Longitudinal analysis of viral suppression before, during, and after pregnancy among women on antiretroviral therapy in Uganda: six-year real-world experience

Abstract

Introduction This study evaluated the effectiveness of antiretroviral therapy (ART) and associated factors on viral suppression before, during, and after pregnancy (maternal timeline). Methods We conducted a cohort study, retrospectively reviewing records of 1291 pregnant and breastfeeding women on ART. Descriptive statistics summarised the demographics and clinical characteristics. Chi-square, Fisher’s exact, and generalised estimating equations were used to assess variations in viral suppression across the maternal timeline. Results ART regimens comprised 62.5% dolutegravir (DTG)-, 28.8% efavirenz (EFV)-, 4.5% nevirapine (NVP)-, and 4.2% protease inhibitor (PI)-based therapy. Viral suppression rates before, during, and after pregnancy were DTG- (95.0%, 94.6%, 95.7%), EFV- (94.9%, 94.2%, 93.6%), NVP- (93.1%, 94.7%, 93.5%), and PI-based (79.6%, 88.0%, 85.7%). ART regimens varied in effectiveness, with statistical significance observed before (p < 0.001) and after (p = 0.018), but not during pregnancy (p = 0.678). PI-based regimens showed higher risk of non-suppression in the non-adjusted model (IRR = 3.20, 95% CI: 1.63–6.30, p = 0.001). In the adjusted model, poor adherence (aIRR = 7.80, 95% CI: 2.54–23.90, p < 0.001), fair adherence (aIRR = 5.03, 95% CI: 1.11–22.86, p = 0.036), second-line ART (aIRR = 3.14, 95% CI: 1.75–5.62, p < 0.001), and third-line ART (aIRR = 8.48, 95% CI: 1.82–39.43, p = 0.006) remained significant. Conclusion ART effectiveness showed variation before and after, but not during pregnancy. EFV- and NVP-based regimens achieved suppression rates comparable to DTG across maternal timelines, with the exception of PI-based regimens. Adherence and ART drugs influence outcomes more than regimen choice alone, with good adherence essential for optimal maternal outcomes.