Background
People with HIV (PWH) are equally likely to have aborted pregnancies as people without HIV (PWoH) but may be at risk of worse outcomes. Abortion-related infections are a significant cause of morbidity and mortality in regions with restricted access to elective abortion. We aimed to determine if HIV increased the risk of an abortion-related infection.
Methods
We conducted a retrospective chart review of women of all ages, <28 weeks’ gestation, admitted to the gynecology unit at a referral hospital in Gaborone, Botswana, for post-abortion care in 2019. Induced and spontaneous abortions were included. Self-reported HIV status was abstracted from a standardized admission form. The admitting diagnosis, including the presence of an abortion-related infection, was captured from the admitting physician’s assessment. Bivariate and multivariate models were used to determine factors associated with an abortion-related infection.
Results
We reviewed 847 medical charts. Participants aged between 16 and 45 years. Table 1 describes demographics and medical exposures by the outcome of interest, abortion-related infection. Nine percent (74/847) of women were diagnosed with an abortion-related infection. Thirty-six (17%) PWH were diagnosed with an abortion-related infection compared to 29 (6%) women without HIV and 9 (7%) with unknown HIV status. Mechanical injury on vaginal exam was more common in PWH, with 21(10%) PWH compared to 21 (4%) PWoH (P-value: 0.012) found to have evidence of vaginal trauma. PWH were also more likely to report a history of sexually transmitted infection (STI), with 116 (55%) of PWH compared to 232 (46%) of PWoH reporting a prior diagnosis (P-value: 0.001). In adjusted analysis (adjusted odds ratio [confidence interval]), PWH were 3.91 [2.00-7.76] more likely to be diagnosed with septic abortion than PWoH. Other independent risk factors included vaginal mechanical injury (60.58 [27.98-131.17]), non-infectious chronic co-morbidities (i.e., diabetes, hypertension, seizure disorder) (4.39 [1.91-10.10]), and being a decade younger (2.20 [1.19-4.09]).
Conclusions
PWH were at higher risk of an abortion-related infection relative to PWoH. Whether this complication is due to immunocompromise or behavioral characteristics in PWH requires additional investigation. The high rate of infection in those with mechanical injury underscores the importance of reducing unsafe abortion practices in settings with extensive abortion restrictions.