Enhanced Detection of Early HIV Infections Using a Combined Diagnostic Algorithm in the Central African Republic: A Cross-Sectional Diagnostic Accuracy Study

Abstract

Background: Human immunodeficiency virus (HIV) infection remains a major global public health challenge, particularly in resource-limited settings such as the Central African Republic. Early and accurate diagnosis is essential to ensure timely initiation of antiretroviral therapy and to reduce ongoing transmission. However, conventional serological assays may fail to detect recent infections during the acute phase due to the diagnostic window period. This study evaluated the performance of an optimized HIV screening algorithm combining a fourth-generation enzyme-linked immunosorbent assay (ELISA), confirmatory rapid diagnostic tests, and HIV-1 polymerase chain reaction (PCR) for discordant cases, compared with a standard serological algorithm. Methods: A descriptive and analytical cross-sectional study was conducted over nine months (December 2024-August 2025) at the National Laboratory of Clinical Biology and Public Health in Bangui. A total of 500 participants aged 18 months and older were included using exhaustive sampling. Initial screening was performed using a fourth-generation ELISA. Reactive samples were confirmed using Determine TM and Uni-Gold TM rapid diagnostic tests. Discordant results were further investigated using HIV-1 PCR for viral RNA detection. Diagnostic performance was assessed by calculating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Diagnostic gain and reduction in median diagnostic delay were also evaluated. Results: Among the 500 participants, 120 (24.0%) were reactive by fourth-generation ELISA. Serological confirmation identified 25 discordant cases (20.8%). HIV-1 RNA was detected in 18 of these cases (72.0%), indicating early HIV infection. The optimized algorithm confirmed 113 HIV infections (22.6%), compared with 95 cases identified using the standard algorithm. The median diagnostic delay decreased from 28 to 14 days. Overall, the optimized algorithm achieved a sensitivity of 98.2% and a specificity of 99.2%, with an 18.9% increase in case detection compared with the standard approach. Conclusions: The optimized HIV screening algorithm improves early and reliable detection of HIV infection, particularly during the acute phase, by reducing false-negative results and shortening diagnostic delay. Its implementation in resource-limited settings may enhance early case detection, improve patient management, and strengthen public health strategies for HIV control.