BACKGROUND: To estimate HIV incidence several methods have been used to discriminate recent HIV infections from long-standing infections using a single serum sample. OBJECTIVE: To evaluate the performance of the anti-HIV avidity index (AI) for identifying recent HIV infections in individuals with a known date of seroconversion from Uganda, where the predominant HIV subtypes are A and D. STUDY DESIGN: We selected 149 repository serum samples from Ugandan HIV-positive individuals and evaluated the AI. Specimens collected < or =6 months after seroconversion were considered as recent infections, and those collected >6 months as long-standing infections. All specimens were serotyped using a V3 peptide enzyme immunoassay. RESULTS: The mean AI was 0.55+/-0.21 among the 108 patients with recent infections and 0.93+/-0.14 among the 41 samples from long-standing infections (p<0.0001). The AI test showed a sensitivity of 85.2% and a specificity of 85.4% at a cutoff of 0.80. No significant association was observed between serotype and the misclassification of samples by AI. CONCLUSIONS: The AI, which is inexpensive and easy-to-perform, can be useful in identifying recent HIV infections in countries where HIV-1 non-B subtypes are prevalent.
BACKGROUND: To estimate HIV incidence several methods have been used to discriminate recent HIV infections from long-standing infections using a single serum sample. OBJECTIVE: To evaluate the performance of the anti-HIV avidity index (AI) for identifying recent HIV infections in individuals with a known date of seroconversion from Uganda, where the predominant HIV subtypes are A and D. STUDY DESIGN: We selected 149 repository serum samples from Ugandan HIV-positive individuals and evaluated the AI. Specimens collected < or =6 months after seroconversion were considered as recent infections, and those collected >6 months as long-standing infections. All specimens were serotyped using a V3 peptide enzyme immunoassay. RESULTS: The mean AI was 0.55+/-0.21 among the 108 patients with recent infections and 0.93+/-0.14 among the 41 samples from long-standing infections (p<0.0001). The AI test showed a sensitivity of 85.2% and a specificity of 85.4% at a cutoff of 0.80. No significant association was observed between serotype and the misclassification of samples by AI. CONCLUSIONS: The AI, which is inexpensive and easy-to-perform, can be useful in identifying recent HIV infections in countries where HIV-1 non-B subtypes are prevalent.
Authors: Timothy D Mastro; Andrea A Kim; Timothy Hallett; Thomas Rehle; Alex Welte; Oliver Laeyendecker; Tom Oluoch; Jesus M Garcia-Calleja Journal: J HIV AIDS Surveill Epidemiol Date: 2010-01-01
Authors: Oliver Laeyendecker; Ron Brookmeyer; Matthew M Cousins; Caroline E Mullis; Jacob Konikoff; Deborah Donnell; Connie Celum; Susan P Buchbinder; George R Seage; Gregory D Kirk; Shruti H Mehta; Jacquie Astemborski; Lisa P Jacobson; Joseph B Margolick; Joelle Brown; Thomas C Quinn; Susan H Eshleman Journal: J Infect Dis Date: 2012-11-05 Impact factor: 5.226
Authors: Giovanni Lorenzin; Franco Gargiulo; Arnaldo Caruso; Francesca Caccuri; Emanuele Focà; Anna Celotti; Eugenia Quiros-Roldan; Ilaria Izzo; Francesco Castelli; Maria A De Francesco Journal: Microorganisms Date: 2019-12-23