The authors describe a patient who demonstrated positive blood responsiveness to the nuclear antigens of human immunodeficiency virus (HIV) (p17, p31 and p55), observed steadily for 1 year and 4 months. The donor's disease history consideration made it impossible to include him in one of the known groups at risk for HIV infection whereas the lack of any changes in immunoblotting enabled one to exclude the diagnosis of HIV infection. The given case and other similar cases form the basis for introducing the second parallel screening during blood testing for HIV infection to bar the use of such blood for transfusion.