Association of HIV status with infections complicating abortion in Botswana.

Association of HIV Status with Infections Complicating Abortion in Botswana - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice AIDS . Author manuscript; available in PMC: 2026 Jun 1. Published in final edited form as: AIDS. 2025 May 8;39(7):923–926. doi: 10.1097/QAD.0000000000004176 Association of HIV Status with Infections Complicating Abortion in Botswana Bogadi Loabile 1 , Warren B Bilker 2, 3 , Khumo Khunou 5 , Ngwao Ngwako 7 , Billy Tsima 5 , Merrian Brooks 4 , Thabo Moloi 6 , Chelsea Morroni 8, 9 , Robert Gross 1, 2, 3 Author information Article notes Copyright and License information 1 Division of Infectious Diseases, Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA 2 Center for Clinical Epidemiology and Biostatistics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA 3 Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA 4 Department of Pediatrics, Craig Dalsimer Division of Adolescent Medicine, Perelman School of Medicine and Children’s Hospital of Philadelphia, Philadelphia, USA. 5 Department of Family Medicine and Public Health, University of Botswana, Gaborone, Botswana 6 Department of Obstetrics and Gynecology, University of Botswana, Gaborone, Botswana 7 Botswana-University of Pennsylvania Partnership (BUP), Gaborone, Botswana 8 Botswana Sexual and Reproductive Health Initiative, Botswana Harvard Health Partnership, Gaborone, Botswana 9 Centre for Reproductive Health, University of Edinburgh, Edinburgh, UK Authors contribution The study concept and design were developed by B.L, R.G, W.B, B.T, M.B, C.M, T.M. Data collection was carried out by B.L, K.K, N.N. Data analysis and interpretation was done by B.L, R.G. The draft manuscript was prepared by B.L. The draft manuscript was reviewed with feedback provided by all authors. All authors reviewed the results and approved the final version of the manuscript. Issue date 2025 Jun 1. PMC Copyright notice PMCID: PMC12074578  NIHMSID: NIHMS2062314  PMID: 40338852 The publisher's version of this article is available at AIDS Summary Women with HIV (WWH) were found to have an increased risk of abortion-related infection relative to women without HIV (WWoH) and those without HIV status documented. This relationship persisted even after accounting for factors such as mechanical injury on pelvic exam, a marker of the least safe methods of self-induction. Whether this complication in WWH is due to immunocompromise or behavioral characteristics requires additional investigation. INTRODUCTION In sub-Saharan Africa (SSA), 75% of abortions occur without appropriate supervision or by dangerous methods [ 1 , 2 ] and cause ~10% of maternal deaths [ 3 ]. Infections are common post-abortion and confer significant mortality risk [ 4 ]. While WWH are equally likely to seek abortions as women without HIV (WWoH), [ 5 , 6 ] little is known about their outcomes. Two small studies in high-income settings found WWH had more post-surgical complications [ 7 , 8 ]. In contrast, in Uganda, no differences in abortion-related infections were found between WWH and WWoH [ 9 ]. Black market-obtained misoprostol is reportedly the most used method of self-termination in Botswana [ 11 ]; if women present to the hospital for post-abortion care (PAC), they undergo surgical uterine evacuation and receive antibiotics [ 12 , 13 ]. In 2019, infection-complicating abortion caused 13% of maternal deaths in Botswana [ 14 ]. We aimed to determine if HIV status was associated with the diagnosis of an abortion-related infection. METHODS We did a retrospective chart review of women admitted to Princess Marina Hospital, a referral hospital in Gaborone, Botswana, for PAC in 2019 using the abortion register to establish a pool of potentially eligible participants. Both spontaneous and self-induced abortions were included since we could not distinguish between them. We excluded women with 1) estimated gestations ≥ 28 weeks, 2) ectopic or molar pregnancies, or 3) presenting for elective abortion. We selected every 3 rd medical chart by month and extracted relevant data into REDCap. Abortion-related infection was determined using the admitting physician’s assessment. We categorized as “unknown HIV status” both those without documentation and those documented as unknown HIV status. Bivariate analyses using chi-squared tests determined if there was an association between abortion-related infection and each categorical exposure. For continuous exposures, t-tests determined differences between the two groups. Multivariable logistic regression controlled for potential confounders of the association between HIV status and abortion-related infection. Potential confounders were selected from our conceptual framework ( Supplementary Figure 1 ) and retained in the models if they changed the association between HIV and infection by 15%. A P -value <0.05 was considered statistically significant, and we targeted a sample size of 810 to detect an OR of ≥2 between HIV and infection. A sample size of 810 was targeted based on the main exposure variable of interest, HIV status. Individuals with missing data on included confounders were excluded from the final multivariable analysis. RESULTS Of the 2369 potentially eligible admissions, 359 charts were unavailable and were replaced until we achieved a sample of 847 women. The age range was 16 to 45 years, and admission diagnoses were: 1) 541/847 (64 %) with an incomplete abortion, 2) 194/847 (23 %) with a complete abortion, 3) 74/847 (9%) with an infection-complicating abortion 4) 23/847 (3%) with an inevitable abortion and 5) 15/847 (2%) with a missed abortion. HIV status was missing or documented as unknown in 130 (15%) charts. Table 1 displays the demographic and medical exposures by infection status. Eighty-one participants (10%) had documentation of a non-communicable chronic illness, including diabetes in 48 (60%), hypertension in 12 (15%), and seizure disorders in 21 (25%). Table 1: Unadjusted and Adjusted Associations between Participant Characteristics and Abortion-related Infection. Variable Overall N=847 No Abortion-related infection n=773 Abortion- related infection n=74 Unadjusted OR (95% CI) Adjusted OR (95% CI) Age in years (mean, SD) 28 (5.11) 28 (5.06) 27 (5.53) 0.95 (0.91–0.99) 0.92 (0.87–0.98) Single (marital status) Missing=5 669 (79%) 606 (91%) 63 (9%) 1.72 (0.87–3.44) - Education (completed secondary school) Missing=4 822 (97%) 750 (91%) 72 (9%) 1.92 (0.25–14.52) - HIV Negative 505 (60%) 476 (94%) 29 (6%) REF - Positive 212 (25%) 176 (83%) 36 (17%) 3.36 (1.99–5.64) 3.91 (2.00–7.65) Unknown 130 (15%) 121 (93%) 9 (7%) 1.22 (0.56–2.64) 0.99 (0.37–2.68) Presence of non-communicable disease Missing=1 81 (10%) 69 (85%) 12 (15%) 1.97 (1.01–3.84) 4.39 (1.91–10.10) Prior STI diagnosis Missing= 3 392 (46%) 351 (90%) 41 (10%) 1.48 (0.92–2.40) - Late pregnancy loss (≥12 weeks) Missing=15 128 (15%) 116 (91%) 12 (9%) 1.26 (0.47–3.36) - Mechanical injury on vaginal exam Missing=7 49 (6%) 13 (27%) 36 (73%) 56.43 (27.61–115.35) 60.58 (27.98–131.17) Open in a new tab Abbreviations: CI, Confidence Intervals; OR, Odds Ratio Mechanical injuries were described in 49 (6%) individuals; descriptions included lacerations of the vulva, puncture wounds, abrasions, bruising, and vaginal tears. This finding was more common in WWH (21/212) (10%)) than either WWoH (21/505 (4%)) or women with unknown HIV status 7/130 (5%)). The final multivariable model included 839 of the 847 admissions and age, non-communicable chronic illnesses, and mechanical injury as covariates. In the adjusted model, HIV-positive status, younger age, the presence of a chronic illness, and mechanical injury on pelvic exam were all independent risk factors for the diagnosis of an abortion-related infection ( Table 1 ). No in-hospital mortality was documented for any women included. DISCUSSION We found abortion-related infections were more common in WWH, consistent with a Uganda study of post-partum women [ 16 ]. However, our results contrast the only other study done in women receiving PAC in SSA, which did not find an association between HIV and abortion-related infection [ 9 ]. These findings may be related to biological or behavioral risk factors that are more prevalent in WWH than WWoH, such as sexually transmitted infections (STIs) [ 17 ]. Prior studies found an association between curable STIs like chlamydia and abortion-related infections [ 18 , 19 ]. Other factors such as stigma, intimate partner violence, and substance abuse are more prevalent in WWH than WWoH [ 20 , 21 ], potentially contributing to riskier methods of termination and delays in seeking care, both of which increase infection risk. Alternatively, WWH presenting for PAC may be immunosuppressed (e.g., from low CD4 counts or uncontrolled viremia), increasing their risk for infection. However, the high rates of ART coverage and viral suppression in PWH in Botswana make immunosuppression a less likely explanation [ 22 ]. This study had strengths, including the large sample size to achieve adequate power for the question. Limitations primarily related to the retrospective nature. It prevented assessment of whether abortions were spontaneous vs induced, the types of STI’s patients had previously had, and HIV specific parameters like ART use and plasma HIV RNA levels at the time of admission. The inability to find medical records on selected participants may have decreased the randomness of our sampling strategy. Finally, this was a single-center analysis that could impact the generalizability of the results. However, this hospital has a large catchment area, providing care for women with a wide range of socio-demographic characteristics, and results are likely representative of other urban and peri-urban settings in the country and potentially other countries in southern Africa with similar restrictions around abortion. Despite these limitations, this study adds to the limited available data that has evaluated HIV as a potential risk factor associated with abortion-related infections, a critical question given the significant population of WWH in SSA, a region where abortion remains an important contributor to maternal mortality [ 3 ]. Future research that attempts to identify biological, behavioral, and HIV-specific factors that may be contributing to this association is imperative to develop strategies to improve sexual and reproductive health outcomes among WWH. Supplementary Material Supplemental Data File (conceptual Framework diagram) NIHMS2062314-supplement-Supplemental_Data_File__conceptual_Framework_diagram_.pdf (310.4KB, pdf) Funding This work was supported by the Penn Center for AIDS Research (CFAR), an NIH-funded program (P30 AI 045008), and the Harris Award from the Division of Infectious Disease at the University of Pennsylvania to B.L. Footnotes Conflicts of interests The authors of this manuscript report no potential conflicts. REFERENCES 1. Bearak J, Popinchalk A, Ganatra B, Moller AB, Tuncalp O, Beavin C, et al. Unintended pregnancy and abortion by income, region, and the legal status of abortion: estimates from a comprehensive model for 1990–2019. Lancet Glob Health. 2020;8(9):e1152–e61. [ DOI ] [ PubMed ] [ Google Scholar ] 2. Ganatra B, Gerdts C, Rossier C, Johnson BR Jr., Tuncalp O, Assifi A, et al. Global, regional, and subregional classification of abortions by safety, 2010–14: estimates from a Bayesian hierarchical model. Lancet. 2017;390(10110):2372–81. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Say L, Chou D, Gemmill A, Tuncalp O, Moller AB, Daniels J, et al. Global causes of maternal death: a WHO systematic analysis. Lancet Glob Health. 2014;2(6):e323–33. [ DOI ] [ PubMed ] [ Google Scholar ] 4. Baguiya A, Mehrtash H, Bonet M, Adu-Bonsaffoh K, Compaore R, Bello FA, et al. Abortion-related infections across 11 countries in Sub-Saharan Africa: Prevalence, severity, and management. Int J Gynaecol Obstet. 2022;156 Suppl 1:36–43. [ DOI ] [ PubMed ] [ Google Scholar ] 5. Bankole A, Keogh S, Akinyemi O, Dzekedzeke K, Awolude O, Adewole I. Differences in unintended pregnancy, contraceptive use and abortion by HIV status among women in Nigeria and Zambia. Int Perspect Sex Reprod Health. 2014;40(1):28–38. [ DOI ] [ PubMed ] [ Google Scholar ] 6. Manski R, Dennis A, Blanchard K, Lince N, Grossman D. Bolstering the Evidence Base for Integrating Abortion and HIV Care: A Literature Review. AIDS Res Treat. 2012;2012:802389. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Stuart GS, Sheffield JS, Hill JB, McIntire DD, McElwee B, Wendel GD. Morbidity that is associated with curettage for the management of spontaneous and induced abortion in women who are infected with HIV. Am J Obstet Gynecol. 2004;191(3):993–7. [ DOI ] [ PubMed ] [ Google Scholar ] 8. Grubert TA, Reindell D, Kastner R, Belohradsky BH, Gurtler L, Stauber M, et al. Rates of postoperative complications among human immunodeficiency virus-infected women who have undergone obstetric and gynecologic surgical procedures. Clin Infect Dis. 2002;34(6):822–30. [ DOI ] [ PubMed ] [ Google Scholar ] 9. Okong P, Biryahwaho B, Bergstrom S. Post-abortion endometritis-myometritis and HIV infection. Int J STD AIDS. 2002;13(11):729–32. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Botswana Government. Penal Code (Amendment) Act, 1991. Government Gazette; 1991. [ Google Scholar ] 11. Ngwako K, Banke-Thomas A. ‘I guess we have to treat them, but … ‘: health care provider perspectives on management of women presenting with unsafe abortion in Botswana. Glob Public Health. 2020;15(9):1308–21. [ DOI ] [ PubMed ] [ Google Scholar ] 12. Melese T, Habte D, Tsima BM, Mogobe KD, Nassali MN. Management of post abortion complications in Botswana -The need for a standardized approach. PLoS One. 2018;13(2):e0192438. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Sinvula MI M. Botswana Maternal Mortality Reduction Initiative. USAID; 2015. [ Google Scholar ] 14. Statistics Botswana. Botswana Maternal Mortality Ratio 2019. Botswana Statistics; 2021. [ Google Scholar ] 15. Melese T, Habte D, Tsima BM, Mogobe KD, Chabaesele K, Rankgoane G, et al. High Levels of Post-Abortion Complication in a Setting Where Abortion Service Is Not Legalized. PLoS One. 2017;12(1):e0166287. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Okong P, Biryahwaho B, Bergstrom S. Intrauterine infection after delivery: a marker of HIV-1 seropositivity among puerperal women in Uganda? Int J STD AIDS. 2004;15(10):669–72. [ DOI ] [ PubMed ] [ Google Scholar ] 17. Jarolimova J, Platt LR, Curtis MR, Philpotts LL, Bekker LG, Morroni C, et al. Curable sexually transmitted infections among women with HIV in sub-Saharan Africa. AIDS. 2022;36(5):697–709. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Wein P, Kloss M, Garland SM. Postabortal pelvic sepsis in association with Chlamydia trachomatis. Aust N Z J Obstet Gynaecol. 1990;30(4):347–50. [ DOI ] [ PubMed ] [ Google Scholar ] 19. Stevenson MM, Radcliffe KW. Preventing pelvic infection after abortion. Int J STD AIDS. 1995;6(5):305–12. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Kouyoumdjian FG, Findlay N, Schwandt M, Calzavara LM. A systematic review of the relationships between intimate partner violence and HIV/AIDS. PLoS One. 2013;8(11):e81044. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Joseph Davey DL, Le Roux SM, Brittain K, Dovell K, Shoptaw S, Miller AP, et al. Alcohol use and intimate partner violence in HIV-uninfected pregnant women in Cape Town, South Africa. AIDS Care. 2022;34(2):214–9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. UNAIDS. Botswana: UNAIDS; 2024. [cited 2024. Country Report of HIV statistics]. Available from: https://www.unaids.org/en/regionscountries/countries/botswana . Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplemental Data File (conceptual Framework diagram) NIHMS2062314-supplement-Supplemental_Data_File__conceptual_Framework_diagram_.pdf (310.4KB, pdf) ACTIONS View on publisher site PDF (56.5 KB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Factors associated with HIV infection among infants born to mothers with HIV in Blantyre, Malawi: a mixed methods study.#368348
• Elizabeth Glaser Pediatric AIDS Foundation Maternal Events and Pregnancy Outcomes (MEP Project)A Multi-site Protocol#1125191
• Integration of HIV Care and Treatment Into Antenatal Care in Migori District, Kenya#847185
• Maternal and birth outcomes in pregnant people with and without HIV in the Western Cape, South Africa#359113
• Study to Improve Survival Among HIV-Exposed Infants in Botswana#906904