Performance of an evidence-based risk algorithm to diagnose chlamydia and gonorrhea among pregnant Rwandan women

Reproductive tract infections (RTI) are associated with adverse outcomes in pregnant African women. However, many diagnostic strategies are unaffordable or perform poorly. Here, we assess RTI prevalence and predictors of chlamydia/gonorrhea (CT/NG) in pregnant women reporting vaginal discharge and the performance of our previously published CT/NG risk algorithm in this population versus Rwandan National Guidelines (RNG). From 2017-2020, free sexually transmitted infections (STI) services were provided to residents in Kigali, Rwanda. Medical history and gynecologic examination were done. Laboratory assessments included HIV; syphilis; microscopy for trichomoniasis, bacterial vaginosis (BV), and candida; and PCR for CT/NG. Eighty-seven pregnant women received STI services. Prevalence was 28% for CT/NG, 15% for trichomoniasis, 24% for BV, 39% for candida, and 79% for any RTI. Predictors of CT/NG were age <=25 (adjusted prevalence odds ratio [aPOR]=4.92; 95% confidence interval [CI]: 1.52-15.90; p=0.008), inconsistent condom use (aPOR=4.86; 95%CI: 0.98-24.10; p=0.053), absence of candida (aPOR=4.23; 95%CI: 1.13-15.82; p=0.032), and endocervical inflammation/discharge (aPOR=4.91; 95%CI: 1.40-17.20; p=0.013). Our algorithm outperformed the 2019 and 2024 RNG (sensitivity: 92% versus 46% and 35% respectively). Pregnant women seeking STI services had high RTI prevalence. Our algorithm performed well. Algorithms tailored for pregnant women and including male partner risk factors should be explored.