Abstract
In resource-limited settings, use of dried blood spots (DBS) could be a pragmatic alternative to plasma for VL monitoring in people living with HIV (PLWH). We compared results from DBS to standard plasma VL testing under field conditions in patients receiving antiretroviral therapy (ART). DBS cards were prepared from venous blood (V-DBS), finger-pricks using micro-capillary tubes (M-DBS), and direct spotting (D-DBS). DBS and matched EDTA plasma were tested on the Abbott m2000 platform using the appropriate RealTime HIV-1 quantitative CE protocol. Matched plasma samples were also tested on the Roche COBAS Ampliprep/COBAS TaqMan version 2.0. Diagnostic accuracy indicators (sensitivity, specificity, misclassification rate, and kappa coefficient) for viral failure (VF) based on different VL threshold levels and agreement of absolute VL were calculated. A total of 669 participants provided 2676 samples. V-DBS had a peak sensitivity for VF of 89.1 % [95 % CI: 85.5-92.7] at the 1000 copies/mL threshold and a peak specificity of 97.4 % [95 % CI: 95.9-99.0] at the 5000 copies/mL threshold. The lowest proportion of upward misclassification (patients classified with VF who actually had viral suppression) for V-DBS was 3.1 % [95 % CI: 1.4-4.8] at the 5000 copies/mL threshold, whereas the lowest proportion of downward misclassification (patients classified as undetectable who actually had VF) was 10.9 % [95 % CI: 7.2-14.5] at the 1000 copies/mL threshold. Abbott RealTime HIV-1 VL results from all 3 DBS types for adults and children showed strong correlation with the gold standard plasma-based assay. DBS could be useful for monitoring VL in resource limited settings such as Nigeria.