Rapid diagnostic testing is more accurate with whole blood compared with serum

HIV testing accuracy is very important because both negative or positive testing that is wrong can have bad consequences.  Rapid diagnostic testing (RDT) is frequently done for HIV testing in Ghana because it is easy to use, cheaper and produces results in about 15 minutes.  RDT is often used for patient testing as well as screening of blood by blood donation. Blood samples that are used for testing can either be whole blood or blood serum. There have been reports that depending on the type of blood sample use, some RDTs can show a positive and negative test for the same HIV person depending on whether whole or serum blood is used. The researchers wanted to determine which of the two types of blood samples (whole or serum) when used is better at detecting HIV in the HIV test kits used in Ghana.  

This study tested 280 people for HIV using the RDT “First-Response HIV-1-2 kits”.  Based on lab tests and medical records, some were known to be infected with HIV and others were not. Blood samples were drawn from patients and fed into the RDT kits either in a whole or serum blood form. The participants of the study were from three different hospitals in the Northern Region of Ghana were recruited from outpatient and HIV treatment wards. 

Whole blood samples are more accurate than serum samples when using RDT.  Some of the serum samples were found to have positive test results where the person did not actually have HIV.  100% of those who tested positive using whole blood were correctly identified as having HIV.  Similarly, whole blood samples that were negative were determined to be negative 100% of the time.  Unfortunately, serum testing missed about 1 in 5 people who had negative HIV testing.
Recommendations and Conclusions
Whole blood should be used during RDT.  Serum testing with RDT is problematic because the risk of not catching a person with HIV and potentially transfusing blood that has HIV is a real risk when only using serum testing.  This study had big limitations.  Of the samples that produced differing HIV results, almost 30% were removed from the analysis because they could not determine HIV status after repeated testing.  This could have skewed the results which show 100% accuracy of RDT in determining those with HIV.  Lastly, results were only determined for HIV-1, so testing for HIV-2 is not determined by this study.

Article Title:  Assessing the sensitivity and specificity of First Response HIV-1-2 test kit with whole blood and serum samples: a cross-sectional study
Author (s) (FN, LN): Raymond Boadu, George Darko, Priscilla Nortey, Patricia Akweongo and Bismark Sarfo