The concept of hemodilution after blood loss and crystalloid infusion is a surgical maxim that remains unproven in humans. We sought to quantify the effect of hemodilution after crystalloid administration in voluntary blood donors as a model for acute hemorrhage.

A prospective, randomized control trial was conducted in conjunction with community blood drives. Donors were randomized to receive no IV fluid (noIVF), 2 liters of normal saline (NS), or 2 liters lactated Ringer's (LR) after blood donation. Blood samples were taken before donation of 500 mL of blood, immediately after donation, and after IV fluid administration. Hemoglobin (Hgb) was measured at each time point. Hemoglobin measurements between time points were compared between groups using standard statistical tests and the Bonferroni correction for multiple comparisons. Statistical significance was set at p ≤ 0.0167.

Of 165 patients consented, 157 patients completed the study. Average pre-donation Hgb was 14.3 g/dL. There was no difference in the mean Hgb levels after blood donation between the 3 groups (p > 0.05). Compared with the control group, there was a significant drop in Hgb in the crystalloid infused groups from the post-donation level to post-resuscitation (13.2 vs 12.1 vs 12.2 g/dL, p < 0.0001). A formula was created to predict hemoglobin levels from a given estimated blood loss (EBL) and volume replacement (VR): Hemodilution Hgb = (mean pre-donation Hgb – hemorrhage Hgb drop – equilibration hemoglobin drop - resuscitation Hgb drop) = Mean pre-donation Hgb – [(EBL/TBV)*l] – [(EBL/TBV)*h] – [(VR/TBV)*r], l = 5.111g/dL = blood loss coefficient, h = 6.722 g/dL = equilibration coefficient, r = 2.617g/dL = resuscitation coefficient.

This study proves the concept of hemodilution and derived a mathematical relationship between blood loss and resuscitation. These data may help to estimate response of hemoglobin levels to blood loss and fluid resuscitation in clinical practice.