Objective To study the effects of three different infusion pumps on red blood cell (RBC) integrity. Setting Laboratory. Interventions Transfusion of packed RBCs using three different pumps (syringe pump, conventional peristaltic pump, and a new volumetric pump with shuttle mechanism). Flow rate was set at 20 mL/hr and duration was 2.5 hrs, simulating a neonatal transfusion. Experiments were repeated in each pump with eight different units of packed RBCs. Measurements and Main Results Plasma hemoglobin, potassium, lactate dehydrogenase, bilirubin, and osmotic fragility were assessed before and after transfusion. There was a significant degree of hemolysis in the RBC bags before transfusion (mean ± sd for plasma hemoglobin, 5.9 ± 3.0 g/L; potassium, 40.0 ± 11.8 mmol/L; lactate dehydrogenase, 64 ± 38 units/L; total bilirubin,: 1.5 ± 0.8 μmol/L). Overall increase between samples before and after transfusion was 12% for plasma hemoglobin (p = .01, paired Student's t-test), 2% for potassium (p = .03), 20% for lactate dehydrogenase (p = .0001), and 47% for total bilirubin (p = .04). Plasma hemoglobin and lactate dehydrogenase changes were significantly influenced by the type of infusion pump (p = .006 and .01, multiple regression analysis). Regarding these two variables, the new volumetric pump was less hemolytic than the other two pumps. The degree of the transfusion-related RBC changes was also significantly and independently influenced by the storage time of the blood bags with plasma hemoglobin, bilirubin, and osmotic fragility changing less in old bags. Conclusions The new pump's shuttle mechanism seems to be less injurious to RBCs compared with a conventional syringe and a peristaltic infusion pump. However, the clinical significance must be related to the hemolytic changes in the RBC units before they are transfused, which are a major source of the overall hemolytic load.