A four-part experiment was undertaken to determine the feasibility of adding normal saline warmed to 45 C to units of packed red blood cells (PRBCs) to warm them and increase blood flow in transfusions. In Part One, the flows of fresh and outdated PRBCs through a modified infusion system were compared. The results showed that there was no significant difference in flow rates and between fresh (11.5 gm/min +/- 0.98 SD) and outdated (10.08 gm/min +/- 1.45 SD) PRBCs using the Student t test for uneven groups. This allowed outdated PRBCs to be used in place of fresh for a good portion of the experiment. In Part Two, PRBCs were run through a Travenol blood warmer coil (Travenol Laboratories, Deerfield, Illinois), and flow rate was 5.17 gm/min (+/- 0.61 SD) with delivery temperature of 29 C (+/- 1.58 SD). In Part Three normal saline warmed to 45 C was added to units of PRBCs, and flow rate was 57.72 gm/min (+/- 0.61 SD) with delivery temperature of 26 C (+/- 1.49 SD). In Part Four normal saline warmed to 45 C was added to units of PRBCs as in Part Three and then run through the blood warmer. The flow rate was 18.23 gm/min (+/- 3.44 SD) with delivery temperature of 35.3 C (+/- 1.45 SD). Plasma hemoglobins also were determined on fresh PRBCs before and after addition of warmed normal saline and passage through the blood warmer. There was no evidence of significant hemolysis of the blood due to this technique.(ABSTRACT TRUNCATED AT 250 WORDS)