A high burden of asymptomatic genital tract infections undermines the syndromic management approach among adolescents and young adults in South Africa: implications for HIV prevention efforts

Adolescents and young adults (aged 15–24 years) account for one-third of incident adult HIV infections in South Africa. Girls and young women are particularly at risk, with nearly 2000 young women acquiring HIV every week [1, 2]. Accordingly, reducing HIV risk among youth remains a pressing global public health priority [3].

Untreated genital tract infections (GTIs) increase risk of HIV acquisition and transmission through several organism-specific pathways [4, 5]. Untreated HIV can also increase GTI severity and duration through immune suppression pathways. For women, untreated GTIs are a significant cause of reproductive morbidity, including pelvic inflammatory disease, tubal factor infertility, adverse birth outcomes, and infertility [6].

The success of GTI control programs is predicated on timely and effective diagnosis and treatment [7]. South African GTI treatment guidelines employ syndromic management [8], consistent with WHO recommendations [9]. Syndromic management is based on the identification of syndromes (i.e., consistent groups of symptoms and clinical signs) associated with infection by a defined pathogen, followed by syndromic treatment for the most common causative GTI organisms.

While syndromic GTI management can be used to treat and manage symptomatic GTIs at the primary healthcare level without the need for laboratory facilities or diagnostic tests, such an approach can underestimate GTI cases, particularly among women for whom an estimated 50–75% of those with a GTI are asymptomatic [10, 11]. Few studies from high HIV-burden settings have examined GTI prevalence or diagnostic performance of syndromic management among adolescents and young adults [11]. This is concerning given the dual burdens and synergies between HIV and GTIs among youth and the potential for long-term sequelae. There are also concerns about the ability of syndromic treatment to differentiate between the etiologies of the disease, leading to the potential for over-prescription of antibiotics and attendant risks of antimicrobial resistance [5].

This analysis measured and compared GTI prevalence by syndromic-based and laboratory assessment among sexually-experienced South African adolescents and young adults, stratified by sex. We assessed the performance of syndromic GTI management compared with the ‘gold standard’ of laboratory diagnosis (including measurement of sensitivity, specificity, positive predictive value [PPV], and negative predictive value [NPV]. In pursuing these objectives, we seek to inform GTI control programs for youth in the context of emerging biomedical HIV prevention efforts.

GTIs, including asymptomatic infections, induce genital tract inflammation among young women, increasing HIV acquisition risk [14]. We found a high and frequently asymptomatic burden of GTI infection among South African youth, with significant differences by sex. While only 16% of young women and < 1% of young men reported GTI-related symptoms, laboratory tests identified at least one GTI in 70% of females and 10% of males. Syndromic GTI management revealed very poor sensitivity. The high burden of asymptomatic GTIs contributes to a silent GTI epidemic in youth, which requires additional attention, innovations, and resources to control. We strongly recommend policy changes to replace syndromic management with innovative public health approaches, inclusive of widespread diagnostic testing to reduce GTI burden and HIV-associated risks among youth.

The high GTI prevalence observed here is particularly notable, given that participants are young (16–24 years of age), primarily recruited from community settings, and were not enrolled due to high GTI risk. Nonetheless, over one-third of young women had at least one non-ulcerative STI and half had laboratory-detected BV. The prevalence of each non-ulcerative STI (CT: 18.2%; NG: 7.1%; MG: 9.6%; TV: 8.1%) was higher or similar to the prevalence observed among older women (18–40 years) presenting for STI care in Durban (CT: 18.4%; NG: 5.2%; MG: not assessed; TV: 3.0%) [16], high HIV-risk women enrolling in the CAPRISA 002 Acute HIV Infection study in Durban (overall STI prevalence = 25.3%; CT: 4.2%; NG: 5.4%; MG: 1.2%; TV: 20.3%) [17], women enrolled in three large clinical trials in Durban (STI prevalence = 13%) [18], “high HIV risk” Tanzanian women (CT: 12%; NG: 4%; MG: not assessed; TV: 19%) [19], and women attending antenatal care in Kenya (CT: 14.9%; NG: 1.0%; MG: not assessed; TV: 7.4%) [20]. The exception to this pattern was TV, which was lower in female youth in our study relative to older women in both the Tanzanian and South African studies. Such patterns are consistent with known biological reasons for increases in TV prevalence with age (i.e., protozoal accumulation over time). The high BV prevalence (53.0%) mirrors levels reported in other South African studies [4, 17, 18, 21, 22].

Fewer (10%) young men in our study had a laboratory-confirmed non-ulcerative STI compared to women. STI prevalence among men was driven largely by CT (7.8%). Our findings are consistent with the limited available data regarding STI prevalence for young and older heterosexual men in South Africa [23, 24].

We observed a 3.7% prevalence of HIV sero-positivity at baseline, despite study eligibility criteria aimed at enrolling HIV-negative participants. Moreover, low median CD4 cell counts at HIV diagnosis signal advanced HIV disease progression. HIV prevalence was higher among Durban participants compared to Soweto, consistent with findings from surveillance data [1]. Given the known synergistic relationship between HIV and STI co-infection, the higher prevalence of non-ulcerative STIs but lower prevalence of HIV in Soweto is notable.

Consistent with previous reports, syndromic management was unable to detect the majority of GTIs due to their asymptomatic nature [17, 25]. Without laboratory testing, approximately 80% of prevalent GTIs among young women and 100% among young men in our study would have remained untreated. Poor sexual health knowledge and a lack of familiarity with GTI symptoms among youth may have exacerbated under-reporting of symptoms [26‐28]. Both VDS and LAP had poor sensitivity (13.7% and 7.0% respectively). This is a well-documented limitation to the VDS algorithm, especially for cervical infection [17, 21, 29]. VDS had a high PPV (90.5%) indicating that most women who reported VDS-related symptoms did have a laboratory-confirmed GTI and would have been appropriately treated under syndromic management. That is, syndromic management performed appropriately in its role of detecting and managing symptomatic GTIs. However, the PPV for LAP was poor (30.8%) highlighting that a majority of young women who reported LAP-associated symptoms did not have a laboratory-confirmed GTI and may have been inappropriately treated under syndromic management guidelines. Insufficient symptomatic infection in young men in our study precluded robust sensitivity analyses of syndromic management; no men with an STI reported symptoms and no men who reported symptoms had an STI.

There are important advantages to syndromic management on a national scale; the primary being cost, and a second being that it requires a single visit, decreasing the likelihood of treatment loss-to-follow-up at a second visit, typically required with diagnostic testing [9]. These advantages must be considered against the performance of syndromic approaches for high-priority populations. Our findings suggest that in South Africa, both male and female youth have a high burden of undetected GTIs. Reliance on a syndromic management approach will continue to yield low detection of prevalent GTIs and thus poor GTI control.

Our findings suggest a need to evolve sexual health services to benefit youth. Although a test-and-treat approach to GTIs has historically posed financial and logistical challenges, recent innovations in affordable real-time STI testing, such as point-of-care (POC) diagnostics, warrant further investigation for wide-scale implementation [16, 30]. By providing immediate testing results and facilitating treatment without delay, POC tests can help reduce secondary transmission and the consequences of infection progression [31]. POC testing also presents an opportunity to screen outside of traditional clinic settings, which young people infrequently access [32], and move testing into community-based and youth-friendly settings that offer comprehensive HIV and GTI services. POC testing quality is growing with improving technology [33], with good performance across several prevalent STIs, including CT, NG, TV, and syphilis [34, 35]. Recent work also suggests that POC testing for STIs is acceptable to providers [33] and to young women and their partners [16]. Additional studies are warranted to assess the cost-effectiveness of POC STI testing for youth in low- and middle-income countries.

GTI control will not be solved by diagnostic innovation alone. In addition to GTI burden, and consistent with previous studies [1], youth in our study report high behavioral, social, and structural vulnerability for GTI and HIV acquisition. Notably, over half of adolescent and young adults in this HIV hyper-endemic setting self-perceived themselves as being at low or no risk of HIV acquisition. These findings underscore the need for efforts to move beyond individual-level risk factors for promoting healthy sexuality for young people, and address the social and structural risk environments, such as gender and power inequity in relationships, coercive sex, economic inequity, violence, and poverty, which shape how young people navigate their sexual lives.

This study has limitations. First, HSV-2 and syphilis prevalence are likely to be under-estimated as the study protocol specified NAAT testing of genital ulcerative lesions only. Future work should incorporate serological testing to assess the true prevalence of both HSV-2 and syphilis in youth. Second, not all GTIs were tested (for example, anorectal samples were not taken), thus the reported GTI prevalence may be an underestimate. Third, participants were asked directly about GTI symptoms and their genital areas carefully examined by a study nurse with expertise in youth-focused care delivery, and may have been more willing to report symptoms or undergo genital examination. Thus, our assessment of the sensitivity and PPV of syndromic management of GTIs among youth, although very low, is likely higher than would be observed in a public care setting.

Maintaining the status quo of syndromic GTI management for youth is not acceptable. We observed a substantial burden of subclinical GTI infection alongside HIV risk behaviors and low risk-perception for HIV acquisition among South African youth. Sensitivity of symptom-based reporting was poor and offered no value in identifying the highly prevalent asymptomatic GTI infections. Syndromic management does not meet the needs of young people in this and similar settings and leaves them at increased risk of HIV acquisition. In the context of evolving and less costly GTI diagnostic capabilities, exploring new approaches to implementing youth-focused GTI services is warranted.