1Department of Anesthesiology, Söder Hospital, S-118 83 Stockholm, Sweden, 2Department of Numeric Analysis and Computer Science, Royal Institute of Technology, Stockholm, Sweden and 3Karolinska Institutet, Stockholm, Sweden*Corresponding author
Presented as a Poster at the International Anesthesia Research Society 74th Clinical and Scientific Congress in Honululu, Hawaii, March 1014, 2000.
Accepted for publication: July 29, 2001
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Br J Anaesth 2001; 87: 83443
Keywords: fluids, i.v.; blood, haemodilution; pharmacokinetics; metabolism, glucose; blood, haemoglobin
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The purpose of the present report was to evaluate whether volume kinetic principles3 can be used to study the disposition of the volume component of glucose solutions given by intravenous (i.v.) infusion. Volume kinetics has previously been used to analyse and simulate the effects of isotonic salt solutions in normo- and hypovolaemic volunteers46 and also after surgery7 and trauma.8 These studies report the size of body volumes that become expanded and the rate at which fluid becomes distributed and eliminated. However, there was a need for a new kinetic model, which considers an additional volume located more peripherally. Successful application of such a model would make it possible to analyse the disposition of solutions that generate osmotic fluid shifts, which is what happens when glucose is taken up to the cells.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Glucose 2.5%
Twelve subjects aged between 24 and 36 (mean 28) yr of age and with a body weight of 62113 (mean 79) kg participated. After an overnight fast, the volunteers rested comfortably on a bed for at least 20 min of equilibration before the experiments started at 08:00. Before any fluid was administered, a cubital vein of each arm was cannulated for the purpose of withdrawing (including sampling) blood and for infusing fluid, respectively. The volunteers were given an i.v. infusion of 12.5 ml kg1 of glucose 2.5% with electrolytes (Na 70, Cl 45 and acetate 25 mmol litre1; Rehydrex, Pharmacia, Uppsala, Sweden) at a constant rate over 45 min via an infusion pump (Flo-Gard 6201, Baxter Healthcare Ltd, Deerfield, IL, USA). Venous blood was collected every 5 min for 75 min and thereafter every 10 min up to 195 min. The plasma glucose concentration was measured in single samples except for the baseline, which was in duplicate. The blood haemoglobin (Hbb) concentration, the red blood cell count (RBC), and the mean corpuscular volume (MCV) were measured in duplicate samples, and the baseline in quadruplicate samples. The heart rate and arterial pressure were measured using an automatic device (Propaq 104, Protocol Systems Inc., Beaverton, OR, USA) immediately after each blood sampling procedure. Venous blood was also withdrawn at 0, 45, 125, and 195 min to determine the serum concentrations of sodium, potassium, and insulin.
The subjects voided just before the infusions started and, in the recumbent position, whenever necessary during the study. The urine volume and its concentration of glucose, sodium, and potassium were measured.
Glucose 5% and Ringers acetate
Nine subjects (mean age 27 yr, range 2438, and body weight 79 kg, range 5988) underwent two i.v. infusion experiments on separate days at least 1 week apart. On one of these occasions they received 1000 ml of glucose 5% without electrolytes over 45 min via the infusion pump. The same amount of Ringers acetate solution (Na 130, K 4, Ca 2, Mg 1, acetate 30, and Cl 110 mmol litre1) was given on the other occasion (Pharmacia). Blood was sampled every 5 min for 135 min to measure RBC, MCV, and the Hbb and plasma glucose concentrations. Venous blood was also withdrawn at 0, 45, 75, and 135 min to measure the serum concentrations of sodium, potassium, and insulin. Monitoring of haemodynamics and urine was the same as in the study of glucose 2.5%.
Blood chemistry
The plasma glucose (P-glucose) concentration was measured with the GLU Gluco-quant reagent (Boehringer Mannheim) on a Hitachi 917 (Hitachi Co., Naka, Japan). The Hbb concentration was measured on a Technicon H.2 (Bayer, Tarrytown, NY, USA) using colourimetry at 546 nm. RBC and MCV were measured with the same equipment, but by light dispersion at two angles using a helium neon laser. Before each infusion, one sample was drawn in duplicate or quadruplicate and the mean value was used in the calculations. The coefficients of variation for the samples were 1.2% for plasma glucose, 1.0% for Hbb, 1.2% for RBC, and 0.5% for MCV. The serum concentration of insulin was determined by radioimmunoassay (Insulin RIA 100, Pharmacia) and the serum concentrations of sodium and potassium using an Ektachem 950IRC System (Johnson & Johnson, Inc., NY, USA).
Glucose kinetics
The volume of distribution for the glucose load (Vd) was used as the scaling factor between the P-glucose concentration and the amount of glucose that readily equilibrated with venous plasma. There is abundant evidence that the intracellular glucose belongs to another pool. First, the intracellular concentration of glucose is much lower than the extracellular concentration.9 Second, glucose is rapidly phosphorylated when translocated into, for example, muscle cells.10 Third, and probably most important, glucose requires an active transport mechanism to penetrate the cell membrane.9
To obtain Vd, the timeconcentration profile of P-glucose for each glucose solution experiment was analysed according to a mono-exponential washout equation in which the plasma concentration (C) at any time (t) after a bolus injection of glucose is expressed as:
C = (C0 Cbaseline)ek·t(1)
where k is the elimination rate constant, C0 is the concentration when the elimination function is extrapolated back to t=0, and is the mean of the duplicate determinations of P-glucose just before the infusion started. During a constant-rate infusion, the equation used for the curve-fitting procedure was:11
C = (C0 Cbaseline) (1 ek·t) / (kT)(2)
where T is the infusion time. After infusion, it was:
C = (C0 Cbaseline) [(ek·t 1) / (kT)]ek·t(3)
This model was fitted to the data using the Model-PK non-linear regression program for PC (McPherson Scientific, Rosanna, Australia). Weights inversely proportional to the predicted concentrations were applied. Further calculations included the area under the curve (AUC), which was obtained by the linear trapezoid method for the concentrationtime profile of glucose in plasma. The clearance for the glucose load was obtained as the infused dose of glucose divided by the AUC and, in turn, the volume of distribution (Vd) as the clearance divided by k. The half-life (T1/2) was obtained as the natural logarithm of 2 divided by k.
The following F test indicated whether it was statistically justified to fit the data to an equation that contained one more exponent (n+1) than the simpler equation (n):12
F = [(SSQn SSQn+1) / SSQn+1]x[dfn+1 / (dfn dfn+1)]
(4)
where df is the degrees of freedom and SSQ is the sum of squares for the difference between the measured dilution of the plasma and the optimal curve fit. A high F value makes it more likely that the curve is best described by the more complicated model, and significance testing is done by consulting a standard statistical table.
According to Equation 4, the one-compartment open model was consistently justified for analysing the kinetics of the infused glucose load. As glucose is a small molecule, which easily diffuses across the plasma membrane13 but requires active transport to enter the cells, a decreasing amount of glucose in Vd corresponds to the uptake of glucose into a peripheral compartment, which does not equilibrate with venous plasma. Hence, the net uptake of glucose into this remote compartment was calculated for each interval between the earlier (time 1) and the later (time 2) sampling point. During infusion, it was obtained as:
Uptake of glucose = infused glucose
Vd (P-glucose2 P-glucose1)(5)
After infusion, it was:
Uptake of glucose = Vd (P-glucose1 P-glucose2)(6)
Volume kinetic models
In the model for kinetic analysis of the distribution and elimination of infused fluids (Fig. 1, upper), an i.v. infusion is given at a constant rate (ki) and enters a central body fluid space having the volume (v1). The volume (v1) strives to be maintained at the baseline volume V1 by allowing fluid to leave the space at a controlled rate proportional by a constant kr to the deviation of v1 from the target volume V1 (this may be considered as dilution-dependent urinary excretion) and at a basal rate (kb, perspiration and baseline diuresis, fixed rate). The net rate of fluid exchange between v1 and v2 is considered to occur at a rate proportional to the relative difference in deviation from the target values (V1 and V2) by a constant (kt).
|
Furthermore, V2 was only reported if it was statistically justified by comparing a bi-exponential and tri-exponential curve by means of the F test (Equation 4). If V2 was not statistically justified, the constant governing the diffusion of fluid from V3 to V1 was called k31 (Fig. 1, middle).
In order for the kinetic program to calculate the net balance of fluid between V1 and V3 (Fig. 1, lower), it required a starting estimate for V3. This was set at 40% of the body weight.4 The accumulation and elimination (k31) of fluid in V3 could then be calculated and was presented as k31/V3 (Table 1), that is slope for the dilution of V3.
|
Calculations
Volume kinetics
The dilution of the plasma in the cubital vein was used to quantitate the water load. As the sampled plasma is a part of V, we obtain:
(v(t) V) / V = [baseline Hbb / Hbb(t) 1 ] /
(1 baseline haematocrit)(7)
at any time (t). The dilution of the RBC count was calculated in the same way as for Hbb and the mean value of the two was used, after correction for changes in cell volume as indicated by MCV. Furthermore, a correction was always made for the losses of erythrocytes in connection with the blood sampling procedure based on the baseline blood volume as estimated according to a regression formula based on the height and weight of subjects.14 A kb of 0.8 ml min1 was used, which represents the sum of the insensible fluid loss of 10 ml kg1 day1 (0.5 ml min1)15 and the withdrawn amount of extracellular fluid during blood sampling.
The model parameters were calculated on a computer using Matlab version 4.2 (Math Works Inc., Natick, MA, USA), in which a non-linear least-squares regression routine based on a modified GaussNewton method was used. Weights inversely proportional to the predicted dilution+0.1 were applied. The factor 0.1 was added to the denominator to avoid division by zero at baseline.
Sodium dilution method
The diffusion of fluid into the intracellular space was also estimated from a comparison between the distribution volume corresponding to the dilution of the serum sodium level and the actual amount of infused fluid and the urinary excretion of water and sodium.16 17 More details about the calculations are given in the Appendix.
Statistics
The results are expressed as the mean and the standard error of the mean (SEM). Differences between the experiments were evaluated by analysis of variance (ANOVA). Correl ations between parameters were studied by simple and multiple linear regression; P<0.05 was considered significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Plots of the dilution of venous plasma, with a correction for blood sampling, showed that dilution resulting from the three solutions had a similar time course (Fig. 3). Some of the volunteers receiving glucose showed a negative dilution at the end of the experiment.
|
The calculated uptake of glucose into V3 increased progressively during the infusions, but it had returned to baseline 90 min after they were completed (Fig. 4). The calculated uptake of glucose for each time interval was then entered as the driving force for fluid uptake (3.6 ml mmol1 glucose) into V3 in the subsequent kinetic analysis.
|
|
The two-volume model was statistically justified in six of the experiments with Ringers acetate solution. The sizes of V1 and V2 were 2.39 and 7.12 litres, respectively (Table 2). The remaining three volunteers, who handled Ringers solution according to the one-volume model, had rates of fluid elimination, expressed as kr/V1, nearly three times higher than the others (0.092 vs 0.032 min1; P<0.03).
|
|
The calculations of fluid shifts, based on the dilution of the serum sodium concentration, indicated the existence of a slight cellular accumulation of fluid at the end of the experiments with glucose 5% (Table 3). This fluid induced a smaller urinary loss of sodium and potassium than the other ones, while the volume of urine was larger in response to both glucose solutions than to Ringers acetate.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The kinetic analysis further indicates that some of the infused fluid remained in the cells at the very end of the experiments when glucose 5% had been given. This view receives support from the sodium-based calculations of the net fluid over the cell membrane.15 16 The fact that 25 mmol of sodium was lost by urinary excretion, while no sodium was infused, might account for some of the cell accumulation of fluid. Furthermore, a fraction of the glucose might have been stored as glycogen, a process augmented by high insulin concentrations, before being metabolized into carbon dioxide and water. The fluid remaining in the cells might serve to explain why the total measured urinary excretion was smaller in response to glucose 5% than to glucose 2.5%. The urinary excretion after infusion of Ringers acetate was, however, only half as large as after the glucose infusions.
In our volunteers, the strong insulin response even promoted slight hypoglycaemia and hypovolaemia and probably also physical stress, as the heart rate increased only at the end of the experiments with glucose 5%. These adverse effects suggest that a lower infusion rate of glucose 5% should be recommended than the one we used in the present study. A similar post-infusion rebound hypoglycaemia has been reported after sudden withdrawal of total parenteral nutrition,1 and also in newborns after glucose solution has been given at a high rate to their mothers just before delivery by Caesarean section.19 20
The slight glucosuria that occurred in some of the experiments with glucose 5% might have reduced k31, as the kinetic models handle such events as if the infused glucose enters the cells while the accompanying fluid never returns to V1. Simulations showed that the impact of the glucosuria was small, however; the lost glucose only constituted 3.6% of the infused amount in the most pronounced case. Another uncertainty is how precise the kinetic models are when the dilution is negative. In such cases, we set the dilution-dependent elimination of fluid to zero, although dilution-dependent mechanisms for the generation of urine probably operate with a delay and do not shut off as soon as zero dilution is reached. Hence, the measured urinary excretion during the period of hypovolaemia sometimes exceeded the volume indicated by kb alone.
The derived volume kinetic models may be used to understand the relationship between the disposition of fluid given i.v. and the glucose metabolism. We hope that there will be a useful asset in studies of fluid balance during and after surgery, such as when evaluating the effects of surgery-induced insulin resistance on the disposition of infused fluid. When both the glucose kinetics and the volume kinetics have been analysed, a simulation program can predict, for example, the effect of a reduction in the glucose clearance on the disposition of infused fluid. Such programs have been constructed. Before they are used, however, more volunteer experiments are needed to outline how much kr and k31 normally change at different infusion rates and glucose concentrations, as these parameters varied much more than the size of V1 in the present study. The volume kinetic parameters have previously been found to be quite similar when Ringers acetate is given at different infusion rates.4 18
The kinetic models derived here are also valid for other fluids, which induce a fluid shift to or from V3 as a result of osmotic forces, such as hypertonic saline with and without dextran added. Their usefulness is not changed by the fact that glucose and hypertonic solutions shift fluid in opposite directions. In the case of hypertonic saline, calculating f(t) is easy as it can be deduced directly from the infused amount of sodium.
In conclusion, kinetic models showing the relationships between the disposition of infused fluid and the glucose metabolism were developed and successfully applied to data from volunteers. Glucose 2.5 and 5% expanded a functional body fluid space of similar size as the plasma volume when account had been taken of the fluid shifts associated with the metabolism of glucose.
![]() |
Appendix |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Introduce
and we obtain:
These differential equations cover the volume kinetics model in the following three cases, infusion (ki>0), no infusion (ki=0), and w1<0, in which case kr=0. The initial values of the process are w1(0)=0 and w2(0)=0. At time t=T1, the infusion stops, and the solution then has the values w1(T1)=w1T1 and w2(T1)=w2T1. We continue with these values as initial values and ki=0. If, at time t=T2, w1(T2)=0 and dw1/dt (T2)<0, we set kr=0 and continue the solution of (2) with the initial values w1(T2)=0, w2(T2)=w2T2
Introduce vector and matrix notation:
The differential equations in Equation 9 can be written as:
The solution of this linear system of differential equations is:
where eAt is the exponential matrix, T is the initial time, and (T) the corresponding initial value. The integral can be evaluated if
(t) is approximated by a constant
k in the time interval [tk, tk+1]. The numerical solution
k+1 at t=tk+1 is then computed recursively from
where
where =V3/k31.
The three-volume model is described by Equation 9 with
The solution is given by Equation 12, and after approximating (t) with piecewise constant values as in the two-volume model, the numerical solution is obtained from Equation 13; the matrix A is non-singular also in the special case kr=0.
Sodium dilution method
As sodium ions (Na) are distributed throughout the extracellular fluid (ECF) space, the serum sodium concentration at any time (t) during or after i.v. infusion of fluid (S-Nat) equals the amount of Na in the ECF volume divided by the current ECF volume. This relationship can be expressed as:
where S-Nao and ECFo are the serum sodium concentration and ECF volume at baseline, respectively, Naloss is the natriuresis (in mmol), and ICF is the change in the water content of the intracellular fluid compartment. AS ECFo corresponds to 20% of the body weight,8
ICF could be calculated by rearranging Equation 16:
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Stoneham MD, Hill EL. Variability in post-operative fluid and electrolyte prescription. Br J Clin Practice 1997; 51: 824[ISI][Medline]
3 Ståhle L, Nilsson A, Hahn RG. Modelling the kinetics of changes in volume of body fluid spaces. Br J Anaesth 1997; 78: 13843
4 Hahn RG, Drobin D, Ståhle L. Volume kinetics of Ringers solution in female volunteers. Br J Anaesth 1997; 78: 1448
5 Svensén C, Hahn RG. Volume kinetics of Ringer solution, dextran 70 and hypertonic saline in male volunteers. Anesthesiology 1997; 87: 20412[ISI][Medline]
6 Drobin D, Hahn RG. Volume kinetics of Ringers solution in hypovolemic volunteers. Anesthesiology 1999; 90: 8191[ISI][Medline]
7 Svensén C, Ponzer S, Hahn RG. Volume kinetics of Ringers solution after surgery for hip fracture. Can J Anaesth 1999: 46: 13341
8 Svensén C, Sjöstrand F, Hahn RG. Volume kinetics of intravenous fluid therapy in the prehospital setting. Prehosp Dis Med 2001; 16: 913
9 Guyton AC, Hall JE. Textbook of Medical Physiology, 9th Edn. Philadelphia: WB Saunders Company, 1996; 4355
10 Zierler K. Whole body glucose metabolism. Am J Physiol 1999; 276: E409E426
11 Wagner JG. Linear pharmacokinetic equations allowing direct calculation of many needed pharmacokinetic parameters from the coefficients and exponents of polyexponential equations which have been fitted to the data. J Pharmacokin Biopharm 1976; 4: 44367[ISI][Medline]
12 Boxenbaum HG, Riegelman S, Elashoff RM. Statistical estimations in pharmacokinetics. J Pharmacokin Biopharm 1974; 2: 12348[ISI][Medline]
13 Guyton AC, Hall JE. Textbook of Medical Physiology, 9th Edn. Philadelphia: WB Saunders Company, 1996; 1856, 298302
14 Nadler SB, Hidalgo JU, Bloch T. Prediction of blood volume in normal human adults. Surgery 1962; 51: 22432[ISI]
15 Cox P. Insensible water loss and its assessment in adult patients: a review. Acta Anaesthesiol Scand 1987; 31: 7716[ISI][Medline]
16 Hahn R, Stalberg H, Carlström K, Hjelmqvist H, Ullman J, Rundgren M. Plasma atrial natriuretic peptide concentration and renin activity during overhydration with 1.5% glycine solution in conscious sheep. Prostate 1994; 24: 5561[ISI][Medline]
17 Hahn RG, Nilsson A, Hjelmqvist H, Zhang W, Rundgren M. Renal function during intravenous infusion of irrigating fluids in the sheep. Acta Anaesthesiol Scand 1996; 40: 6718[ISI][Medline]
18 Svensén C, Drobin D, Olsson, J, Hahn RG. Stability of the interstitial matrix after crystalloid fluid loading studied by volume kinetic analysis. Br J Anaesth 1999; 82: 496502
19 Philipson EH, Kalham SC, Riha MM, Pimentel R. Effects of maternal glucose infusion on fetal acidbase status in human pregnancy. Am J Obstet Gynecol 1987; 157: 86673[ISI][Medline]
20 Kenepp NB, Shelley WC, Gabbe SG, Kumar S, Stanley CA, Gutsche BB. Neonatal hazards of maternal hydration with 5% dextrose before Caesarean section. Lancet 1982; i: 11502