OBJECT: Children with craniosynostosis may require cranial vault remodeling to prevent or relieve elevated intracranial pressure and to correct the underlying craniofacial abnormalities. The procedure is typically associated with significant blood loss and high transfusion rates. The risks associated with transfusions are well documented and include transmission of infectious agents, bacterial contamination, acute hemolytic reactions, transfusion-related lung injury, and transfusion-related immune modulation. This study presents the Children's Hospital of Richmond (CHoR) protocol, which was developed to reduce the rate of blood transfusion in infants undergoing primary craniosynostosis repair. METHODS: A retrospective chart review of pediatric patients treated between January 2003 and Febuary 2012 was performed. The CHoR protocol was instituted in November 2008, with the following 3 components; 1) the use of preoperative erythropoietin and iron therapy, 2) the use of an intraoperative blood recycling device, and 3) acceptance of a lower level of hemoglobin as a trigger for transfusion (< 7 g/dl). Patients who underwent surgery prior to the protocol implementation served as controls. RESULTS: A total of 60 children were included in the study, 32 of whom were treated with the CHoR protocol. The control (C) and protocol (P) groups were comparable with respect to patient age (7 vs 8.4 months, p = 0.145). Recombinant erythropoietin effectively raised the mean preoperative hemoglobin level in the P group (12 vs 9.7 g/dl, p < 0.001). Although adoption of more aggressive surgical vault remodeling in 2008 resulted in a higher estimated blood loss (212 vs 114.5 ml, p = 0.004) and length of surgery (4 vs 2.8 hours, p < 0.001), transfusion was performed in significantly fewer cases in the P group (56% vs 96%, p < 0.001). The mean length of stay in the hospital was shorter for the P group (2.6 vs 3.4 days, p < 0.001). CONCLUSIONS: A protocol that includes preoperative administration of recombinant erythropoietin, intraoperative autologous blood recycling, and accepting a lower transfusion trigger significantly decreased transfusion utilization (p < 0.001). A decreased length of stay (p < 0.001) was seen, although the authors did not investigate whether composite transfusion complication reductions led to better outcomes.