RESUMO
STUDY DESIGN: A case-control study. OBJECTIVES: 1) To determine if hemodilution adequately meets the transfusion needs in children who undergo posterior spinal fusion for idiopathic scoliosis and 2) to compare the efficacy of the various methods used to reduce the risk of allogeneic blood transfusion at the authors' institution. SUMMARY OF BACKGROUND DATA: Methods to reduce blood loss and avoid allogeneic blood transfusion caused by extensive spinal surgery in adolescents include 1) autologous blood predonation, 2) controlled hypotensive anesthesia, 3) intraoperative salvage of shed blood (cell saver), 4) acute normovolemic hemodilution, and 5) transfusion decisions by clinical judgment rather than by a preset value of hemoglobin. Although all methods have some efficacy, it is not clear which methods, separate or combined, are best in the adolescent scoliosis population. METHODS: Hemodilution, hypotensive anesthesia, and cell saver were used in 43 children between June 1996 and July 1997. A comparison group (43 children) underwent similar surgery with hypotensive anesthesia and cell saver, but no hemodilution (between July 1995 and December 1996). These two groups were similar with respect to means of age, levels of instrumentation, magnitude of curvature, estimated blood volume, mean arterial pressure, duration of surgery, duration of anesthesia, estimated blood loss, volume returned from cell saver, volume in the hemovac drain, and length of hospitalization. The groups differed in preoperative hemoglobin and hematocrit and in volume of crystalloid used. RESULTS: Transfusions were given to 34 of 43 patients (79%) in the nonhemodilution group. These patients received 61 units of packed cells (57 autologous, 2 donor directed, and 2 allogeneic). In comparison, 16 of 43 patients (37%) in the hemodilution group required transfusion. They received 16 units of packed cells (15 autologous and 1 allogeneic). There was no significant difference between the groups with respect to postoperative hemoglobin and hematocrit immediately after surgery (hemodilution, 10.2/29.2; nonhemodilution, 10.0/29.1), postoperative day 1 (hemodilution, 9.2/26.9; nonhemodilution, 9.2/27.3), and postoperative day 2 (hemodilution 9.0/26.4; nonhemodilution, 9.2/27.1). There were non complications related to the technique of hemodilution in the 43 patients of this group. Cell saver was used in all patients, but sufficient volume to return blood to the patient was available in only 23 hemodilution patients (mean volume, 230 mL) and 25 nonhemodilution patients (mean volume, 215 mL). In only two patients of each group (< 5%) did the volume returned prevent the absolute need for additional transfusions. CONCLUSIONS: Hemodilution was safely used as a method to satisfy the perioperative transfusion requirements of adolescents undergoing extensive spinal surgery. By allowing patients to arrive at surgery with a higher preoperative hemoglobin and hematocrit, and by decreasing the quantity of predonated autologous blood-collected and therefore used, the hemodilution method may indirectly decrease the quantity of postoperative autologous transfusions in this population. Cell saver was not shown to be effective, and its selective use is recommended.