The aim of the study was to determine whether a single unit of red blood cell (RBC) transfusion is a true criterion of poor patient care. Medical records of 148 patients, 76 (51.4%) males, 72 (48.6%) females, mean age 66.88 +/- 14.56 years who have received a single unit RBC transfusion at the Department of Medicine in the period from 1997 to 2000 were retrospectively studied. Pretransfusion mean hemoglobin (Hb) value (Hb1) was 85.0 +/- 12.0 (mod 78) g/L, increasing to (Hb2) 96.9 +/- 11.8 (mod 90) g/L after transfusion and the last measured value (Hb3) was 98.2 +/- 13.4 (mod 102) g/L. Six patients (66.6%) of 9 with decrease in Hb3 value (group A) died, 13 patients (35.1%) of 37 with raise in Hb3 level < 5 g/L (group B) and 5-10 g/L (group C) died during the hospital stay vs. 7 patients (6.8%) of 102 with Hb3 level raise > 10 g/L (group D). The differences between the Hb2 and Hb1 values were significant considerably from the differences between the Hb3 and Hb1 values (p = 0.001), as well as the difference in the distribution of differences between Hb2 and Hb1 (p 0.028) and Hb3 and Hb1 (p = 0.001) according to the disease outcome and the difference in mortality of men and women (p = 0.021). 38.4% of the deceased patients were from the group of coronary syndrome and/or heart failure. Statistically significant were the differences between the arterial oxygen tension (PaO2) at Hb1 and Hb2 (p = 0.0001), as well as between the arterial oxygen saturation (SaO2) at the Hb1 and Hb2 values (p = 0.0001). There is no strict clinical indication for a single unit of RBC transfusion. Determinants of the need and amount of RBC transfusion except hemoglobin value include the etiology and duration of the anemia, the patient's ability to compensate for decreased oxygen carrying capacity and tissue oxygen requirements. The majority of a single unit RBC transfusion was appropriate for the clinical situation. Our data indicate that the use of smaller volumes of allogenic RBC transfusion was not a bad clinical practice.