BACKGROUND: Commercially-available kits for HIV-1 detection include instructions for detecting HIV-1 in plasma and DBS, but don't support other specimen types.
OBJECTIVES: Show quantitative stability of HIV-1 total nucleic acid (TNA) in blood and improved HIV-1 detection in alternative specimen types.
STUDY DESIGN: Whole blood and DBS specimens, tested as part of an external quality assurance program for qualitative HIV-1 detection, were used to evaluated error rates (false negative [FN], false positive [FP] and indeterminant [IND] results) across assays (internally developed [IH], Roche Amplicor [RA], and Roche TaqMan Qual [TQ]) and specimen types (frozen whole blood [BLD], DBS and cell pellets [PEL]). A modified Roche TaqMan HIV-1 assay was used to quantify HIV-1 TNA.
RESULTS: Significantly higher error rates were noted in DBS across all of the assays (4% vs. 0% for DBS and PEL, IH, p = 0.005; 4% vs. 0.1% for DBS and PEL, RA, p < 0.001; 10% vs. 1% for DBS and PEL or BLD, TQ, p < 0.001). HIV TNA concentration is stable in BLD (day 1 vs. day 10, p = 0.39) and higher than DBS (p < 0.001).
CONCLUSIONS: Transporting refrigerated whole blood for centralized processing into alternative specimen types will improve the sensitivitiy of HIV-1 detection in samples with low virus loads.