Abstract Body

As countries work towards attaining UNAIDS 90-90-90 targets, challenges related to Early Infant Diagnosis (EID) of HIV should be addressed. In Malawi, EID programmes use dried blood spot (DBS) and HIV PCR with turn-around-times (TAT) of 2-3 months with 33% of exposed infants lost to follow-up. There is an urgent need for point-of-care tests (POCTs) which can improve TAT and reduce loss to follow-up. We evaluated the feasibility, sensitivity and specificity, turn-around-time, acceptability and usability of Cepheid Xpert HIV-1 Qual assay (Xpert HIV) whole blood protocol in a rural district hospital compared to HIV PCR.

This prospective diagnostic study consecutively recruited children aged 0-14 years attending Mulanje District Hospital (MDH) in Malawi between July-September 2018. All POCT were done on site using Xpert HIV. DBS were prepared for HIV PCR testing at a central facility, Queen Elizabeth Central Hospital (QECH). As a standard procedure for EID testing DBS were also sent to Thyolo District Hospital (TDH) for testing by PCR. We compared the sensitivity and specificity between Xpert HIV and PCR. We also compared the median TAT between Xpert HIV from MDH and PCR from QECH and TDH. Acceptability of Xpert HIV was evaluated among caregivers and nurses.

Of 600 participants, 324 (54%) were female. 272 (45.3%) were aged over 5 years, 227 (37.83) between 1-5 years and 101 (16.83%) <1 year. Most of the participants 585 (97.5%) were HIV non-infected. A total of 15 participants were diagnosed with HIV. Most HIV positives aged >1 year (11/13 (85%)) started antiretroviral therapy in 1 day and 4/15 (26%) of all HIV positives were lost to follow up. Sensitivity and specificity of Xpert HIV versus PCR at QECH were 100% (95% CI: 78.2 – 100%) and 100% (95% CI: 99.4 – 100%), respectively. Xpert HIV had the shortest median TAT from time of blood test (median = 5.34 hours (interquartile range: 2.45–10.19) compared to PCR performed at QECH (median = 3 days (IQR: 1.05–4.19)), p<0.001 and PCR performed at TDH (median = 24 days (IQR:  20–28)), p<0.001. Results for only 17 of the 32 samples sent to TDH were received. Xpert HIV was well accepted by caregivers and nurses; and was deemed easy to use by laboratory technicians in comparison to PCR.

These results suggest that implementing Xpert HIV may improve EID and linkage into HIV care.