Poor predictive value of hematocrit and hemodynamic parameters for erythrocyte deficits after extensive elective...

Fluid resuscitation and transfusion therapy are particularly critical in patients undergoing extensive vascular operations because of diffuse atherosclerosis and the risk of perioperative myocardial infarction. Sophisticated perioperative monitoring has reduced the mortality rate substantially, but indications for transfusion remain controversial. We determined erythrocyte volume, (EV), total blood volume (TBV) and plasma volume (PV) preoperatively and 18 to 24 hours postoperatively in 41 elderly patients (68.8 +/- 1.3 years) undergoing elective vascular operations (30 abdominal aortic aneurysmorrhaphy, ten aortofemoral bypass and one carotid endarterectomy). EV was measured using 51chromium-labeled autologous erythrocytes; TBV and PV were calculated from EV and total body hematocrit (peripheral venous hematocrit [HCT] x 0.89). Ideal blood volumes were calculated from nomograms based on body surface area and gender. Relationships between body volumes (percentage of ideal), simultaneously measured peripheral venous HCT and hemodynamic parameters heart rate, mean arterial pressure, central venous pressure, pulmonary capillary wedge pressure, cardiac index and systemic vascular resistance index were studied by stepwise regression. In 24 patients, blood volumes and hemodynamic parameters were also measured in the recovery room. HCT significantly correlated with EV at all three time periods (p less than 0.001), but the ability of HCT to predict EV in an individual patient was relatively poor (r = 0.50 preoperatively; r = 0.54 in recovery room and r = 0.66 24 hour postoperatively). By 24 hours postoperatively, EV had decreased to 78.3 +/- 2.4 percent of ideal EV (range of 47 to 112 percent). However, only two patients had HCT less than 30 despite the fact that 13 of 41 patients had an EV deficit of greater than 30 percent. No patient had a HCT of less than 25 percent. Hemodynamic parameters did not contribute to the prediction of EV, PV or TBV at any time. Two patients had myocardial infarctions postoperatively associated with 24 hour EV deficits of 18.5 and 29.6 percent. One patient died of a pulmonary embolus. Because of these findings, the concept of a "transfusion trigger" must be viewed with caution, since many patients undergoing vascular operations will have considerable EV deficits despite an "acceptable" HCT.