This is a descriptive and analytical cross-sectional study of HIV-infected patients admitted after being victims of either rape or atrocities committed by armed groups in the eastern Democratic Republic of Congo, admitted to the Panzi General Reference Hospital (South Kivu, Democratic Republic of Congo) between May 1, 2024, and June 28, 2025.

The research was conducted at Panzi General Reference Hospital. This tertiary and university-affiliated facility provides holistic care to women who are victims of atrocities, as well as treating HIV infections and gynecological conditions within the Department of Gynecology, Obstetrics, and Infectious Diseases. This study included patients who had tested or were known to be positive for HIV infection but had been exposed to the atrocities of armed conflict. This enabled epidemiological and cytopathological data to be collected. The analysis was carried out in the laboratory of the Panzi General Reference Hospital and evaluated by two specialist pathologists.

This cross-sectional study included all HIV-positive patients (Unigold and determined tests) receiving outpatient or inpatient care at the Panzi General Reference Hospital, within the infectious diseases department of the internal medicine department and the gynecology-obstetrics department for the follow-up of rape victims. These patients had undergone a cervical smear in gynecology, regardless of the stage of their disease. Convenience sampling was performed. The sample was taken from the cervix by a doctor specializing in gynecology. Patients were examined, and samples were taken by a gynecology resident.

In this study, 103 patients were recruited, but only 56 patients obtained results from their cytopathological tests. However, due to stigma, discomfort, and fear, 47 patients refused to undergo cervical cytology sampling.

The samples were taken directly to the laboratory by a qualified laboratory technician, in accordance with ethical standards. Pregnant patients were excluded from the study.

Independent sociodemographic variables were examined, and their proportions were articulated as percentages. These variables include: age, the number of previous sexual partners, the number of current sexual partners, the age at first pregnancy, the age at first sexual intercourse, the age of menarche, the duration of HIV infection, socioeconomic status (determined by salary in dollars), parity, occupation, living environment, marital status, educational attainment, religion, contraception methods, and the duration of HIV/AIDS disease.

The dependent variables involved the cervico-uterine smear outcomes of HIV-positive patients who consented to participate in the study. The cytopathological findings that made up our dependent variables included normal smear, inflammatory smear, atypical squamous cells of undetermined significance (ASC-US), low-grade squamous intraepithelial lesion (LSIL), and high-grade squamous intraepithelial lesion (HSIL). These findings were classified into two categories: atypical, low-grade, and high-grade squamous intraepithelial lesions, which are regarded as precancerous (pathological) lesions, and normal lesions.

Samples were taken by trained and qualified gynecology residents, accompanied by a laboratory technician also trained in sampling. Sample transport complied with Clinical and Laboratory Standards Institute (CLSI) guidelines [11][12]. The smear was taken at the Panzi general referral hospital and described the various cytological aspects of cervical lesions. To limit the risk of bias in the interpretation of cytopathological results, the results were analyzed and interpreted by two specialists in anatomopathology, one of whom was a professor.

Sampling was by convenience, and participants who gave their consent were allowed to participate in the study.

Biais:

In order to avoid selection bias and enable readers to interpret the results correctly, we performed linear and logistic regressions, excluding any incomplete observations.

Data collection was conducted via Kobocollect with a predefined and refined form, filled out by a medical specialist skilled in using the software. The data were cleaned in Microsoft Excel 2019 and analyzed with SPSS 26. Both descriptive statistics and linear regression analyses were carried out. To minimize bias, multinomial logistic regression was employed. The significance level was set at a p-value of 0.05.

Confidentiality and anonymity of patients and their information were preserved. Informed consent was obtained from each patient before any information was collected. The scientific committee of the Université Évangélique en Afrique (UEA) had granted authorization for the research, and the provincial ethics committee registered under the number CNES 001/DPSK/224PP/2025.

In this study, 103 patients infected with HIV (human immunodeficiency virus) were recruited. All of these women had been exposed to armed conflict and its atrocities, but 37 of them had also been victims of rape. Due to stigma, discomfort, and fear, 47 patients refused to undergo cervical cytology sampling. Cytopathological tests were performed on cervical samples taken from 56 patients, revealing 14 cases of precancerous lesions. The description of sociodemographic factors in Table 1, and the averages of quantitative variables, showed that the mean age of the patients was 41 ± 10.3 years. Table 1 suggests that patients from urban areas were predominantly represented, i.e., 74.8% of cases.

Table 1. Sociodemographic data.

The most common age range was between 31 and 40 (35.9%). The age of first sexual intercourse and first pregnancy among patients was between 10 and 20 years (85.4%) and 10 to 20 years in 72.8% of cases, respectively. 88.3% of patients had between 1 and 5 previous sexual partners; however, 96.1% of them currently have between 1 and 3 sexual partners. HIV infection was present in 38.8% of patients more than 10 years ago, and 45.1% of these patients were on antiretroviral treatment within 1 to 5 years.

Table 1shows that widows accounted for 41.7%, with the predominant level of education being secondary school. However, 50.5% of these patients were housewives.

Protestants made up 64.1% of the cases. Contraception was used by 17% of the study participants, with depot contraception accounting for 38.9% of occurrences.

Table 2.Distribution of means for quantitative variables of sociodemographic factors.

SD: standard deviation; Min: minimum; Max: maximum; HIV: human immunodeficiency virus.

Table2 shows that the average age of first pregnancy was 18.6 ± 4.3 years, and the average age of sexual intercourse was 17.5 ± 2.6 years. The average number of former partners was 2.05 ± 2.3 years. The average duration of HIV infection in our patients was 9.1 ± 6.3 years, with a viral load of 52619.0 ± 374234.3 copies/milliliter. This table also suggests that the average economic level of the patients was 27.9 ± 53.9 USD. The mean parity among patients in this study was 5 ± 2.7. The mean duration of anti-viral (HIV) treatment was 8.3 ± 6.1 years.

Table 3.Distribution of proportions of cytopathological lesions.

ASC-US: Atypical Squamous Cells of Undetermined Significance; LSIL: low-grade squamous intraepithelial lesions; ASC-H: high-grade Atypical Squamous Cells.

The precancerous lesions observed in this study were low-grade ASC-US (12.5%), low-grade squamous intraepithelial lesions (LSIL) (10.7%), and high-grade atypical squamous cells (ASC-H) (3.6%) (Table3). Table 4 shows the association between cytopathological findings and sociodemographic factors.

Table 4.Linear regressions of sociodemographic factors and cytopathological findings.

There was a significant association between the presence of precancerous lesions on cervical smear and the number of former sexual partners (p = 0.029, OR [0.020; 0.357]; IC95%).

This table also suggests that there was no statistically significant association between most of the sociodemographic factors studied in HIV-infected women. However, the level of education (p = 0.036, [0.015; 0.442]; CI95%) significantly influenced the occurrence of precancerous lesions. Table 5 presents the logistic regression according to high or low grade lesions observed on cytopathological examination.

Table 5.Multinomial logistic regression of precancerous lesions and sociodemographic factors.

This study found that having fewer sexual partners protects against low-grade ASC-US lesions (p < 0.05; ORa at 2.4E−16, 95% CI [8.8E−18 to 6.7E−15]. The no-conception impact increases the probability of high-grade cervical precancerous lesions (ASC-H) (p < 0.05; ORa = 3977917.8, 95% CI [225149.6 to 70281411.9]).

The study’s strength comes in providing the epidemiological aspects of cytopathological results of cervical lesions in resource-limited situations where the frequency of these lesions is unknown. The study does not use genotyping, which is a limitation given the wide range of co-infections and viruses that may be present in HIV-infected patients; thus, future researchers should conduct a randomized genotypic study comparing the various viruses and co-infections, particularly their impact on precancerous lesions and the occurrence of cervical cancer.

HIV infection and precancerous cervical lesions are underdiagnosed, despite their high frequency among women in the Democratic Republic of Congo. As a result, early screening campaigns remain the best way to control the occurrence of cervical cancer in this category of the population, requiring the cervical smear to be included among mandatory examinations for women with declining immunity linked to HIV infection in resource-limited countries. In epidemiological terms, primary prevention will therefore be the key element in cancer detection.

Most women are familiar with the risk factors following unprotected intercourse or early conception. However, conceiving a pregnancy reduces the risk of high-grade lesions, whereas a high number of former sexual partners increases this risk. Systematic cervical smear screening of all sexually active girls, with or without HIV infection, would improve long-term follow-up of cancerous lesions if detected early.