Background:  Collecting whole blood on filter paper simplifies collection, transport, and storage of specimens. As dried blood spots (DBS) are frequently collected in tropical environments we examined the stability of HIV-1 DNA in DBS stored in high heat and humidity to assist in designing reliable methods for polymerase chain reaction (PCR) testing to diagnose and study HIV-1 infection in infants.

Methods:  We compared HIV-1 DNA recovery from multiple sets of DBS from 2 HIV-1 infected adult donors spotted onto 903 filter paper and stored at –20°C or 37°C. The DBS were made in the VQA laboratory and sent to 8 laboratories at ambient temperature. DNA was extracted from 10 to 50 µL DBS per donor within a day of receipt. We assayed 2 extracts at full concentration and 2 each diluted 1:2, 1:4, 1:8, or 1:16 using the Roche Amplicor HIV-1 Test version 1.5 for DNA (each in duplicate). Half the remaining DBS were stored at –20°C and half at 37°C with high humidity. After storage for 2, 6, or 12 months, neat or diluted DNA from DBS were tested. Each PCR was scored: (+) = optical density (OD) ≥0.8; (–) = OD <0.2; indeterminate = OD 0.2 to 0.8; invalid = specimen and internal control OD were <0.2. PCR results were compared between storage conditions and over time, using general estimating equation to account for repeat measures.

Results:  After exclusion of invalid and incomplete data, 1832 PCR from 916 DBS were analyzed, including 100 DBS evaluated on receipt at the laboratories, 418 DBS were stored at –20°C, and 398 at 37°C. The 200 baseline PCR resulted in 164 (82%) positives, 0 (0%) indeterminate, and 36 (18%) negatives. Combining data from subsequent time points, the 836 PCR from DBS stored at –20°C resulted in 651 (78%) positives, 6 (0.8%) indeterminates, and 179 (21.4%) negatives. The 796 PCR from DBS stored at 37°C resulted in 439 (55%) positives, 17 (2.1%) indeterminates, and 340 (42.7%) negatives (p <0.0001). DBS stored at –20°C had similar rates of positive PCR over time:  2 months 75%, 6 months 78%, and 12 months 80%. One lab had a significant drop in recovered DNA stored at 37°C (17.5% positives at 2 months, 0% at 6 months), but in the majority of labs DNA recovery from DBS stored at 37°C decreased between baseline and 2 months (p = 0.0030), and was stable thereafter:  2 months, 60% positives; 6 months, 61.7%; 12 months, 59.3% (p = 0.8810).

Conclusions: Exposure of DBS to 37°C and high humidity for 2 to 12 months impairs recovery of HIV-1 DNA from DBS, whereas DNA recovery is preserved when DBS are stored frozen.