical analyses; and the collaborators of the Clinical Research Center, Geneva University Hospitals and Faculty of Medicine, Geneva, for their contributions in data management. We also would like to thank the clinical staff of the Dschang District Hospital (now Dschang Regional Hospital Annex), as well as all women who participated in the study. Open access funding provided by Universite de Geneve.
Funding: This study was funded by the Nuovo Soldati Foundation for Cancer Research, Swiss Society for Gynecology and Obstetrics, Humanitarian Affairs Commission of Geneva University Hospitals, and Swiss Cancer Research.
Bruno Kenfack and Patrick Petignat joint senior authorship.
Data Availability Statement
Individual participant data that underlie the results reported in this article (text, tables, figures and appendices) will be made available in a data repository after de‐identification, along with a data dictionary defining each field in the set. The data will be available upon publication of the article and for an indefinite period. The study protocol will be made available upon request to the principal investigator (PP).
References
- 1. Sung H., Ferlay J., Siegel R. L., et al., “Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: A Cancer Journal for Clinicians 71, no. 3 (2021): 209–249. [DOI] [PubMed] [Google Scholar]
- 2. World Health Organization , Global Strategy to Accelerate the Elimination of Cervical Cancer as a Public Health Problem (Geneva, Switzerland: World Health Organization, 2020), ISBN 978‐92‐4‐001410‐7. [Google Scholar]
- 3. World Health Organization , “WHO Guidelines for the Use of Thermal Ablation for Cervical Pre‐Cancer Lesions,” 2019, https://www.who.int/publications‐detail‐redirect/guidelines‐use‐of‐thermal‐ablation‐for‐cervical‐pre‐cancer‐lesions. [PubMed]
- 4. Pinder L. F., Parham G. P., Basu P., et al., “Thermal Ablation Versus Cryotherapy or Loop Excision to Treat Women Positive for Cervical Precancer on Visual Inspection With Acetic Acid Test: Pilot Phase of a Randomised Controlled Trial,” Lancet Oncology 21, no. 1 (2020): 175–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Randall T. C., Sauvaget C., Muwonge R., Trimble E. L., and Jeronimo J., “Worthy of Further Consideration: An Updated Meta‐Analysis to Address the Feasibility, Acceptability, Safety and Efficacy of Thermal Ablation in the Treatment of Cervical Cancer Precursor Lesions,” Preventive Medicine 118 (2019): 81–91. [DOI] [PubMed] [Google Scholar]
- 6. Sauvaget C., Bazikamwe S., Lucas E., Ndayikengurukiye A., Harerimana S., and Barango P., “Evaluation of Effectiveness, Acceptability and Safety of Thermal Ablation in the Treatment of Cervical Neoplasia in Burundi,” International Journal of Cancer 151, no. 7 (2022): 1120–1126. [DOI] [PubMed] [Google Scholar]
- 7. World Health Organization , WHO Guideline for Screening and Treatment of Cervical Pre‐Cancer Lesions for Cervical Cancer Prevention, 2nd ed. (Geneva, Switzerland: WHO, 2021), https://www.who.int/publications‐detail‐redirect/9789240030824. [PubMed] [Google Scholar]
- 8. Chigbu C. O. and Aniebue U. U., “Non‐Uptake of Colposcopy in a Resource‐Poor Setting,” International Journal of Gynecology & Obstetrics 113, no. 2 (2011): 100–102. [DOI] [PubMed] [Google Scholar]
- 9. Buchanan Lunsford N., Ragan K., Lee Smith J., Saraiya M., and Aketch M., “Environmental and Psychosocial Barriers to and Benefits of Cervical Cancer Screening in Kenya,” Oncologist 22, no. 2 (2017): 173–181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Cuzick J., Szarewski A., Cubie H., et al., “Management of Women Who Test Positive for High‐Risk Types of Human Papillomavirus: The HART Study,” Lancet 362, no. 9399 (2003): 1871–1876. [DOI] [PubMed] [Google Scholar]
- 11. Broquet C., Vassilakos P., Ndam Nsangou F. M., et al., “Utility of Extended HPV Genotyping for the Triage of Self‐Sampled HPV‐Positive Women in a Screen‐and‐Treat Strategy for Cervical Cancer Prevention in Cameroon: A Prospective Study of Diagnostic Accuracy,” BMJ Open 12, no. 12 (2022): e057234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kyrgiou M., Mitra A., Arbyn M., et al., “Fertility and Early Pregnancy Outcomes After Treatment for Cervical Intraepithelial Neoplasia: Systematic Review and Meta‐Analysis,” BMJ 349 (2014): g6192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Spracklen C. N., Harland K. K., Stegmann B. J., and Saftlas A. F., “Cervical Surgery for Cervical Intraepithelial Neoplasia and Prolonged Time to Conception of a Live Birth: A Case–Control Study,” BJOG: An International Journal of Obstetrics & Gynaecology 120, no. 8 (2013): 960–965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kalliala I., Anttila A., Dyba T., Hakulinen T., Halttunen M., and Nieminen P., “Pregnancy Incidence and Outcome Among Patients With Cervical Intraepithelial Neoplasia: A Retrospective Cohort Study,” BJOG: An International Journal of Obstetrics & Gynaecology 119, no. 2 (2012): 227–235. [DOI] [PubMed] [Google Scholar]
- 15. Wise L. A., Willis S. K., Perkins R. B., et al., “A Prospective Study of Treatments for Cervical Intraepithelial Neoplasia and Fecundability,” American Journal of Obstetrics and Gynecology 223, no. 1 (2020): 96.e1–96.e15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kunckler M., Schumacher F., Kenfack B., et al., “Cervical Cancer Screening in a Low‐Resource Setting: A Pilot Study on an HPV‐Based Screen‐and‐Treat Approach,” Cancer Medicine 6, no. 7 (2017): 1752–1761. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Levy J., de Preux M., Kenfack B., et al., “Implementing the 3T‐Approach for Cervical Cancer Screening in Cameroon: Preliminary Results on Program Performance,” Cancer Medicine 9, no. 19 (2020): 7293–7300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Petignat P., Kenfack B., Wisniak A., et al., “ABCD Criteria to Improve Visual Inspection With Acetic Acid (VIA) Triage in HPV‐Positive Women: A Prospective Study of Diagnostic Accuracy,” BMJ Open 12, no. 4 (2022): e052504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Thermo Coagulation , “Cervical Cancer Prevention” WISAP, cited Mar 14, 2024, https://www.thermo‐coagulation.com/.
- 20. de Fouw M., Oosting R. M., Eijkel B. I. M., et al., “Comparison of the Tissue Interaction Between Thermal Ablation and Cryotherapy as Treatment for Cervical Precancerous Lesions in an Ex‐Vivo Model,” Health Technology 10, no. 5 (2020): 1275–1281. [Google Scholar]
- 21. Zhang J. and Yu K. F., “What's the Relative Risk? A Method of Correcting the Odds Ratio in Cohort Studies of Common Outcomes,” Journal of the American Medical Association 280, no. 19 (1998): 1690–1691. [DOI] [PubMed] [Google Scholar]
- 22. Zhuang H., Hong S., Zheng L., et al., “Effects of Cervical Conisation on Pregnancy Outcome: A Meta‐Analysis,” Journal of Obstetrics and Gynaecology 39, no. 1 (2019): 74–81. [DOI] [PubMed] [Google Scholar]
- 23. Spitzer M., Herman J., Krumholz B. A., and Lesser M., “The Fertility of Women After Cervical Laser Surgery,” Obstetrics and Gynecology 86, no. 4 Pt 1 (1995): 504–508. [DOI] [PubMed] [Google Scholar]
- 24. Naleway A. L., Weinmann S., Krishnarajah G., et al., “Pregnancy After Treatment for Cervical Cancer Precursor Lesions in a Retrospective Matched Cohort,” PLoS One 10, no. 2 (2015): e0117525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. WHO Guidelines: Use of Cryotherapy for Cervical Intraepithelial Neoplasia (Geneva, Switzerland: World Health Organization, 2011), WHO Guidelines Approved by the Guidelines Review Committee, http://www.ncbi.nlm.nih.gov/books/NBK138476/. [Google Scholar]
- 26. Hammond R. H. and Edmonds D. K., “Does Treatment for Cervical Intraepithelial Neoplasia Affect Fertility and Pregnancy?,” BMJ 301, no. 6765 (1990): 1344–1345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Papoutsis D., Underwood M., Parry‐Smith W., and Panikkar J., “Early and Late Pregnancy Outcomes in Women Treated With Cold‐Coagulation Versus LLETZ Cervical Treatment for Cervical Intraepithelial Neoplasia; a Retrospective Cohort Study,” Archives of Gynecology and Obstetrics 297, no. 4 (2018): 1015–1025. [DOI] [PubMed] [Google Scholar]
- 28. Gordon H. K. and Duncan I. D., “Effective Destruction of Cervical Intraepithelial Neoplasia (CIN) 3 at 100°C Using the Semm Cold Coagulator: 14 Years Experience,” BJOG: An International Journal of Obstetrics & Gynaecology 98, no. 1 (1991): 14–20. [DOI] [PubMed] [Google Scholar]
- 29. Popescu S. D., Boiangiu A. G., Sima R. M., Bilteanu L., Vladareanu S., and Vladareanu R., “Maternal HPV Infection and the Estimated Risks for Adverse Pregnancy Outcomes—A Systematic Review,” Diagnostics 12, no. 6 (2022): 1471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Basonidis A., Liberis A., Daniilidis A., Petousis S., and Dinas K., “Human Papilloma Virus Infection and Miscarriage: Is There an Association?,” Taiwanese Journal of Obstetrics & Gynecology 59, no. 5 (2020): 656–659. [DOI] [PubMed] [Google Scholar]
- 31. Bruni L., Diaz M., Barrionuevo‐Rosas L., et al., “Global Estimates of Human Papillomavirus Vaccination Coverage by Region and Income Level: A Pooled Analysis,” Lancet Global Health 4, no. 7 (2016): e453–e463. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1.
Figure S2.
Figure S3.
Table S1.
Table S2.
Table S3.
Table S4.
Data Availability Statement
Individual participant data that underlie the results reported in this article (text, tables, figures and appendices) will be made available in a data repository after de‐identification, along with a data dictionary defining each field in the set. The data will be available upon publication of the article and for an indefinite period. The study protocol will be made available upon request to the principal investigator (PP).