The use of volume kinetics as a method to optimise fluid therapy

Background: This thesis presents a new application of pharmacokinetic analysis, usually applied to drugs, to the disposition of intravenous fluid. One-volume and two-volume models were developed and non-linear regression was used to characterise the blood marker dilution versus time, relative to the administered dose of intravenous fluid. The pharmacokinetic, or rather volume kinetic, behaviour of fluids were examined under different conditions. This model makes it possible for the scientist to distinguish between expendable and non-expandable fluid spaces in the body.

Methods: I. Eight healthy male volunteers were given Ringer's solution 25 ml kg-1, dextran 5 ml kg-1 and hypertonic saline 3 ml kg-1 for 30 minutes during three different sessions. Plasma dilution was measured by three different indices: blood haemoglobin, blood water and serum albunlin. The results were fitted to the volume kinetic models. II. Fifteen healthy male volunteers and six female volunteers were given Ringer's solution 25 ml kg-1 for 30 minutes (males) and over 30, 45 and 80 min (females). The plasma dilution following various other infusion rates were simulated by computer using the mean values of the kinetic analysis. III. The volume kinetics of patients with a standardized trauma were studied. Ten patients suffering from a hamate fracture were compared to healthy controls. A fluid load of 12.5 ml kg-1 of Ringer's solution was given the day after closed reduction and surgical repair. Bioelectrical impedance values were measured to establish the fluid balance status. IV. The effect of endotoxin on the volume kinetic parameters was examined. Ten rabbits, which were their own controls, were used. In the first session the rabbits were given a fluid load of 25 ml kg-'. In the next session the rabbits were given the same fluid load but were preinjected with endotoxins 20 µg kg-1. V. To investigate whether fluid therapy changes the prerequisites for the development of oedema, four intravenous infusions of Ringer's solution of 25 n-d kg-'were given over 15 or 30 niin in a randomised cross-over study to 10 healthy male volunteers.

Results: I. Ringer's solution gave the most pronounced dilution. Blood haemoglobin and blood water correlated closely. The two-volume space model was statistically justified especially when Ringer's solution was given. II. Nomograms for obtaining and maintains a certain plasma dilution for both men and women could be constructed. III. The results show that the elimination of fluid is slower in a trauma population. The bioelectrical impedance analysis showed that the extracellular fluid space and the total body water volumes did not differ between the two groups at the time of the study. IV. The dilution of the plasma volume had a markedly variable volume effect after endotoxin had been given. The expanded fluid space was smaller after the injection of endotoxin. V. The volume kinetics analysis shows that the elimination rate was higher when another infusion had been given earlier on the same day. The size of the primary fluid space was slightly larger during the 15-min infusions.

Conclusions: It is possible to use a volume kinetic model to study the behaviour of intravenous fluids. The dilution profiles have an exponential appearance and could therefore be handled according to pharmacokinetic principles.