Relationship between intraoperative fluid administration and perioperative outcome after pancreaticoduodenectomy: results of a prospective randomized trial of acute normovolemic hemodilution compared with standard intraoperative management.

Fischer M, Matsuo K, Gonen M, et al.
Ann Surg 2010;252(6):952-958.
NATA Rating :
Review by : O. Habler
NATA Review

In their prospective, controlled, randomized study, Fischer and coworkers investigated the effects of the performance of acute normovolemic hemodilution (ANH, target hemoglobin concentration 8 g/dL) on the 30-day perioperative transfusion rate (primary endpoint) and the quality of intestinal anastomoses (secondary endpoint) in 130 patients undergoing pancreaticoduodenectomy (ANH group: n = 65; standard procedure STD-group: n = 65). During ANH whole blood is collected from the patient in the direct preoperative period and replaced by crystalloids and/or colloids in order to maintain a constant circulating intravascular volume (normovolemia). The allogeneic transfusion-sparing effect of ANH is mainly based on the fact that during a blood loss the hemodiluted patient will lose diluted blood, i.e. less red blood cells per mL of blood loss as compared to a patient with a normal hematocrit. The lower the targetet post-ANH Hct of the patient, the more pronounced this blood-sparing effect. In case of an intra- or perioperative transfusion indication, the hemodiluted patient receives first his autologous ANH whole blood (including coagulation factors and platelets).

The main problem with the study of Fisher et al. is the fact that there was no intraoperative transfusion indication at all in the patients described, as reflected by the 0% intraoperative transfusion rate in both groups (in fact one patient in each study arm was transfused intraoperatively). Given the median blood loss of 700 (ANH group) and 500 mL (control group), this comes as no surprise. However, the patients of the ANH group donated up to 3000 mL autologous whole-blood (median 2250 mL) and were therefore infused with up to 1500 mL 5% albumin solution and 4500 mL crystalloidal solution, i.e. up to 6 liters more than the patients in the control group. Moreover, in all cases unused ANH whole blood was returned to the patient after completion of surgery representing another additional substantial volume load. In a subgroup analysis of patients with blood losses > 700 mL, ANH decreased allogeneic transfusion by 39% – unfortunately without statistical significance. It is absolutely no surprise that there was no advantage of ANH observed with regard to lowering the allogeneic transfusion rate which by the way was with 18.5% already very low in the STD group. It is also no surprise that the extreme volume overload of patients in the ANH group led to higher insufficiency rates of their intestinal anastomoses. The failure of ANH in the present study is, in my opinion, not explained by the insufficiency of ANH itself but by some inherent methodologic problems.

– Oliver Habler