Evaluation of HIV antigen and antibody combination elisas for diagnosis of HIV infection in Dar es salaam, Tanzania

Abstract

Background: Fourth generation HIV screening assays permit a simultaneous detection of HIV antigens and antibodies and have reduced the diagnostic period on average by seven days. New antigen/antibody combination ELISAs have been introduced for screening of blood donors and clinical samples and are commercially available. The Enzygnost® anti- HIV-1/2 Plus that has been used in the previous HIV ELISAs algorithm for more than fifteen years was withdrawn from the market since 2009. A new antigen/antibody combination HIV ELISA, Enzygnost HIV Integral II, has been introduced and is now available in the market since 2009, and has never been evaluated. Currently, there is validated national HIV ELISA testing algorithm in place. Objectives: To evaluate the performance and operational characteristics of HIVantigen/ antibody combination ELISAs as well as to formulate HIV ELISA testing algorithms that could be used as an alternative HIV testing strategy for laboratory diagnosis of HIV infection in Tanzania. Materials and Methods: This was a laboratory-based evaluation of three HIV antigen/antibody combination ELISAs conducted between October 2011 and May 2012. The Enzygnost HIV Integral II Antibody/ Antigen (Siemens, Germany), Murex HIV antigen/antibody (Abbott Murex, UK) and Vironostika HIV Uniform II antigen/antibody (Biomerieux, the Netherlands) were evaluated using blood samples from blood donors, voluntary counselling and testing, pregnant women and HIV care and treatment centre. INNO-LIA HIV I/II immunoblot kit (Innogenetics, Belgium) was used as a confirmatory test. Panel AU PRB945 was used to evaluate the seroconversion activity of the assays. Data were double entered and checked for accuracy before analysis using Epi info 3.5.1 and Microsoft excel software. Sensitivity, specificity, PPV and NPV with 95% CI of HIV antigen/antibody combination ELISAs were determined. Operational characteristic of HIV antigen/antibody ELISAs were compared using a defined scoring system. Results: A total of 600 blood samples were included in the evaluation. Four samples gave indeterminate results on immunoblot assay (1 positive from ANC attendee and 3 negative from one blood donor and two ANC attendees) and were excluded in the analysis. A total of 209/596 (35.1%) serum samples were confirmed HIV positive. Of these, the prevalence of HIV infection was 2.3% (3/130), 2.3% (3/127), 2.2% (3/139) and 100% (200/200) for VCT clients, ANC attendees, blood donors and CTC patients, respectively. Three hundredand eighty seven (64.9%) were HIV negative samples. Sensitivity was 100% (95% CI; 98.3-100%) for all the three HIV ELISAs. The specificity for the Enzygnost HIV Integral II and Murex was 100% (95% CI; 99.1-100%). The final specificity at repeat testing was 99.5% (95% CI; 98.2-99.9%) for Vironostika. Enzygnost HIV Integral II detected HIV infection seven days since first bleed. The three assays had same operational aspects in terms of storage conditions (2-8 0C), incubation temperature (370C) and need of special equipment. Enzygnost HIV Integral II is more expensive as compared to Murex HIV Ag/Ab and Vironostika HIV Ag/Ab. Conclusions: The initial testing using either Vironostika or Murex HIV antigen/antibody combination ELISA followed by testing of reactive samples on the Enzygnost HIV Integral II gave a sensitivity and specificity of 100% with reduced window period. Recommendations: Combination of two HIV antigen/antibody combination ELISAs can be used as an alternative confirmatory testing strategy for screening of donated blood at the National and Zonal blood transfusion centres and in lab diagnosis of HIV infection at the Regional, National and Zonal Hospital Laboratories.