Use of pulse pressure variation to estimate changes in preload during experimental acute normovolemic hemodilution

Acute normovolemic hemodilution (ANH) is an alternative to blood transfusion in surgeries involving blood loss. This experimental study was designed to evaluate whether pulse pressure variation (PPV) would be an adequate tool for monitoring changes in preload during ANH, as assessed by transesophageal echocardiography.

Twenty-one anesthetized and mechanically ventilated pigs were randomized into three groups: CTL (control), HES (hemodilution with 6% hydroxyethyl starch at a 1:1 ratio) or NS (hemodilution with saline 0.9% at a 3:1 ratio). Hemodilution was performed in animals of groups NS and HES in two stages, with target hematocrits 22% and 15%, achieved at 30-minute intervals. After two hours, 50% of the blood volume withdrawn was transfused and animals were monitored for another hour. Statistical analysis was based on ANOVA for repeated measures followed by multiple comparison test (P<0.05). Pearson's correlations were performed between changes in left ventricular end-diastolic volume (LVEDV) and PPV, central venous pressure (CVP) and pulmonary artery occlusion pressure (PAOP).

Group NS received a significantly greater amount of fluids during ANH (NS, 900 ± 168 mL vs. HES, 200 ± 50 mL, P<0.05) and presented greater urine output (NS, 2643 ± 1097 mL vs. HES, 641 ± 338 mL, P<0.001). Significant decreases in LVEDV were observed in group NS from completion of ANH until transfusion. In group HES, only increases in LVEDV were observed, at the end of ANH and at transfusion. Such changes in LVEDV (∆LVEDV) were better reflected by changes in PPV (∆PPV, R=-0.62) than changes in CVP (∆CVP, R=0.32) or in PAOP (∆PAOP, R=0.42, respectively).

Changes in preload during ANH were detected by changes in PPV. ∆PPV was superior to ∆PAOP and ∆CVP to this end.