par Stoffels, Karolien ;Vanroye, Fien;Mortier, Virginie;Debaisieux, Laurent ;Delforge, Marie-Luce ;Depypere, Melissa;Dessilly, Géraldine;Vaira, Dolores;Vancutsem, Ellen;Van den Wijngaert, Sigi;Van Laethem, Kristel;Vercauteren, Koen K.O.A.;Verhofstede, Chris;Fransen, Katrien
Référence The Journal of infectious diseases, 222, 10, page (1660-1669)
Publication Publié, 2020-10
Référence The Journal of infectious diseases, 222, 10, page (1660-1669)
Publication Publié, 2020-10
Article révisé par les pairs
Résumé : | This retrospective study evaluated the reactivity of 3 human immunodeficiency virus (HIV) confirmatory assays (INNO-LIA, Geenius, and MP) and 7 HIV rapid tests on samples from 2 different study populations in Belgium. For the early-treated cohort (83 HIV-1 adult patients treated within 3 months after infection), HIV-1 diagnosis was not obtained in at least 1 confirmatory assay in 12.0% (10/83) and in an HIV rapid test in 31.3% (26/83). Confirmation assay sensitivities ranged from 87.5% to 95.2%, whereas rapid test assay sensitivities ranged from 75.9% to 100%. The time to treatment initiation or the length of time on treatment did not have a statistical influence on the probability to obtain a false-negative test result. The fastest reversion was demonstrated after 4 months of treatment. Among the long-term treated cohort (390 HIV-1 patients with ≥ 9 years of undetectable viral load), false-negative test results were found in at least 1 HIV confirmatory assay for 2.1% (8/390) of the patients and in a HIV rapid test for 4.9% (19/390). Confirmation assay sensitivities ranged from 98.1% to 99.5%, whereas rapid test sensitivities ranged from 96.2% to 100%. Longer treatment increased nonreactivity of the HIV rapid tests (P = .033). Undetectable viral load decreases the sensitivities of HIV diagnostic tests, and further monitoring of the performance of serological assays is advised. |