Time until emergence of HIV test reactivity following infection with HIV-1: Implications for interpreting test results and retesting after exposure

Kevin P. Delaney, Debra L. Hanson, Silvina Masciotra, Steven F. Ethridge, Laura Wesolowski, S. Michele Owen

 Understanding the period of time between an exposure resulting in infection with HIV and when a test can reliably detect the presence of that infection, i.e. the test window period, may benefit testing programs and clinicians in counseling patients about when the clinician and the patient can be confident a suspected exposure did not result in HIV infection.  We evaluated the intervals between reactivity of the Aptima HIV-1 RNA nucleic acid test (Aptima) and 20 FDA-approved HIV immunoassays using 222 longitudinally collected plasma specimens from HIV-1 seroconverters from the United States. A multi-model framework based upon two general approaches, interval-censored survival and binomial regression, was implemented to estimate the relative emergence of test reactivity, referred to in this report as an inter-test reactivity interval (ITRI). We then combined ITRI results with simulated data for the eclipse period, the time between exposure and detection of HIV virus by Aptima, to develop estimates of the window period for each test.  The estimated ITRIs were shorter with each new class of HIV tests, ranging from 5.9 to 24.8 days. The 99(th) percentiles of the window period probability distribution ranged from 44 days for laboratory screening tests that detect both antigen and antibody to 65 days for the Western blot test.  Our directly comparable estimates of the emergence of reactivity for 20 immunoassays are valuable to testing providers for interpreting negative HIV test results obtained shortly after exposure, and for counseling individuals on when to retest after an exposure.