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.

Background: Pancreaticoduodenectomy (PD) can be associated with significant blood loss and transfusion requirements, with potential adverse short- and long-term consequences. The aim of this study was to determine whether acute normovolemic hemodilution (ANH), an established blood conservation technique, reduces perioperative allogeneic transfusions in patients undergoing PD.

Methods: One hundred thirty patients undergoing PD were randomized to ANH or standard management (STDM). In the ANH group, intraoperative blood collection was performed to a target hemoglobin of 8.0 g/dL; crystalloid and colloid were used for volume replacement. Strict transfusion triggers were applied during and after operation. Perioperative complications were prospectively assessed and graded for severity.

Results: From July 2005 to May 2009, 209 patients were registered, 79 excluded, 65 were randomized to ANH, and 65 to STD. The groups were well matched for demographic, operative, and histopathologic variables. Patients undergoing ANH received over 2 L more fluid intraoperatively (6250 mL, range 2000-11850) compared with patients undergoing STD (3900 mL, range 2000-9000) (P < 0.001). Transfusion rates were similar (ANH = 16.9%, 30 units vs STD = 18.5%, 33 units; P = 0.82), as was overall perioperative morbidity (ANH = 49.2% vs STD = 47%, P = 0.86). There was, however, a trend toward more grade-3 complications in patients undergoing ANH (32% vs 23.1% STD, P = 0.17), and complications related to the pancreatic anastomosis (leak/fistula/abscess) were significantly higher in the ANH group (21.5% vs 7.7%, P = 0.045). The intraoperative fluid volume was higher for all patients with pancreatic anastomotic complications (n = 19), regardless of randomization arm (ANH 6000 mL, range 2800-11350 mL vs STD 5000 mL, range 2000-11850 mL, P < 0.042).

Conclusion: In this randomized trial of patients undergoing PD, ANH did not reduce allogeneic transfusions and resulted in more pancreatic anastomotic complications, likely related to greater intraoperative fluid administration. The benefits of ANH do not necessarily extend to all procedures, and restrictive intravenous fluid management during PD may help improve postoperative outcome.