Statistical evaluation of plasma substitutes based on 10 variables.

Inasmuch as no single variable is able to predict the outcome or evaluate the effect of various treatments used for resuscitation of shock, a statistical method was applied to evaluate several plasma substitutes by a numerical system based on 10 different hemodynamic and metabolic variables. Shock was induced in 60 dogs by laparatomy and exteriorization of the small intestines. After 3 h, the intestine was returned and the abdomen closed. Fluid infusion was then given during a 20-min period. Measurements were carried out during the next 4 h. Seven therapeutic agents were studied in 7 groups: control (no infusion); Ringer's acetate; gelatin; dextran-40; dextran-70; ACD-plasma; albumin. Variables included: cardiac output, oxygen consumption, plasma volume, hematocrit, skeletal muscle capillary blood flow and permeability surface area, arterial blood pH and base excess, mean arterial blood pressure, and the extravascular water gain. Dextran-40, dextran-70, and albumin restored an average of 7 variables to values above the preshock median value. Gelatin and plasma restored 4 variables and Ringer's acetate restored 3 variables. When no infusion was given, no variable was restored and the animals remained in shock. It is concluded that 3.5% colloid solutions of dextran-40, dextran-70, and albumin are more effective than plasma, gelatin, and Ringer's acetate in this order.