Background: Liver resection is a major operation for which, even with the improvements in surgical and anesthetic techniques, the reported rate of blood transfusion was rarely less than 30%. About 60% of transfused patients require only 1 or 2 units of blood, a blood requirement that may be accommodated by the use of acute normovolemic hemodilution (ANH).
Methods: The efficacy, hemodynamic effects, and safety of ANH were investigated in a randomized, active-control study in patients with American Society of Anesthesiologists status I-II who were undergoing major liver resection with fentanyl-nitrous oxide-isoflurane anesthesia. Patients were randomized to the ANH (n = 39) or control group (n = 39). Patients in the ANH group underwent hemodilution to a target hematocrit of 24%. The indication for blood transfusion was standardized. In both groups transfusion was started at a hematocrit of 20%. The primary efficacy endpoint was the avoidance of allogeneic blood transfusion in the intraoperative period and first 72 h after surgery. Various laboratory and hemodynamic parameters as well as postoperative morbidity were monitored to define the safety of ANH in this patient population.
Results: During the perioperative period, 14 control patients (36%) received at least one unit of allogeneic blood compared with 4 patients (10%) in the ANH group ( < 0.05). The hemodilution process was not associated with significant changes in patients' hemodynamics. Morbidity was similar between the control and the ANH groups. Postoperative hematocrit levels and biochemical liver, renal, and standard coagulation test results were similar in both groups.
Conclusions: Acute normovolemic hemodilution in patients with American Society of Anesthesiologists status I-II undergoing major liver resection may allow a significant number of patients to avoid exposure to allogeneic blood.