ObjectiveTo investigate the effect of acute non-isovolemic hemodilution (ANIH) on spinal surgery. MethodsFrom January 2012 to July 2013, 80 patients scheduled for spinal surgery were enrolled. The patients were randomized into four groups, with 20 in each group. Patients in group A were infused with Ringer's injection and polygeline for fluid loss, physiological needs and blood loss. In group B, the patients were infused with acute hypervolemic hemodilution. In group C, patients were infused with acute nomovolemic hemodilution. In group D, patients were infused with ANIH. The hemodynamics and arterial blood gas indexes were detected in all the patients. ResultsThe mean blood pressure in group A and C was significantly more stable than that in group B and D. The central vein pressure in group B and D was significantly higher than that in group A and C after hemodilution (P<0.05), while the hematocrit in group B and D was significantly lower than that in group A and C (P<0.05). Allogeneic blood transfusion was performed in group A and B, while it was not performed in group C and D. ConclusionANIH can reduce the volume of blood loss during spinal surgery, and it is safe.