HIV-2 infection in HIV-1 indeterminate sera in Spain [letter]

We read with interest the article of Myers et al. who claimed the identification of 4 human immunodeficiency virus type 2 (HIV-2) cases in a nonendemic area i.e. as the US through the reevaluation of HIV-1 indeterminate sera. We agree that targeting these inconclusive sera may represent a more appropriate mechanism to monitor the spread of HIV-2 in non-African countries than testing blood donors. In a previous study conducted in Madrid at the beginning of 1991 we identified 3 subjects carrying HIV-2 by testing 26 samples with discordant HIV-1 serological results which represents 11.5% of them. Since that time we decided to exclude specific HIV-2 antibodies in all HIV-1 indeterminate sera. IN this form 7 more cases of HIV-2 infection have been detected in our institute until January 1992. Thus the use of this algorithm allows us to detect > 20% of the 35 cases of HIV-2 infection reported in Spain until June 1992. All HIV-2 infected subjects identified after testing HIV-1 indeterminate sera in our institution are African immigrants. They were referred to the Clinic of Tropical Medicine for several reasons but to date no one has been diagnosed with an AIDS-defining condition. Since January 1992 we used a combination viral lysate HIV-1/HIV-2 enzyme immunoassay (EIA) for the screening of AIDS viruses. Furthermore we used a HIV-1 viral lysate Western blot supplemented by one HIV-2 specific peptide (BioKit Barcelona Spain) to confirm EIA-reactive samples. In this form we were able to recognize 5 HIV-2 and 2 HIV-1 plus HIV-2-infected subjects. Polymerase chain reaction (PCR) using gag and long terminal repeat primers was done in peripheral blood mononuclear cells collected from 2 available patients showing dual HIV-1 and HIV-2 seroreactivity and PCR results confirm it (V. Soriano et al. unpublished data). In conclusion results from Myers and us pointed out that by performing HIV-2-specific testing on HIV-1 indeterminate specimens many more HIV-2 infected individuals could be identified providing valuable insights into the extent of HIV-2 infection in nonendemic areas. Moreover because some HIV-1 infected subjects may be also HIV-2 carriers we suggest that for epidemiological purposes confirmatory assays should be able to detect both HIV-1 and HIV-2 antibodies. (full text)