Distribution of human T-lymphotropic virus type I among blood donors: a nationwide Brazilian study.

Previous studies have demonstrated the presence of human T-lymphotropic virus type I (HTLV-I) andlor HTLV type I1 (HTLV-11) in certain populations in Data on qualified blood donors are limited primarily to those in the southeast region of Brazil, while the country's 8,400,000 square miles include diverse geographic, climatic, and sociodemographic characteristics. To assist in the formulation of public health policies for blood donors, we conducted the following seroprevalence study. Between July and September 1993, blood samples were collected from 5842 blood donors in five state capitals. All individuals fulfilled criteria for blood donation (i.e., they were 18-60 years old and they had no reported risk behavior for sexually transmitted diseases). Samples were not linked to donor identification. To avoid repeated sampling of the same donor, the duration of the study was limited to 60 days, which is the permissible interval between donations. Sera were screened for antibodies to HTLV-1/11 by using a recombinant HTLV-I enu enzyme-linked immunosorbent assay and were confirmed by an HTLV-I recombinant p21 enu-enhanced Western blot (Cambridge Biotech Corp., Worcester, MA). Serologic discrimination between HTLV-I and HTLV-I1 was performed by using specific synthetic peptides (Select HTLV, Coulter Corp., Hialeah, FL). All Western blot samples reacted with a synthetic peptide sequence derived from HTLV-1. Results were interpreted according to manufacturers' instructions. The overall HTLV-I seroprevalence was 0.41 percent (95% CI, 0.25-0.58) (Table 1). The city of Salvador, which has socio-demographic characteristics like those of some sub-Saharan cities, had the highest seroprevalence (1.35%). Unexpectedly, a relatively low HTLV-I seroprevalence (0.08%) was found in Manaus (capital of the Amazonas State) in the Amazon Basin. None of the 5842 samples was confirmed positive for anti-HTLV-11, although a previous report described a relatively high HTLV-I1 seroprevalence among Ge-speaking Cayapo and Kraho Indians in Central Brazil,' as well as a low HTLV-I1 seroprevalence (0.03%) in Sad Pau10.~ The discrepancy between the absence of anti-HTLV-I1 in our study of blood donors in state capitals and the presence of anti-HTLV-I1 reported by others may be explained by 1) the possibility that we screened an insufficient number of donors in our study, as suggested by the low HTLV-I1 seroprevalence (0.03%) found by other^,^ and 2) the fact that individuals living in Amazon Basin metropolitan areas have not mixed, recently or in the past, with the relatively isolated Brazilian Indians.' Our findings indicate that a more comprehensive sero-prevalence study is needed to determine the prevalence of HTLV-I1 …