Understanding risk factors for incident maternal HIV-1 infection

Screening for HIV-1 infection in pregnant women is the cornerstone of all prevention of mother-to-child transmission (MTCT) programs. It should ideally allow the timely administration of antiretroviral agents to both the mother and the newborn, and implementation of comprehensive prevention of MTCT protocols that may or may not include avoidance of breastfeeding to minimize the risk of postnatal transmission [1]. In this context, care should be exercised to take into account possible incident HIV-1 infection in the mother during pregnancy and the postpartum period. In that regard, the serologically silent window period associated with acute HIV infection poses a significant challenge to healthcare providers [2]. In spite of conflicting evidence, it is generally thought that rates of incident HIV-1 infection in women are heightened during pregnancy and the postpartum period [3–6]. In addition, acute HIV infection during pregnancy or shortly thereafter may lead to increased risk of MTCT in utero, during delivery or via breastfeeding [5–9], possibly as a result of the high viral load typically associated with primary HIV-1 infection [10]. A solid understanding of the biological and sociobehavioral risk factors associated with incident HIV infection in pregnancy or during the postpartum period would provide additional opportunities for intervention directed at mitigating said risk factors and/or for implementation of targeted HIV screening protocols. In this issue of AIDS, Kinuthia et al.[11] present the results of a prospective cohort study of HIV-1 acquisition during pregnancy and the postpartum period, as well as associated risk factors. Based on an enrolment of 1304 pregnant women attending the Ahero and Bondo maternal child healthcare clinics (Western Kenya) and 1232 person-years of follow-up, the study revealed an incidence rate of 2.31 per 100 person-years, with no statistically significant difference between the rate of acquisition of HIV during pregnancy and during the postpartum period. In addition, Kinuthia et al.[11] reported associations between incident HIV and syphilis, Chlamydia trachomatis infection, bacterial vaginosis, yeast infection, history of sexually transmitted infections, partner age discordance, and shorter duration of marriage. However, when adjusted hazard ratios (HRs) were examined, statistically significant associations were restricted to partner age difference (adjusted HR = 1.07), C. trachomatis infection (adjusted HR = 4.77), and yeast infection (adjusted HR = 2.99). As the authors noted, partner age difference could act as a proxy for ‘unknown partner HIV positive status or increased likelihood of external partnerships’ [11]. Indeed, none of the 25 women who became infected during the course of the study reported having an HIV-seropositive partner. Identification of serologically discordant status among couples has been recently underlined as a key challenge for the elimination of perinatal HIV infection [12], although risk stratification based on an HR of 1.07 may not be of practical use in real life. The issue of diagnosis and management of sexually transmitted diseases and genital infections, in general, in pregnant women is important, as these can lead to loss of integrity of the vaginal epithelium. Consistent with known relationships between genital ulcer disease and prevalence of HIV-1 infection [13,14], Kinuthia et al. observed that positive syphilis serology was strongly associated with incident HIV-1 infection in their study population (crude HR = 9.18; P = 0.003). The reason why statistical significance was lost in the adjusted model is unclear from a biological standpoint but may have to do with multivariate imputation for independent variables with missing data (highest at 21.1% in the case of syphilis). In addition to weakening of barrier integrity, sexually transmitted diseases and genital infections, including bacterial vaginosis, can promote the establishment of an inflammatory environment in the vaginal tract that is thought to enhance susceptibility to acquire HIV-1 infection [15,16]. This environment is characterized by increased expression of a broad portfolio of soluble inflammatory mediators, including proinflammatory cytokines (interleukin-1α, interleukin-1β, interleukin-6, tumor necrosis factor-α) and chemokines [interleukin-8, macrophage inhibitory protein-1β, regulated on activation, normal T-cell expressed and secreted, interferon gamma-induced protein (IP-10)] [17]. In contrast, resistance to HIV-1 infection in highly exposed seronegative commercial sex workers was characterized by the presence of low levels of interleukin-1α and IP-10 in the cervicovaginal lavage fluids, compatible with the ‘immune quiescence’ model of protection from HIV-1 infection [18]. It is interesting to note that C. trachomatis infection, for which the highest adjusted HR for incident HIV-1 infection (4.77) was observed by Kinuthia et al., and for which comparatively little missing data (0.3%) were imputed, was associated with the highest levels of cytokines in the female genital tract [17]. Implementation of rapid point-of-care testing for C. trachomatis infection, which was not performed in the context of the study by Kinuthia et al., would therefore appear to be desirable for timely return of results and possible intervention, that is, treatment with appropriate antibiotics, use of microbicides, and/or introduction of HIV preexposure prophylaxis. Although the study by Kinuthia et al. would have certainly benefited from a larger enrolment – on the whole, it is based on 25 incident cases of maternal HIV-1 infection, the message comes out loud and clear: monitoring of incident HIV-1 infection in pregnancy and the postpartum period and identification of risk factors to guide healthcare providers throughout the screening process are shaping up as key issues in reaching the goal of the Joint United Nations Programme on HIV/AIDS (UNAIDS) to reduce the number of new HIV infections among children by 90% by the end of 2015 [19]. Acknowledgements Conflicts of interest There are no conflicts of interest.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.