Balanced crystalloid compared with balanced colloid solution using a goal-directed haemodynamic algorithm

Controversy exists regarding the optimal i.v. fluids for use with a goal-directed haemodynamic algorithm. In a double-blind pilot study, we randomly assigned 50 patients with primary ovarian cancer undergoing cytoreductive surgery to receive either balanced crystalloid or balanced starch (HES, 130/0.4, 6%) solutions up to the dose limit (50 ml kg−1). Fluids were administered to optimize stroke volume measured by oesophageal Doppler within a goal-directed haemodynamic algorithm. Baseline subject characteristics were similar in both groups. The balanced HES solution maintained stroke volume (P=0.012) better with administration of less fluid. Subjects in the colloid group reached the dose limits of the study medication less frequently (92% vs 62%, P=0.036) and later (2:26 vs 3:33 h, P=0.006) and also required less transfusion of fresh-frozen plasma units (6.0 vs 3.5 units, P=0.035) compared with the crystalloid group. Intra- and postoperative urine output and perioperative plasma levels of creatinine and neutrophil gelatinase-associated lipocalin as renal injury marker were similar in both groups. No differences in the length of intensive care unit and hospital stay were found. Using a goal-directed haemodynamic algorithm to optimize stroke volume, a balanced HES solution is associated with better haemodynamic stability and reduced need for fresh-frozen plasma. There were no signs of renal impairment by colloid solutions when fluid administration is targeted to optimize cardiac preload.